I realized this while having a map containing both a struct bpf_timer and
a struct bpf_wq: the third argument provided to the bpf_wq callback is
not the struct bpf_wq pointer itself, but the pointer to the value in
the map.
Which means that the users need to double cast the provided "value" as
this is not a struct bpf_wq *.
This is a change of API, but there doesn't seem to be much users of bpf_wq
right now, so we should be able to go with this right now.
Signed-off-by: Benjamin Tissoires <bentiss(a)kernel.org>
---
Benjamin Tissoires (2):
bpf: helpers: fix bpf_wq_set_callback_impl signature
selftests/bpf: amend for wrong bpf_wq_set_callback_impl signature
kernel/bpf/helpers.c | 2 +-
tools/testing/selftests/bpf/bpf_experimental.h | 2 +-
tools/testing/selftests/bpf/progs/wq.c | 8 ++++----
tools/testing/selftests/bpf/progs/wq_failures.c | 4 ++--
4 files changed, 8 insertions(+), 8 deletions(-)
---
base-commit: fd8db07705c55a995c42b1e71afc42faad675b0b
change-id: 20240705-fix-wq-f069c7fb36c3
Best regards,
--
Benjamin Tissoires <bentiss(a)kernel.org>
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v9:
- new patches 5-7, new struct member expect_errno for network_helper_opts.
- patches 1-4, 8-9 unchanged.
- update patches 10-11 to make sure all tests pass.
v8:
- only patch 8 updated, to fix errors reported by CI.
v7:
- address Martin's comments in v6. (thanks)
- use MAX(opts->backlog, 0) instead of opts->backlog.
- use connect_to_fd_opts instead connect_to_fd.
- more ASSERT_* to check errors.
v6:
- update patch 6 as Daniel suggested. (thanks)
v5:
- keep make_server and make_client as Eduard suggested.
v4:
- a new patch to use make_sockaddr in sockmap_ktls.
- a new patch to close fd in error path in drop_on_reuseport.
- drop make_server() in patch 7.
- drop make_client() too in patch 9.
v3:
- a new patch to add backlog for network_helper_opts.
- use start_server_str in sockmap_ktls now, not start_server.
v2:
- address Eduard's comments in v1. (thanks)
- fix errors reported by CI.
This patch set uses network helpers in sockmap_ktls and sk_lookup, and
drop three local helpers tcp_server(), inetaddr_len() and make_socket()
in them.
Geliang Tang (11):
selftests/bpf: Add backlog for network_helper_opts
selftests/bpf: Use start_server_str in sockmap_ktls
selftests/bpf: Use connect_to_fd_opts in sockmap_ktls
selftests/bpf: Use make_sockaddr in sockmap_ktls
selftests/bpf: Add network_helper_opts for connect_fd_to_fd
selftests/bpf: Add expect_errno for network_helper_opts
selftests/bpf: Set expect_errno for cgroup_skb_sk_lookup
selftests/bpf: Close fd in error path in drop_on_reuseport
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Use connect_to_addr in sk_lookup
selftests/bpf: Drop make_socket in sk_lookup
tools/testing/selftests/bpf/network_helpers.c | 23 ++-
tools/testing/selftests/bpf/network_helpers.h | 8 +-
.../testing/selftests/bpf/prog_tests/bpf_nf.c | 5 +-
.../bpf/prog_tests/cgroup_skb_sk_lookup.c | 8 +-
.../selftests/bpf/prog_tests/cgroup_tcp_skb.c | 4 +-
.../selftests/bpf/prog_tests/cgroup_v1v2.c | 1 +
.../selftests/bpf/prog_tests/sk_lookup.c | 152 +++++++-----------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 53 ++----
8 files changed, 106 insertions(+), 148 deletions(-)
--
2.43.0
** Background **
Currently, OVS supports several packet sampling mechanisms (sFlow,
per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
userspace action that needs to be handled by ovs-vswitchd's handler
threads only to be forwarded to some third party application that
will somehow process the sample and provide observability on the
datapath.
A particularly interesting use-case is controller-driven
per-flow IPFIX sampling where the OpenFlow controller can add metadata
to samples (via two 32bit integers) and this metadata is then available
to the sample-collecting system for correlation.
** Problem **
The fact that sampled traffic share netlink sockets and handler thread
time with upcalls, apart from being a performance bottleneck in the
sample extraction itself, can severely compromise the datapath,
yielding this solution unfit for highly loaded production systems.
Users are left with little options other than guessing what sampling
rate will be OK for their traffic pattern and system load and dealing
with the lost accuracy.
Looking at available infrastructure, an obvious candidated would be
to use psample. However, it's current state does not help with the
use-case at stake because sampled packets do not contain user-defined
metadata.
** Proposal **
This series is an attempt to fix this situation by extending the
existing psample infrastructure to carry a variable length
user-defined cookie.
The main existing user of psample is tc's act_sample. It is also
extended to forward the action's cookie to psample.
Finally, a new OVS action (OVS_SAMPLE_ATTR_PSAMPLE) is created.
It accepts a group and an optional cookie and uses psample to
multicast the packet and the metadata.
--
v8 -> v9:
- Rebased.
v7 -> v8:
- Rebased
- Redirect flow insertion to /dev/null to avoid spat in test.
- Removed inline keyword in stub execute_psample_action function.
v6 -> v7:
- Rebased
- Fixed typo in comment.
v5 -> v6:
- Renamed emit_sample -> psample
- Addressed unused variable and conditionally compilation of function.
v4 -> v5:
- Rebased.
- Removed lefover enum value and wrapped some long lines in selftests.
v3 -> v4:
- Rebased.
- Addressed Jakub's comment on private and unused nla attributes.
v2 -> v3:
- Addressed comments from Simon, Aaron and Ilya.
- Dropped probability propagation in nested sample actions.
- Dropped patch v2's 7/9 in favor of a userspace implementation and
consume skb if emit_sample is the last action, same as we do with
userspace.
- Split ovs-dpctl.py features in independent patches.
v1 -> v2:
- Create a new action ("emit_sample") rather than reuse existing
"sample" one.
- Add probability semantics to psample's sampling rate.
- Store sampling probability in skb's cb area and use it in emit_sample.
- Test combining "emit_sample" with "trunc"
- Drop group_id filtering and tracepoint in psample.
rfc_v2 -> v1:
- Accommodate Ilya's comments.
- Split OVS's attribute in two attributes and simplify internal
handling of psample arguments.
- Extend psample and tc with a user-defined cookie.
- Add a tracepoint to psample to facilitate troubleshooting.
rfc_v1 -> rfc_v2:
- Use psample instead of a new OVS-only multicast group.
- Extend psample and tc with a user-defined cookie.
Adrian Moreno (10):
net: psample: add user cookie
net: sched: act_sample: add action cookie to sample
net: psample: skip packet copy if no listeners
net: psample: allow using rate as probability
net: openvswitch: add psample action
net: openvswitch: store sampling probability in cb.
selftests: openvswitch: add psample action
selftests: openvswitch: add userspace parsing
selftests: openvswitch: parse trunc action
selftests: openvswitch: add psample test
Documentation/netlink/specs/ovs_flow.yaml | 17 ++
include/net/psample.h | 5 +-
include/uapi/linux/openvswitch.h | 31 +-
include/uapi/linux/psample.h | 11 +-
net/openvswitch/Kconfig | 1 +
net/openvswitch/actions.c | 66 ++++-
net/openvswitch/datapath.h | 3 +
net/openvswitch/flow_netlink.c | 32 ++-
net/openvswitch/vport.c | 1 +
net/psample/psample.c | 16 +-
net/sched/act_sample.c | 12 +
.../selftests/net/openvswitch/openvswitch.sh | 115 +++++++-
.../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
13 files changed, 566 insertions(+), 16 deletions(-)
--
2.45.2
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v8:
- only patch 8 updated, to fix errors reported by CI.
v7:
- address Martin's comments in v6. (thanks)
- use MAX(opts->backlog, 0) instead of opts->backlog.
- use connect_to_fd_opts instead connect_to_fd.
- more ASSERT_* to check errors.
v6:
- update patch 6 as Daniel suggested. (thanks)
v5:
- keep make_server and make_client as Eduard suggested.
v4:
- a new patch to use make_sockaddr in sockmap_ktls.
- a new patch to close fd in error path in drop_on_reuseport.
- drop make_server() in patch 7.
- drop make_client() too in patch 9.
v3:
- a new patch to add backlog for network_helper_opts.
- use start_server_str in sockmap_ktls now, not start_server.
v2:
- address Eduard's comments in v1. (thanks)
- fix errors reported by CI.
This patch set uses network helpers in sockmap_ktls and sk_lookup, and
drop three local helpers tcp_server(), inetaddr_len() and make_socket()
in them.
Geliang Tang (9):
selftests/bpf: Add backlog for network_helper_opts
selftests/bpf: Use start_server_str in sockmap_ktls
selftests/bpf: Use connect_to_fd_opts in sockmap_ktls
selftests/bpf: Use make_sockaddr in sockmap_ktls
selftests/bpf: Close fd in error path in drop_on_reuseport
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Use connect_to_fd_opts in sk_lookup
selftests/bpf: Use connect_to_addr in sk_lookup
selftests/bpf: Drop make_socket in sk_lookup
tools/testing/selftests/bpf/network_helpers.c | 2 +-
tools/testing/selftests/bpf/network_helpers.h | 4 +
.../selftests/bpf/prog_tests/sk_lookup.c | 152 +++++++-----------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 53 ++----
4 files changed, 76 insertions(+), 135 deletions(-)
--
2.43.0
Hi Shuah,
These are for 6.10, as we just discussed.
Changes since v4:
1) Subject line on patch #2/3: s/mm/vDSO/
2) Added Muhammad's review tag.
Changes since v3:
1. Rebased onto Linux 6.10-rc6+.
Cover letter for v3:
Jason A. Donenfeld, I've added you because I ended up looking through
your latest "implement getrandom() in vDSO" series [1], which also
touches this Makefile, so just a heads up about upcoming (minor) merge
conflicts.
Changes since v2:
1. Added two patches, both of which apply solely to the Makefile.
These provide a smaller, cleaner, and more accurate Makefile.
2. Added Reviewed-by and Tested-by tags for the original patch, which
fixes all of the clang errors and warnings for this selftest.
3. Removed an obsolete blurb from the commit description of the original
patch, now that Valentin Obst LLVM build fix has been merged.
thanks,
John Hubbard
NVIDIA
John Hubbard (3):
selftests/vDSO: fix clang build errors and warnings
selftests/vDSO: remove partially duplicated "all:" target in Makefile
selftests/vDSO: remove duplicate compiler invocations from Makefile
tools/testing/selftests/vDSO/Makefile | 29 ++++++++-----------
tools/testing/selftests/vDSO/parse_vdso.c | 16 ++++++----
.../selftests/vDSO/vdso_standalone_test_x86.c | 18 ++++++++++--
3 files changed, 39 insertions(+), 24 deletions(-)
base-commit: d270dd21bee023ab627f34cfb77a9b89a688492a
--
2.40.1
Hi,
This is basically a resend, with a rebase onto today's latest Linux
main, in order to show that the patches are still relevant and correct.
Changes since v3:
1. Rebased onto Linux 6.10-rc6+.
Cover letter for v3:
Jason A. Donenfeld, I've added you because I ended up looking through
your latest "implement getrandom() in vDSO" series [1], which also
touches this Makefile, so just a heads up about upcoming (minor) merge
conflicts.
Changes since v2:
1. Added two patches, both of which apply solely to the Makefile.
These provide a smaller, cleaner, and more accurate Makefile.
2. Added Reviewed-by and Tested-by tags for the original patch, which
fixes all of the clang errors and warnings for this selftest.
3. Removed an obsolete blurb from the commit description of the original
patch, now that Valentin Obst LLVM build fix has been merged.
John Hubbard (3):
selftests/vDSO: fix clang build errors and warnings
selftests/mm: remove partially duplicated "all:" target in Makefile
selftests/vDSO: remove duplicate compiler invocations from Makefile
tools/testing/selftests/vDSO/Makefile | 29 ++++++++-----------
tools/testing/selftests/vDSO/parse_vdso.c | 16 ++++++----
.../selftests/vDSO/vdso_standalone_test_x86.c | 18 ++++++++++--
3 files changed, 39 insertions(+), 24 deletions(-)
base-commit: 8a9c6c40432e265600232b864f97d7c675e8be52
--
2.45.2
When building with clang, via:
make LLVM=1 -C tools/testing/selftest
...clang warns about an unused irqcount variable. clang is correct: the
variable is incremented and then ignored.
Fix this by deleting the irqcount variable.
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
Changes since v2:
1) Rebased onto Linux 6.10-rc6+
Changes since the first version:
1) Rebased onto Linux 6.10-rc1
thanks,
John Hubbard
tools/testing/selftests/timers/rtcpie.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/tools/testing/selftests/timers/rtcpie.c b/tools/testing/selftests/timers/rtcpie.c
index 4ef2184f1558..7c07edd0d450 100644
--- a/tools/testing/selftests/timers/rtcpie.c
+++ b/tools/testing/selftests/timers/rtcpie.c
@@ -29,7 +29,7 @@ static const char default_rtc[] = "/dev/rtc0";
int main(int argc, char **argv)
{
- int i, fd, retval, irqcount = 0;
+ int i, fd, retval;
unsigned long tmp, data, old_pie_rate;
const char *rtc = default_rtc;
struct timeval start, end, diff;
@@ -120,7 +120,6 @@ int main(int argc, char **argv)
fprintf(stderr, " %d",i);
fflush(stderr);
- irqcount++;
}
/* Disable periodic interrupts */
base-commit: 8a9c6c40432e265600232b864f97d7c675e8be52
--
2.45.2
These patches aim to make using the openvswitch testsuite more reliable.
These should address the major sources of flakiness in the openvswitch
test suite allowing the CI infrastructure to exercise the openvswitch
module for patch series. There should be no change for users who simply
run the tests (except that patch 3/3 does make some of the debugging a bit
easier by making some output more verbose).
Aaron Conole (3):
selftests: openvswitch: Bump timeout to 15 minutes.
selftests: openvswitch: Attempt to autoload module.
selftests: openvswitch: Be more verbose with selftest debugging.
.../selftests/net/openvswitch/openvswitch.sh | 23 ++++++++++++-------
.../selftests/net/openvswitch/settings | 1 +
2 files changed, 16 insertions(+), 8 deletions(-)
create mode 100644 tools/testing/selftests/net/openvswitch/settings
--
2.45.1
On Fri, Jul 05, 2024 at 11:49:28AM GMT, Adrián Moreno wrote:
> On Thu, Jul 04, 2024 at 10:56:51AM GMT, Adrian Moreno wrote:
> > ** Background **
> > Currently, OVS supports several packet sampling mechanisms (sFlow,
> > per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
> > userspace action that needs to be handled by ovs-vswitchd's handler
> > threads only to be forwarded to some third party application that
> > will somehow process the sample and provide observability on the
> > datapath.
> >
> > A particularly interesting use-case is controller-driven
> > per-flow IPFIX sampling where the OpenFlow controller can add metadata
> > to samples (via two 32bit integers) and this metadata is then available
> > to the sample-collecting system for correlation.
> >
> > ** Problem **
> > The fact that sampled traffic share netlink sockets and handler thread
> > time with upcalls, apart from being a performance bottleneck in the
> > sample extraction itself, can severely compromise the datapath,
> > yielding this solution unfit for highly loaded production systems.
> >
> > Users are left with little options other than guessing what sampling
> > rate will be OK for their traffic pattern and system load and dealing
> > with the lost accuracy.
> >
> > Looking at available infrastructure, an obvious candidated would be
> > to use psample. However, it's current state does not help with the
> > use-case at stake because sampled packets do not contain user-defined
> > metadata.
> >
> > ** Proposal **
> > This series is an attempt to fix this situation by extending the
> > existing psample infrastructure to carry a variable length
> > user-defined cookie.
> >
> > The main existing user of psample is tc's act_sample. It is also
> > extended to forward the action's cookie to psample.
> >
> > Finally, a new OVS action (OVS_SAMPLE_ATTR_PSAMPLE) is created.
> > It accepts a group and an optional cookie and uses psample to
> > multicast the packet and the metadata.
> >
> > --
> > v8 -> v9:
> > - Rebased.
> >
> > v7 -> v8:
> > - Rebased
> > - Redirect flow insertion to /dev/null to avoid spat in test.
> > - Removed inline keyword in stub execute_psample_action function.
> >
> > v6 -> v7:
> > - Rebased
> > - Fixed typo in comment.
> >
> > v5 -> v6:
> > - Renamed emit_sample -> psample
> > - Addressed unused variable and conditionally compilation of function.
> >
> > v4 -> v5:
> > - Rebased.
> > - Removed lefover enum value and wrapped some long lines in selftests.
> >
> > v3 -> v4:
> > - Rebased.
> > - Addressed Jakub's comment on private and unused nla attributes.
> >
> > v2 -> v3:
> > - Addressed comments from Simon, Aaron and Ilya.
> > - Dropped probability propagation in nested sample actions.
> > - Dropped patch v2's 7/9 in favor of a userspace implementation and
> > consume skb if emit_sample is the last action, same as we do with
> > userspace.
> > - Split ovs-dpctl.py features in independent patches.
> >
> > v1 -> v2:
> > - Create a new action ("emit_sample") rather than reuse existing
> > "sample" one.
> > - Add probability semantics to psample's sampling rate.
> > - Store sampling probability in skb's cb area and use it in emit_sample.
> > - Test combining "emit_sample" with "trunc"
> > - Drop group_id filtering and tracepoint in psample.
> >
> > rfc_v2 -> v1:
> > - Accommodate Ilya's comments.
> > - Split OVS's attribute in two attributes and simplify internal
> > handling of psample arguments.
> > - Extend psample and tc with a user-defined cookie.
> > - Add a tracepoint to psample to facilitate troubleshooting.
> >
> > rfc_v1 -> rfc_v2:
> > - Use psample instead of a new OVS-only multicast group.
> > - Extend psample and tc with a user-defined cookie.
> >
> > Adrian Moreno (10):
> > net: psample: add user cookie
> > net: sched: act_sample: add action cookie to sample
> > net: psample: skip packet copy if no listeners
> > net: psample: allow using rate as probability
> > net: openvswitch: add psample action
> > net: openvswitch: store sampling probability in cb.
> > selftests: openvswitch: add psample action
> > selftests: openvswitch: add userspace parsing
> > selftests: openvswitch: parse trunc action
> > selftests: openvswitch: add psample test
> >
> > Documentation/netlink/specs/ovs_flow.yaml | 17 ++
> > include/net/psample.h | 5 +-
> > include/uapi/linux/openvswitch.h | 31 +-
> > include/uapi/linux/psample.h | 11 +-
> > net/openvswitch/Kconfig | 1 +
> > net/openvswitch/actions.c | 66 ++++-
> > net/openvswitch/datapath.h | 3 +
> > net/openvswitch/flow_netlink.c | 32 ++-
> > net/openvswitch/vport.c | 1 +
> > net/psample/psample.c | 16 +-
> > net/sched/act_sample.c | 12 +
> > .../selftests/net/openvswitch/openvswitch.sh | 115 +++++++-
> > .../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
> > 13 files changed, 566 insertions(+), 16 deletions(-)
> >
> > --
> > 2.45.2
> >
>
> Hi,
>
> Simon Horman has spotted that openvswitch.sh tests are failing in the
> debug executor:
>
> https://netdev.bots.linux.dev/contest.html?test=openvswitch-sh
>
> The failing tests are two: psample and upcall_interfaces. These two
> tests have a known source of instability (they use "sleep") that make
> them specially unreliable in slow systems.
>
> Aaron and I already discussed this and I'm working on a patch to make
> both tests more robust by adding a wait-and-retry mechanism.
>
> I hope this series can be considered regardless of this flaky tests.
>
Adding more context to explain our situation.
This series has a counterpart in OVS [1]. The state of this other series
is still RFC just because the kernel bits have not yet been merged.
OVS 3.4 "softfreeze" was declared last monday, which excludes from the
release any series that is stil in RFC state.
Given the kernel parts seemed very close to be merged, an exception was
given to the series so we can consider it for inclusion [2].
I hate to put any pressure on already busy maintainers but I would also
dislike missing this OVS release by just one or two days and having
to wait 6 months (OVS release cadence) for it to be available.
Again, I don't want to put pressure on maintainers. If it's not
possible, that's it. I just wanted to voice our timeline constraints.
Thanks for your understanding.
Adrián
[1] https://patchwork.ozlabs.org/project/openvswitch/cover/20240704085710.35384…
[2] https://mail.openvswitch.org/pipermail/ovs-dev/2024-July/415261.html
in the do_setcpu, this function does not need to have a return value,
which is meaningless
Signed-off-by: Liu Jing <liujing(a)cmss.chinamobile.com>
---
tools/testing/selftests/net/msg_zerocopy.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/tools/testing/selftests/net/msg_zerocopy.c b/tools/testing/selftests/net/msg_zerocopy.c
index bdc03a2097e8..0b54f2011449 100644
--- a/tools/testing/selftests/net/msg_zerocopy.c
+++ b/tools/testing/selftests/net/msg_zerocopy.c
@@ -118,7 +118,7 @@ static uint16_t get_ip_csum(const uint16_t *start, int num_words)
return ~sum;
}
-static int do_setcpu(int cpu)
+static void do_setcpu(int cpu)
{
cpu_set_t mask;
@@ -129,7 +129,6 @@ static int do_setcpu(int cpu)
else if (cfg_verbose)
fprintf(stderr, "cpu: %u\n", cpu);
- return 0;
}
static void do_setsockopt(int fd, int level, int optname, int val)
--
2.33.0
Recent CI failures brought my attention to the fact that pmu_counters_test
sometimes fails because it doesn't get any LLC cache misses.
It apparently happens because CLFLUSH can race with CPU prediction.
To attempt to fix this, implement a more aggressive cache flushing - now it is flushed
on each iteration of the measured loop which should at least reduce by order
of magnitude the chance of this happening.
This patch survived more that a day of running in a loop on a Comet Lake machine,
where the test used to fail after about 10-20 minites.
Best regards,
Maxim Levitsky
Maxim Levitsky (1):
KVM: selftests: pmu_counters_test: increase robustness of LLC cache
misses
.../selftests/kvm/x86_64/pmu_counters_test.c | 20 +++++++++----------
1 file changed, 9 insertions(+), 11 deletions(-)
--
2.26.3
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v7:
- address Martin's comments in v6. (thanks)
- use MAX(opts->backlog, 0) instead of opts->backlog.
- use connect_to_fd_opts instead connect_to_fd.
- more ASSERT_* to check errors.
v6:
- update patch 6 as Daniel suggested. (thanks)
v5:
- keep make_server and make_client as Eduard suggested.
v4:
- a new patch to use make_sockaddr in sockmap_ktls.
- a new patch to close fd in error path in drop_on_reuseport.
- drop make_server() in patch 7.
- drop make_client() too in patch 9.
v3:
- a new patch to add backlog for network_helper_opts.
- use start_server_str in sockmap_ktls now, not start_server.
v2:
- address Eduard's comments in v1. (thanks)
- fix errors reported by CI.
This patch set uses network helpers in sockmap_ktls and sk_lookup, and
drop three local helpers tcp_server(), inetaddr_len() and make_socket()
in them.
Geliang Tang (9):
selftests/bpf: Add backlog for network_helper_opts
selftests/bpf: Use start_server_str in sockmap_ktls
selftests/bpf: Use connect_to_fd_opts in sockmap_ktls
selftests/bpf: Use make_sockaddr in sockmap_ktls
selftests/bpf: Close fd in error path in drop_on_reuseport
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Use connect_to_fd_opts in sk_lookup
selftests/bpf: Use connect_to_addr in sk_lookup
selftests/bpf: Drop make_socket in sk_lookup
tools/testing/selftests/bpf/network_helpers.c | 2 +-
tools/testing/selftests/bpf/network_helpers.h | 4 +
.../selftests/bpf/prog_tests/sk_lookup.c | 150 ++++++++----------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 53 ++-----
4 files changed, 77 insertions(+), 132 deletions(-)
--
2.43.0
Hi Linus,
This PR fixes a few kselftests [1]. This has been in linux-next for a week and
rebased to add Mark Brown's Tested-by. The race condition found while writing
this fix is not new and seems specific to UML's hostfs (I also tested against
ext4 and btrfs without being able to trigger this issue).
Feel free to take this PR if you see fit.
Regards,
Mickaël
[1] https://lore.kernel.org/r/9341d4db-5e21-418c-bf9e-9ae2da7877e1@sirena.org.uk
--
The following changes since commit f2661062f16b2de5d7b6a5c42a9a5c96326b8454:
Linux 6.10-rc5 (2024-06-23 17:08:54 -0400)
are available in the Git repository at:
https://git.kernel.org/pub/scm/linux/kernel/git/mic/linux.git tags/kselftest-fix-2024-07-04
for you to fetch changes up to 130e42806773013e9cf32d211922c935ae2df86c:
selftests/harness: Fix tests timeout and race condition (2024-06-28 16:06:03 +0200)
----------------------------------------------------------------
Fix Kselftests timeout and race condition
----------------------------------------------------------------
Mickaël Salaün (1):
selftests/harness: Fix tests timeout and race condition
tools/testing/selftests/kselftest_harness.h | 43 ++++++++++++++++-------------
1 file changed, 24 insertions(+), 19 deletions(-)
Current practice in the selftests Makefiles is to use $(LLVM) as a way
to decide if clang is being used as the compiler (and/or the linker
front end). Unfortunately, this does not cover all of the use cases:
1) CC could have been set within selftests/lib.mk, by inferring it from
LLVM==1, or
2) CC could have been set externally, such as when cross compiling.
Solution: In order to allow subsystem selftests to more accurately
control clang-specific behavior, such as compiler options, provide a new
Makefile variable: SELFTESTS_CC_IS_CLANG. If $(CC) contains an
invocation of clang in any form, then SELFTESTS_CC_IS_CLANG will be
non-empty.
SELFTESTS_CC_IS_CLANG does not specify which linker is being used.
However, it can still help with linker options, because $(CC) is often
used to do both the compile and link steps (often in the same step).
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
Hi,
If this looks reasonable, I'll break it up into separate patches and
post it as a non-RFC.
thanks,
John Hubbard
tools/testing/selftests/bpf/Makefile | 2 +-
tools/testing/selftests/fchmodat2/Makefile | 12 +++++++-----
tools/testing/selftests/hid/Makefile | 2 +-
tools/testing/selftests/lib.mk | 15 +++++++++++++++
tools/testing/selftests/openat2/Makefile | 16 +++++++++-------
5 files changed, 33 insertions(+), 14 deletions(-)
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index dd49c1d23a60..6b924297ab71 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -55,7 +55,7 @@ progs/test_sk_lookup.c-CFLAGS := -fno-strict-aliasing
progs/timer_crash.c-CFLAGS := -fno-strict-aliasing
progs/test_global_func9.c-CFLAGS := -fno-strict-aliasing
-ifneq ($(LLVM),)
+ifeq ($(SELFTESTS_CC_IS_CLANG),)
# Silence some warnings when compiled with clang
CFLAGS += -Wno-unused-command-line-argument
endif
diff --git a/tools/testing/selftests/fchmodat2/Makefile b/tools/testing/selftests/fchmodat2/Makefile
index 4373cea79b79..d00b01be5d96 100644
--- a/tools/testing/selftests/fchmodat2/Makefile
+++ b/tools/testing/selftests/fchmodat2/Makefile
@@ -2,14 +2,16 @@
CFLAGS += -Wall -O2 -g -fsanitize=address -fsanitize=undefined $(KHDR_INCLUDES)
+TEST_GEN_PROGS := fchmodat2_test
+
+include ../lib.mk
+
# gcc requires -static-libasan in order to ensure that Address Sanitizer's
# library is the first one loaded. However, clang already statically links the
# Address Sanitizer if -fsanitize is specified. Therefore, simply omit
# -static-libasan for clang builds.
-ifeq ($(LLVM),)
+# This check must be done after including ../lib.mk, in order to pick up the
+# correct value of SELFTESTS_CC_IS_CLANG.
+ifeq ($(SELFTESTS_CC_IS_CLANG),)
CFLAGS += -static-libasan
endif
-
-TEST_GEN_PROGS := fchmodat2_test
-
-include ../lib.mk
diff --git a/tools/testing/selftests/hid/Makefile b/tools/testing/selftests/hid/Makefile
index 2b5ea18bde38..734a53dc8ad9 100644
--- a/tools/testing/selftests/hid/Makefile
+++ b/tools/testing/selftests/hid/Makefile
@@ -27,7 +27,7 @@ CFLAGS += -I$(OUTPUT)/tools/include
LDLIBS += -lelf -lz -lrt -lpthread
# Silence some warnings when compiled with clang
-ifneq ($(LLVM),)
+ifeq ($(SELFTESTS_CC_IS_CLANG),)
CFLAGS += -Wno-unused-command-line-argument
endif
diff --git a/tools/testing/selftests/lib.mk b/tools/testing/selftests/lib.mk
index 429535816dbd..f321ad5a1d0c 100644
--- a/tools/testing/selftests/lib.mk
+++ b/tools/testing/selftests/lib.mk
@@ -43,6 +43,21 @@ else
CC := $(CROSS_COMPILE)gcc
endif # LLVM
+# SELFTESTS_CC_IS_CLANG allows subsystem selftests to more accurately control
+# clang-specific behavior, such as compiler options. If CC is an invocation of
+# clang in any form, then SELFTESTS_CC_IS_CLANG will be non-empty. Notes:
+#
+# 1) CC could have been set above, by inferring it from LLVM==1, or externally,
+# from the CC shell environment variable.
+#
+# 2) SELFTESTS_CC_IS_CLANG does not specify which linker is being used. However,
+# it can still help with linker options, if clang or gcc is used for the
+# linker front end.
+SELFTESTS_CC_IS_CLANG :=
+ifeq ($(findstring clang,$(CC)),clang)
+ SELFTESTS_CC_IS_CLANG := 1
+endif
+
ifeq (0,$(MAKELEVEL))
ifeq ($(OUTPUT),)
OUTPUT := $(shell pwd)
diff --git a/tools/testing/selftests/openat2/Makefile b/tools/testing/selftests/openat2/Makefile
index 185dc76ebb5f..7acb85a8f2ac 100644
--- a/tools/testing/selftests/openat2/Makefile
+++ b/tools/testing/selftests/openat2/Makefile
@@ -3,16 +3,18 @@
CFLAGS += -Wall -O2 -g -fsanitize=address -fsanitize=undefined
TEST_GEN_PROGS := openat2_test resolve_test rename_attack_test
+LOCAL_HDRS += helpers.h
+
+include ../lib.mk
+
+$(TEST_GEN_PROGS): helpers.c
+
# gcc requires -static-libasan in order to ensure that Address Sanitizer's
# library is the first one loaded. However, clang already statically links the
# Address Sanitizer if -fsanitize is specified. Therefore, simply omit
# -static-libasan for clang builds.
-ifeq ($(LLVM),)
+# This check must be done after including ../lib.mk, in order to pick up the
+# correct value of SELFTESTS_CC_IS_CLANG.
+ifeq ($(SELFTESTS_CC_IS_CLANG),)
CFLAGS += -static-libasan
endif
-
-LOCAL_HDRS += helpers.h
-
-include ../lib.mk
-
-$(TEST_GEN_PROGS): helpers.c
base-commit: 9a5cd459be8a425d70cda1fa1c89af7875a35d17
--
2.45.2
Clang does not support implicit LMUL in the vset* instruction sequences.
Introduce an explicit LMUL in the vsetivli instruction.
Signed-off-by: Charlie Jenkins <charlie(a)rivosinc.com>
Fixes: 9d5328eeb185 ("riscv: selftests: Add signal handling vector tests")
---
There is one more error that occurs when the test cases for riscv are
compiled with llvm:
ld.lld: error: undefined symbol: putchar
>>> referenced by crt.h:69 (./../../../../include/nolibc/crt.h:69)
>>> /tmp/v_initval_nolibc-5b14c8.o:(dump)
>>> referenced by crt.h:67 (./../../../../include/nolibc/crt.h:67)
>>> /tmp/v_initval_nolibc-5b14c8.o:(dump)
This is fixed in my rework of the vector tests in a different series [1]
Link: https://patchwork.kernel.org/project/linux-riscv/patch/20240619-xtheadvecto… [1]
---
tools/testing/selftests/riscv/sigreturn/sigreturn.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/riscv/sigreturn/sigreturn.c b/tools/testing/selftests/riscv/sigreturn/sigreturn.c
index 62397d5934f1..ed351a1cb917 100644
--- a/tools/testing/selftests/riscv/sigreturn/sigreturn.c
+++ b/tools/testing/selftests/riscv/sigreturn/sigreturn.c
@@ -51,7 +51,7 @@ static int vector_sigreturn(int data, void (*handler)(int, siginfo_t *, void *))
asm(".option push \n\
.option arch, +v \n\
- vsetivli x0, 1, e32, ta, ma \n\
+ vsetivli x0, 1, e32, m1, ta, ma \n\
vmv.s.x v0, %1 \n\
# Generate SIGSEGV \n\
lw a0, 0(x0) \n\
---
base-commit: f2661062f16b2de5d7b6a5c42a9a5c96326b8454
change-id: 20240701-fix_sigreturn_test-47d7063ac8e6
--
- Charlie
** Background **
Currently, OVS supports several packet sampling mechanisms (sFlow,
per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
userspace action that needs to be handled by ovs-vswitchd's handler
threads only to be forwarded to some third party application that
will somehow process the sample and provide observability on the
datapath.
A particularly interesting use-case is controller-driven
per-flow IPFIX sampling where the OpenFlow controller can add metadata
to samples (via two 32bit integers) and this metadata is then available
to the sample-collecting system for correlation.
** Problem **
The fact that sampled traffic share netlink sockets and handler thread
time with upcalls, apart from being a performance bottleneck in the
sample extraction itself, can severely compromise the datapath,
yielding this solution unfit for highly loaded production systems.
Users are left with little options other than guessing what sampling
rate will be OK for their traffic pattern and system load and dealing
with the lost accuracy.
Looking at available infrastructure, an obvious candidated would be
to use psample. However, it's current state does not help with the
use-case at stake because sampled packets do not contain user-defined
metadata.
** Proposal **
This series is an attempt to fix this situation by extending the
existing psample infrastructure to carry a variable length
user-defined cookie.
The main existing user of psample is tc's act_sample. It is also
extended to forward the action's cookie to psample.
Finally, a new OVS action (OVS_SAMPLE_ATTR_PSAMPLE) is created.
It accepts a group and an optional cookie and uses psample to
multicast the packet and the metadata.
--
v7 -> v8:
- Rebased
- Redirect flow insertion to /dev/null to avoid spat in test.
- Removed inline keyword in stub execute_psample_action function.
v6 -> v7:
- Rebased
- Fixed typo in comment.
v5 -> v6:
- Renamed emit_sample -> psample
- Addressed unused variable and conditionally compilation of function.
v4 -> v5:
- Rebased.
- Removed lefover enum value and wrapped some long lines in selftests.
v3 -> v4:
- Rebased.
- Addressed Jakub's comment on private and unused nla attributes.
v2 -> v3:
- Addressed comments from Simon, Aaron and Ilya.
- Dropped probability propagation in nested sample actions.
- Dropped patch v2's 7/9 in favor of a userspace implementation and
consume skb if emit_sample is the last action, same as we do with
userspace.
- Split ovs-dpctl.py features in independent patches.
v1 -> v2:
- Create a new action ("emit_sample") rather than reuse existing
"sample" one.
- Add probability semantics to psample's sampling rate.
- Store sampling probability in skb's cb area and use it in emit_sample.
- Test combining "emit_sample" with "trunc"
- Drop group_id filtering and tracepoint in psample.
rfc_v2 -> v1:
- Accommodate Ilya's comments.
- Split OVS's attribute in two attributes and simplify internal
handling of psample arguments.
- Extend psample and tc with a user-defined cookie.
- Add a tracepoint to psample to facilitate troubleshooting.
rfc_v1 -> rfc_v2:
- Use psample instead of a new OVS-only multicast group.
- Extend psample and tc with a user-defined cookie.
Adrian Moreno (10):
net: psample: add user cookie
net: sched: act_sample: add action cookie to sample
net: psample: skip packet copy if no listeners
net: psample: allow using rate as probability
net: openvswitch: add psample action
net: openvswitch: store sampling probability in cb.
selftests: openvswitch: add psample action
selftests: openvswitch: add userspace parsing
selftests: openvswitch: parse trunc action
selftests: openvswitch: add psample test
Documentation/netlink/specs/ovs_flow.yaml | 17 ++
include/net/psample.h | 5 +-
include/uapi/linux/openvswitch.h | 31 +-
include/uapi/linux/psample.h | 11 +-
net/openvswitch/Kconfig | 1 +
net/openvswitch/actions.c | 66 ++++-
net/openvswitch/datapath.h | 3 +
net/openvswitch/flow_netlink.c | 32 ++-
net/openvswitch/vport.c | 1 +
net/psample/psample.c | 16 +-
net/sched/act_sample.c | 12 +
.../selftests/net/openvswitch/openvswitch.sh | 115 +++++++-
.../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
13 files changed, 566 insertions(+), 16 deletions(-)
--
2.45.2
Hi,
Dave Hansen, Muhammad Usama Anjum, here is the combined series that we
discussed yesterday [1].
As I mentioned then, this is a bit intrusive--but no more than
necessary, IMHO. Specifically, it moves some clang-un-inlineable things
out to "pure" assembly code files.
I've tested this by building with clang, then running each binary on my
x86_64 test system with today's 6.10-rc1, and comparing the console and
dmesg output to a gcc-based build without these patches applied. Aside
from timestamps and virtual addresses, it looks identical.
Earlier cover letter:
Just a bunch of build and warnings fixes that show up when building with
clang. Some of these depend on each other, so I'm sending them as a
series.
Changes since v2:
1) Dropped my test_FISTTP.c patch, and picked up Muhammad's fix instead,
seeing as how that was posted first.
2) Updated patch descriptions to reflect that Valentin Obst's build fix
for LLVM [1] has already been merged into Linux main.
3) Minor wording and typo corrections in the commit logs throughout.
Changes since the first version:
1) Rebased onto Linux 6.10-rc1
Enjoy!
[1] https://lore.kernel.org/44428518-4d21-4de7-8587-04eceefb330d@nvidia.com
thanks,
John Hubbard
John Hubbard (6):
selftests/x86: fix Makefile dependencies to work with clang
selftests/x86: build fsgsbase_restore.c with clang
selftests/x86: build sysret_rip.c with clang
selftests/x86: avoid -no-pie warnings from clang during compilation
selftests/x86: remove (or use) unused variables and functions
selftests/x86: fix printk warnings reported by clang
Muhammad Usama Anjum (1):
selftests: x86: test_FISTTP: use fisttps instead of ambiguous fisttp
tools/testing/selftests/x86/Makefile | 31 +++++++++++++++----
tools/testing/selftests/x86/amx.c | 16 ----------
.../testing/selftests/x86/clang_helpers_32.S | 11 +++++++
.../testing/selftests/x86/clang_helpers_64.S | 28 +++++++++++++++++
tools/testing/selftests/x86/fsgsbase.c | 6 ----
.../testing/selftests/x86/fsgsbase_restore.c | 11 +++----
tools/testing/selftests/x86/sigreturn.c | 2 +-
.../testing/selftests/x86/syscall_arg_fault.c | 1 -
tools/testing/selftests/x86/sysret_rip.c | 20 ++++--------
tools/testing/selftests/x86/test_FISTTP.c | 8 ++---
tools/testing/selftests/x86/test_vsyscall.c | 15 +++------
tools/testing/selftests/x86/vdso_restorer.c | 2 ++
12 files changed, 87 insertions(+), 64 deletions(-)
create mode 100644 tools/testing/selftests/x86/clang_helpers_32.S
create mode 100644 tools/testing/selftests/x86/clang_helpers_64.S
base-commit: 4a4be1ad3a6efea16c56615f31117590fd881358
--
2.45.1
The watchdog selftest script supports various parameters for testing
different IOCTLs. The watchdog ping functionality is validated by starting
a loop where the watchdog device is periodically pet, which can only be
stopped by the user interrupting the script.
This results in a timeout when running this test using the kselftest runner
with no non-oneshot parameters (or no parameters at all):
TAP version 13
1..1
# timeout set to 45
# selftests: watchdog: watchdog-test
# Watchdog Ticking Away!
# .............................................#
not ok 1 selftests: watchdog: watchdog-test # TIMEOUT 45 seconds
To address this issue, the first patch in this series limits the loop to 5
iterations by default and adds support for a new '-c' option to customize
the number of pings as required.
The second patch conforms the test output to the KTAP format.
Laura Nao (2):
selftests/watchdog: limit ping loop and allow configuring the number
of pings
selftests/watchdog: convert the test output to KTAP format
.../selftests/watchdog/watchdog-test.c | 166 +++++++++++-------
1 file changed, 101 insertions(+), 65 deletions(-)
--
2.30.2
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v6:
- update patch 6 as Daniel suggested. (thanks)
v5:
- keep make_server and make_client as Eduard suggested.
v4:
- a new patch to use make_sockaddr in sockmap_ktls.
- a new patch to close fd in error path in drop_on_reuseport.
- drop make_server() in patch 7.
- drop make_client() too in patch 9.
v3:
- a new patch to add backlog for network_helper_opts.
- use start_server_str in sockmap_ktls now, not start_server.
v2:
- address Eduard's comments in v1. (thanks)
- fix errors reported by CI.
This patch set uses network helpers in sockmap_ktls and sk_lookup, and
drop three local helpers tcp_server(), inetaddr_len() and make_socket()
in them.
Geliang Tang (9):
selftests/bpf: Add backlog for network_helper_opts
selftests/bpf: Use start_server_str in sockmap_ktls
selftests/bpf: Use connect_to_fd in sockmap_ktls
selftests/bpf: Use make_sockaddr in sockmap_ktls
selftests/bpf: Close fd in error path in drop_on_reuseport
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Use connect_to_fd in sk_lookup
selftests/bpf: Use connect_to_addr in sk_lookup
selftests/bpf: Drop make_socket in sk_lookup
tools/testing/selftests/bpf/network_helpers.c | 2 +-
tools/testing/selftests/bpf/network_helpers.h | 1 +
.../selftests/bpf/prog_tests/sk_lookup.c | 141 +++++++-----------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 51 ++-----
4 files changed, 61 insertions(+), 134 deletions(-)
--
2.43.0
in main_loop_s function, when the open(cfg_input, O_RDONLY) function is run,
the last fd is not closed if the "--cfg_repeat > 0" branch is not taken.
Fixes: 05be5e273c84("selftests: mptcp: add disconnect tests").
Signed-off-by: Liu Jing <liujing(a)cmss.chinamobile.com>
---
Changes from v1
- add close function in main_loop_s function
---
tools/testing/selftests/net/mptcp/mptcp_connect.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/mptcp/mptcp_connect.c b/tools/testing/selftests/net/mptcp/mptcp_connect.c
index d2043ec3bf6d..48b7389ae75b 100644
--- a/tools/testing/selftests/net/mptcp/mptcp_connect.c
+++ b/tools/testing/selftests/net/mptcp/mptcp_connect.c
@@ -1119,7 +1119,8 @@ int main_loop_s(int listensock)
if (cfg_input)
close(fd);
goto again;
- }
+ } else
+ close(fd);
return 0;
}
--
2.33.0
xtheadvector is a custom extension that is based upon riscv vector
version 0.7.1 [1]. All of the vector routines have been modified to
support this alternative vector version based upon whether xtheadvector
was determined to be supported at boot.
vlenb is not supported on the existing xtheadvector hardware, so a
devicetree property thead,vlenb is added to provide the vlenb to Linux.
There is a new hwprobe key RISCV_HWPROBE_KEY_VENDOR_EXT_THEAD_0 that is
used to request which thead vendor extensions are supported on the
current platform. This allows future vendors to allocate hwprobe keys
for their vendor.
Support for xtheadvector is also added to the vector kselftests.
Signed-off-by: Charlie Jenkins <charlie(a)rivosinc.com>
[1] https://github.com/T-head-Semi/thead-extension-spec/blob/95358cb2cca9489361…
---
This series is a continuation of a different series that was fragmented
into two other series in an attempt to get part of it merged in the 6.10
merge window. The split-off series did not get merged due to a NAK on
the series that added the generic riscv,vlenb devicetree entry. This
series has converted riscv,vlenb to thead,vlenb to remedy this issue.
The original series is titled "riscv: Support vendor extensions and
xtheadvector" [3].
The series titled "riscv: Extend cpufeature.c to detect vendor
extensions" is still under development and this series is based on that
series! [4]
I have tested this with an Allwinner Nezha board. I ran into issues
booting the board after 6.9-rc1 so I applied these patches to 6.8. There
are a couple of minor merge conflicts that do arrise when doing that, so
please let me know if you have been able to boot this board with a 6.9
kernel. I used SkiffOS [1] to manage building the image, but upgraded
the U-Boot version to Samuel Holland's more up-to-date version [2] and
changed out the device tree used by U-Boot with the device trees that
are present in upstream linux and this series. Thank you Samuel for all
of the work you did to make this task possible.
[1] https://github.com/skiffos/SkiffOS/tree/master/configs/allwinner/nezha
[2] https://github.com/smaeul/u-boot/commit/2e89b706f5c956a70c989cd31665f1429e9…
[3] https://lore.kernel.org/all/20240503-dev-charlie-support_thead_vector_6_9-v…
[4] https://lore.kernel.org/linux-riscv/20240609-support_vendor_extensions-v2-0…
---
Changes in v4:
- Replace inline asm with C (Samuel)
- Rename VCSRs to CSRs (Samuel)
- Replace .insn directives with .4byte directives
- Link to v3: https://lore.kernel.org/r/20240619-xtheadvector-v3-0-bff39eb9668e@rivosinc.…
Changes in v3:
- Add back Heiko's signed-off-by (Conor)
- Mark RISCV_HWPROBE_KEY_VENDOR_EXT_THEAD_0 as a bitmask
- Link to v2: https://lore.kernel.org/r/20240610-xtheadvector-v2-0-97a48613ad64@rivosinc.…
Changes in v2:
- Removed extraneous references to "riscv,vlenb" (Jess)
- Moved declaration of "thead,vlenb" into cpus.yaml and added
restriction that it's only applicable to thead cores (Conor)
- Check CONFIG_RISCV_ISA_XTHEADVECTOR instead of CONFIG_RISCV_ISA_V for
thead,vlenb (Jess)
- Fix naming of hwprobe variables (Evan)
- Link to v1: https://lore.kernel.org/r/20240609-xtheadvector-v1-0-3fe591d7f109@rivosinc.…
---
Charlie Jenkins (12):
dt-bindings: riscv: Add xtheadvector ISA extension description
dt-bindings: cpus: add a thead vlen register length property
riscv: dts: allwinner: Add xtheadvector to the D1/D1s devicetree
riscv: Add thead and xtheadvector as a vendor extension
riscv: vector: Use vlenb from DT for thead
riscv: csr: Add CSR encodings for CSR_VXRM/CSR_VXSAT
riscv: Add xtheadvector instruction definitions
riscv: vector: Support xtheadvector save/restore
riscv: hwprobe: Add thead vendor extension probing
riscv: hwprobe: Document thead vendor extensions and xtheadvector extension
selftests: riscv: Fix vector tests
selftests: riscv: Support xtheadvector in vector tests
Heiko Stuebner (1):
RISC-V: define the elements of the VCSR vector CSR
Documentation/arch/riscv/hwprobe.rst | 10 +
Documentation/devicetree/bindings/riscv/cpus.yaml | 19 ++
.../devicetree/bindings/riscv/extensions.yaml | 10 +
arch/riscv/Kconfig.vendor | 26 ++
arch/riscv/boot/dts/allwinner/sun20i-d1s.dtsi | 3 +-
arch/riscv/include/asm/cpufeature.h | 2 +
arch/riscv/include/asm/csr.h | 15 ++
arch/riscv/include/asm/hwprobe.h | 5 +-
arch/riscv/include/asm/switch_to.h | 2 +-
arch/riscv/include/asm/vector.h | 224 ++++++++++++----
arch/riscv/include/asm/vendor_extensions/thead.h | 42 +++
.../include/asm/vendor_extensions/thead_hwprobe.h | 18 ++
.../include/asm/vendor_extensions/vendor_hwprobe.h | 37 +++
arch/riscv/include/uapi/asm/hwprobe.h | 3 +-
arch/riscv/include/uapi/asm/vendor/thead.h | 3 +
arch/riscv/kernel/cpufeature.c | 51 +++-
arch/riscv/kernel/kernel_mode_vector.c | 8 +-
arch/riscv/kernel/process.c | 4 +-
arch/riscv/kernel/signal.c | 6 +-
arch/riscv/kernel/sys_hwprobe.c | 5 +
arch/riscv/kernel/vector.c | 25 +-
arch/riscv/kernel/vendor_extensions.c | 10 +
arch/riscv/kernel/vendor_extensions/Makefile | 2 +
arch/riscv/kernel/vendor_extensions/thead.c | 18 ++
.../riscv/kernel/vendor_extensions/thead_hwprobe.c | 19 ++
tools/testing/selftests/riscv/vector/.gitignore | 3 +-
tools/testing/selftests/riscv/vector/Makefile | 17 +-
.../selftests/riscv/vector/v_exec_initval_nolibc.c | 93 +++++++
tools/testing/selftests/riscv/vector/v_helpers.c | 67 +++++
tools/testing/selftests/riscv/vector/v_helpers.h | 7 +
tools/testing/selftests/riscv/vector/v_initval.c | 22 ++
.../selftests/riscv/vector/v_initval_nolibc.c | 68 -----
.../selftests/riscv/vector/vstate_exec_nolibc.c | 20 +-
.../testing/selftests/riscv/vector/vstate_prctl.c | 295 ++++++++++++---------
34 files changed, 888 insertions(+), 271 deletions(-)
---
base-commit: 11cc01d4d2af304b7288251aad7e03315db8dffc
change-id: 20240530-xtheadvector-833d3d17b423
--
- Charlie
Hi,
Here is the v2 patch to support polling on event 'hist' file.
The previous version is here;
https://lore.kernel.org/all/171932861260.584123.15653284949837094747.stgit@…
This version updates the test program, because previous version will
return fail on stable kernels which does not have this feature.
This checks whether the poll(POLLIN) on hist is timeout or not without
sending event. If poll() is implemented, it should timed out. If not,
poll(POLLIN) retuns soon.
And it tests both of POLLIN and POLLPRI in this version.
Background
----------
There has been interest in allowing user programs to monitor kernel
events in real time. Ftrace provides `trace_pipe` interface to wait
on events in the ring buffer, but it is needed to wait until filling
up a page with events in the ring buffer. We can also peek the
`trace` file periodically, but that is inefficient way to monitor
a randomely happening event.
Overview
--------
This patch set allows user to `poll`(or `select`, `epoll`) on event
histogram interface. As you know each event has its own `hist` file
which shows histograms generated by trigger action. So user can set
a new hist trigger on any event you want to monitor, and poll on the
`hist` file until it is updated.
There are 2 poll events are supported, POLLIN and POLLPRI. POLLIN
means that there are any readable update on `hist` file and this
event will be flashed only when you call read(). So, this is
useful if you want to read the histogram periodically.
The other POLLPRI event is for monitoring trace event. Like the
POLLIN, this will be returned when the histogram is updated, but
you don't need to read() the file and use poll() again.
Note that this waits for histogram update (not event arrival), thus
you must set a histogram on the event at first.
Usage
-----
Here is an example usage:
----
TRACEFS=/sys/kernel/tracing
EVENT=$TRACEFS/events/sched/sched_process_free
# setup histogram trigger and enable event
echo "hist:key=comm" >> $EVENT/trigger
echo 1 > $EVENT/enable
# Wait for update
poll pri $EVENT/hist
# Event arrived.
echo "process free event is comming"
tail $TRACEFS/trace
----
The 'poll' command is in the selftest patch.
You can take this series also from here;
https://git.kernel.org/pub/scm/linux/kernel/git/mhiramat/linux.git/log/?h=t…
Thank you,
---
Masami Hiramatsu (Google) (3):
tracing/hist: Add poll(POLLIN) support on hist file
tracing/hist: Support POLLPRI event for poll on histogram
selftests/tracing: Add hist poll() support test
include/linux/trace_events.h | 5 +
kernel/trace/trace_events.c | 18 ++++
kernel/trace/trace_events_hist.c | 101 +++++++++++++++++++-
tools/testing/selftests/ftrace/Makefile | 2
tools/testing/selftests/ftrace/poll.c | 62 ++++++++++++
.../ftrace/test.d/trigger/trigger-hist-poll.tc | 74 +++++++++++++++
6 files changed, 259 insertions(+), 3 deletions(-)
create mode 100644 tools/testing/selftests/ftrace/poll.c
create mode 100644 tools/testing/selftests/ftrace/test.d/trigger/trigger-hist-poll.tc
--
Masami Hiramatsu (Google) <mhiramat(a)kernel.org>
Hi,
Here is an RFC patch to support polling on event 'hist' file.
There has been interest in allowing user programs to monitor kernel
events in real time. Ftrace provides `trace_pipe` interface to wait
on events in the ring buffer, but it is needed to wait until filling
up a page with events in the ring buffer. We can also peek the
`trace` file periodically, but that is inefficient way to monitor
a randomely happening event.
This patch set allows user to `poll`(or `select`, `epoll`) on event
histogram interface. As you know each event has its own `hist` file
which shows histograms generated by trigger action. So user can set
a new hist trigger on any event you want to monitor, and poll on the
`hist` file until it is updated.
There are 2 poll events are supported, POLLIN and POLLPRI. POLLIN
means that there are any readable update on `hist` file and this
event will be flashed only when you call read(). So, this is
useful if you want to read the histogram periodically.
The other POLLPRI event is for monitoring trace event. Like the
POLLIN, this will be returned when the histogram is updated, but
you don't need to read() the file and use poll() again.
Note that this waits for histogram update (not event arrival), thus
you must set a histogram on the event at first.
Here is an example usage:
----
TRACEFS=/sys/kernel/tracing
EVENT=$TRACEFS/events/sched/sched_process_free
# setup histogram trigger and enable event
echo "hist:key=comm" >> $EVENT/trigger
echo 1 > $EVENT/enable
# Wait for update
poll $EVENT/hist
# Event arrived.
echo "process free event is comming"
tail $TRACEFS/trace
----
The 'poll' command is in the selftest patch.
You can take this series also from here;
https://git.kernel.org/pub/scm/linux/kernel/git/mhiramat/linux.git/log/?h=t…
Thank you,
---
Masami Hiramatsu (Google) (3):
tracing/hist: Add poll(POLLIN) support on hist file
tracing/hist: Support POLLPRI event for poll on histogram
selftests/tracing: Add hist poll() support test
include/linux/trace_events.h | 5 +
kernel/trace/trace_events.c | 18 ++++
kernel/trace/trace_events_hist.c | 101 +++++++++++++++++++-
tools/testing/selftests/ftrace/Makefile | 3 +
tools/testing/selftests/ftrace/poll.c | 34 +++++++
.../ftrace/test.d/trigger/trigger-hist-poll.tc | 46 +++++++++
6 files changed, 204 insertions(+), 3 deletions(-)
create mode 100644 tools/testing/selftests/ftrace/poll.c
create mode 100644 tools/testing/selftests/ftrace/test.d/trigger/trigger-hist-poll.tc
--
Masami Hiramatsu (Google) <mhiramat(a)kernel.org>
This patch series adds unit tests for the clk fixed rate basic type and
the clk registration functions that use struct clk_parent_data. To get
there, we add support for loading device tree overlays onto the live DTB
along with probing platform drivers to bind to device nodes in the
overlays. With this series, we're able to exercise some of the code in
the common clk framework that uses devicetree lookups to find parents
and the fixed rate clk code that scans device tree directly and creates
clks. Please review.
I Cced everyone to all the patches so they get the full context. I'm
hoping I can take the whole pile through the clk tree as they all build
upon each other. Or the DT part can be merged through the DT tree to
reduce the dependencies.
Changes from v4: https://lore.kernel.org/r/20240422232404.213174-1-sboyd@kernel.org
* Picked up reviewed-by tags
* Check for non-NULL device pointers before calling put_device()
* Fix CFI issues with kunit actions
* Introduce platform_device_prepare_wait_for_probe() helper to wait for
a platform device to probe
* Move platform code to lib/kunit and rename functions to have kunit
prefix
* Fix issue with platform wrappers messing up reference counting
because they used kunit actions
* New patch to populate overlay devices on root node for powerpc
* Make fixed-rate binding generic single clk consumer binding
Changes from v3: https://lore.kernel.org/r/20230327222159.3509818-1-sboyd@kernel.org
* No longer depend on Frank's series[1] because it was merged upstream[2]
* Use kunit_add_action_or_reset() to shorten code
* Skip tests properly when CONFIG_OF_OVERLAY isn't set
Changes from v2: https://lore.kernel.org/r/20230315183729.2376178-1-sboyd@kernel.org
* Overlays don't depend on __symbols__ node
* Depend on Frank's always create root node if CONFIG_OF series[1]
* Added kernel-doc to KUnit API doc
* Fixed some kernel-doc on functions
* More test cases for fixed rate clk
Changes from v1: https://lore.kernel.org/r/20230302013822.1808711-1-sboyd@kernel.org
* Don't depend on UML, use unittest data approach to attach nodes
* Introduce overlay loading API for KUnit
* Move platform_device KUnit code to drivers/base/test
* Use #define macros for constants shared between unit tests and
overlays
* Settle on "test" as a vendor prefix
* Make KUnit wrappers have "_kunit" postfix
[1] https://lore.kernel.org/r/20230317053415.2254616-1-frowand.list@gmail.com
[2] https://lore.kernel.org/r/20240308195737.GA1174908-robh@kernel.org
Stephen Boyd (11):
of/platform: Allow overlays to create platform devices from the root
node
of: Add test managed wrappers for of_overlay_apply()/of_node_put()
dt-bindings: vendor-prefixes: Add "test" vendor for KUnit and friends
dt-bindings: test: Add KUnit empty node binding
of: Add a KUnit test for overlays and test managed APIs
platform: Add test managed platform_device/driver APIs
dt-bindings: test: Add single clk consumer
clk: Add test managed clk provider/consumer APIs
clk: Add KUnit tests for clk fixed rate basic type
dt-bindings: clk: Add clk_parent_data test
clk: Add KUnit tests for clks registered with struct clk_parent_data
Documentation/dev-tools/kunit/api/clk.rst | 10 +
Documentation/dev-tools/kunit/api/index.rst | 21 +
Documentation/dev-tools/kunit/api/of.rst | 13 +
.../dev-tools/kunit/api/platformdevice.rst | 10 +
.../bindings/clock/test,clk-parent-data.yaml | 47 ++
.../devicetree/bindings/test/test,empty.yaml | 30 ++
.../test/test,single-clk-consumer.yaml | 34 ++
.../devicetree/bindings/vendor-prefixes.yaml | 2 +
drivers/clk/.kunitconfig | 2 +
drivers/clk/Kconfig | 9 +
drivers/clk/Makefile | 9 +-
drivers/clk/clk-fixed-rate_test.c | 379 +++++++++++++++
drivers/clk/clk-fixed-rate_test.h | 8 +
drivers/clk/clk_kunit_helpers.c | 204 ++++++++
drivers/clk/clk_parent_data_test.h | 10 +
drivers/clk/clk_test.c | 451 +++++++++++++++++-
drivers/clk/kunit_clk_fixed_rate_test.dtso | 19 +
drivers/clk/kunit_clk_parent_data_test.dtso | 28 ++
drivers/of/.kunitconfig | 1 +
drivers/of/Kconfig | 10 +
drivers/of/Makefile | 2 +
drivers/of/kunit_overlay_test.dtso | 9 +
drivers/of/of_kunit_helpers.c | 74 +++
drivers/of/overlay_test.c | 116 +++++
drivers/of/platform.c | 9 +-
include/kunit/clk.h | 28 ++
include/kunit/of.h | 115 +++++
include/kunit/platform_device.h | 20 +
lib/kunit/Makefile | 4 +-
lib/kunit/platform-test.c | 223 +++++++++
lib/kunit/platform.c | 302 ++++++++++++
31 files changed, 2193 insertions(+), 6 deletions(-)
create mode 100644 Documentation/dev-tools/kunit/api/clk.rst
create mode 100644 Documentation/dev-tools/kunit/api/of.rst
create mode 100644 Documentation/dev-tools/kunit/api/platformdevice.rst
create mode 100644 Documentation/devicetree/bindings/clock/test,clk-parent-data.yaml
create mode 100644 Documentation/devicetree/bindings/test/test,empty.yaml
create mode 100644 Documentation/devicetree/bindings/test/test,single-clk-consumer.yaml
create mode 100644 drivers/clk/clk-fixed-rate_test.c
create mode 100644 drivers/clk/clk-fixed-rate_test.h
create mode 100644 drivers/clk/clk_kunit_helpers.c
create mode 100644 drivers/clk/clk_parent_data_test.h
create mode 100644 drivers/clk/kunit_clk_fixed_rate_test.dtso
create mode 100644 drivers/clk/kunit_clk_parent_data_test.dtso
create mode 100644 drivers/of/kunit_overlay_test.dtso
create mode 100644 drivers/of/of_kunit_helpers.c
create mode 100644 drivers/of/overlay_test.c
create mode 100644 include/kunit/clk.h
create mode 100644 include/kunit/of.h
create mode 100644 include/kunit/platform_device.h
create mode 100644 lib/kunit/platform-test.c
create mode 100644 lib/kunit/platform.c
base-commit: 1613e604df0cd359cf2a7fbd9be7a0bcfacfabd0
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
Hi Linus,
Please pull this kselftest fixes update for Linux 6.10-rc7.
This kselftest fixes update for Linux 6.10-rc7 consists of one single
patch to fix the non-contiguous CBM resctrl:
- AMD supports non-contiguous CBM but does not report it via CPUID. This
test should not use CPUID on AMD to detect non-contiguous CBM support.
Fix the problem so the test uses CPUID to discover non-contiguous CBM
support only on Intel.
diff is attached.
thanks,
-- Shuah
----------------------------------------------------------------
The following changes since commit ed3994ac847e0d6605f248e7f6776b1d4f445f4b:
selftests/fchmodat2: fix clang build failure due to -static-libasan (2024-06-11 15:05:05 -0600)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest tags/linux_kselftest-fixes-6.10-rc7
for you to fetch changes up to 48236960c06d32370bfa6f2cc408e786873262c8:
selftests/resctrl: Fix non-contiguous CBM for AMD (2024-06-26 13:22:34 -0600)
----------------------------------------------------------------
linux_kselftest-fixes-6.10-rc7
This kselftest fixes update for Linux 6.10-rc7 consists of one single
patch to fix the non-contiguous CBM resctrl:
- AMD supports non-contiguous CBM but does not report it via CPUID. This
test should not use CPUID on AMD to detect non-contiguous CBM support.
Fix the problem so the test uses CPUID to discover non-contiguous CBM
support only on Intel.
----------------------------------------------------------------
Babu Moger (1):
selftests/resctrl: Fix non-contiguous CBM for AMD
tools/testing/selftests/resctrl/cat_test.c | 32 ++++++++++++++++++++----------
1 file changed, 22 insertions(+), 10 deletions(-)
----------------------------------------------------------------
Hi,
Jason A. Donenfeld, I've added you because I ended up looking through
your latest "implement getrandom() in vDSO" series [1], which also
touches this Makefile, so just a heads up about upcoming (minor) merge
conflicts.
Changes since v2:
1. Added two patches, both of which apply solely to the Makefile.
These provide a smaller, cleaner, and more accurate Makefile.
2. Added Reviewed-by and Tested-by tags for the original patch, which
fixes all of the clang errors and warnings for this selftest.
3. Removed an obsolete blurb from the commit description of the original
patch, now that Valentin Obst LLVM build fix has been merged.
[1] https://lore.kernel.org/20240614190646.2081057-1-Jason@zx2c4.com
John Hubbard (3):
selftests/vDSO: fix clang build errors and warnings
selftests/mm: remove partially duplicated "all:" target in Makefile
selftests/vDSO: remove duplicate compiler invocations from Makefile
tools/testing/selftests/vDSO/Makefile | 29 ++++++++-----------
tools/testing/selftests/vDSO/parse_vdso.c | 16 ++++++----
.../selftests/vDSO/vdso_standalone_test_x86.c | 18 ++++++++++--
3 files changed, 39 insertions(+), 24 deletions(-)
base-commit: 2ccbdf43d5e758f8493a95252073cf9078a5fea5
--
2.45.2
The open() function returns -1 on error. openat() and open() initialize
'from' and 'to', and only 'from' validated with 'if' statement. If the
initialization of variable 'to' fails, we should better check the value
of 'to' and close 'from' to avoid possible file leak. Improve the checking
of 'from' additionally.
Fixes: 32ae976ed3b5 ("selftests/capabilities: Add tests for capability evolution")
Signed-off-by: Ma Ke <make24(a)iscas.ac.cn>
---
Changes in v3:
- Thank you for your interest in our vulnerability detection method. We
extract vulnerability characteristics from a known vulnerability and match
the same characteristics in the project code. As our work is still in
progress, we are not able to disclose it at this time. Appreciate your
understanding, we could better focus on the potential vulnerability itself.
Reference link: https://lore.kernel.org/all/20240510003424.2016914-1-samasth.norway.ananda@…
Changes in v2:
- modified the patch according to suggestions;
- found by customized static analysis tool.
---
tools/testing/selftests/capabilities/test_execve.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/capabilities/test_execve.c b/tools/testing/selftests/capabilities/test_execve.c
index 47bad7ddc5bc..6406ab6aa1f5 100644
--- a/tools/testing/selftests/capabilities/test_execve.c
+++ b/tools/testing/selftests/capabilities/test_execve.c
@@ -145,10 +145,14 @@ static void chdir_to_tmpfs(void)
static void copy_fromat_to(int fromfd, const char *fromname, const char *toname)
{
int from = openat(fromfd, fromname, O_RDONLY);
- if (from == -1)
+ if (from < 0)
ksft_exit_fail_msg("open copy source - %s\n", strerror(errno));
int to = open(toname, O_CREAT | O_WRONLY | O_EXCL, 0700);
+ if (to < 0) {
+ close(from);
+ ksft_exit_fail_msg("open copy destination - %s\n", strerror(errno));
+ }
while (true) {
char buf[4096];
--
2.25.1
** Background **
Currently, OVS supports several packet sampling mechanisms (sFlow,
per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
userspace action that needs to be handled by ovs-vswitchd's handler
threads only to be forwarded to some third party application that
will somehow process the sample and provide observability on the
datapath.
A particularly interesting use-case is controller-driven
per-flow IPFIX sampling where the OpenFlow controller can add metadata
to samples (via two 32bit integers) and this metadata is then available
to the sample-collecting system for correlation.
** Problem **
The fact that sampled traffic share netlink sockets and handler thread
time with upcalls, apart from being a performance bottleneck in the
sample extraction itself, can severely compromise the datapath,
yielding this solution unfit for highly loaded production systems.
Users are left with little options other than guessing what sampling
rate will be OK for their traffic pattern and system load and dealing
with the lost accuracy.
Looking at available infrastructure, an obvious candidated would be
to use psample. However, it's current state does not help with the
use-case at stake because sampled packets do not contain user-defined
metadata.
** Proposal **
This series is an attempt to fix this situation by extending the
existing psample infrastructure to carry a variable length
user-defined cookie.
The main existing user of psample is tc's act_sample. It is also
extended to forward the action's cookie to psample.
Finally, a new OVS action (OVS_SAMPLE_ATTR_PSAMPLE) is created.
It accepts a group and an optional cookie and uses psample to
multicast the packet and the metadata.
--
v6 -> v7:
- Rebased
- Fixed typo in comment.
v5 -> v6:
- Renamed emit_sample -> psample
- Addressed unused variable and conditionally compilation of function.
v4 -> v5:
- Rebased.
- Removed lefover enum value and wrapped some long lines in selftests.
v3 -> v4:
- Rebased.
- Addressed Jakub's comment on private and unused nla attributes.
v2 -> v3:
- Addressed comments from Simon, Aaron and Ilya.
- Dropped probability propagation in nested sample actions.
- Dropped patch v2's 7/9 in favor of a userspace implementation and
consume skb if emit_sample is the last action, same as we do with
userspace.
- Split ovs-dpctl.py features in independent patches.
v1 -> v2:
- Create a new action ("emit_sample") rather than reuse existing
"sample" one.
- Add probability semantics to psample's sampling rate.
- Store sampling probability in skb's cb area and use it in emit_sample.
- Test combining "emit_sample" with "trunc"
- Drop group_id filtering and tracepoint in psample.
rfc_v2 -> v1:
- Accommodate Ilya's comments.
- Split OVS's attribute in two attributes and simplify internal
handling of psample arguments.
- Extend psample and tc with a user-defined cookie.
- Add a tracepoint to psample to facilitate troubleshooting.
rfc_v1 -> rfc_v2:
- Use psample instead of a new OVS-only multicast group.
- Extend psample and tc with a user-defined cookie.
Adrian Moreno (10):
net: psample: add user cookie
net: sched: act_sample: add action cookie to sample
net: psample: skip packet copy if no listeners
net: psample: allow using rate as probability
net: openvswitch: add psample action
net: openvswitch: store sampling probability in cb.
selftests: openvswitch: add psample action
selftests: openvswitch: add userspace parsing
selftests: openvswitch: parse trunc action
selftests: openvswitch: add psample test
Documentation/netlink/specs/ovs_flow.yaml | 17 ++
include/net/psample.h | 5 +-
include/uapi/linux/openvswitch.h | 31 +-
include/uapi/linux/psample.h | 11 +-
net/openvswitch/Kconfig | 1 +
net/openvswitch/actions.c | 65 ++++-
net/openvswitch/datapath.h | 3 +
net/openvswitch/flow_netlink.c | 32 ++-
net/openvswitch/vport.c | 1 +
net/psample/psample.c | 16 +-
net/sched/act_sample.c | 12 +
.../selftests/net/openvswitch/openvswitch.sh | 115 +++++++-
.../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
13 files changed, 565 insertions(+), 16 deletions(-)
--
2.45.2
Adrian Moreno (10):
net: psample: add user cookie
net: sched: act_sample: add action cookie to sample
net: psample: skip packet copy if no listeners
net: psample: allow using rate as probability
net: openvswitch: add psample action
net: openvswitch: store sampling probability in cb.
selftests: openvswitch: add psample action
selftests: openvswitch: add userspace parsing
selftests: openvswitch: parse trunc action
selftests: openvswitch: add psample test
Documentation/netlink/specs/ovs_flow.yaml | 17 ++
include/net/psample.h | 5 +-
include/uapi/linux/openvswitch.h | 31 +-
include/uapi/linux/psample.h | 11 +-
net/openvswitch/Kconfig | 1 +
net/openvswitch/actions.c | 65 ++++-
net/openvswitch/datapath.h | 3 +
net/openvswitch/flow_netlink.c | 32 ++-
net/openvswitch/vport.c | 1 +
net/psample/psample.c | 16 +-
net/sched/act_sample.c | 12 +
.../selftests/net/openvswitch/openvswitch.sh | 115 +++++++-
.../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
13 files changed, 565 insertions(+), 16 deletions(-)
--
2.45.2
From: Geliang Tang <tanggeliang(a)kylinos.cn>
Run this BPF selftests (./test_progs -t sockmap_basic) on a Loongarch
platform, a kernel panic occurs:
'''
Oops[#1]:
CPU: 22 PID: 2824 Comm: test_progs Tainted: G OE 6.10.0-rc2+ #18
Hardware name: LOONGSON Dabieshan/Loongson-TC542F0, BIOS Loongson-UDK2018
... ...
ra: 90000000048bf6c0 sk_msg_recvmsg+0x120/0x560
ERA: 9000000004162774 copy_page_to_iter+0x74/0x1c0
CRMD: 000000b0 (PLV0 -IE -DA +PG DACF=CC DACM=CC -WE)
PRMD: 0000000c (PPLV0 +PIE +PWE)
EUEN: 00000007 (+FPE +SXE +ASXE -BTE)
ECFG: 00071c1d (LIE=0,2-4,10-12 VS=7)
ESTAT: 00010000 [PIL] (IS= ECode=1 EsubCode=0)
BADV: 0000000000000040
PRID: 0014c011 (Loongson-64bit, Loongson-3C5000)
Modules linked in: bpf_testmod(OE) xt_CHECKSUM xt_MASQUERADE xt_conntrack
Process test_progs (pid: 2824, threadinfo=0000000000863a31, task=...)
Stack : ...
...
Call Trace:
[<9000000004162774>] copy_page_to_iter+0x74/0x1c0
[<90000000048bf6c0>] sk_msg_recvmsg+0x120/0x560
[<90000000049f2b90>] tcp_bpf_recvmsg_parser+0x170/0x4e0
[<90000000049aae34>] inet_recvmsg+0x54/0x100
[<900000000481ad5c>] sock_recvmsg+0x7c/0xe0
[<900000000481e1a8>] __sys_recvfrom+0x108/0x1c0
[<900000000481e27c>] sys_recvfrom+0x1c/0x40
[<9000000004c076ec>] do_syscall+0x8c/0xc0
[<9000000003731da4>] handle_syscall+0xc4/0x160
Code: ...
---[ end trace 0000000000000000 ]---
Kernel panic - not syncing: Fatal exception
Kernel relocated by 0x3510000
.text @ 0x9000000003710000
.data @ 0x9000000004d70000
.bss @ 0x9000000006469400
---[ end Kernel panic - not syncing: Fatal exception ]---
'''
This crash happens every time when running sockmap_skb_verdict_shutdown
subtest in sockmap_basic.
This crash is because a NULL pointer is passed to page_address() in
sk_msg_recvmsg(). Due to the difference implementations depending on the
architecture, page_address(NULL) will trigger a panic on Loongarch
platform but not on X86 platform. So this bug was hidden on X86 platform
for a while, but now it is exposed on Loongarch platform.
The root cause is an empty skb (skb->len == 0) is put on the queue.
This empty skb is a TCP FIN package, which is sent by shutdown(), invoked
in test_sockmap_skb_verdict_shutdown():
shutdown(p1, SHUT_WR);
In this case, in sk_psock_skb_ingress_enqueue(), num_sge is zero, and no
page is put to this sge (see sg_set_page in sg_set_page), but this empty
sge is queued into ingress_msg list.
And in sk_msg_recvmsg(), this empty sge is used, and a NULL page is got by
sg_page(sge). Pass this NULL-page to copy_page_to_iter(), it passed to
kmap_local_page() and page_address(), then kernel panics.
To solve this, we should skip the empty sge on the queue. So in
sk_msg_recvmsg(), if msg_rx->sg.end is zero, that means it's an empty sge,
skip it.
Fixes: 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface")
Signed-off-by: Geliang Tang <tanggeliang(a)kylinos.cn>
---
v4:
- skmsg: skip empty sge in sk_msg_recvmsg
v3:
- skmsg: prevent empty ingress skb from enqueuing
v2:
- skmsg: null check for sg_page in sk_msg_recvmsg
---
net/core/skmsg.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/core/skmsg.c b/net/core/skmsg.c
index fd20aae30be2..66db1631852b 100644
--- a/net/core/skmsg.c
+++ b/net/core/skmsg.c
@@ -421,7 +421,7 @@ int sk_msg_recvmsg(struct sock *sk, struct sk_psock *psock, struct msghdr *msg,
while (copied != len) {
struct scatterlist *sge;
- if (unlikely(!msg_rx))
+ if (unlikely(!msg_rx || !msg_rx->sg.end))
break;
i = msg_rx->sg.start;
--
2.43.0
From: Quan Zhou <zhouquan(a)iscas.ac.cn>
Due to the path that modifies a0 in syscall_enter_from_user_mode before the
actual execution of syscall_handler [1], the kernel currently saves a0 to
orig_a0 at the entry point of do_trap_ecall_u as an original copy of a0.
Once the syscall is interrupted and later resumed, the restarted syscall
will use orig_a0 to continue execution.
The above rules generally apply except for ptrace(PTRACE_SETREGSET,),
where the kernel will ignore the tracer's setting of tracee/a0 and
will restart with the tracee/orig_a0. For the current kernel implementation
of ptrace, projects like CRIU/Proot will encounter issues where the a0
setting becomes ineffective when performing ptrace(PTRACE_SETREGSET,).
Here is a suggested solution, expose orig_a0 to userspace so that ptrace
can choose whether to set orig_a0 based on the actual scenario. In fact,
x86/orig_eax and loongArch/orig_a0 have adopted similar solutions.
[1] link:
https://lore.kernel.org/lkml/20230403-crisping-animosity-04ed8a45c625@spud/…
---
Changes from RFC->v1:
- Rebased on Linux 6.10-rc5.
- Updated the patch description.
- Adjust MAX_REG_OFFSET to match the new bottom of pt_regs (Charlie).
- Simplify selftest to verify if a0 can be set (Charlie).
- Fix .gitignore error (Charlie).
---
RFC link:
https://lore.kernel.org/all/cover.1718693532.git.zhouquan@iscas.ac.cn/
Quan Zhou (2):
riscv: Expose orig_a0 in the user_regs_struct structure
riscv: selftests: Add a ptrace test to verify syscall parameter
modification
arch/riscv/include/asm/ptrace.h | 7 +-
arch/riscv/include/uapi/asm/ptrace.h | 2 +
tools/testing/selftests/riscv/Makefile | 2 +-
tools/testing/selftests/riscv/abi/.gitignore | 1 +
tools/testing/selftests/riscv/abi/Makefile | 12 ++
tools/testing/selftests/riscv/abi/ptrace.c | 124 +++++++++++++++++++
6 files changed, 144 insertions(+), 4 deletions(-)
create mode 100644 tools/testing/selftests/riscv/abi/.gitignore
create mode 100644 tools/testing/selftests/riscv/abi/Makefile
create mode 100644 tools/testing/selftests/riscv/abi/ptrace.c
base-commit: f2661062f16b2de5d7b6a5c42a9a5c96326b8454
--
2.34.1
In the TEST_F(epoll_busy_poll, test_get_params), the initialized value of 'ret' is unused,
because it will be assigned by the ioctl.thus remove it.
Signed-off-by: Liu Jing <liujing(a)cmss.chinamobile.com>
---
tools/testing/selftests/net/epoll_busy_poll.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/epoll_busy_poll.c b/tools/testing/selftests/net/epoll_busy_poll.c
index 16e457c2f877..652b0957b6c5 100644
--- a/tools/testing/selftests/net/epoll_busy_poll.c
+++ b/tools/testing/selftests/net/epoll_busy_poll.c
@@ -130,7 +130,7 @@ TEST_F(epoll_busy_poll, test_get_params)
* the default should be default and all fields should be zero'd by the
* kernel, so set params fields to garbage to test this.
*/
- int ret = 0;
+ int ret;
self->params.busy_poll_usecs = 0xff;
self->params.busy_poll_budget = 0xff;
--
2.33.0
We had several complains in linux-next that there were warnings:
CKI was not happy: it was the same situation than in an early report
when HID-BPF was initially included: the automatically generated
vmlinux.h doesn't contain all of the required structs and the
compilation of the bpf program fails.
We have multiple pointer to int cast complains and some docs that were
not rendered properly.
Include everything here.
Signed-off-by: Benjamin Tissoires <bentiss(a)kernel.org>
---
Changes in v2:
- Also fix the pointer to int casts
- Also fix the docs complains
- Link to v1: https://lore.kernel.org/r/20240627-fix-cki-v1-1-2b47ceac116a@kernel.org
---
Benjamin Tissoires (4):
selftests/hid: ensure CKI can compile our new tests on old kernels
HID: bpf: fix gcc warning and unify __u64 into u64
HID: bpf: doc fixes for hid_hw_request() hooks
HID: bpf: doc fixes for hid_hw_request() hooks
drivers/hid/bpf/hid_bpf_dispatch.c | 8 +++---
drivers/hid/bpf/hid_bpf_struct_ops.c | 2 +-
drivers/hid/hid-core.c | 4 +--
drivers/hid/hidraw.c | 6 ++---
include/linux/hid_bpf.h | 31 +++++++++++++---------
.../testing/selftests/hid/progs/hid_bpf_helpers.h | 16 +++++++++++
6 files changed, 44 insertions(+), 23 deletions(-)
---
base-commit: d3e15189bfd4d0a9d3a7ad8bd0e6ebb1c0419f93
change-id: 20240627-fix-cki-f372855cbf6f
Best regards,
--
Benjamin Tissoires <bentiss(a)kernel.org>
Centralizes the definition of _GNU_SOURCE into lib.mk and addresses all
resulting macro redefinition warnings.
The initial attempt at this patch was abandoned because it affected
lines in many source files and caused a large amount of churn. However,
from earlier discussions, centralizing _GNU_SOURCE is still desireable.
This attempt limits the changes to 1 source file and 14 Makefiles.
This is condensed into a single commit to avoid redefinition warnings
from partial merges.
v1: https://lore.kernel.org/linux-kselftest/20240430235057.1351993-1-edliaw@goo…
v2: https://lore.kernel.org/linux-kselftest/20240507214254.2787305-1-edliaw@goo…
- Add -D_GNU_SOURCE to KHDR_INCLUDES so that it is in a single
location.
- Remove #define _GNU_SOURCE from source code to resolve redefinition
warnings.
v3: https://lore.kernel.org/linux-kselftest/20240509200022.253089-1-edliaw@goog…
- Rebase onto linux-next 20240508.
- Split patches by directory.
- Add -D_GNU_SOURCE directly to CFLAGS in lib.mk.
- Delete additional _GNU_SOURCE definitions from source code in
linux-next.
- Delete additional -D_GNU_SOURCE flags from Makefiles.
v4: https://lore.kernel.org/linux-kselftest/20240510000842.410729-1-edliaw@goog…
- Rebase onto linux-next 20240509.
- Remove Fixes tag from patches that drop _GNU_SOURCE definition.
- Restore space between comment and includes for selftests/damon.
v5: https://lore.kernel.org/linux-kselftest/20240522005913.3540131-1-edliaw@goo…
- Rebase onto linux-next 20240521
- Drop initial patches that modify KHDR_INCLUDES.
- Incorporate Mark Brown's patch to replace static_assert with warning.
- Don't drop #define _GNU_SOURCE from nolibc and wireguard.
- Change Makefiles for x86 and vDSO to append to CFLAGS.
v6: https://lore.kernel.org/linux-kselftest/20240624232718.1154427-1-edliaw@goo…
- Rewrite patch to use -D_GNU_SOURCE= form in lib.mk.
- Reduce the amount of churn significantly by allowing definition to
coexist with source code macro defines.
v7:
- Squash patch into a single commit.
Edward Liaw (1):
selftests: Centralize -D_GNU_SOURCE= to CFLAGS in lib.mk
tools/testing/selftests/exec/Makefile | 1 -
tools/testing/selftests/futex/functional/Makefile | 2 +-
tools/testing/selftests/intel_pstate/Makefile | 2 +-
tools/testing/selftests/iommu/Makefile | 2 --
tools/testing/selftests/kvm/Makefile | 2 +-
tools/testing/selftests/lib.mk | 3 +++
tools/testing/selftests/mm/thuge-gen.c | 2 +-
tools/testing/selftests/net/Makefile | 2 +-
tools/testing/selftests/net/tcp_ao/Makefile | 2 +-
tools/testing/selftests/proc/Makefile | 1 -
tools/testing/selftests/resctrl/Makefile | 2 +-
tools/testing/selftests/ring-buffer/Makefile | 1 -
tools/testing/selftests/riscv/mm/Makefile | 2 +-
tools/testing/selftests/sgx/Makefile | 2 +-
tools/testing/selftests/tmpfs/Makefile | 1 -
15 files changed, 12 insertions(+), 15 deletions(-)
--
2.45.2.803.g4e1b14247a-goog
From: Adrian Moreno <amorenoz(a)redhat.com>
[ Upstream commit a8763466669d21b570b26160d0a5e0a2ee529d22 ]
Netlink flags, although they don't have payload at the netlink level,
are represented as having "True" as value in pyroute2.
Without it, trying to add a flow with a flag-type action (e.g: pop_vlan)
fails with the following traceback:
Traceback (most recent call last):
File "[...]/ovs-dpctl.py", line 2498, in <module>
sys.exit(main(sys.argv))
^^^^^^^^^^^^^^
File "[...]/ovs-dpctl.py", line 2487, in main
ovsflow.add_flow(rep["dpifindex"], flow)
File "[...]/ovs-dpctl.py", line 2136, in add_flow
reply = self.nlm_request(
^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/nlsocket.py", line 822, in nlm_request
return tuple(self._genlm_request(*argv, **kwarg))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/generic/__init__.py", line 126, in
nlm_request
return tuple(super().nlm_request(*argv, **kwarg))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/nlsocket.py", line 1124, in nlm_request
self.put(msg, msg_type, msg_flags, msg_seq=msg_seq)
File "[...]/pyroute2/netlink/nlsocket.py", line 389, in put
self.sendto_gate(msg, addr)
File "[...]/pyroute2/netlink/nlsocket.py", line 1056, in sendto_gate
msg.encode()
File "[...]/pyroute2/netlink/__init__.py", line 1245, in encode
offset = self.encode_nlas(offset)
^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/__init__.py", line 1560, in encode_nlas
nla_instance.setvalue(cell[1])
File "[...]/pyroute2/netlink/__init__.py", line 1265, in setvalue
nlv.setvalue(nla_tuple[1])
~~~~~~~~~^^^
IndexError: list index out of range
Signed-off-by: Adrian Moreno <amorenoz(a)redhat.com>
Acked-by: Aaron Conole <aconole(a)redhat.com>
Signed-off-by: David S. Miller <davem(a)davemloft.net>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/net/openvswitch/ovs-dpctl.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 5e0e539a323d5..8b120718768ec 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -531,7 +531,7 @@ class ovsactions(nla):
for flat_act in parse_flat_map:
if parse_starts_block(actstr, flat_act[0], False):
actstr = actstr[len(flat_act[0]):]
- self["attrs"].append([flat_act[1]])
+ self["attrs"].append([flat_act[1], True])
actstr = actstr[strspn(actstr, ", ") :]
parsed = True
--
2.43.0
From: Adrian Moreno <amorenoz(a)redhat.com>
[ Upstream commit a8763466669d21b570b26160d0a5e0a2ee529d22 ]
Netlink flags, although they don't have payload at the netlink level,
are represented as having "True" as value in pyroute2.
Without it, trying to add a flow with a flag-type action (e.g: pop_vlan)
fails with the following traceback:
Traceback (most recent call last):
File "[...]/ovs-dpctl.py", line 2498, in <module>
sys.exit(main(sys.argv))
^^^^^^^^^^^^^^
File "[...]/ovs-dpctl.py", line 2487, in main
ovsflow.add_flow(rep["dpifindex"], flow)
File "[...]/ovs-dpctl.py", line 2136, in add_flow
reply = self.nlm_request(
^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/nlsocket.py", line 822, in nlm_request
return tuple(self._genlm_request(*argv, **kwarg))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/generic/__init__.py", line 126, in
nlm_request
return tuple(super().nlm_request(*argv, **kwarg))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/nlsocket.py", line 1124, in nlm_request
self.put(msg, msg_type, msg_flags, msg_seq=msg_seq)
File "[...]/pyroute2/netlink/nlsocket.py", line 389, in put
self.sendto_gate(msg, addr)
File "[...]/pyroute2/netlink/nlsocket.py", line 1056, in sendto_gate
msg.encode()
File "[...]/pyroute2/netlink/__init__.py", line 1245, in encode
offset = self.encode_nlas(offset)
^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/__init__.py", line 1560, in encode_nlas
nla_instance.setvalue(cell[1])
File "[...]/pyroute2/netlink/__init__.py", line 1265, in setvalue
nlv.setvalue(nla_tuple[1])
~~~~~~~~~^^^
IndexError: list index out of range
Signed-off-by: Adrian Moreno <amorenoz(a)redhat.com>
Acked-by: Aaron Conole <aconole(a)redhat.com>
Signed-off-by: David S. Miller <davem(a)davemloft.net>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/net/openvswitch/ovs-dpctl.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 5e0e539a323d5..8b120718768ec 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -531,7 +531,7 @@ class ovsactions(nla):
for flat_act in parse_flat_map:
if parse_starts_block(actstr, flat_act[0], False):
actstr = actstr[len(flat_act[0]):]
- self["attrs"].append([flat_act[1]])
+ self["attrs"].append([flat_act[1], True])
actstr = actstr[strspn(actstr, ", ") :]
parsed = True
--
2.43.0
Correctable memory errors are very common on servers with large
amount of memory, and are corrected by ECC, but with two
pain points to users:
1. Correction usually happens on the fly and adds latency overhead
2. Not-fully-proved theory states excessive correctable memory
errors can develop into uncorrectable memory error.
Soft offline is kernel's additional solution for memory pages
having (excessive) corrected memory errors. Impacted page is migrated
to healthy page if it is in use, then the original page is discarded
for any future use.
The actual policy on whether (and when) to soft offline should be
maintained by userspace, especially in case of an 1G HugeTLB page.
Soft-offline dissolves the HugeTLB page, either in-use or free, into
chunks of 4K pages, reducing HugeTLB pool capacity by 1 hugepage.
If userspace has not acknowledged such behavior, it may be surprised
when later mmap hugepages MAP_FAILED due to lack of hugepages.
In case of a transparent hugepage, it will be split into 4K pages
as well; userspace will stop enjoying the transparent performance.
In addition, discarding the entire 1G HugeTLB page only because of
corrected memory errors sounds very costly and kernel better not
doing under the hood. But today there are at least 2 such cases:
1. GHES driver sees both GHES_SEV_CORRECTED and
CPER_SEC_ERROR_THRESHOLD_EXCEEDED after parsing CPER.
2. RAS Correctable Errors Collector counts correctable errors per
PFN and when the counter for a PFN reaches threshold
In both cases, userspace has no control of the soft offline performed
by kernel's memory failure recovery.
This patch series give userspace the control of softofflining any page:
kernel only soft offlines raw page / transparent hugepage / HugeTLB
hugepage if userspace has agreed to. The interface to userspace is a
new sysctl called enable_soft_offline under /proc/sys/vm. By default
enable_soft_line is 1 to preserve existing behavior in kernel.
Changelog
v6 => v7
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com> and David
Rientjes <rientjes(a)google.com>
* remove PFN value from enable_soft_offline log
* save/restore enable_soft_offline in run_vmtests.sh
* v7 is based on commit 7c89bdbd3778 ("khugepaged: simplify the
allocation of slab caches")
v5 => v6:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>
* add a ':' in soft offline log.
* close hugetlbfs file descriptor in selftest.
* no need to "return" after ksft_exit_fail_msg.
v4 => v5:
* incorporate feedbacks from Muhammad Usama Anjum
<usama.anjum(a)collabora.com>
* refactor selftest to use what available in kselftest.h
v3 => v4:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>,
Andrew Morton <akpm(a)linux-foundation.org>, and
Oscar Salvador <osalvador(a)suse.de>.
* insert a refactor commit to unify soft offline's logs to follow
"Soft offline: 0x${pfn}: ${message}" format.
* some rewords in document: fail => will not perform.
* v4 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3"),
akpm/mm-stable.
v2 => v3:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>,
Lance Yang <ioworker0(a)gmail.com>, Oscar Salvador <osalvador(a)suse.de>,
and David Rientjes <rientjes(a)google.com>.
* release potential refcount if enable_soft_offline is 0.
* soft_offline_page() returns EOPNOTSUPP if enable_soft_offline is 0.
* refactor hugetlb-soft-offline.c, for example, introduce
test_soft_offline_common to reduce repeated code.
* rewrite enable_soft_offline's documentation, adds more details about
the cost of soft-offline for transparent and hugetlb hugepages, and
components that are impacted when enable_soft_offline becomes 0.
* fix typos in commit messages.
* v3 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3").
v1 => v2:
* incorporate feedbacks from both Miaohe Lin <linmiaohe(a)huawei.com> and
Jane Chu <jane.chu(a)oracle.com>.
* make the switch to control all pages, instead of HugeTLB specific.
* change the API from
/sys/kernel/mm/hugepages/hugepages-${size}kB/softoffline_corrected_errors
to /proc/sys/vm/enable_soft_offline.
* minor update to test code.
* update documentation of the user control API.
* v2 is based on commit 83a7eefedc9b ("Linux 6.10-rc3").
Jiaqi Yan (4):
mm/memory-failure: refactor log format in soft offline code
mm/memory-failure: userspace controls soft-offlining pages
selftest/mm: test enable_soft_offline behaviors
docs: mm: add enable_soft_offline sysctl
Documentation/admin-guide/sysctl/vm.rst | 32 +++
mm/memory-failure.c | 37 ++-
tools/testing/selftests/mm/.gitignore | 1 +
tools/testing/selftests/mm/Makefile | 1 +
.../selftests/mm/hugetlb-soft-offline.c | 228 ++++++++++++++++++
tools/testing/selftests/mm/run_vmtests.sh | 6 +
6 files changed, 297 insertions(+), 8 deletions(-)
create mode 100644 tools/testing/selftests/mm/hugetlb-soft-offline.c
--
2.45.2.803.g4e1b14247a-goog
Currently, if a user wants to run pmtu.sh and cover all the provided test
cases, they need to install the Open vSwitch userspace utilities. This
dependency is difficult for users as well as CI environments, because the
userspace build and setup may require lots of support and devel packages
to be installed, system setup to be correct, and things like permissions
and selinux policies to be properly configured.
The kernel selftest suite includes an ovs-dpctl.py utility which can
interact with the openvswitch module directly. This lets developers and
CI environments run without needing too many extra dependencies - just
the pyroute2 python package.
This series enhances the ovs-dpctl utility to provide support for set()
and tunnel() flow specifiers, better ipv6 handling support, and the
ability to add tunnel vports, and LWT interfaces. Finally, it modifies
the pmtu.sh script to call the ovs-dpctl.py utility rather than the
typical OVS userspace utilities. The pmtu.sh can still fall back on
the Open vSwitch userspace utilities if the ovs-dpctl.py script can't
be used.
Aaron Conole (7):
selftests: openvswitch: Support explicit tunnel port creation.
selftests: openvswitch: Refactor actions parsing.
selftests: openvswitch: Add set() and set_masked() support.
selftests: openvswitch: Add support for tunnel() key.
selftests: openvswitch: Support implicit ipv6 arguments.
selftests: net: Use the provided dpctl rather than the vswitchd for
tests.
selftests: net: add config for openvswitch
tools/testing/selftests/net/config | 5 +
.../selftests/net/openvswitch/ovs-dpctl.py | 368 +++++++++++++++---
tools/testing/selftests/net/pmtu.sh | 145 +++++--
3 files changed, 451 insertions(+), 67 deletions(-)
--
2.45.1
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v5:
- keep make_server and make_client as Eduard suggested.
v4:
- a new patch to use make_sockaddr in sockmap_ktls.
- a new patch to close fd in error path in drop_on_reuseport.
- drop make_server() in patch 7.
- drop make_client() too in patch 9.
v3:
- a new patch to add backlog for network_helper_opts.
- use start_server_str in sockmap_ktls now, not start_server.
v2:
- address Eduard's comments in v1. (thanks)
- fix errors reported by CI.
This patch set uses network helpers in sockmap_ktls and sk_lookup, and
drop three local helpers tcp_server(), inetaddr_len() and make_socket()
in them.
Geliang Tang (9):
selftests/bpf: Add backlog for network_helper_opts
selftests/bpf: Use start_server_str in sockmap_ktls
selftests/bpf: Use connect_to_fd in sockmap_ktls
selftests/bpf: Use make_sockaddr in sockmap_ktls
selftests/bpf: Close fd in error path in drop_on_reuseport
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Use connect_to_fd in sk_lookup
selftests/bpf: Use connect_to_addr in sk_lookup
selftests/bpf: Drop make_socket in sk_lookup
tools/testing/selftests/bpf/network_helpers.c | 2 +-
tools/testing/selftests/bpf/network_helpers.h | 1 +
.../selftests/bpf/prog_tests/sk_lookup.c | 141 +++++++-----------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 51 ++-----
4 files changed, 61 insertions(+), 134 deletions(-)
--
2.43.0
Adds a simple implementation of strerror() and makes use of it in
kselftests.
Shuah, could you Ack patch 3?
Willy, this should work *without* your Ack.
Signed-off-by: Thomas Weißschuh <linux(a)weissschuh.net>
---
Thomas Weißschuh (3):
selftests/nolibc: introduce condition to run tests only on nolibc
tools/nolibc: implement strerror()
selftests: kselftest: also use strerror() on nolibc
tools/include/nolibc/stdio.h | 10 ++++++++
tools/testing/selftests/kselftest.h | 8 -------
tools/testing/selftests/nolibc/nolibc-test.c | 36 ++++++++++++++++++----------
3 files changed, 33 insertions(+), 21 deletions(-)
---
base-commit: a3063ba97f31e0364379a3ffc567203e3f79e877
change-id: 20240425-nolibc-strerror-67f4bfa03035
Best regards,
--
Thomas Weißschuh <linux(a)weissschuh.net>
** Background **
Currently, OVS supports several packet sampling mechanisms (sFlow,
per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
userspace action that needs to be handled by ovs-vswitchd's handler
threads only to be forwarded to some third party application that
will somehow process the sample and provide observability on the
datapath.
A particularly interesting use-case is controller-driven
per-flow IPFIX sampling where the OpenFlow controller can add metadata
to samples (via two 32bit integers) and this metadata is then available
to the sample-collecting system for correlation.
** Problem **
The fact that sampled traffic share netlink sockets and handler thread
time with upcalls, apart from being a performance bottleneck in the
sample extraction itself, can severely compromise the datapath,
yielding this solution unfit for highly loaded production systems.
Users are left with little options other than guessing what sampling
rate will be OK for their traffic pattern and system load and dealing
with the lost accuracy.
Looking at available infrastructure, an obvious candidated would be
to use psample. However, it's current state does not help with the
use-case at stake because sampled packets do not contain user-defined
metadata.
** Proposal **
This series is an attempt to fix this situation by extending the
existing psample infrastructure to carry a variable length
user-defined cookie.
The main existing user of psample is tc's act_sample. It is also
extended to forward the action's cookie to psample.
Finally, a new OVS action (OVS_SAMPLE_ATTR_EMIT_SAMPLE) is created.
It accepts a group and an optional cookie and uses psample to
multicast the packet and the metadata.
--
v5 -> v6:
- Renamed emit_sample -> psample
- Addressed unused variable and conditionally compilation of function.
v4 -> v5:
- Rebased.
- Removed lefover enum value and wrapped some long lines in selftests.
v3 -> v4:
- Rebased.
- Addressed Jakub's comment on private and unused nla attributes.
v2 -> v3:
- Addressed comments from Simon, Aaron and Ilya.
- Dropped probability propagation in nested sample actions.
- Dropped patch v2's 7/9 in favor of a userspace implementation and
consume skb if emit_sample is the last action, same as we do with
userspace.
- Split ovs-dpctl.py features in independent patches.
v1 -> v2:
- Create a new action ("emit_sample") rather than reuse existing
"sample" one.
- Add probability semantics to psample's sampling rate.
- Store sampling probability in skb's cb area and use it in emit_sample.
- Test combining "emit_sample" with "trunc"
- Drop group_id filtering and tracepoint in psample.
rfc_v2 -> v1:
- Accommodate Ilya's comments.
- Split OVS's attribute in two attributes and simplify internal
handling of psample arguments.
- Extend psample and tc with a user-defined cookie.
- Add a tracepoint to psample to facilitate troubleshooting.
rfc_v1 -> rfc_v2:
- Use psample instead of a new OVS-only multicast group.
- Extend psample and tc with a user-defined cookie.
Adrian Moreno (10):
net: psample: add user cookie
net: sched: act_sample: add action cookie to sample
net: psample: skip packet copy if no listeners
net: psample: allow using rate as probability
net: openvswitch: add psample action
net: openvswitch: store sampling probability in cb.
selftests: openvswitch: add psample action
selftests: openvswitch: add userspace parsing
selftests: openvswitch: parse trunc action
selftests: openvswitch: add psample test
Documentation/netlink/specs/ovs_flow.yaml | 17 ++
include/net/psample.h | 5 +-
include/uapi/linux/openvswitch.h | 31 +-
include/uapi/linux/psample.h | 11 +-
net/openvswitch/Kconfig | 1 +
net/openvswitch/actions.c | 65 ++++-
net/openvswitch/datapath.h | 3 +
net/openvswitch/flow_netlink.c | 32 ++-
net/openvswitch/vport.c | 1 +
net/psample/psample.c | 16 +-
net/sched/act_sample.c | 12 +
.../selftests/net/openvswitch/openvswitch.sh | 115 +++++++-
.../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
13 files changed, 565 insertions(+), 16 deletions(-)
--
2.45.2
Correctable memory errors are very common on servers with large
amount of memory, and are corrected by ECC, but with two
pain points to users:
1. Correction usually happens on the fly and adds latency overhead
2. Not-fully-proved theory states excessive correctable memory
errors can develop into uncorrectable memory error.
Soft offline is kernel's additional solution for memory pages
having (excessive) corrected memory errors. Impacted page is migrated
to healthy page if it is in use, then the original page is discarded
for any future use.
The actual policy on whether (and when) to soft offline should be
maintained by userspace, especially in case of an 1G HugeTLB page.
Soft-offline dissolves the HugeTLB page, either in-use or free, into
chunks of 4K pages, reducing HugeTLB pool capacity by 1 hugepage.
If userspace has not acknowledged such behavior, it may be surprised
when later mmap hugepages MAP_FAILED due to lack of hugepages.
In case of a transparent hugepage, it will be split into 4K pages
as well; userspace will stop enjoying the transparent performance.
In addition, discarding the entire 1G HugeTLB page only because of
corrected memory errors sounds very costly and kernel better not
doing under the hood. But today there are at least 2 such cases:
1. GHES driver sees both GHES_SEV_CORRECTED and
CPER_SEC_ERROR_THRESHOLD_EXCEEDED after parsing CPER.
2. RAS Correctable Errors Collector counts correctable errors per
PFN and when the counter for a PFN reaches threshold
In both cases, userspace has no control of the soft offline performed
by kernel's memory failure recovery.
This patch series give userspace the control of softofflining any page:
kernel only soft offlines raw page / transparent hugepage / HugeTLB
hugepage if userspace has agreed to. The interface to userspace is a
new sysctl called enable_soft_offline under /proc/sys/vm. By default
enable_soft_line is 1 to preserve existing behavior in kernel.
Changelog
v5=> v6:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>
* add a ':' in soft offline log.
* close hugetlbfs file descriptor in selftest.
* no need to "return" after ksft_exit_fail_msg.
v4 => v5:
* incorporate feedbacks from Muhammad Usama Anjum
<usama.anjum(a)collabora.com>
* refactor selftest to use what available in kselftest.h
v3 => v4:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>,
Andrew Morton <akpm(a)linux-foundation.org>, and
Oscar Salvador <osalvador(a)suse.de>.
* insert a refactor commit to unify soft offline's logs to follow
"Soft offline: 0x${pfn}: ${message}" format.
* some rewords in document: fail => will not perform.
* v4 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3"),
akpm/mm-stable.
v2 => v3:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>,
Lance Yang <ioworker0(a)gmail.com>, Oscar Salvador <osalvador(a)suse.de>,
and David Rientjes <rientjes(a)google.com>.
* release potential refcount if enable_soft_offline is 0.
* soft_offline_page() returns EOPNOTSUPP if enable_soft_offline is 0.
* refactor hugetlb-soft-offline.c, for example, introduce
test_soft_offline_common to reduce repeated code.
* rewrite enable_soft_offline's documentation, adds more details about
the cost of soft-offline for transparent and hugetlb hugepages, and
components that are impacted when enable_soft_offline becomes 0.
* fix typos in commit messages.
* v3 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3").
v1 => v2:
* incorporate feedbacks from both Miaohe Lin <linmiaohe(a)huawei.com> and
Jane Chu <jane.chu(a)oracle.com>.
* make the switch to control all pages, instead of HugeTLB specific.
* change the API from
/sys/kernel/mm/hugepages/hugepages-${size}kB/softoffline_corrected_errors
to /proc/sys/vm/enable_soft_offline.
* minor update to test code.
* update documentation of the user control API.
* v2 is based on commit 83a7eefedc9b ("Linux 6.10-rc3").
Jiaqi Yan (4):
mm/memory-failure: refactor log format in soft offline code
mm/memory-failure: userspace controls soft-offlining pages
selftest/mm: test enable_soft_offline behaviors
docs: mm: add enable_soft_offline sysctl
Documentation/admin-guide/sysctl/vm.rst | 32 +++
mm/memory-failure.c | 38 ++-
tools/testing/selftests/mm/.gitignore | 1 +
tools/testing/selftests/mm/Makefile | 1 +
.../selftests/mm/hugetlb-soft-offline.c | 228 ++++++++++++++++++
tools/testing/selftests/mm/run_vmtests.sh | 4 +
6 files changed, 296 insertions(+), 8 deletions(-)
create mode 100644 tools/testing/selftests/mm/hugetlb-soft-offline.c
--
2.45.2.741.gdbec12cfda-goog
This patch series introduces a new user namespace capability set, as
well as some plumbing around it (i.e. sysctl, secbit, lsm support).
First patch goes over the motivations for this as well as prior art.
In summary, while user namespaces are a great success today in that they
avoid running a lot of code as root, they also expand the attack surface
of the kernel substantially which is often abused by attackers.
Methods exist to limit the creation of such namespaces [1], however,
application developers often need to assume that user namespaces are
available for various tasks such as sandboxing. Thus, instead of
restricting the creation of user namespaces, we offer ways for userspace
to limit the capabilities granted to them.
Why a new capability set and not something specific to the userns (e.g.
ioctl_ns)?
1. We can't really expect userspace to patch every single callsite
and opt-in this new security mechanism.
2. We don't necessarily want policies enforced at said callsites.
For example a service like systemd-machined or a PAM session need to
be able to place restrictions on any namespace spawned under it.
3. We would need to come up with inheritance rules, querying
capabilities, etc. At this point we're just reinventing capability
sets.
4. We can easily define interactions between capability sets, thus
helping with adoption (patch 2 is an example of this)
Some examples of how this could be leveraged in userspace:
- Prevent user from getting CAP_NET_ADMIN in user namespaces under SSH:
echo "auth optional pam_cap.so" >> /etc/pam.d/sshd
echo "!cap_net_admin $USER" >> /etc/security/capability.conf
capsh --secbits=$((1 << 8)) -- -c /usr/sbin/sshd
- Prevent containers from ever getting CAP_DAC_OVERRIDE:
systemd-run -p CapabilityBoundingSet=~CAP_DAC_OVERRIDE \
-p SecureBits=userns-strict-caps \
/usr/bin/dockerd
systemd-run -p UserNSCapabilities=~CAP_DAC_OVERRIDE \
/usr/bin/incusd
- Kernel could be vulnerable to CAP_SYS_RAWIO exploits, prevent it:
sysctl -w cap_bound_userns_mask=0x1fffffdffff
- Drop CAP_SYS_ADMIN for this shell and all the user namespaces below it:
bwrap --unshare-user --cap-drop CAP_SYS_ADMIN /bin/sh
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?…
---
Changes since v1:
- Add documentation
- Change commit wording
- Cleanup various aspects of the code based on feedback
- Add new CAP_SYS_CONTROL capability for sysctl check
- Add BPF-LSM support for modifying userns capabilities
---
Jonathan Calmels (4):
capabilities: Add user namespace capabilities
capabilities: Add securebit to restrict userns caps
capabilities: Add sysctl to mask off userns caps
bpf,lsm: Allow editing capabilities in BPF-LSM hooks
Documentation/filesystems/proc.rst | 1 +
Documentation/security/credentials.rst | 6 ++
fs/proc/array.c | 9 +++
include/linux/cred.h | 3 +
include/linux/lsm_hook_defs.h | 2 +-
include/linux/securebits.h | 1 +
include/linux/security.h | 4 +-
include/linux/user_namespace.h | 7 ++
include/uapi/linux/capability.h | 6 +-
include/uapi/linux/prctl.h | 7 ++
include/uapi/linux/securebits.h | 11 ++-
kernel/bpf/bpf_lsm.c | 55 +++++++++++++
kernel/cred.c | 3 +
kernel/sysctl.c | 10 +++
kernel/umh.c | 15 ++++
kernel/user_namespace.c | 80 +++++++++++++++++--
security/apparmor/lsm.c | 2 +-
security/commoncap.c | 62 +++++++++++++-
security/keys/process_keys.c | 3 +
security/security.c | 6 +-
security/selinux/hooks.c | 2 +-
security/selinux/include/classmap.h | 5 +-
.../selftests/bpf/prog_tests/deny_namespace.c | 12 ++-
.../selftests/bpf/progs/test_deny_namespace.c | 7 +-
24 files changed, 291 insertions(+), 28 deletions(-)
--
2.45.2
We cannot use CLONE_VFORK because we also need to wait for the timeout
signal.
Restore tests timeout by using the original fork() call in __run_test()
but also in __TEST_F_IMPL(). Also fix a race condition when waiting for
the test child process.
Because test metadata are shared between test processes, only the
parent process must set the test PID (child). Otherwise, t->pid may be
set to zero, leading to inconsistent error cases:
# RUN layout1.rule_on_mountpoint ...
# rule_on_mountpoint: Test ended in some other way [127]
# OK layout1.rule_on_mountpoint
ok 20 layout1.rule_on_mountpoint
As safeguards, initialize the "status" variable with a valid exit code,
and handle unknown test exits as errors.
The use of fork() introduces a new race condition in landlock/fs_test.c
which seems to be specific to hostfs bind mounts, but I haven't found
the root cause and it's difficult to trigger. I'll try to fix it with
another patch.
Cc: Christian Brauner <brauner(a)kernel.org>
Cc: Günther Noack <gnoack(a)google.com>
Cc: Jakub Kicinski <kuba(a)kernel.org>
Cc: Kees Cook <keescook(a)chromium.org>
Cc: Mark Brown <broonie(a)kernel.org>
Cc: Shuah Khan <shuah(a)kernel.org>
Cc: Will Drewry <wad(a)chromium.org>
Cc: stable(a)vger.kernel.org
Closes: https://lore.kernel.org/r/9341d4db-5e21-418c-bf9e-9ae2da7877e1@sirena.org.uk
Fixes: a86f18903db9 ("selftests/harness: Fix interleaved scheduling leading to race conditions")
Fixes: 24cf65a62266 ("selftests/harness: Share _metadata between forked processes")
Signed-off-by: Mickaël Salaün <mic(a)digikod.net>
Link: https://lore.kernel.org/r/20240621180605.834676-1-mic@digikod.net
---
tools/testing/selftests/kselftest_harness.h | 43 ++++++++++++---------
1 file changed, 24 insertions(+), 19 deletions(-)
diff --git a/tools/testing/selftests/kselftest_harness.h b/tools/testing/selftests/kselftest_harness.h
index b634969cbb6f..40723a6a083f 100644
--- a/tools/testing/selftests/kselftest_harness.h
+++ b/tools/testing/selftests/kselftest_harness.h
@@ -66,8 +66,6 @@
#include <sys/wait.h>
#include <unistd.h>
#include <setjmp.h>
-#include <syscall.h>
-#include <linux/sched.h>
#include "kselftest.h"
@@ -82,17 +80,6 @@
# define TH_LOG_ENABLED 1
#endif
-/* Wait for the child process to end but without sharing memory mapping. */
-static inline pid_t clone3_vfork(void)
-{
- struct clone_args args = {
- .flags = CLONE_VFORK,
- .exit_signal = SIGCHLD,
- };
-
- return syscall(__NR_clone3, &args, sizeof(args));
-}
-
/**
* TH_LOG()
*
@@ -437,7 +424,7 @@ static inline pid_t clone3_vfork(void)
} \
if (setjmp(_metadata->env) == 0) { \
/* _metadata and potentially self are shared with all forks. */ \
- child = clone3_vfork(); \
+ child = fork(); \
if (child == 0) { \
fixture_name##_setup(_metadata, self, variant->data); \
/* Let setup failure terminate early. */ \
@@ -1016,7 +1003,14 @@ void __wait_for_test(struct __test_metadata *t)
.sa_flags = SA_SIGINFO,
};
struct sigaction saved_action;
- int status;
+ /*
+ * Sets status so that WIFEXITED(status) returns true and
+ * WEXITSTATUS(status) returns KSFT_FAIL. This safe default value
+ * should never be evaluated because of the waitpid(2) check and
+ * SIGALRM handling.
+ */
+ int status = KSFT_FAIL << 8;
+ int child;
if (sigaction(SIGALRM, &action, &saved_action)) {
t->exit_code = KSFT_FAIL;
@@ -1028,7 +1022,15 @@ void __wait_for_test(struct __test_metadata *t)
__active_test = t;
t->timed_out = false;
alarm(t->timeout);
- waitpid(t->pid, &status, 0);
+ child = waitpid(t->pid, &status, 0);
+ if (child == -1 && errno != EINTR) {
+ t->exit_code = KSFT_FAIL;
+ fprintf(TH_LOG_STREAM,
+ "# %s: Failed to wait for PID %d (errno: %d)\n",
+ t->name, t->pid, errno);
+ return;
+ }
+
alarm(0);
if (sigaction(SIGALRM, &saved_action, NULL)) {
t->exit_code = KSFT_FAIL;
@@ -1083,6 +1085,7 @@ void __wait_for_test(struct __test_metadata *t)
WTERMSIG(status));
}
} else {
+ t->exit_code = KSFT_FAIL;
fprintf(TH_LOG_STREAM,
"# %s: Test ended in some other way [%u]\n",
t->name,
@@ -1218,6 +1221,7 @@ void __run_test(struct __fixture_metadata *f,
struct __test_xfail *xfail;
char test_name[1024];
const char *diagnostic;
+ int child;
/* reset test struct */
t->exit_code = KSFT_PASS;
@@ -1236,15 +1240,16 @@ void __run_test(struct __fixture_metadata *f,
fflush(stdout);
fflush(stderr);
- t->pid = clone3_vfork();
- if (t->pid < 0) {
+ child = fork();
+ if (child < 0) {
ksft_print_msg("ERROR SPAWNING TEST CHILD\n");
t->exit_code = KSFT_FAIL;
- } else if (t->pid == 0) {
+ } else if (child == 0) {
setpgrp();
t->fn(t, variant);
_exit(t->exit_code);
} else {
+ t->pid = child;
__wait_for_test(t);
}
ksft_print_msg(" %4s %s\n",
base-commit: 83a7eefedc9b56fe7bfeff13b6c7356688ffa670
--
2.45.2
The mirroring selftests work by sending ICMP traffic between two hosts.
Along the way, this traffic is mirrored to a gretap netdevice, and counter
taps are then installed strategically along the path of the mirrored
traffic to verify the mirroring took place.
The problem with this is that besides mirroring the primary traffic, any
other service traffic is mirrored as well. At the same time, because the
tests need to work in HW-offloaded scenarios, the ability of the device to
do arbitrary packet inspection should not be taken for granted. Most tests
therefore simply use matchall, one uses flower to match on IP address.
As a result, the selftests are noisy.
mirror_test() accommodated this noisiness by giving the counters an
allowance of several packets. But that only works up to a point, and on
busy systems won't be always enough.
In this patch set, clean up and stabilize the mirroring selftests. The
original intention was to port the tests over to UDP, but the logic of
ICMP ends up being so entangled in the mirroring selftests that the
changes feel overly invasive. Instead, ICMP is kept, but where possible,
we match on ICMP message type, thus filtering out hits by other ICMP
messages.
Where this is not practical (where the counter tap is put on a device
that carries encapsulated packets), switch the counter condition to _at
least_ X observed packets. This is less robust, but barely so --
probably the only scenario that this would not catch is something like
erroneous packet duplication, which would hopefully get caught by the
numerous other tests in this extensive suite.
- Patches #1 to #3 clean up parameters at various helpers.
- Patches #4 to #6 stabilize the mirroring selftests as described above.
- Mirroring tests currently allow testing SW datapath even on HW
netdevices by trapping traffic to the SW datapath. This complicates
the tests a bit without a good reason: to test SW datapath, just run
the selftests on the veth topology. Thus in patch #7, drop support for
this dual SW/HW testing.
- At this point, some cleanups were either made possible by the previous
patches, or were always possible. In patches #8 to #11, realize these
cleanups.
- In patch #12, fix mlxsw mirror_gre selftest to respect setting TESTS.
Petr Machata (12):
selftests: libs: Expand "$@" where possible
selftests: mirror: Drop direction argument from several functions
selftests: lib: tc_rule_stats_get(): Move default to argument
definition
selftests: mirror_gre_lag_lacp: Check counters at tunnel
selftests: mirror: do_test_span_dir_ips(): Install accurate taps
selftests: mirror: mirror_test(): Allow exact count of packets
selftests: mirror: Drop dual SW/HW testing
selftests: mlxsw: mirror_gre: Simplify
selftests: mirror_gre_lag_lacp: Drop unnecessary code
selftests: libs: Drop slow_path_trap_install()/_uninstall()
selftests: libs: Drop unused functions
selftests: mlxsw: mirror_gre: Obey TESTS
.../selftests/drivers/net/mlxsw/mirror_gre.sh | 71 ++++++---------
.../drivers/net/mlxsw/mirror_gre_scale.sh | 18 +---
tools/testing/selftests/net/forwarding/lib.sh | 83 +++++++++++------
.../selftests/net/forwarding/mirror_gre.sh | 45 +++-------
.../net/forwarding/mirror_gre_bound.sh | 23 +----
.../net/forwarding/mirror_gre_bridge_1d.sh | 21 +----
.../forwarding/mirror_gre_bridge_1d_vlan.sh | 21 +----
.../net/forwarding/mirror_gre_bridge_1q.sh | 21 +----
.../forwarding/mirror_gre_bridge_1q_lag.sh | 29 ++----
.../net/forwarding/mirror_gre_changes.sh | 73 ++++++---------
.../net/forwarding/mirror_gre_flower.sh | 43 ++++-----
.../net/forwarding/mirror_gre_lag_lacp.sh | 65 ++++++--------
.../net/forwarding/mirror_gre_lib.sh | 90 ++++++++++++++-----
.../net/forwarding/mirror_gre_neigh.sh | 39 +++-----
.../selftests/net/forwarding/mirror_gre_nh.sh | 35 ++------
.../net/forwarding/mirror_gre_vlan.sh | 21 +----
.../forwarding/mirror_gre_vlan_bridge_1q.sh | 69 ++++++--------
.../selftests/net/forwarding/mirror_lib.sh | 79 +++++++++++-----
.../selftests/net/forwarding/mirror_vlan.sh | 43 +++------
tools/testing/selftests/net/lib.sh | 4 +-
20 files changed, 355 insertions(+), 538 deletions(-)
--
2.45.0
Hello,
KernelCI is hosting a bi-weekly call on Thursday to discuss improvements
to existing upstream tests, the development of new tests to increase
kernel testing coverage, and the enablement of these tests in KernelCI.
In recent months, we at Collabora have focused on various kernel areas,
assessing the tests already available upstream and contributing patches
to make them easily runnable in CIs.
Below is a list of the tests we've been working on and their latest
status updates, as discussed in the last meeting held on 2024-06-27:
*USB/PCI devices kselftest*
- Upstream test to detect unprobed devices on discoverable buses:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?…
- Kernel patches to allow running the test on more platforms on KernelCI
were merged:
https://lore.kernel.org/all/20240613-kselftest-discoverable-probe-mt8195-kc…
- Waiting for KernelCI PRs to be merged:
https://github.com/kernelci/kernelci-core/pull/2577 and https://github.com/kernelci/kernelci-pipeline/pull/642
*Error log test*
- Proposing new kselftest to report device log errors:
https://lore.kernel.org/all/20240423-dev-err-log-selftest-v1-0-690c1741d68b…
- Currently fixing test failures in KernelCI
*Suspend/resume in cpufreq kselftest*
- Enabling suspend/resume test within the cpufreq kselftest in KernelCI
- Parameter support for running subtests in a kselftest was merged:
https://github.com/Linaro/test-definitions/pull/511
- Added rtcwake support in the test to enable automated resume, currently
testing/debugging solution
*Boot time test*
- Investigating possibility of adding new test upstream to measure the
kernel boot time and detect regressions
- Currently looking into boot tracing with ftrace events and kprobes
(see: https://www.kernel.org/doc/html/latest/trace/boottime-trace.html)
- Idea for potential kselftest: insert explicit tracepoints in strategic
places, let the user configure which times to measure. The test could
provide a bootconfig file and a fragment to enable the required configs.
This could be an alternative to using external tools (e.g. grabserial
w/ early serial port init).
- Need a list of functions to track in order to measure key metrics
(e.g. device tree overhead, probe overhead, module load overhead)
- Identify key drivers that need to be loaded early, for potentially
supporting a two-phase boot: (1) time-critical, and (2) rest of the
system
*Other interesting updates*
- Flaky serial on sc7180 was recently fixed:
https://github.com/kernelci/kernelci-project/issues/380 and https://lore.kernel.org/all/20240610222515.3023730-1-dianders@chromium.org/…
*Strategy for test enablement in KernelCI*
- Guidance on test quality: KernelCI should set the standard for test
quality, providing guidance on which tests to enable from various
projects (e.g., kselftest, LTP). By doing so, KernelCI can serve as a
model for other CI systems.
- Develop mechanisms to automatically detect which tests should run on a
specific platform
- Embed metadata in the test themselves to facilitate the test selection
process
- Leverage device tree info to determine the appropriate tests for each
platform
Please reply to this thread if you'd like to join the call or discuss
any of the topics further. We look forward to collaborating with the
community to improve upstream tests and expand coverage to more areas
of interest within the kernel.
Best regards,
Laura Nao
Changes v2:
- Removed patches 2 and 3 since now this part will be supported by the
kernel.
Sub-Numa Clustering (SNC) allows splitting CPU cores, caches and memory
into multiple NUMA nodes. When enabled, NUMA-aware applications can
achieve better performance on bigger server platforms.
SNC support in the kernel is currently in review [1]. With SNC enabled
and kernel support in place all the tests will function normally. There
might be a problem when SNC is enabled but the system is still using an
older kernel version without SNC support. Currently the only message
displayed in that situation is a guess that SNC might be enabled and is
causing issues. That message also is displayed whenever the test fails
on an Intel platform.
Add a mechanism to discover kernel support for SNC which will add more
meaning and certainty to the error message.
Series was tested on Ice Lake server platforms with SNC disabled, SNC-2
and SNC-4. The tests were also ran with and without kernel support for
SNC.
Series applies cleanly on kselftest/next.
[1] https://lore.kernel.org/all/20240503203325.21512-1-tony.luck@intel.com/
Previous versions of this series:
[v1] https://lore.kernel.org/all/cover.1709721159.git.maciej.wieczor-retman@inte…
Maciej Wieczor-Retman (2):
selftests/resctrl: Adjust effective L3 cache size with SNC enabled
selftests/resctrl: Adjust SNC support messages
tools/testing/selftests/resctrl/cat_test.c | 2 +-
tools/testing/selftests/resctrl/cmt_test.c | 6 +-
tools/testing/selftests/resctrl/mba_test.c | 2 +
tools/testing/selftests/resctrl/mbm_test.c | 4 +-
tools/testing/selftests/resctrl/resctrl.h | 8 +-
tools/testing/selftests/resctrl/resctrlfs.c | 131 +++++++++++++++++++-
6 files changed, 144 insertions(+), 9 deletions(-)
--
2.45.0
Allow userspace to change the guest-visible value of the register with
some severe limitation:
- No changes to features not virtualized by KVM (MPAM_frac, RAS_frac,
SME, RNDP_trap).
- No changes to features (CSV2_frac, NMI, MTE_frac, GCS, THE, MTEX,
DF2, PFAR) which haven't been added into the ftr_id_aa64pfr1[].
Because the struct arm64_ftr_bits definition for each feature in the
ftr_id_aa64pfr1[] is used by arm64_check_features. If they're not
existing in the ftr_id_aa64pfr1[], the for loop won't check the if
the new_val is safe for those features.
For the question why can't those fields be hidden depending on the VM
configuration? I don't find there is the related VM configuration, maybe we
should add the new VM configuration?
I'm not sure I'm right, so if there're any problems please help to point out and
I will fix them.
Also add the selftest for it.
Changelog:
----------
v2 -> v3:
* Give more description about why only part of the fields can be writable.
* Updated the writable mask by referring the latest ARM spec.
v1 -> v2:
* Tackling the full register instead of single field.
* Changing the patch title and commit message.
RFCv1 -> v1:
* Fix the compilation error.
* Delete the machine specific information and make the description more
generable.
RFCv1: https://lore.kernel.org/all/20240612023553.127813-1-shahuang@redhat.com/
v1: https://lore.kernel.org/all/20240617075131.1006173-1-shahuang@redhat.com/
v2: https://lore.kernel.org/all/20240618063808.1040085-1-shahuang@redhat.com/
Shaoqin Huang (2):
KVM: arm64: Allow userspace to change ID_AA64PFR1_EL1
KVM: selftests: aarch64: Add writable test for ID_AA64PFR1_EL1
arch/arm64/kvm/sys_regs.c | 4 +++-
tools/testing/selftests/kvm/aarch64/set_id_regs.c | 8 ++++++++
2 files changed, 11 insertions(+), 1 deletion(-)
--
2.40.1
Currently, we can run string-stream and assertion tests only when they
are built into the kernel (with config options = y), since some of the
symbols (string-stream functions and functions from assert.c) are not
exported into any of the namespaces, therefore they are not accessible
for the modules.
This patch series exports the required symbols into the KUnit namespace.
Also, it makes the string-stream test a separate module and removes the
log test stub from kunit-test since now we can access the string-stream
symbols even if the test which uses it is built as a module.
Additionally, this patch series merges the assertion test suite into the
kunit-test, since assert.c (and all of the assertion formatting
functions in it) is a part of the KUnit core.
V1 -> V2:
- Patch which exports the non-static assert.c functions is replaced with
the patch which prepares assert_test.c to be merged into kunit-test.c
- Also, David Gow <davidgow(a)google.com> suggested merging 4th and 5th
patches together, but since now the 4th patch does more than it used to
do, I send it separately
Ivan Orlov (5):
kunit: string-stream: export non-static functions
kunit: kunit-test: Remove stub for log tests
kunit: string-stream-test: Make it a separate module
kunit: assert_test: Prepare to be merged into kunit-test.c
kunit: Merge assertion test into kunit-test.c
include/kunit/assert.h | 4 +-
lib/kunit/Kconfig | 8 +
lib/kunit/Makefile | 7 +-
lib/kunit/assert.c | 19 +-
lib/kunit/assert_test.c | 388 --------------------------------
lib/kunit/kunit-test.c | 397 +++++++++++++++++++++++++++++++--
lib/kunit/string-stream-test.c | 2 +
lib/kunit/string-stream.c | 12 +-
8 files changed, 416 insertions(+), 421 deletions(-)
delete mode 100644 lib/kunit/assert_test.c
--
2.34.1
v14: https://patchwork.kernel.org/project/netdevbpf/list/?series=865135&archive=…
====
No material changes in this version. Only rebase and re-verification on
top of net-next. v13, I think, raced with commit ebad6d0334793
("net/ipv4: Use nested-BH locking for ipv4_tcp_sk.") being merged to
net-next that caused a patchwork failure to apply. This series should
apply cleanly on commit c4532232fa2a4 ("selftests: net: remove unneeded
IP_GRE config").
I did not wait the customary 24hr as Jakub said it's OK to repost as soon
as I build test the rebased version:
https://lore.kernel.org/netdev/20240625075926.146d769d@kernel.org/
v13: https://patchwork.kernel.org/project/netdevbpf/list/?series=861406&archive=…
====
Major changes:
--------------
This iteration addresses Pavel's review comments, applies his
reviewed-by's, and seeks to fix the patchwork build error (sorry!).
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v13/
v12: https://patchwork.kernel.org/project/netdevbpf/list/?series=859747&state=*
====
Major changes:
--------------
This iteration only addresses one minor comment from Pavel with regards
to the trace printing of netmem, and the patchwork build error
introduced in v11 because I missed doing an allmodconfig build, sorry.
Other than that v11, AFAICT, received no feedback. There is one
discussion about how the specifics of plugging io uring memory through
the page pool, but not relevant to content in this particular patchset,
AFAICT.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v12/
v11: https://patchwork.kernel.org/project/netdevbpf/list/?series=857457&state=*
====
Major Changes:
--------------
v11 addresses feedback received in v10. The major change is the removal
of the memory provider ops as requested by Christoph. We still
accomplish the same thing, but utilizing direct function calls with if
statements rather than generic ops.
Additionally address sparse warnings, bugs and review comments from
folks that reviewed.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v11/
Detailed changelog:
-------------------
- Fixes in netdev_rx_queue_restart() from Pavel & David.
- Remove commit e650e8c3a36f5 ("net: page_pool: create hooks for
custom page providers") from the series to address Christoph's
feedback and rebased other patches on the series on this change.
- Fixed build errors with CONFIG_DMA_SHARED_BUFFER &&
!CONFIG_GENERIC_ALLOCATOR build.
- Fixed sparse warnings pointed out by Paolo.
- Drop unnecessary gro_pull_from_frag0 checks.
- Added Bagas reviewed-by to docs.
Cc: Bagas Sanjaya <bagasdotme(a)gmail.com>
Cc: Steven Rostedt <rostedt(a)goodmis.org>
Cc: Christoph Hellwig <hch(a)infradead.org>
Cc: Nikolay Aleksandrov <razor(a)blackwall.org>
v10: https://patchwork.kernel.org/project/netdevbpf/list/?series=852422&state=*
====
Major Changes:
--------------
v9 was sent right before the merge window closed (sorry!). v10 is almost
a re-send of the series now that the merge window re-opened. Only
rebased to latest net-next and addressed some minor iterative comments
received on v9.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v10/
Detailed changelog:
-------------------
- Fixed tokens leaking in DONTNEED setsockopt (Nikolay).
- Moved net_iov_dma_addr() to devmem.c and made it a devmem specific
helpers (David).
- Rename hook alloc_pages to alloc_netmems as alloc_pages is now
preprocessor macro defined and causes a build error.
v9:
===
Major Changes:
--------------
GVE queue API has been merged. Submitting this version as non-RFC after
rebasing on top of the merged API, and dropped the out of tree queue API
I was carrying on github. Addressed the little feedback v8 has received.
Detailed changelog:
------------------
- Added new patch from David Wei to this series for
netdev_rx_queue_restart()
- Fixed sparse error.
- Removed CONFIG_ checks in netmem_is_net_iov()
- Flipped skb->readable to skb->unreadable
- Minor fixes to selftests & docs.
RFC v8:
=======
Major Changes:
--------------
- Fixed build error generated by patch-by-patch build.
- Applied docs suggestions from Randy.
RFC v7:
=======
Major Changes:
--------------
This revision largely rebases on top of net-next and addresses the feedback
RFCv6 received from folks, namely Jakub, Yunsheng, Arnd, David, & Pavel.
The series remains in RFC because the queue-API ndos defined in this
series are not yet implemented. I have a GVE implementation I carry out
of tree for my testing. A upstreamable GVE implementation is in the
works. Aside from that, in my estimation all the patches are ready for
review/merge. Please do take a look.
As usual the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v7/
Detailed changelog:
- Use admin-perm in netlink API.
- Addressed feedback from Jakub with regards to netlink API
implementation.
- Renamed devmem.c functions to something more appropriate for that
file.
- Improve the performance seen through the page_pool benchmark.
- Fix the value definition of all the SO_DEVMEM_* uapi.
- Various fixes to documentation.
Perf - page-pool benchmark:
---------------------------
Improved performance of bench_page_pool_simple.ko tests compared to v6:
https://pastebin.com/raw/v5dYRg8L
net-next base: 8 cycle fast path.
RFC v6: 10 cycle fast path.
RFC v7: 9 cycle fast path.
RFC v7 with CONFIG_DMA_SHARED_BUFFER disabled: 8 cycle fast path,
same as baseline.
Perf - Devmem TCP benchmark:
---------------------
Perf is about the same regardless of the changes in v7, namely the
removal of the static_branch_unlikely to improve the page_pool benchmark
performance:
189/200gbps bi-directional throughput with RX devmem TCP and regular TCP
TX i.e. ~95% line rate.
RFC v6:
=======
Major Changes:
--------------
This revision largely rebases on top of net-next and addresses the little
feedback RFCv5 received.
The series remains in RFC because the queue-API ndos defined in this
series are not yet implemented. I have a GVE implementation I carry out
of tree for my testing. A upstreamable GVE implementation is in the
works. Aside from that, in my estimation all the patches are ready for
review/merge. Please do take a look.
As usual the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v6/
This version also comes with some performance data recorded in the cover
letter (see below changelog).
Detailed changelog:
- Rebased on top of the merged netmem_ref changes.
- Converted skb->dmabuf to skb->readable (Pavel). Pavel's original
suggestion was to remove the skb->dmabuf flag entirely, but when I
looked into it closely, I found the issue that if we remove the flag
we have to dereference the shinfo(skb) pointer to obtain the first
frag to tell whether an skb is readable or not. This can cause a
performance regression if it dirties the cache line when the
shinfo(skb) was not really needed. Instead, I converted the skb->dmabuf
flag into a generic skb->readable flag which can be re-used by io_uring
0-copy RX.
- Squashed a few locking optimizations from Eric Dumazet in the RX path
and the DEVMEM_DONTNEED setsockopt.
- Expanded the tests a bit. Added validation for invalid scenarios and
added some more coverage.
Perf - page-pool benchmark:
---------------------------
bench_page_pool_simple.ko tests with and without these changes:
https://pastebin.com/raw/ncHDwAbn
AFAIK the number that really matters in the perf tests is the
'tasklet_page_pool01_fast_path Per elem'. This one measures at about 8
cycles without the changes but there is some 1 cycle noise in some
results.
With the patches this regresses to 9 cycles with the changes but there
is 1 cycle noise occasionally running this test repeatedly.
Lastly I tried disable the static_branch_unlikely() in
netmem_is_net_iov() check. To my surprise disabling the
static_branch_unlikely() check reduces the fast path back to 8 cycles,
but the 1 cycle noise remains.
Perf - Devmem TCP benchmark:
---------------------
189/200gbps bi-directional throughput with RX devmem TCP and regular TCP
TX i.e. ~95% line rate.
Major changes in RFC v5:
========================
1. Rebased on top of 'Abstract page from net stack' series and used the
new netmem type to refer to LSB set pointers instead of re-using
struct page.
2. Downgraded this series back to RFC and called it RFC v5. This is
because this series is now dependent on 'Abstract page from net
stack'[1] and the queue API. Both are removed from the series to
reduce the patch # and those bits are fairly independent or
pre-requisite work.
3. Reworked the page_pool devmem support to use netmem and for some
more unified handling.
4. Reworked the reference counting of net_iov (renamed from
page_pool_iov) to use pp_ref_count for refcounting.
The full changes including the dependent series and GVE page pool
support is here:
https://github.com/mina/linux/commits/tcpdevmem-rfcv5/
[1] https://patchwork.kernel.org/project/netdevbpf/list/?series=810774
Major changes in v1:
====================
1. Implemented MVP queue API ndos to remove the userspace-visible
driver reset.
2. Fixed issues in the napi_pp_put_page() devmem frag unref path.
3. Removed RFC tag.
Many smaller addressed comments across all the patches (patches have
individual change log).
Full tree including the rest of the GVE driver changes:
https://github.com/mina/linux/commits/tcpdevmem-v1
Changes in RFC v3:
==================
1. Pulled in the memory-provider dependency from Jakub's RFC[1] to make the
series reviewable and mergeable.
2. Implemented multi-rx-queue binding which was a todo in v2.
3. Fix to cmsg handling.
The sticking point in RFC v2[2] was the device reset required to refill
the device rx-queues after the dmabuf bind/unbind. The solution
suggested as I understand is a subset of the per-queue management ops
Jakub suggested or similar:
https://lore.kernel.org/netdev/20230815171638.4c057dcd@kernel.org/
This is not addressed in this revision, because:
1. This point was discussed at netconf & netdev and there is openness to
using the current approach of requiring a device reset.
2. Implementing individual queue resetting seems to be difficult for my
test bed with GVE. My prototype to test this ran into issues with the
rx-queues not coming back up properly if reset individually. At the
moment I'm unsure if it's a mistake in the POC or a genuine issue in
the virtualization stack behind GVE, which currently doesn't test
individual rx-queue restart.
3. Our usecases are not bothered by requiring a device reset to refill
the buffer queues, and we'd like to support NICs that run into this
limitation with resetting individual queues.
My thought is that drivers that have trouble with per-queue configs can
use the support in this series, while drivers that support new netdev
ops to reset individual queues can automatically reset the queue as
part of the dma-buf bind/unbind.
The same approach with device resets is presented again for consideration
with other sticking points addressed.
This proposal includes the rx devmem path only proposed for merge. For a
snapshot of my entire tree which includes the GVE POC page pool support &
device memory support:
https://github.com/torvalds/linux/compare/master...mina:linux:tcpdevmem-v3
[1] https://lore.kernel.org/netdev/f8270765-a27b-6ccf-33ea-cda097168d79@redhat.…
[2] https://lore.kernel.org/netdev/CAHS8izOVJGJH5WF68OsRWFKJid1_huzzUK+hpKbLcL4…
Changes in RFC v2:
==================
The sticking point in RFC v1[1] was the dma-buf pages approach we used to
deliver the device memory to the TCP stack. RFC v2 is a proof-of-concept
that attempts to resolve this by implementing scatterlist support in the
networking stack, such that we can import the dma-buf scatterlist
directly. This is the approach proposed at a high level here[2].
Detailed changes:
1. Replaced dma-buf pages approach with importing scatterlist into the
page pool.
2. Replace the dma-buf pages centric API with a netlink API.
3. Removed the TX path implementation - there is no issue with
implementing the TX path with scatterlist approach, but leaving
out the TX path makes it easier to review.
4. Functionality is tested with this proposal, but I have not conducted
perf testing yet. I'm not sure there are regressions, but I removed
perf claims from the cover letter until they can be re-confirmed.
5. Added Signed-off-by: contributors to the implementation.
6. Fixed some bugs with the RX path since RFC v1.
Any feedback welcome, but specifically the biggest pending questions
needing feedback IMO are:
1. Feedback on the scatterlist-based approach in general.
2. Netlink API (Patch 1 & 2).
3. Approach to handle all the drivers that expect to receive pages from
the page pool (Patch 6).
[1] https://lore.kernel.org/netdev/dfe4bae7-13a0-3c5d-d671-f61b375cb0b4@gmail.c…
[2] https://lore.kernel.org/netdev/CAHS8izPm6XRS54LdCDZVd0C75tA1zHSu6jLVO8nzTLX…
==================
* TL;DR:
Device memory TCP (devmem TCP) is a proposal for transferring data to and/or
from device memory efficiently, without bouncing the data to a host memory
buffer.
* Problem:
A large amount of data transfers have device memory as the source and/or
destination. Accelerators drastically increased the volume of such transfers.
Some examples include:
- ML accelerators transferring large amounts of training data from storage into
GPU/TPU memory. In some cases ML training setup time can be as long as 50% of
TPU compute time, improving data transfer throughput & efficiency can help
improving GPU/TPU utilization.
- Distributed training, where ML accelerators, such as GPUs on different hosts,
exchange data among them.
- Distributed raw block storage applications transfer large amounts of data with
remote SSDs, much of this data does not require host processing.
Today, the majority of the Device-to-Device data transfers the network are
implemented as the following low level operations: Device-to-Host copy,
Host-to-Host network transfer, and Host-to-Device copy.
The implementation is suboptimal, especially for bulk data transfers, and can
put significant strains on system resources, such as host memory bandwidth,
PCIe bandwidth, etc. One important reason behind the current state is the
kernel’s lack of semantics to express device to network transfers.
* Proposal:
In this patch series we attempt to optimize this use case by implementing
socket APIs that enable the user to:
1. send device memory across the network directly, and
2. receive incoming network packets directly into device memory.
Packet _payloads_ go directly from the NIC to device memory for receive and from
device memory to NIC for transmit.
Packet _headers_ go to/from host memory and are processed by the TCP/IP stack
normally. The NIC _must_ support header split to achieve this.
Advantages:
- Alleviate host memory bandwidth pressure, compared to existing
network-transfer + device-copy semantics.
- Alleviate PCIe BW pressure, by limiting data transfer to the lowest level
of the PCIe tree, compared to traditional path which sends data through the
root complex.
* Patch overview:
** Part 1: netlink API
Gives user ability to bind dma-buf to an RX queue.
** Part 2: scatterlist support
Currently the standard for device memory sharing is DMABUF, which doesn't
generate struct pages. On the other hand, networking stack (skbs, drivers, and
page pool) operate on pages. We have 2 options:
1. Generate struct pages for dmabuf device memory, or,
2. Modify the networking stack to process scatterlist.
Approach #1 was attempted in RFC v1. RFC v2 implements approach #2.
** part 3: page pool support
We piggy back on page pool memory providers proposal:
https://github.com/kuba-moo/linux/tree/pp-providers
It allows the page pool to define a memory provider that provides the
page allocation and freeing. It helps abstract most of the device memory
TCP changes from the driver.
** part 4: support for unreadable skb frags
Page pool iovs are not accessible by the host; we implement changes
throughput the networking stack to correctly handle skbs with unreadable
frags.
** Part 5: recvmsg() APIs
We define user APIs for the user to send and receive device memory.
Not included with this series is the GVE devmem TCP support, just to
simplify the review. Code available here if desired:
https://github.com/mina/linux/tree/tcpdevmem
This series is built on top of net-next with Jakub's pp-providers changes
cherry-picked.
* NIC dependencies:
1. (strict) Devmem TCP require the NIC to support header split, i.e. the
capability to split incoming packets into a header + payload and to put
each into a separate buffer. Devmem TCP works by using device memory
for the packet payload, and host memory for the packet headers.
2. (optional) Devmem TCP works better with flow steering support & RSS support,
i.e. the NIC's ability to steer flows into certain rx queues. This allows the
sysadmin to enable devmem TCP on a subset of the rx queues, and steer
devmem TCP traffic onto these queues and non devmem TCP elsewhere.
The NIC I have access to with these properties is the GVE with DQO support
running in Google Cloud, but any NIC that supports these features would suffice.
I may be able to help reviewers bring up devmem TCP on their NICs.
* Testing:
The series includes a udmabuf kselftest that show a simple use case of
devmem TCP and validates the entire data path end to end without
a dependency on a specific dmabuf provider.
** Test Setup
Kernel: net-next with this series and memory provider API cherry-picked
locally.
Hardware: Google Cloud A3 VMs.
NIC: GVE with header split & RSS & flow steering support.
Cc: Pavel Begunkov <asml.silence(a)gmail.com>
Cc: David Wei <dw(a)davidwei.uk>
Cc: Jason Gunthorpe <jgg(a)ziepe.ca>
Cc: Yunsheng Lin <linyunsheng(a)huawei.com>
Cc: Shailend Chand <shailend(a)google.com>
Cc: Harshitha Ramamurthy <hramamurthy(a)google.com>
Cc: Shakeel Butt <shakeel.butt(a)linux.dev>
Cc: Jeroen de Borst <jeroendb(a)google.com>
Cc: Praveen Kaligineedi <pkaligineedi(a)google.com>
Mina Almasry (13):
netdev: add netdev_rx_queue_restart()
net: netdev netlink api to bind dma-buf to a net device
netdev: support binding dma-buf to netdevice
netdev: netdevice devmem allocator
page_pool: convert to use netmem
page_pool: devmem support
memory-provider: dmabuf devmem memory provider
net: support non paged skb frags
net: add support for skbs with unreadable frags
tcp: RX path for devmem TCP
net: add SO_DEVMEM_DONTNEED setsockopt to release RX frags
net: add devmem TCP documentation
selftests: add ncdevmem, netcat for devmem TCP
Documentation/netlink/specs/netdev.yaml | 57 +++
Documentation/networking/devmem.rst | 258 +++++++++++
Documentation/networking/index.rst | 1 +
arch/alpha/include/uapi/asm/socket.h | 6 +
arch/mips/include/uapi/asm/socket.h | 6 +
arch/parisc/include/uapi/asm/socket.h | 6 +
arch/sparc/include/uapi/asm/socket.h | 6 +
include/linux/skbuff.h | 61 ++-
include/linux/skbuff_ref.h | 11 +-
include/linux/socket.h | 1 +
include/net/devmem.h | 124 ++++++
include/net/mp_dmabuf_devmem.h | 44 ++
include/net/netdev_rx_queue.h | 5 +
include/net/netmem.h | 208 ++++++++-
include/net/page_pool/helpers.h | 124 ++++--
include/net/page_pool/types.h | 22 +-
include/net/sock.h | 2 +
include/net/tcp.h | 5 +-
include/trace/events/page_pool.h | 30 +-
include/uapi/asm-generic/socket.h | 6 +
include/uapi/linux/netdev.h | 19 +
include/uapi/linux/uio.h | 17 +
net/bpf/test_run.c | 5 +-
net/core/Makefile | 3 +-
net/core/datagram.c | 6 +
net/core/dev.c | 6 +-
net/core/devmem.c | 376 ++++++++++++++++
net/core/gro.c | 3 +-
net/core/netdev-genl-gen.c | 23 +
net/core/netdev-genl-gen.h | 6 +
net/core/netdev-genl.c | 103 +++++
net/core/netdev_rx_queue.c | 74 ++++
net/core/page_pool.c | 362 +++++++++-------
net/core/skbuff.c | 83 +++-
net/core/sock.c | 61 +++
net/ipv4/esp4.c | 3 +-
net/ipv4/tcp.c | 261 +++++++++++-
net/ipv4/tcp_input.c | 13 +-
net/ipv4/tcp_ipv4.c | 16 +
net/ipv4/tcp_minisocks.c | 2 +
net/ipv4/tcp_output.c | 5 +-
net/ipv6/esp6.c | 3 +-
net/packet/af_packet.c | 4 +-
tools/include/uapi/linux/netdev.h | 19 +
tools/testing/selftests/net/.gitignore | 1 +
tools/testing/selftests/net/Makefile | 5 +
tools/testing/selftests/net/ncdevmem.c | 542 ++++++++++++++++++++++++
47 files changed, 2753 insertions(+), 251 deletions(-)
create mode 100644 Documentation/networking/devmem.rst
create mode 100644 include/net/devmem.h
create mode 100644 include/net/mp_dmabuf_devmem.h
create mode 100644 net/core/devmem.c
create mode 100644 net/core/netdev_rx_queue.c
create mode 100644 tools/testing/selftests/net/ncdevmem.c
--
2.45.2.741.gdbec12cfda-goog
Hi,
This builds on the proposal[1] from Mark and lets me convert the
existing usercopy selftest to KUnit. Besides adding this basic test to
the KUnit collection, it also opens the door for execve testing (which
depends on having a functional current->mm), and should provide the
basic infrastructure for adding Mark's much more complete usercopy tests.
v3:
- use MEMEQ KUnit helper (David)
- exclude pathological address confusion test for systems with separate
address spaces, noticed by David
- add KUnit-conditional exports for alloc_mm() and arch_pick_mmap_layout()
noticed by 0day
v2: https://lore.kernel.org/lkml/20240610213055.it.075-kees@kernel.org/
v1: https://lore.kernel.org/lkml/20240519190422.work.715-kees@kernel.org/
-Kees
[1] https://lore.kernel.org/lkml/20230321122514.1743889-2-mark.rutland@arm.com/
Kees Cook (2):
kunit: test: Add vm_mmap() allocation resource manager
usercopy: Convert test_user_copy to KUnit test
MAINTAINERS | 1 +
include/kunit/test.h | 17 ++
kernel/fork.c | 3 +
lib/Kconfig.debug | 21 +-
lib/Makefile | 2 +-
lib/kunit/Makefile | 1 +
lib/kunit/user_alloc.c | 113 +++++++++
lib/{test_user_copy.c => usercopy_kunit.c} | 282 ++++++++++-----------
mm/util.c | 3 +
9 files changed, 288 insertions(+), 155 deletions(-)
create mode 100644 lib/kunit/user_alloc.c
rename lib/{test_user_copy.c => usercopy_kunit.c} (46%)
--
2.34.1
Add support for (yet again) more RVA23U64 missing extensions. Add
support for Zimop, Zcmop, Zca, Zcf, Zcd and Zcb extensions ISA string
parsing, hwprobe and kvm support. Zce, Zcmt and Zcmp extensions have
been left out since they target microcontrollers/embedded CPUs and are
not needed by RVA23U64.
Since Zc* extensions states that C implies Zca, Zcf (if F and RV32), Zcd
(if D), this series modifies the way ISA string is parsed and now does
it in two phases. First one parses the string and the second one
validates it for the final ISA description.
Link: https://lore.kernel.org/linux-riscv/20240404103254.1752834-1-cleger@rivosin… [1]
Link: https://lore.kernel.org/all/20240409143839.558784-1-cleger@rivosinc.com/ [2]
---
v7:
- Rebased on riscv/for-next to fix conflicts
v6:
- Rebased on riscv/for-next
- Remove ternary operator to use 'if()' instead in extension checks
- v5: https://lore.kernel.org/all/20240517145302.971019-1-cleger@rivosinc.com/
v5:
- Merged in Zimop to avoid any uneeded series dependencies
- Rework dependency resolution loop to loop on source isa first rather
than on all extension.
- Disabled extensions in source isa once set in resolved isa
- Rename riscv_resolve_isa() parameters
- v4: https://lore.kernel.org/all/20240429150553.625165-1-cleger@rivosinc.com/
v4:
- Modify validate() callbacks to return 0, -EPROBEDEFER or another
error.
- v3: https://lore.kernel.org/all/20240423124326.2532796-1-cleger@rivosinc.com/
v3:
- Fix typo "exists" -> "exist"
- Remove C implies Zca, Zcd, Zcf, dt-bindings rules
- Rework ISA string resolver to handle dependencies
- v2: https://lore.kernel.org/all/20240418124300.1387978-1-cleger@rivosinc.com/
v2:
- Add Zc* dependencies validation in dt-bindings
- v1: https://lore.kernel.org/lkml/20240410091106.749233-1-cleger@rivosinc.com/
Clément Léger (16):
dt-bindings: riscv: add Zimop ISA extension description
riscv: add ISA extension parsing for Zimop
riscv: hwprobe: export Zimop ISA extension
RISC-V: KVM: Allow Zimop extension for Guest/VM
KVM: riscv: selftests: Add Zimop extension to get-reg-list test
dt-bindings: riscv: add Zca, Zcf, Zcd and Zcb ISA extension
description
riscv: add ISA extensions validation callback
riscv: add ISA parsing for Zca, Zcf, Zcd and Zcb
riscv: hwprobe: export Zca, Zcf, Zcd and Zcb ISA extensions
RISC-V: KVM: Allow Zca, Zcf, Zcd and Zcb extensions for Guest/VM
KVM: riscv: selftests: Add some Zc* extensions to get-reg-list test
dt-bindings: riscv: add Zcmop ISA extension description
riscv: add ISA extension parsing for Zcmop
riscv: hwprobe: export Zcmop ISA extension
RISC-V: KVM: Allow Zcmop extension for Guest/VM
KVM: riscv: selftests: Add Zcmop extension to get-reg-list test
Documentation/arch/riscv/hwprobe.rst | 28 ++
.../devicetree/bindings/riscv/extensions.yaml | 95 ++++++
arch/riscv/include/asm/cpufeature.h | 1 +
arch/riscv/include/asm/hwcap.h | 6 +
arch/riscv/include/uapi/asm/hwprobe.h | 6 +
arch/riscv/include/uapi/asm/kvm.h | 6 +
arch/riscv/kernel/cpufeature.c | 277 ++++++++++++------
arch/riscv/kernel/sys_hwprobe.c | 6 +
arch/riscv/kvm/vcpu_onereg.c | 12 +
.../selftests/kvm/riscv/get-reg-list.c | 24 ++
10 files changed, 375 insertions(+), 86 deletions(-)
--
2.45.2
The open() function returns -1 on error. openat() and open() initialize
'from' and 'to', and only 'from' validated with 'if' statement. If the
initialization of variable 'to' fails, we should better check the value
of 'to' and close 'from' to avoid possible file leak. Improve the checking
of 'from' additionally.
Fixes: 32ae976ed3b5 ("selftests/capabilities: Add tests for capability evolution")
Signed-off-by: Ma Ke <make24(a)iscas.ac.cn>
---
Changes in v2:
- modified the patch according to suggestions;
- found by customized static analysis tool.
---
tools/testing/selftests/capabilities/test_execve.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/capabilities/test_execve.c b/tools/testing/selftests/capabilities/test_execve.c
index 47bad7ddc5bc..6406ab6aa1f5 100644
--- a/tools/testing/selftests/capabilities/test_execve.c
+++ b/tools/testing/selftests/capabilities/test_execve.c
@@ -145,10 +145,14 @@ static void chdir_to_tmpfs(void)
static void copy_fromat_to(int fromfd, const char *fromname, const char *toname)
{
int from = openat(fromfd, fromname, O_RDONLY);
- if (from == -1)
+ if (from < 0)
ksft_exit_fail_msg("open copy source - %s\n", strerror(errno));
int to = open(toname, O_CREAT | O_WRONLY | O_EXCL, 0700);
+ if (to < 0) {
+ close(from);
+ ksft_exit_fail_msg("open copy destination - %s\n", strerror(errno));
+ }
while (true) {
char buf[4096];
--
2.25.1
In the same way than commit ae7487d112cf ("selftests/hid: ensure we can
compile the tests on kernels pre-6.3") we should expose struct hid_bpf_ops
when it's not available in vmlinux.h.
So unexpose an eventual struct hid_bpf_ops, include vmlinux.h, and
re-export struct hid_bpf_ops.
Fixes: d7696738d66b ("selftests/hid: convert the hid_bpf selftests with struct_ops")
Reported-by: kernel test robot <lkp(a)intel.com>
Closes: https://lore.kernel.org/r/202406270328.bscLN1IF-lkp@intel.com/
Signed-off-by: Benjamin Tissoires <bentiss(a)kernel.org>
---
Same situation than in an early report when HID-BPF was initially
included: the automatically generated vmlinux.h doesn't contain all of
the required structs and the compilation of the bpf program fails.
---
tools/testing/selftests/hid/progs/hid_bpf_helpers.h | 16 ++++++++++++++++
1 file changed, 16 insertions(+)
diff --git a/tools/testing/selftests/hid/progs/hid_bpf_helpers.h b/tools/testing/selftests/hid/progs/hid_bpf_helpers.h
index c72e44321764..5a911f0e8625 100644
--- a/tools/testing/selftests/hid/progs/hid_bpf_helpers.h
+++ b/tools/testing/selftests/hid/progs/hid_bpf_helpers.h
@@ -7,6 +7,7 @@
/* "undefine" structs and enums in vmlinux.h, because we "override" them below */
#define hid_bpf_ctx hid_bpf_ctx___not_used
+#define hid_bpf_ops hid_bpf_ops___not_used
#define hid_report_type hid_report_type___not_used
#define hid_class_request hid_class_request___not_used
#define hid_bpf_attach_flags hid_bpf_attach_flags___not_used
@@ -24,6 +25,7 @@
#include "vmlinux.h"
#undef hid_bpf_ctx
+#undef hid_bpf_ops
#undef hid_report_type
#undef hid_class_request
#undef hid_bpf_attach_flags
@@ -68,6 +70,20 @@ enum hid_class_request {
HID_REQ_SET_PROTOCOL = 0x0B,
};
+struct hid_bpf_ops {
+ int hid_id;
+ u32 flags;
+ struct list_head list;
+ int (*hid_device_event)(struct hid_bpf_ctx *ctx, enum hid_report_type report_type,
+ __u64 source);
+ int (*hid_rdesc_fixup)(struct hid_bpf_ctx *ctx);
+ int (*hid_hw_request)(struct hid_bpf_ctx *ctx, unsigned char reportnum,
+ enum hid_report_type rtype, enum hid_class_request reqtype,
+ __u64 source);
+ int (*hid_hw_output_report)(struct hid_bpf_ctx *ctx, __u64 source);
+ struct hid_device *hdev;
+};
+
#ifndef BPF_F_BEFORE
#define BPF_F_BEFORE (1U << 3)
#endif
---
base-commit: d3e15189bfd4d0a9d3a7ad8bd0e6ebb1c0419f93
change-id: 20240627-fix-cki-f372855cbf6f
Best regards,
--
Benjamin Tissoires <bentiss(a)kernel.org>
This series is a followup of the struct_ops conversion.
Therefore, it is based on top of the for-6.11/bpf branch of the hid.git
tree:
https://git.kernel.org/pub/scm/linux/kernel/git/hid/hid.git/log/?h=for-6.11…
The first patch should go in ASAP, it's a fix that was detected by Dan
and which is actually breaking some use cases.
The rest is adding new capabilities to HID-BPF: being able to intercept
hid_hw_raw_request() and hid_hw_ouptut_report(). Both operations are
write operations to the device.
Having those new hooks allows to implement the "firewall" of HID
devices: this way a bpf program can selectively authorize an hidraw
client to write or not to the device depending on what is requested.
This also allows to completely emulate new behavior: we can now create a
"fake" feature on a HID device, and when we receive a request on this
feature, we can emulate the answer by either statically answering or
even by communicating with the device from bpf, as those new hooks are
sleepable.
Last, there is one change in the kfunc hid_bpf_input_report, in which it
actually waits for the device to be ready. This will not break any
potential users as the function was already declared as sleepable.
Cheers,
Benjamin
Signed-off-by: Benjamin Tissoires <bentiss(a)kernel.org>
---
Changes in v2:
- made use of srcu, for sleepable users
- Link to v1: https://lore.kernel.org/r/20240621-hid_hw_req_bpf-v1-0-d7ab8b885a0b@kernel.…
---
Benjamin Tissoires (13):
HID: bpf: fix dispatch_hid_bpf_device_event uninitialized ret value
HID: add source argument to HID low level functions
HID: bpf: protect HID-BPF prog_list access by a SRCU
HID: bpf: add HID-BPF hooks for hid_hw_raw_requests
HID: bpf: prevent infinite recursions with hid_hw_raw_requests hooks
selftests/hid: add tests for hid_hw_raw_request HID-BPF hooks
HID: bpf: add HID-BPF hooks for hid_hw_output_report
selftests/hid: add tests for hid_hw_output_report HID-BPF hooks
HID: bpf: make hid_bpf_input_report() sleep until the device is ready
selftests/hid: add wq test for hid_bpf_input_report()
HID: bpf: allow hid_device_event hooks to inject input reports on self
selftests/hid: add another test for injecting an event from an event hook
selftests/hid: add an infinite loop test for hid_bpf_try_input_report
Documentation/hid/hid-bpf.rst | 2 +-
drivers/hid/bpf/hid_bpf_dispatch.c | 165 ++++++++++-
drivers/hid/bpf/hid_bpf_dispatch.h | 1 +
drivers/hid/bpf/hid_bpf_struct_ops.c | 6 +-
drivers/hid/hid-core.c | 118 +++++---
drivers/hid/hidraw.c | 10 +-
include/linux/hid.h | 7 +
include/linux/hid_bpf.h | 80 ++++-
tools/testing/selftests/hid/hid_bpf.c | 326 +++++++++++++++++++++
tools/testing/selftests/hid/progs/hid.c | 292 ++++++++++++++++++
.../testing/selftests/hid/progs/hid_bpf_helpers.h | 13 +
11 files changed, 955 insertions(+), 65 deletions(-)
---
base-commit: 33c0fb85b571b0f1bbdbf466e770eebeb29e6f41
change-id: 20240614-hid_hw_req_bpf-df0b95aeb425
Best regards,
--
Benjamin Tissoires <bentiss(a)kernel.org>
** Background **
Currently, OVS supports several packet sampling mechanisms (sFlow,
per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
userspace action that needs to be handled by ovs-vswitchd's handler
threads only to be forwarded to some third party application that
will somehow process the sample and provide observability on the
datapath.
A particularly interesting use-case is controller-driven
per-flow IPFIX sampling where the OpenFlow controller can add metadata
to samples (via two 32bit integers) and this metadata is then available
to the sample-collecting system for correlation.
** Problem **
The fact that sampled traffic share netlink sockets and handler thread
time with upcalls, apart from being a performance bottleneck in the
sample extraction itself, can severely compromise the datapath,
yielding this solution unfit for highly loaded production systems.
Users are left with little options other than guessing what sampling
rate will be OK for their traffic pattern and system load and dealing
with the lost accuracy.
Looking at available infrastructure, an obvious candidated would be
to use psample. However, it's current state does not help with the
use-case at stake because sampled packets do not contain user-defined
metadata.
** Proposal **
This series is an attempt to fix this situation by extending the
existing psample infrastructure to carry a variable length
user-defined cookie.
The main existing user of psample is tc's act_sample. It is also
extended to forward the action's cookie to psample.
Finally, a new OVS action (OVS_SAMPLE_ATTR_EMIT_SAMPLE) is created.
It accepts a group and an optional cookie and uses psample to
multicast the packet and the metadata.
--
v4 -> v5:
- Rebased.
- Removed lefover enum value and wrapped some long lines in selftests.
v3 -> v4:
- Rebased.
- Addressed Jakub's comment on private and unused nla attributes.
v2 -> v3:
- Addressed comments from Simon, Aaron and Ilya.
- Dropped probability propagation in nested sample actions.
- Dropped patch v2's 7/9 in favor of a userspace implementation and
consume skb if emit_sample is the last action, same as we do with
userspace.
- Split ovs-dpctl.py features in independent patches.
v1 -> v2:
- Create a new action ("emit_sample") rather than reuse existing
"sample" one.
- Add probability semantics to psample's sampling rate.
- Store sampling probability in skb's cb area and use it in emit_sample.
- Test combining "emit_sample" with "trunc"
- Drop group_id filtering and tracepoint in psample.
rfc_v2 -> v1:
- Accommodate Ilya's comments.
- Split OVS's attribute in two attributes and simplify internal
handling of psample arguments.
- Extend psample and tc with a user-defined cookie.
- Add a tracepoint to psample to facilitate troubleshooting.
rfc_v1 -> rfc_v2:
- Use psample instead of a new OVS-only multicast group.
- Extend psample and tc with a user-defined cookie.
Adrian Moreno (10):
net: psample: add user cookie
net: sched: act_sample: add action cookie to sample
net: psample: skip packet copy if no listeners
net: psample: allow using rate as probability
net: openvswitch: add emit_sample action
net: openvswitch: store sampling probability in cb.
selftests: openvswitch: add emit_sample action
selftests: openvswitch: add userspace parsing
selftests: openvswitch: parse trunc action
selftests: openvswitch: add emit_sample test
Documentation/netlink/specs/ovs_flow.yaml | 17 ++
include/net/psample.h | 5 +-
include/uapi/linux/openvswitch.h | 31 +-
include/uapi/linux/psample.h | 11 +-
net/openvswitch/Kconfig | 1 +
net/openvswitch/actions.c | 63 +++-
net/openvswitch/datapath.h | 3 +
net/openvswitch/flow_netlink.c | 33 ++-
net/openvswitch/vport.c | 1 +
net/psample/psample.c | 16 +-
net/sched/act_sample.c | 12 +
.../selftests/net/openvswitch/openvswitch.sh | 114 +++++++-
.../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
13 files changed, 563 insertions(+), 16 deletions(-)
--
2.45.1
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v2:
- add patch 2, a new fix for sk_msg_memcopy_from_iter.
- update patch 3, only test "sk->sk_prot->close" as Eric suggested.
- update patch 4, use "goto err" instead of "return" as Eduard
suggested.
- add "fixes" tag for patch 1-3.
- change subject prefixes as "bpf-next" to trigger BPF CI.
- cc Loongarch maintainers too.
BPF selftests seem to have not been fully tested on Loongarch. When I
ran these tests on Loongarch recently, some errors occur. This patch set
contains some null-check related fixes for these errors.
Geliang Tang (4):
skmsg: null check for sg_page in sk_msg_recvmsg
skmsg: null check for sg_page in sk_msg_memcopy_from_iter
inet: null check for close in inet_release
selftests/bpf: Null checks for link in bpf_tcp_ca
net/core/skmsg.c | 4 ++++
net/ipv4/af_inet.c | 3 ++-
.../selftests/bpf/prog_tests/bpf_tcp_ca.c | 16 ++++++++++++----
3 files changed, 18 insertions(+), 5 deletions(-)
--
2.43.0
serial_test_fexit_stress() has a non-robust handling of file descriptor
closure. If an error occurs, the function may exit without closing open
file descriptors, potentially causing resource leaks.
Fix the issue by closing file descriptors in reverse order and starting
from the last opened. Ensure proper closure even if an error occurs early.
Fixes: 8fb9fb2f1728 ("selftests/bpf: Query BPF_MAX_TRAMP_LINKS using BTF")
Signed-off-by: Ma Ke <make24(a)iscas.ac.cn>
---
tools/testing/selftests/bpf/prog_tests/fexit_stress.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/bpf/prog_tests/fexit_stress.c b/tools/testing/selftests/bpf/prog_tests/fexit_stress.c
index 596536def43d..b1980bd61583 100644
--- a/tools/testing/selftests/bpf/prog_tests/fexit_stress.c
+++ b/tools/testing/selftests/bpf/prog_tests/fexit_stress.c
@@ -49,11 +49,14 @@ void serial_test_fexit_stress(void)
ASSERT_OK(err, "bpf_prog_test_run_opts");
out:
- for (i = 0; i < bpf_max_tramp_links; i++) {
+ if (i >= bpf_max_tramp_links)
+ i = bpf_max_tramp_links - 1;
+ while (i >= 0) {
if (link_fd[i])
close(link_fd[i]);
if (fexit_fd[i])
close(fexit_fd[i]);
+ i--;
}
free(fd);
}
--
2.25.1
The selftest noncont_cat_run_test fails on AMD with the warnings. Reason
is, AMD supports non contiguous CBM masks but does not report it via CPUID.
Update noncont_cat_run_test to check for the vendor when verifying CPUID.
Fixes: ae638551ab64 ("selftests/resctrl: Add non-contiguous CBMs CAT test")
Signed-off-by: Babu Moger <babu.moger(a)amd.com>
---
This was part of the series
https://lore.kernel.org/lkml/cover.1708637563.git.babu.moger@amd.com/
Sending this as a separate fix per review comments.
---
tools/testing/selftests/resctrl/cat_test.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/resctrl/cat_test.c b/tools/testing/selftests/resctrl/cat_test.c
index d4dffc934bc3..b2988888786e 100644
--- a/tools/testing/selftests/resctrl/cat_test.c
+++ b/tools/testing/selftests/resctrl/cat_test.c
@@ -308,7 +308,7 @@ static int noncont_cat_run_test(const struct resctrl_test *test,
else
return -EINVAL;
- if (sparse_masks != ((ecx >> 3) & 1)) {
+ if ((get_vendor() == ARCH_INTEL) && sparse_masks != ((ecx >> 3) & 1)) {
ksft_print_msg("CPUID output doesn't match 'sparse_masks' file content!\n");
return 1;
}
--
2.34.1
The open() function returns -1 on error. openat() and open() initialize
'from' and 'to', and only 'from' validated with 'if' statement. If the
initialization of variable 'to' fails, we should better check the value
of 'to' and close 'from' to avoid possible file leak.
Fixes: 32ae976ed3b5 ("selftests/capabilities: Add tests for capability evolution")
Signed-off-by: Ma Ke <make24(a)iscas.ac.cn>
---
Found this error through static analysis.
---
tools/testing/selftests/capabilities/test_execve.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/tools/testing/selftests/capabilities/test_execve.c b/tools/testing/selftests/capabilities/test_execve.c
index 47bad7ddc5bc..de187eff177d 100644
--- a/tools/testing/selftests/capabilities/test_execve.c
+++ b/tools/testing/selftests/capabilities/test_execve.c
@@ -149,6 +149,10 @@ static void copy_fromat_to(int fromfd, const char *fromname, const char *toname)
ksft_exit_fail_msg("open copy source - %s\n", strerror(errno));
int to = open(toname, O_CREAT | O_WRONLY | O_EXCL, 0700);
+ if (to == -1) {
+ close(from);
+ ksft_exit_fail_msg("open copy destination - %s\n", strerror(errno));
+ }
while (true) {
char buf[4096];
--
2.25.1
This patch series is motivated by the following observation:
Raise a signal, jump to signal handler. The ucontext_t structure dumped
by kernel to userspace has a uc_sigmask field having the mask of blocked
signals. If you run a fresh minimalistic program doing this, this field
is empty, even if you block some signals while registering the handler
with sigaction().
Here is what the man-pages have to say:
sigaction(2): "sa_mask specifies a mask of signals which should be blocked
(i.e., added to the signal mask of the thread in which the signal handler
is invoked) during execution of the signal handler. In addition, the
signal which triggered the handler will be blocked, unless the SA_NODEFER
flag is used."
signal(7): Under "Execution of signal handlers", (1.3) implies:
"The thread's current signal mask is accessible via the ucontext_t
object that is pointed to by the third argument of the signal handler."
But, (1.4) states:
"Any signals specified in act->sa_mask when registering the handler with
sigprocmask(2) are added to the thread's signal mask. The signal being
delivered is also added to the signal mask, unless SA_NODEFER was
specified when registering the handler. These signals are thus blocked
while the handler executes."
There clearly is no distinction being made in the man pages between
"Thread's signal mask" and ucontext_t; this logically should imply
that a signal blocked by populating struct sigaction should be visible
in ucontext_t.
Here is what the kernel code does (for Aarch64):
do_signal() -> handle_signal() -> sigmask_to_save(), which returns
¤t->blocked, is passed to setup_rt_frame() -> setup_sigframe() ->
__copy_to_user(). Hence, ¤t->blocked is copied to ucontext_t
exposed to userspace. Returning back to handle_signal(),
signal_setup_done() -> signal_delivered() -> sigorsets() and
set_current_blocked() are responsible for using information from
struct ksignal ksig, which was populated through the sigaction()
system call in kernel/signal.c:
copy_from_user(&new_sa.sa, act, sizeof(new_sa.sa)),
to update ¤t->blocked; hence, the set of blocked signals for the
current thread is updated AFTER the kernel dumps ucontext_t to
userspace.
Assuming that the above is indeed the intended behaviour, because it
semantically makes sense, since the signals blocked using sigaction()
remain blocked only till the execution of the handler, and not in the
context present before jumping to the handler (but nothing can be
confirmed from the man-pages), the series introduces a test for
mangling with uc_sigmask. I will send a separate series to fix the
man-pages.
The proposed selftest has been tested out on Aarch32, Aarch64 and x86_64.
v2->v3:
- ucontext describes current state -> ucontext describes interrupted context
- Add a comment for blockage of USR2 even after return from handler
- Describe blockage of signals in a better way
v1->v2:
- Replace all occurrences of SIGPIPE with SIGSEGV
- Fixed a mismatch between code comment and ksft log
- Add a testcase: Raise the same signal again; it must not be queued
- Remove unneeded <assert.h>, <unistd.h>
- Give a detailed test description in the comments; also describe the
exact meaning of delivered and blocked
- Handle errors for all libc functions/syscalls
- Mention tests in Makefile and .gitignore in alphabetical order
v1:
- https://lore.kernel.org/all/20240607122319.768640-1-dev.jain@arm.com/
Dev Jain (2):
selftests: Rename sigaltstack to generic signal
selftests: Add a test mangling with uc_sigmask
tools/testing/selftests/Makefile | 2 +-
.../{sigaltstack => signal}/.gitignore | 3 +-
.../{sigaltstack => signal}/Makefile | 3 +-
.../current_stack_pointer.h | 0
.../selftests/signal/mangle_uc_sigmask.c | 194 ++++++++++++++++++
.../sas.c => signal/sigaltstack.c} | 0
6 files changed, 199 insertions(+), 3 deletions(-)
rename tools/testing/selftests/{sigaltstack => signal}/.gitignore (57%)
rename tools/testing/selftests/{sigaltstack => signal}/Makefile (53%)
rename tools/testing/selftests/{sigaltstack => signal}/current_stack_pointer.h (100%)
create mode 100644 tools/testing/selftests/signal/mangle_uc_sigmask.c
rename tools/testing/selftests/{sigaltstack/sas.c => signal/sigaltstack.c} (100%)
--
2.34.1
Currently if we request a feature that is not set in the Kernel
config we fail silently and return all the available features. However,
the man page indicates we should return an EINVAL.
We need to fix this issue since we can end up with a Kernel warning
should a program request the feature UFFD_FEATURE_WP_UNPOPULATED on
a kernel with the config not set with this feature.
[ 200.812896] WARNING: CPU: 91 PID: 13634 at mm/memory.c:1660 zap_pte_range+0x43d/0x660
[ 200.820738] Modules linked in:
[ 200.869387] CPU: 91 PID: 13634 Comm: userfaultfd Kdump: loaded Not tainted 6.9.0-rc5+ #8
[ 200.877477] Hardware name: Dell Inc. PowerEdge R6525/0N7YGH, BIOS 2.7.3 03/30/2022
[ 200.885052] RIP: 0010:zap_pte_range+0x43d/0x660
Fixes: e06f1e1dd499 ("userfaultfd: wp: enabled write protection in userfaultfd API")
Signed-off-by: Audra Mitchell <audra(a)redhat.com>
---
fs/userfaultfd.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index eee7320ab0b0..17e409ceaa33 100644
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -2057,7 +2057,7 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
goto out;
features = uffdio_api.features;
ret = -EINVAL;
- if (uffdio_api.api != UFFD_API || (features & ~UFFD_API_FEATURES))
+ if (uffdio_api.api != UFFD_API)
goto err_out;
ret = -EPERM;
if ((features & UFFD_FEATURE_EVENT_FORK) && !capable(CAP_SYS_PTRACE))
@@ -2081,6 +2081,11 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
uffdio_api.features &= ~UFFD_FEATURE_WP_UNPOPULATED;
uffdio_api.features &= ~UFFD_FEATURE_WP_ASYNC;
#endif
+
+ ret = -EINVAL;
+ if (features & ~uffdio_api.features)
+ goto err_out;
+
uffdio_api.ioctls = UFFD_API_IOCTLS;
ret = -EFAULT;
if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api)))
--
2.44.0
v13: https://patchwork.kernel.org/project/netdevbpf/list/?series=861406&archive=…
====
Major changes:
--------------
This iteration addresses Pavel's review comments, applies his
reviewed-by's, and seeks to fix the patchwork build error (sorry!).
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v13/
v12: https://patchwork.kernel.org/project/netdevbpf/list/?series=859747&state=*
====
Major changes:
--------------
This iteration only addresses one minor comment from Pavel with regards
to the trace printing of netmem, and the patchwork build error
introduced in v11 because I missed doing an allmodconfig build, sorry.
Other than that v11, AFAICT, received no feedback. There is one
discussion about how the specifics of plugging io uring memory through
the page pool, but not relevant to content in this particular patchset,
AFAICT.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v12/
v11: https://patchwork.kernel.org/project/netdevbpf/list/?series=857457&state=*
====
Major Changes:
--------------
v11 addresses feedback received in v10. The major change is the removal
of the memory provider ops as requested by Christoph. We still
accomplish the same thing, but utilizing direct function calls with if
statements rather than generic ops.
Additionally address sparse warnings, bugs and review comments from
folks that reviewed.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v11/
Detailed changelog:
-------------------
- Fixes in netdev_rx_queue_restart() from Pavel & David.
- Remove commit e650e8c3a36f5 ("net: page_pool: create hooks for
custom page providers") from the series to address Christoph's
feedback and rebased other patches on the series on this change.
- Fixed build errors with CONFIG_DMA_SHARED_BUFFER &&
!CONFIG_GENERIC_ALLOCATOR build.
- Fixed sparse warnings pointed out by Paolo.
- Drop unnecessary gro_pull_from_frag0 checks.
- Added Bagas reviewed-by to docs.
Cc: Bagas Sanjaya <bagasdotme(a)gmail.com>
Cc: Steven Rostedt <rostedt(a)goodmis.org>
Cc: Christoph Hellwig <hch(a)infradead.org>
Cc: Nikolay Aleksandrov <razor(a)blackwall.org>
v10: https://patchwork.kernel.org/project/netdevbpf/list/?series=852422&state=*
====
Major Changes:
--------------
v9 was sent right before the merge window closed (sorry!). v10 is almost
a re-send of the series now that the merge window re-opened. Only
rebased to latest net-next and addressed some minor iterative comments
received on v9.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v10/
Detailed changelog:
-------------------
- Fixed tokens leaking in DONTNEED setsockopt (Nikolay).
- Moved net_iov_dma_addr() to devmem.c and made it a devmem specific
helpers (David).
- Rename hook alloc_pages to alloc_netmems as alloc_pages is now
preprocessor macro defined and causes a build error.
v9:
===
Major Changes:
--------------
GVE queue API has been merged. Submitting this version as non-RFC after
rebasing on top of the merged API, and dropped the out of tree queue API
I was carrying on github. Addressed the little feedback v8 has received.
Detailed changelog:
------------------
- Added new patch from David Wei to this series for
netdev_rx_queue_restart()
- Fixed sparse error.
- Removed CONFIG_ checks in netmem_is_net_iov()
- Flipped skb->readable to skb->unreadable
- Minor fixes to selftests & docs.
RFC v8:
=======
Major Changes:
--------------
- Fixed build error generated by patch-by-patch build.
- Applied docs suggestions from Randy.
RFC v7:
=======
Major Changes:
--------------
This revision largely rebases on top of net-next and addresses the feedback
RFCv6 received from folks, namely Jakub, Yunsheng, Arnd, David, & Pavel.
The series remains in RFC because the queue-API ndos defined in this
series are not yet implemented. I have a GVE implementation I carry out
of tree for my testing. A upstreamable GVE implementation is in the
works. Aside from that, in my estimation all the patches are ready for
review/merge. Please do take a look.
As usual the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v7/
Detailed changelog:
- Use admin-perm in netlink API.
- Addressed feedback from Jakub with regards to netlink API
implementation.
- Renamed devmem.c functions to something more appropriate for that
file.
- Improve the performance seen through the page_pool benchmark.
- Fix the value definition of all the SO_DEVMEM_* uapi.
- Various fixes to documentation.
Perf - page-pool benchmark:
---------------------------
Improved performance of bench_page_pool_simple.ko tests compared to v6:
https://pastebin.com/raw/v5dYRg8L
net-next base: 8 cycle fast path.
RFC v6: 10 cycle fast path.
RFC v7: 9 cycle fast path.
RFC v7 with CONFIG_DMA_SHARED_BUFFER disabled: 8 cycle fast path,
same as baseline.
Perf - Devmem TCP benchmark:
---------------------
Perf is about the same regardless of the changes in v7, namely the
removal of the static_branch_unlikely to improve the page_pool benchmark
performance:
189/200gbps bi-directional throughput with RX devmem TCP and regular TCP
TX i.e. ~95% line rate.
RFC v6:
=======
Major Changes:
--------------
This revision largely rebases on top of net-next and addresses the little
feedback RFCv5 received.
The series remains in RFC because the queue-API ndos defined in this
series are not yet implemented. I have a GVE implementation I carry out
of tree for my testing. A upstreamable GVE implementation is in the
works. Aside from that, in my estimation all the patches are ready for
review/merge. Please do take a look.
As usual the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v6/
This version also comes with some performance data recorded in the cover
letter (see below changelog).
Detailed changelog:
- Rebased on top of the merged netmem_ref changes.
- Converted skb->dmabuf to skb->readable (Pavel). Pavel's original
suggestion was to remove the skb->dmabuf flag entirely, but when I
looked into it closely, I found the issue that if we remove the flag
we have to dereference the shinfo(skb) pointer to obtain the first
frag to tell whether an skb is readable or not. This can cause a
performance regression if it dirties the cache line when the
shinfo(skb) was not really needed. Instead, I converted the skb->dmabuf
flag into a generic skb->readable flag which can be re-used by io_uring
0-copy RX.
- Squashed a few locking optimizations from Eric Dumazet in the RX path
and the DEVMEM_DONTNEED setsockopt.
- Expanded the tests a bit. Added validation for invalid scenarios and
added some more coverage.
Perf - page-pool benchmark:
---------------------------
bench_page_pool_simple.ko tests with and without these changes:
https://pastebin.com/raw/ncHDwAbn
AFAIK the number that really matters in the perf tests is the
'tasklet_page_pool01_fast_path Per elem'. This one measures at about 8
cycles without the changes but there is some 1 cycle noise in some
results.
With the patches this regresses to 9 cycles with the changes but there
is 1 cycle noise occasionally running this test repeatedly.
Lastly I tried disable the static_branch_unlikely() in
netmem_is_net_iov() check. To my surprise disabling the
static_branch_unlikely() check reduces the fast path back to 8 cycles,
but the 1 cycle noise remains.
Perf - Devmem TCP benchmark:
---------------------
189/200gbps bi-directional throughput with RX devmem TCP and regular TCP
TX i.e. ~95% line rate.
Major changes in RFC v5:
========================
1. Rebased on top of 'Abstract page from net stack' series and used the
new netmem type to refer to LSB set pointers instead of re-using
struct page.
2. Downgraded this series back to RFC and called it RFC v5. This is
because this series is now dependent on 'Abstract page from net
stack'[1] and the queue API. Both are removed from the series to
reduce the patch # and those bits are fairly independent or
pre-requisite work.
3. Reworked the page_pool devmem support to use netmem and for some
more unified handling.
4. Reworked the reference counting of net_iov (renamed from
page_pool_iov) to use pp_ref_count for refcounting.
The full changes including the dependent series and GVE page pool
support is here:
https://github.com/mina/linux/commits/tcpdevmem-rfcv5/
[1] https://patchwork.kernel.org/project/netdevbpf/list/?series=810774
Major changes in v1:
====================
1. Implemented MVP queue API ndos to remove the userspace-visible
driver reset.
2. Fixed issues in the napi_pp_put_page() devmem frag unref path.
3. Removed RFC tag.
Many smaller addressed comments across all the patches (patches have
individual change log).
Full tree including the rest of the GVE driver changes:
https://github.com/mina/linux/commits/tcpdevmem-v1
Changes in RFC v3:
==================
1. Pulled in the memory-provider dependency from Jakub's RFC[1] to make the
series reviewable and mergeable.
2. Implemented multi-rx-queue binding which was a todo in v2.
3. Fix to cmsg handling.
The sticking point in RFC v2[2] was the device reset required to refill
the device rx-queues after the dmabuf bind/unbind. The solution
suggested as I understand is a subset of the per-queue management ops
Jakub suggested or similar:
https://lore.kernel.org/netdev/20230815171638.4c057dcd@kernel.org/
This is not addressed in this revision, because:
1. This point was discussed at netconf & netdev and there is openness to
using the current approach of requiring a device reset.
2. Implementing individual queue resetting seems to be difficult for my
test bed with GVE. My prototype to test this ran into issues with the
rx-queues not coming back up properly if reset individually. At the
moment I'm unsure if it's a mistake in the POC or a genuine issue in
the virtualization stack behind GVE, which currently doesn't test
individual rx-queue restart.
3. Our usecases are not bothered by requiring a device reset to refill
the buffer queues, and we'd like to support NICs that run into this
limitation with resetting individual queues.
My thought is that drivers that have trouble with per-queue configs can
use the support in this series, while drivers that support new netdev
ops to reset individual queues can automatically reset the queue as
part of the dma-buf bind/unbind.
The same approach with device resets is presented again for consideration
with other sticking points addressed.
This proposal includes the rx devmem path only proposed for merge. For a
snapshot of my entire tree which includes the GVE POC page pool support &
device memory support:
https://github.com/torvalds/linux/compare/master...mina:linux:tcpdevmem-v3
[1] https://lore.kernel.org/netdev/f8270765-a27b-6ccf-33ea-cda097168d79@redhat.…
[2] https://lore.kernel.org/netdev/CAHS8izOVJGJH5WF68OsRWFKJid1_huzzUK+hpKbLcL4…
Changes in RFC v2:
==================
The sticking point in RFC v1[1] was the dma-buf pages approach we used to
deliver the device memory to the TCP stack. RFC v2 is a proof-of-concept
that attempts to resolve this by implementing scatterlist support in the
networking stack, such that we can import the dma-buf scatterlist
directly. This is the approach proposed at a high level here[2].
Detailed changes:
1. Replaced dma-buf pages approach with importing scatterlist into the
page pool.
2. Replace the dma-buf pages centric API with a netlink API.
3. Removed the TX path implementation - there is no issue with
implementing the TX path with scatterlist approach, but leaving
out the TX path makes it easier to review.
4. Functionality is tested with this proposal, but I have not conducted
perf testing yet. I'm not sure there are regressions, but I removed
perf claims from the cover letter until they can be re-confirmed.
5. Added Signed-off-by: contributors to the implementation.
6. Fixed some bugs with the RX path since RFC v1.
Any feedback welcome, but specifically the biggest pending questions
needing feedback IMO are:
1. Feedback on the scatterlist-based approach in general.
2. Netlink API (Patch 1 & 2).
3. Approach to handle all the drivers that expect to receive pages from
the page pool (Patch 6).
[1] https://lore.kernel.org/netdev/dfe4bae7-13a0-3c5d-d671-f61b375cb0b4@gmail.c…
[2] https://lore.kernel.org/netdev/CAHS8izPm6XRS54LdCDZVd0C75tA1zHSu6jLVO8nzTLX…
==================
* TL;DR:
Device memory TCP (devmem TCP) is a proposal for transferring data to and/or
from device memory efficiently, without bouncing the data to a host memory
buffer.
* Problem:
A large amount of data transfers have device memory as the source and/or
destination. Accelerators drastically increased the volume of such transfers.
Some examples include:
- ML accelerators transferring large amounts of training data from storage into
GPU/TPU memory. In some cases ML training setup time can be as long as 50% of
TPU compute time, improving data transfer throughput & efficiency can help
improving GPU/TPU utilization.
- Distributed training, where ML accelerators, such as GPUs on different hosts,
exchange data among them.
- Distributed raw block storage applications transfer large amounts of data with
remote SSDs, much of this data does not require host processing.
Today, the majority of the Device-to-Device data transfers the network are
implemented as the following low level operations: Device-to-Host copy,
Host-to-Host network transfer, and Host-to-Device copy.
The implementation is suboptimal, especially for bulk data transfers, and can
put significant strains on system resources, such as host memory bandwidth,
PCIe bandwidth, etc. One important reason behind the current state is the
kernel’s lack of semantics to express device to network transfers.
* Proposal:
In this patch series we attempt to optimize this use case by implementing
socket APIs that enable the user to:
1. send device memory across the network directly, and
2. receive incoming network packets directly into device memory.
Packet _payloads_ go directly from the NIC to device memory for receive and from
device memory to NIC for transmit.
Packet _headers_ go to/from host memory and are processed by the TCP/IP stack
normally. The NIC _must_ support header split to achieve this.
Advantages:
- Alleviate host memory bandwidth pressure, compared to existing
network-transfer + device-copy semantics.
- Alleviate PCIe BW pressure, by limiting data transfer to the lowest level
of the PCIe tree, compared to traditional path which sends data through the
root complex.
* Patch overview:
** Part 1: netlink API
Gives user ability to bind dma-buf to an RX queue.
** Part 2: scatterlist support
Currently the standard for device memory sharing is DMABUF, which doesn't
generate struct pages. On the other hand, networking stack (skbs, drivers, and
page pool) operate on pages. We have 2 options:
1. Generate struct pages for dmabuf device memory, or,
2. Modify the networking stack to process scatterlist.
Approach #1 was attempted in RFC v1. RFC v2 implements approach #2.
** part 3: page pool support
We piggy back on page pool memory providers proposal:
https://github.com/kuba-moo/linux/tree/pp-providers
It allows the page pool to define a memory provider that provides the
page allocation and freeing. It helps abstract most of the device memory
TCP changes from the driver.
** part 4: support for unreadable skb frags
Page pool iovs are not accessible by the host; we implement changes
throughput the networking stack to correctly handle skbs with unreadable
frags.
** Part 5: recvmsg() APIs
We define user APIs for the user to send and receive device memory.
Not included with this series is the GVE devmem TCP support, just to
simplify the review. Code available here if desired:
https://github.com/mina/linux/tree/tcpdevmem
This series is built on top of net-next with Jakub's pp-providers changes
cherry-picked.
* NIC dependencies:
1. (strict) Devmem TCP require the NIC to support header split, i.e. the
capability to split incoming packets into a header + payload and to put
each into a separate buffer. Devmem TCP works by using device memory
for the packet payload, and host memory for the packet headers.
2. (optional) Devmem TCP works better with flow steering support & RSS support,
i.e. the NIC's ability to steer flows into certain rx queues. This allows the
sysadmin to enable devmem TCP on a subset of the rx queues, and steer
devmem TCP traffic onto these queues and non devmem TCP elsewhere.
The NIC I have access to with these properties is the GVE with DQO support
running in Google Cloud, but any NIC that supports these features would suffice.
I may be able to help reviewers bring up devmem TCP on their NICs.
* Testing:
The series includes a udmabuf kselftest that show a simple use case of
devmem TCP and validates the entire data path end to end without
a dependency on a specific dmabuf provider.
** Test Setup
Kernel: net-next with this series and memory provider API cherry-picked
locally.
Hardware: Google Cloud A3 VMs.
NIC: GVE with header split & RSS & flow steering support.
Cc: Pavel Begunkov <asml.silence(a)gmail.com>
Cc: David Wei <dw(a)davidwei.uk>
Cc: Jason Gunthorpe <jgg(a)ziepe.ca>
Cc: Yunsheng Lin <linyunsheng(a)huawei.com>
Cc: Shailend Chand <shailend(a)google.com>
Cc: Harshitha Ramamurthy <hramamurthy(a)google.com>
Cc: Shakeel Butt <shakeel.butt(a)linux.dev>
Cc: Jeroen de Borst <jeroendb(a)google.com>
Cc: Praveen Kaligineedi <pkaligineedi(a)google.com>
Mina Almasry (13):
netdev: add netdev_rx_queue_restart()
net: netdev netlink api to bind dma-buf to a net device
netdev: support binding dma-buf to netdevice
netdev: netdevice devmem allocator
page_pool: convert to use netmem
page_pool: devmem support
memory-provider: dmabuf devmem memory provider
net: support non paged skb frags
net: add support for skbs with unreadable frags
tcp: RX path for devmem TCP
net: add SO_DEVMEM_DONTNEED setsockopt to release RX frags
net: add devmem TCP documentation
selftests: add ncdevmem, netcat for devmem TCP
Documentation/netlink/specs/netdev.yaml | 57 +++
Documentation/networking/devmem.rst | 258 +++++++++++
Documentation/networking/index.rst | 1 +
arch/alpha/include/uapi/asm/socket.h | 6 +
arch/mips/include/uapi/asm/socket.h | 6 +
arch/parisc/include/uapi/asm/socket.h | 6 +
arch/sparc/include/uapi/asm/socket.h | 6 +
include/linux/skbuff.h | 61 ++-
include/linux/skbuff_ref.h | 11 +-
include/linux/socket.h | 1 +
include/net/devmem.h | 124 ++++++
include/net/mp_dmabuf_devmem.h | 44 ++
include/net/netdev_rx_queue.h | 5 +
include/net/netmem.h | 208 ++++++++-
include/net/page_pool/helpers.h | 124 ++++--
include/net/page_pool/types.h | 22 +-
include/net/sock.h | 2 +
include/net/tcp.h | 5 +-
include/trace/events/page_pool.h | 30 +-
include/uapi/asm-generic/socket.h | 6 +
include/uapi/linux/netdev.h | 19 +
include/uapi/linux/uio.h | 17 +
net/bpf/test_run.c | 5 +-
net/core/Makefile | 3 +-
net/core/datagram.c | 6 +
net/core/dev.c | 6 +-
net/core/devmem.c | 376 ++++++++++++++++
net/core/gro.c | 3 +-
net/core/netdev-genl-gen.c | 23 +
net/core/netdev-genl-gen.h | 6 +
net/core/netdev-genl.c | 103 +++++
net/core/netdev_rx_queue.c | 74 ++++
net/core/page_pool.c | 362 +++++++++-------
net/core/skbuff.c | 83 +++-
net/core/sock.c | 61 +++
net/ipv4/esp4.c | 3 +-
net/ipv4/tcp.c | 261 +++++++++++-
net/ipv4/tcp_input.c | 13 +-
net/ipv4/tcp_ipv4.c | 16 +
net/ipv4/tcp_minisocks.c | 2 +
net/ipv4/tcp_output.c | 5 +-
net/ipv6/esp6.c | 3 +-
net/packet/af_packet.c | 4 +-
tools/include/uapi/linux/netdev.h | 19 +
tools/testing/selftests/net/.gitignore | 1 +
tools/testing/selftests/net/Makefile | 5 +
tools/testing/selftests/net/ncdevmem.c | 542 ++++++++++++++++++++++++
47 files changed, 2753 insertions(+), 251 deletions(-)
create mode 100644 Documentation/networking/devmem.rst
create mode 100644 include/net/devmem.h
create mode 100644 include/net/mp_dmabuf_devmem.h
create mode 100644 net/core/devmem.c
create mode 100644 net/core/netdev_rx_queue.c
create mode 100644 tools/testing/selftests/net/ncdevmem.c
--
2.45.2.741.gdbec12cfda-goog
Currently if we request a feature that is not set in the Kernel
config we fail silently and return all the available features. However,
the man page indicates we should return an EINVAL.
We need to fix this issue since we can end up with a Kernel warning
should a program request the feature UFFD_FEATURE_WP_UNPOPULATED on
a kernel with the config not set with this feature.
[ 200.812896] WARNING: CPU: 91 PID: 13634 at mm/memory.c:1660 zap_pte_range+0x43d/0x660
[ 200.820738] Modules linked in:
[ 200.869387] CPU: 91 PID: 13634 Comm: userfaultfd Kdump: loaded Not tainted 6.9.0-rc5+ #8
[ 200.877477] Hardware name: Dell Inc. PowerEdge R6525/0N7YGH, BIOS 2.7.3 03/30/2022
[ 200.885052] RIP: 0010:zap_pte_range+0x43d/0x660
Fixes: e06f1e1dd499 ("userfaultfd: wp: enabled write protection in userfaultfd API")
Signed-off-by: Audra Mitchell <audra(a)redhat.com>
---
fs/userfaultfd.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index eee7320ab0b0..17e409ceaa33 100644
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -2057,7 +2057,7 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
goto out;
features = uffdio_api.features;
ret = -EINVAL;
- if (uffdio_api.api != UFFD_API || (features & ~UFFD_API_FEATURES))
+ if (uffdio_api.api != UFFD_API)
goto err_out;
ret = -EPERM;
if ((features & UFFD_FEATURE_EVENT_FORK) && !capable(CAP_SYS_PTRACE))
@@ -2081,6 +2081,11 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
uffdio_api.features &= ~UFFD_FEATURE_WP_UNPOPULATED;
uffdio_api.features &= ~UFFD_FEATURE_WP_ASYNC;
#endif
+
+ ret = -EINVAL;
+ if (features & ~uffdio_api.features)
+ goto err_out;
+
uffdio_api.ioctls = UFFD_API_IOCTLS;
ret = -EFAULT;
if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api)))
--
2.44.0
Correctable memory errors are very common on servers with large
amount of memory, and are corrected by ECC, but with two
pain points to users:
1. Correction usually happens on the fly and adds latency overhead
2. Not-fully-proved theory states excessive correctable memory
errors can develop into uncorrectable memory error.
Soft offline is kernel's additional solution for memory pages
having (excessive) corrected memory errors. Impacted page is migrated
to healthy page if it is in use, then the original page is discarded
for any future use.
The actual policy on whether (and when) to soft offline should be
maintained by userspace, especially in case of an 1G HugeTLB page.
Soft-offline dissolves the HugeTLB page, either in-use or free, into
chunks of 4K pages, reducing HugeTLB pool capacity by 1 hugepage.
If userspace has not acknowledged such behavior, it may be surprised
when later mmap hugepages MAP_FAILED due to lack of hugepages.
In case of a transparent hugepage, it will be split into 4K pages
as well; userspace will stop enjoying the transparent performance.
In addition, discarding the entire 1G HugeTLB page only because of
corrected memory errors sounds very costly and kernel better not
doing under the hood. But today there are at least 2 such cases:
1. GHES driver sees both GHES_SEV_CORRECTED and
CPER_SEC_ERROR_THRESHOLD_EXCEEDED after parsing CPER.
2. RAS Correctable Errors Collector counts correctable errors per
PFN and when the counter for a PFN reaches threshold
In both cases, userspace has no control of the soft offline performed
by kernel's memory failure recovery.
This patch series give userspace the control of softofflining any page:
kernel only soft offlines raw page / transparent hugepage / HugeTLB
hugepage if userspace has agreed to. The interface to userspace is a
new sysctl called enable_soft_offline under /proc/sys/vm. By default
enable_soft_line is 1 to preserve existing behavior in kernel.
Changelog
v4 => v5:
* incorportate feedbacks from Muhammad Usama Anjum
<usama.anjum(a)collabora.com>
* refactor selftest to use what available in kselftest.h.
* update a comment in soft_offline_page.
v3 => v4:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>,
Andrew Morton <akpm(a)linux-foundation.org>, and
Oscar Salvador <osalvador(a)suse.de>.
* insert a refactor commit to unify soft offline's logs to follow
"Soft offline: 0x${pfn}: ${message}" format.
* some rewords in document: fail => will not perform.
* v4 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3"),
akpm/mm-stable.
v2 => v3:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>,
Lance Yang <ioworker0(a)gmail.com>, Oscar Salvador <osalvador(a)suse.de>,
and David Rientjes <rientjes(a)google.com>.
* release potential refcount if enable_soft_offline is 0.
* soft_offline_page() returns EOPNOTSUPP if enable_soft_offline is 0.
* refactor hugetlb-soft-offline.c, for example, introduce
test_soft_offline_common to reduce repeated code.
* rewrite enable_soft_offline's documentation, adds more details about
the cost of soft-offline for transparent and hugetlb hugepages, and
components that are impacted when enable_soft_offline becomes 0.
* fix typos in commit messages.
* v3 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3").
v1 => v2:
* incorporate feedbacks from both Miaohe Lin <linmiaohe(a)huawei.com> and
Jane Chu <jane.chu(a)oracle.com>.
* make the switch to control all pages, instead of HugeTLB specific.
* change the API from
/sys/kernel/mm/hugepages/hugepages-${size}kB/softoffline_corrected_errors
to /proc/sys/vm/enable_soft_offline.
* minor update to test code.
* update documentation of the user control API.
* v2 is based on commit 83a7eefedc9b ("Linux 6.10-rc3").
Jiaqi Yan (4):
mm/memory-failure: refactor log format in soft offline code
mm/memory-failure: userspace controls soft-offlining pages
selftest/mm: test enable_soft_offline behaviors
docs: mm: add enable_soft_offline sysctl
Documentation/admin-guide/sysctl/vm.rst | 32 +++
mm/memory-failure.c | 38 ++-
tools/testing/selftests/mm/.gitignore | 1 +
tools/testing/selftests/mm/Makefile | 1 +
.../selftests/mm/hugetlb-soft-offline.c | 227 ++++++++++++++++++
tools/testing/selftests/mm/run_vmtests.sh | 4 +
6 files changed, 295 insertions(+), 8 deletions(-)
create mode 100644 tools/testing/selftests/mm/hugetlb-soft-offline.c
--
2.45.2.741.gdbec12cfda-goog
Centralizes the definition of _GNU_SOURCE into lib.mk and addresses all
resulting macro redefinition warnings.
These patches will need to be merged in one shot to avoid redefinition
warnings.
The initial attempt at this patch was abandoned because it affected
lines in many source files and caused a large amount of churn. However,
from earlier discussions, centralizing _GNU_SOURCE is still desireable.
This attempt limits the changes to 1 source file and 12 Makefiles.
v1: https://lore.kernel.org/linux-kselftest/20240430235057.1351993-1-edliaw@goo…
v2: https://lore.kernel.org/linux-kselftest/20240507214254.2787305-1-edliaw@goo…
- Add -D_GNU_SOURCE to KHDR_INCLUDES so that it is in a single
location.
- Remove #define _GNU_SOURCE from source code to resolve redefinition
warnings.
v3: https://lore.kernel.org/linux-kselftest/20240509200022.253089-1-edliaw@goog…
- Rebase onto linux-next 20240508.
- Split patches by directory.
- Add -D_GNU_SOURCE directly to CFLAGS in lib.mk.
- Delete additional _GNU_SOURCE definitions from source code in
linux-next.
- Delete additional -D_GNU_SOURCE flags from Makefiles.
v4: https://lore.kernel.org/linux-kselftest/20240510000842.410729-1-edliaw@goog…
- Rebase onto linux-next 20240509.
- Remove Fixes tag from patches that drop _GNU_SOURCE definition.
- Restore space between comment and includes for selftests/damon.
v5: https://lore.kernel.org/linux-kselftest/20240522005913.3540131-1-edliaw@goo…
- Rebase onto linux-next 20240521
- Drop initial patches that modify KHDR_INCLUDES.
- Incorporate Mark Brown's patch to replace static_assert with warning.
- Don't drop #define _GNU_SOURCE from nolibc and wireguard.
- Change Makefiles for x86 and vDSO to append to CFLAGS.
v6:
- Rewrite patch to use -D_GNU_SOURCE= form in lib.mk.
- Reduce the amount of churn significantly by allowing definition to
coexist with source code macro defines.
Edward Liaw (13):
selftests/mm: Define _GNU_SOURCE to an empty string
selftests: Add -D_GNU_SOURCE= to CFLAGS in lib.mk
selftests/net: Append to lib.mk CFLAGS in Makefile
selftests/exec: Drop redundant -D_GNU_SOURCE CFLAGS in Makefile
selftests/futex: Drop redundant -D_GNU_SOURCE CFLAGS in Makefile
selftests/intel_pstate: Drop redundant -D_GNU_SOURCE CFLAGS in
Makefile
selftests/iommu: Drop redundant -D_GNU_SOURCE CFLAGS in Makefile
selftests/kvm: Drop redundant -D_GNU_SOURCE CFLAGS in Makefile
selftests/proc: Drop redundant -D_GNU_SOURCE CFLAGS in Makefile
selftests/resctrl: Drop redundant -D_GNU_SOURCE CFLAGS in Makefile
selftests/ring-buffer: Drop redundant -D_GNU_SOURCE CFLAGS in Makefile
selftests/riscv: Drop redundant -D_GNU_SOURCE CFLAGS in Makefile
selftests/sgx: Append CFLAGS from lib.mk to HOST_CFLAGS
tools/testing/selftests/exec/Makefile | 1 -
tools/testing/selftests/futex/functional/Makefile | 2 +-
tools/testing/selftests/intel_pstate/Makefile | 2 +-
tools/testing/selftests/iommu/Makefile | 2 --
tools/testing/selftests/kvm/Makefile | 2 +-
tools/testing/selftests/lib.mk | 3 +++
tools/testing/selftests/mm/thuge-gen.c | 2 +-
tools/testing/selftests/net/Makefile | 2 +-
tools/testing/selftests/net/tcp_ao/Makefile | 2 +-
tools/testing/selftests/proc/Makefile | 1 -
tools/testing/selftests/resctrl/Makefile | 2 +-
tools/testing/selftests/ring-buffer/Makefile | 1 -
tools/testing/selftests/riscv/mm/Makefile | 2 +-
tools/testing/selftests/sgx/Makefile | 2 +-
14 files changed, 12 insertions(+), 14 deletions(-)
--
2.45.2.741.gdbec12cfda-goog
This series introduces the selftests/arm directory, which tests 32 and
64-bit kernel compatibility with 32-bit ELFs running on the Aarch platform.
The need for this bucket of tests is that 32 bit applications built on
legacy ARM architecture must not break on the new Aarch64 platforms and
the 64-bit kernel. The kernel must emulate the data structures, system
calls and the registers according to Aarch32, when running a 32-bit
process; this directory fills that testing requirement.
One may find similarity between this directory and selftests/arm64; it is
advisable to refer to that since a lot has been pulled from there itself.
The mm directory includes a test for checking 4GB limit of the virtual
address space of a process.
The signal directory contains two tests, following a common theme: mangle
with arm_cpsr, dumped by the kernel to user space while invoking the signal
handler; kernel must spot this illegal attempt and terminate the program by
SEGV.
The elf directory includes a test for checking the 32-bit status of the ELF.
The abi directory includes two ptrace tests, in the first, a 32-bit parent
debugs a 32-bit child, and in the second, a 64-bit parent debugs a 32-bit
child. The second test will be skipped when running on a 32-bit kernel.
Credits to Mark Brown for suggesting this work.
Testing:
The series has been tested on the Aarch64 kernel. For the Aarch32 kernel,
I used qemu-system-arm with machine 'vexpress-a15', along with a buildroot
rootfs; the individual statically built tests pass on that, but building
the entire test suite on that remains untested, due to my lack of
experience with qemu and rootfses.
Since I have done some changes in selftests/arm64, I have tested that
those tests do not break.
v2->v3:
- mm, elf: Split into multiple testcases
- Eliminate copying in signal/ using ifdeffery and pulling from selftests/arm64
- Delete config file, since it does not make sense for testing a 32-bit kernel
- Split ptrace in selftests/arm64, and pull some stuff from there
- Add abi tests containing ptrace and ptrace_64
- Fix build warnings in selftests/arm64 (can be applied independent of this series)
v1->v2:
- Formatting changes
- Add .gitignore files and config file
v1:
- https://lore.kernel.org/all/20240405084410.256788-1-dev.jain@arm.com/
Dev Jain (9):
selftests/arm: Add mm test
selftests/arm: Add elf test
selftests: arm, arm64: Use ifdeffery to pull signal infrastructure
selftests/arm: Add signal tests
selftests/arm64: Fix build warnings for ptrace
selftests/arm64: Split ptrace, use ifdeffery
selftests/arm: Add ptrace test
selftests/arm: Add ptrace_64 test
selftests: Add build infrastructure along with README
tools/testing/selftests/Makefile | 1 +
tools/testing/selftests/arm/Makefile | 56 ++++++++
tools/testing/selftests/arm/README | 32 +++++
tools/testing/selftests/arm/abi/.gitignore | 4 +
tools/testing/selftests/arm/abi/Makefile | 26 ++++
tools/testing/selftests/arm/abi/ptrace.c | 82 +++++++++++
tools/testing/selftests/arm/abi/ptrace.h | 57 ++++++++
tools/testing/selftests/arm/abi/ptrace_64.c | 91 ++++++++++++
.../selftests/arm/abi/trivial_32bit_program.c | 14 ++
tools/testing/selftests/arm/elf/.gitignore | 2 +
tools/testing/selftests/arm/elf/Makefile | 6 +
tools/testing/selftests/arm/elf/parse_elf.c | 77 ++++++++++
tools/testing/selftests/arm/mm/.gitignore | 2 +
tools/testing/selftests/arm/mm/Makefile | 6 +
tools/testing/selftests/arm/mm/compat_va.c | 89 ++++++++++++
tools/testing/selftests/arm/signal/.gitignore | 3 +
tools/testing/selftests/arm/signal/Makefile | 30 ++++
.../selftests/arm/signal/test_signals.c | 2 +
.../selftests/arm/signal/test_signals.h | 2 +
.../selftests/arm/signal/test_signals_utils.c | 2 +
.../selftests/arm/signal/test_signals_utils.h | 2 +
.../testcases/mangle_cpsr_invalid_aif_bits.c | 33 +++++
.../mangle_cpsr_invalid_compat_toggle.c | 29 ++++
tools/testing/selftests/arm64/abi/ptrace.c | 121 ++--------------
tools/testing/selftests/arm64/abi/ptrace.h | 135 ++++++++++++++++++
.../selftests/arm64/signal/test_signals.h | 12 ++
.../arm64/signal/test_signals_utils.c | 51 +++++--
.../arm64/signal/test_signals_utils.h | 3 +
28 files changed, 850 insertions(+), 120 deletions(-)
create mode 100644 tools/testing/selftests/arm/Makefile
create mode 100644 tools/testing/selftests/arm/README
create mode 100644 tools/testing/selftests/arm/abi/.gitignore
create mode 100644 tools/testing/selftests/arm/abi/Makefile
create mode 100644 tools/testing/selftests/arm/abi/ptrace.c
create mode 100644 tools/testing/selftests/arm/abi/ptrace.h
create mode 100644 tools/testing/selftests/arm/abi/ptrace_64.c
create mode 100644 tools/testing/selftests/arm/abi/trivial_32bit_program.c
create mode 100644 tools/testing/selftests/arm/elf/.gitignore
create mode 100644 tools/testing/selftests/arm/elf/Makefile
create mode 100644 tools/testing/selftests/arm/elf/parse_elf.c
create mode 100644 tools/testing/selftests/arm/mm/.gitignore
create mode 100644 tools/testing/selftests/arm/mm/Makefile
create mode 100644 tools/testing/selftests/arm/mm/compat_va.c
create mode 100644 tools/testing/selftests/arm/signal/.gitignore
create mode 100644 tools/testing/selftests/arm/signal/Makefile
create mode 100644 tools/testing/selftests/arm/signal/test_signals.c
create mode 100644 tools/testing/selftests/arm/signal/test_signals.h
create mode 100644 tools/testing/selftests/arm/signal/test_signals_utils.c
create mode 100644 tools/testing/selftests/arm/signal/test_signals_utils.h
create mode 100644 tools/testing/selftests/arm/signal/testcases/mangle_cpsr_invalid_aif_bits.c
create mode 100644 tools/testing/selftests/arm/signal/testcases/mangle_cpsr_invalid_compat_toggle.c
create mode 100644 tools/testing/selftests/arm64/abi/ptrace.h
--
2.39.2
This patch series fix a minor issue in a program for DAMON selftest, and
implement new functionality selftests for DAMOS tried regions and
{min,max}_nr_regions. The test for max_nr_regions also test the
recovery from online tuning-caused limit violation, which was fixed by a
previous patch [1] titled "mm/damon/core: merge regions aggressively
when max_nr_regions is unmet".
The first patch fixes a minor problem in the articial memory access
pattern generator for tests. Following 3 patches (2-4) implement
schemes tried regions test. Then a couple of patches (5-6) implementing
static setup based {min,max}_nr_regions functionality test follows.
Final two patches (7-8) implement dynamic max_nr_regions update test.
[1] https://lore.kernel.org/20240624210650.53960C2BBFC@smtp.kernel.org
SeongJae Park (8):
selftests/damon/access_memory: use user-defined region size
selftests/damon/_damon_sysfs: support schemes_update_tried_regions
selftests/damon: implement a program for even-numbered memory regions
access
selftests/damon: implement DAMOS tried regions test
selftests/damon/_damon_sysfs: implement kdamonds stop function
selftests/damon: implement test for min/max_nr_regions
_damon_sysfs: implement commit() for online parameters update
selftests/damon/damon_nr_regions: test online-tuned max_nr_regions
tools/testing/selftests/damon/Makefile | 3 +-
tools/testing/selftests/damon/_damon_sysfs.py | 65 +++++++-
tools/testing/selftests/damon/access_memory.c | 2 +-
.../selftests/damon/access_memory_even.c | 42 +++++
.../selftests/damon/damon_nr_regions.py | 145 ++++++++++++++++++
.../selftests/damon/damos_tried_regions.py | 65 ++++++++
6 files changed, 319 insertions(+), 3 deletions(-)
create mode 100644 tools/testing/selftests/damon/access_memory_even.c
create mode 100644 tools/testing/selftests/damon/damon_nr_regions.py
create mode 100644 tools/testing/selftests/damon/damos_tried_regions.py
base-commit: 99348045d11f3bac71146b381f90b0aa39855ee7
--
2.39.2
Currently, if a user wants to run pmtu.sh and cover all the provided test
cases, they need to install the Open vSwitch userspace utilities. This
dependency is difficult for users as well as CI environments, because the
userspace build and setup may require lots of support and devel packages
to be installed, system setup to be correct, and things like permissions
and selinux policies to be properly configured.
The kernel selftest suite includes an ovs-dpctl.py utility which can
interact with the openvswitch module directly. This lets developers and
CI environments run without needing too many extra dependencies - just
the pyroute2 python package.
This series enhances the ovs-dpctl utility to provide support for set()
and tunnel() flow specifiers, better ipv6 handling support, and the
ability to add tunnel vports, and LWT interfaces. Finally, it modifies
the pmtu.sh script to call the ovs-dpctl.py utility rather than the
typical OVS userspace utilities.
Aaron Conole (7):
selftests: openvswitch: Support explicit tunnel port creation.
selftests: openvswitch: Refactor actions parsing.
selftests: openvswitch: Add set() and set_masked() support.
selftests: openvswitch: Add support for tunnel() key.
selftests: openvswitch: Support implicit ipv6 arguments.
selftests: net: Use the provided dpctl rather than the vswitchd for
tests.
selftests: net: add config for openvswitch
tools/testing/selftests/net/config | 5 +
.../selftests/net/openvswitch/ovs-dpctl.py | 366 +++++++++++++++---
tools/testing/selftests/net/pmtu.sh | 145 +++++--
3 files changed, 450 insertions(+), 66 deletions(-)
--
2.45.1
It seems that there is no definition for config IP_GRE, and it is not a
dependency of other configs, so remove it.
linux$ find -name Kconfig | xargs grep "IP_GRE"
<-- nothing
There is a IPV6_GRE config defined in net/ipv6/Kconfig. It only depends
on NET_IPGRE_DEMUX but not IP_GRE.
Fixes: 04fe7c5029cb ("selftests: fill in some missing configs for net")
Reported-by: kernel test robot <lkp(a)intel.com>
Signed-off-by: Yujie Liu <yujie.liu(a)intel.com>
---
tools/testing/selftests/net/config | 1 -
1 file changed, 1 deletion(-)
diff --git a/tools/testing/selftests/net/config b/tools/testing/selftests/net/config
index d4891f7a2bfa..f46e27cd1e70 100644
--- a/tools/testing/selftests/net/config
+++ b/tools/testing/selftests/net/config
@@ -26,7 +26,6 @@ CONFIG_INET_ESP=y
CONFIG_INET_ESP_OFFLOAD=y
CONFIG_NET_FOU=y
CONFIG_NET_FOU_IP_TUNNELS=y
-CONFIG_IP_GRE=m
CONFIG_NETFILTER=y
CONFIG_NETFILTER_ADVANCED=y
CONFIG_NF_CONNTRACK=m
--
2.34.1
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v4:
- a new patch to use make_sockaddr in sockmap_ktls.
- a new patch to close fd in error path in drop_on_reuseport.
- drop make_server() in patch 7.
- drop make_client() too in patch 9.
v3:
- a new patch to add backlog for network_helper_opts.
- use start_server_str in sockmap_ktls now, not start_server.
v2:
- address Eduard's comments in v1. (thanks)
- fix errors reported by CI.
This patch set uses network helpers in sockmap_ktls and sk_lookup, and
drop three local helpers tcp_server(), inetaddr_len(), make_socket(),
make_server() and make_client() in them.
Geliang Tang (9):
selftests/bpf: Add backlog for network_helper_opts
selftests/bpf: Use start_server_str in sockmap_ktls
selftests/bpf: Use connect_to_fd in sockmap_ktls
selftests/bpf: Use make_sockaddr in sockmap_ktls
selftests/bpf: Close fd in error path in drop_on_reuseport
selftests/bpf: Invoke attach_reuseport out of make_server
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Use connect_to_fd in sk_lookup
selftests/bpf: Use connect_to_addr in sk_lookup
tools/testing/selftests/bpf/network_helpers.c | 2 +-
tools/testing/selftests/bpf/network_helpers.h | 1 +
.../selftests/bpf/prog_tests/sk_lookup.c | 263 +++++++++---------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 51 +---
4 files changed, 140 insertions(+), 177 deletions(-)
--
2.43.0
From: Geliang Tang <tanggeliang(a)kylinos.cn>
When I ran BPF selftests on loongarch recently, some errors occur. Fix
them in this set.
Geliang Tang (3):
skmsg: null check for page in sk_msg_recvmsg
inet: null check for close in inet_release
selftests/bpf: Null checks for link in bpf_tcp_ca
net/core/skmsg.c | 2 ++
net/ipv4/af_inet.c | 3 ++-
tools/testing/selftests/bpf/prog_tests/bpf_tcp_ca.c | 12 ++++++++----
3 files changed, 12 insertions(+), 5 deletions(-)
--
2.43.0
Correctable memory errors are very common on servers with large
amount of memory, and are corrected by ECC, but with two
pain points to users:
1. Correction usually happens on the fly and adds latency overhead
2. Not-fully-proved theory states excessive correctable memory
errors can develop into uncorrectable memory error.
Soft offline is kernel's additional solution for memory pages
having (excessive) corrected memory errors. Impacted page is migrated
to healthy page if it is in use, then the original page is discarded
for any future use.
The actual policy on whether (and when) to soft offline should be
maintained by userspace, especially in case of an 1G HugeTLB page.
Soft-offline dissolves the HugeTLB page, either in-use or free, into
chunks of 4K pages, reducing HugeTLB pool capacity by 1 hugepage.
If userspace has not acknowledged such behavior, it may be surprised
when later mmap hugepages MAP_FAILED due to lack of hugepages.
In case of a transparent hugepage, it will be split into 4K pages
as well; userspace will stop enjoying the transparent performance.
In addition, discarding the entire 1G HugeTLB page only because of
corrected memory errors sounds very costly and kernel better not
doing under the hood. But today there are at least 2 such cases:
1. GHES driver sees both GHES_SEV_CORRECTED and
CPER_SEC_ERROR_THRESHOLD_EXCEEDED after parsing CPER.
2. RAS Correctable Errors Collector counts correctable errors per
PFN and when the counter for a PFN reaches threshold
In both cases, userspace has no control of the soft offline performed
by kernel's memory failure recovery.
This patch series give userspace the control of softofflining any page:
kernel only soft offlines raw page / transparent hugepage / HugeTLB
hugepage if userspace has agreed to. The interface to userspace is a
new sysctl called enable_soft_offline under /proc/sys/vm. By default
enable_soft_line is 1 to preserve existing behavior in kernel.
Changelog
v3 => v4:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>,
Andrew Morton <akpm(a)linux-foundation.org>, and
Oscar Salvador <osalvador(a)suse.de>.
* insert a refactor commit to unify soft offline's logs to follow
"Soft offline: 0x${pfn}: ${message}" format.
* some rewords in document: fail => will not perform.
* v4 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3"),
akpm/mm-stable.
v2 => v3:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.com>,
Lance Yang <ioworker0(a)gmail.com>, Oscar Salvador <osalvador(a)suse.de>,
and David Rientjes <rientjes(a)google.com>.
* release potential refcount if enable_soft_offline is 0.
* soft_offline_page() returns EOPNOTSUPP if enable_soft_offline is 0.
* refactor hugetlb-soft-offline.c, for example, introduce
test_soft_offline_common to reduce repeated code.
* rewrite enable_soft_offline's documentation, adds more details about
the cost of soft-offline for transparent and hugetlb hugepages, and
components that are impacted when enable_soft_offline becomes 0.
* fix typos in commit messages.
* v3 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3").
v1 => v2:
* incorporate feedbacks from both Miaohe Lin <linmiaohe(a)huawei.com> and
Jane Chu <jane.chu(a)oracle.com>.
* make the switch to control all pages, instead of HugeTLB specific.
* change the API from
/sys/kernel/mm/hugepages/hugepages-${size}kB/softoffline_corrected_errors
to /proc/sys/vm/enable_soft_offline.
* minor update to test code.
* update documentation of the user control API.
* v2 is based on commit 83a7eefedc9b ("Linux 6.10-rc3").
Jiaqi Yan (4):
mm/memory-failure: refactor log format in soft offline code
mm/memory-failure: userspace controls soft-offlining pages
selftest/mm: test enable_soft_offline behaviors
docs: mm: add enable_soft_offline sysctl
Documentation/admin-guide/sysctl/vm.rst | 32 +++
mm/memory-failure.c | 38 ++-
tools/testing/selftests/mm/.gitignore | 1 +
tools/testing/selftests/mm/Makefile | 1 +
.../selftests/mm/hugetlb-soft-offline.c | 229 ++++++++++++++++++
tools/testing/selftests/mm/run_vmtests.sh | 4 +
6 files changed, 297 insertions(+), 8 deletions(-)
create mode 100644 tools/testing/selftests/mm/hugetlb-soft-offline.c
--
2.45.2.741.gdbec12cfda-goog
From: Emanuele Ghidoli <emanuele.ghidoli(a)toradex.com>
Hi,
I am submitting two patches related to the RX8310 RTC configuration:
1. **Correct RX8310 RTC Features Configuration:**
- Sets one-minute alarm resolution.
- Disables update IRQ.
Without this patch, if the interrupt is configured in the device tree, `hwclock` fails to read the RTC as it waits for the update IRQ.
2. **Avoid rtctest Alarm Test Failures:**
- Ensures rtctest alarm tests do not fail on hardware that only supports one-minute alarm resolution.
I know the second patch creates conflict with this one: https://lore.kernel.org/all/20240524013807.154338-1-jjang@nvidia.com/.
I think that series can reuse the rtc_get_features function I implemented.
Best regards,
Emanuele Ghidoli
Emanuele Ghidoli (2):
rtc: ds1307: set one-minute alarm resolution for rx_8130
selftests: rtc: rtctest: skip alarm test if alarm resolution is one
minute
drivers/rtc/rtc-ds1307.c | 14 +++++++++++--
tools/testing/selftests/rtc/rtctest.c | 30 +++++++++++++++++++++++++++
2 files changed, 42 insertions(+), 2 deletions(-)
--
2.34.1
This series is a followup of the struct_ops conversion.
Therefore, it is based on top of the for-6.11/bpf branch of the hid.git
tree:
https://git.kernel.org/pub/scm/linux/kernel/git/hid/hid.git/log/?h=for-6.11…
The first patch should go in ASAP, it's a fix that was detected by Dan
and which is actually breaking some use cases.
The rest is adding new capabilities to HID-BPF: being able to intercept
hid_hw_raw_request() and hid_hw_ouptut_report(). Both operations are
write operations to the device.
Having those new hooks allows to implement the "firewall" of HID
devices: this way a bpf program can selectively authorize an hidraw
client to write or not to the device depending on what is requested.
This also allows to completely emulate new behavior: we can now create a
"fake" feature on a HID device, and when we receive a request on this
feature, we can emulate the answer by either statically answering or
even by communicating with the device from bpf, as those new hooks are
sleepable.
Last, there is one change in the kfunc hid_bpf_input_report, in which it
actually waits for the device to be ready. This will not break any
potential users as the function was already declared as sleepable.
Cheers,
Benjamin
Signed-off-by: Benjamin Tissoires <bentiss(a)kernel.org>
---
Benjamin Tissoires (12):
HID: bpf: fix dispatch_hid_bpf_device_event uninitialized ret value
HID: add source argument to HID low level functions
HID: bpf: add HID-BPF hooks for hid_hw_raw_requests
HID: bpf: prevent infinite recursions with hid_hw_raw_requests hooks
selftests/hid: add tests for hid_hw_raw_request HID-BPF hooks
HID: bpf: add HID-BPF hooks for hid_hw_output_report
selftests/hid: add tests for hid_hw_output_report HID-BPF hooks
HID: bpf: make hid_bpf_input_report() sleep until the device is ready
selftests/hid: add wq test for hid_bpf_input_report()
HID: bpf: allow hid_device_event hooks to inject input reports on self
selftests/hid: add another test for injecting an event from an event hook
selftests/hid: add an infinite loop test for hid_bpf_try_input_report
Documentation/hid/hid-bpf.rst | 2 +-
drivers/hid/bpf/hid_bpf_dispatch.c | 157 +++++++++-
drivers/hid/bpf/hid_bpf_dispatch.h | 1 +
drivers/hid/bpf/hid_bpf_struct_ops.c | 4 +-
drivers/hid/hid-core.c | 110 ++++---
drivers/hid/hidraw.c | 10 +-
include/linux/hid.h | 7 +
include/linux/hid_bpf.h | 73 ++++-
tools/testing/selftests/hid/hid_bpf.c | 326 +++++++++++++++++++++
tools/testing/selftests/hid/progs/hid.c | 292 ++++++++++++++++++
.../testing/selftests/hid/progs/hid_bpf_helpers.h | 13 +
11 files changed, 934 insertions(+), 61 deletions(-)
---
base-commit: 33c0fb85b571b0f1bbdbf466e770eebeb29e6f41
change-id: 20240614-hid_hw_req_bpf-df0b95aeb425
Best regards,
--
Benjamin Tissoires <bentiss(a)kernel.org>
From: Quan Zhou <zhouquan(a)iscas.ac.cn>
Due to the path that modifies a0 in syscall_enter_from_user_mode before the
actual execution of syscall_handler [1], the kernel currently saves a0 to
orig_a0 at the entry point of do_trap_ecall_u as an original copy of a0.
Once the syscall is interrupted and later resumed, the restarted syscall
will use orig_a0 to continue execution.
The above rules generally apply except for ptrace(PTRACE_SETREGSET,),
where the kernel will ignore the tracer's setting of tracee/a0 and
will restart with the tracee's original a0 value. For the current
kernel implementation of ptrace, projects like CRIU/Proot will
encounter issues where the a0 setting becomes ineffective when
performing ptrace(PTRACE_{SET/GET}REGSET,).
Here is a suggested solution, expose orig_a0 to userspace so that ptrace
can choose whether to set orig_a0 based on the actual scenario. In fact,
x86/orig_eax and loongArch/orig_a0 have adopted similar solutions.
[1] link: https://lore.kernel.org/lkml/20230403-crisping-animosity-04ed8a45c625@spud/…
Quan Zhou (2):
riscv: Expose orig_a0 in the user_regs_struct structure
riscv: selftests: Add a ptrace test to check a0 of restarted syscall
arch/riscv/include/asm/ptrace.h | 4 +-
arch/riscv/include/uapi/asm/ptrace.h | 2 +
tools/testing/selftests/riscv/Makefile | 2 +-
tools/testing/selftests/riscv/abi/.gitignore | 1 +
tools/testing/selftests/riscv/abi/Makefile | 12 ++
.../riscv/abi/ptrace_restart_syscall.c | 148 ++++++++++++++++++
6 files changed, 166 insertions(+), 3 deletions(-)
create mode 100644 tools/testing/selftests/riscv/abi/.gitignore
create mode 100644 tools/testing/selftests/riscv/abi/Makefile
create mode 100644 tools/testing/selftests/riscv/abi/ptrace_restart_syscall.c
base-commit: 6ba59ff4227927d3a8530fc2973b80e94b54d58f
--
2.34.1
v12: https://patchwork.kernel.org/project/netdevbpf/list/?series=859747&state=*
====
Major changes:
--------------
This iteration only addresses one minor comment from Pavel with regards
to the trace printing of netmem, and the patchwork build error
introduced in v11 because I missed doing an allmodconfig build, sorry.
Other than that v11, AFAICT, received no feedback. There is one
discussion about how the specifics of plugging io uring memory through
the page pool, but not relevant to content in this particular patchset,
AFAICT.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v12/
v11: https://patchwork.kernel.org/project/netdevbpf/list/?series=857457&state=*
====
Major Changes:
--------------
v11 addresses feedback received in v10. The major change is the removal
of the memory provider ops as requested by Christoph. We still
accomplish the same thing, but utilizing direct function calls with if
statements rather than generic ops.
Additionally address sparse warnings, bugs and review comments from
folks that reviewed.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v11/
Detailed changelog:
-------------------
- Fixes in netdev_rx_queue_restart() from Pavel & David.
- Remove commit e650e8c3a36f5 ("net: page_pool: create hooks for
custom page providers") from the series to address Christoph's
feedback and rebased other patches on the series on this change.
- Fixed build errors with CONFIG_DMA_SHARED_BUFFER &&
!CONFIG_GENERIC_ALLOCATOR build.
- Fixed sparse warnings pointed out by Paolo.
- Drop unnecessary gro_pull_from_frag0 checks.
- Added Bagas reviewed-by to docs.
Cc: Bagas Sanjaya <bagasdotme(a)gmail.com>
Cc: Steven Rostedt <rostedt(a)goodmis.org>
Cc: Christoph Hellwig <hch(a)infradead.org>
Cc: Nikolay Aleksandrov <razor(a)blackwall.org>
v10: https://patchwork.kernel.org/project/netdevbpf/list/?series=852422&state=*
====
Major Changes:
--------------
v9 was sent right before the merge window closed (sorry!). v10 is almost
a re-send of the series now that the merge window re-opened. Only
rebased to latest net-next and addressed some minor iterative comments
received on v9.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v10/
Detailed changelog:
-------------------
- Fixed tokens leaking in DONTNEED setsockopt (Nikolay).
- Moved net_iov_dma_addr() to devmem.c and made it a devmem specific
helpers (David).
- Rename hook alloc_pages to alloc_netmems as alloc_pages is now
preprocessor macro defined and causes a build error.
v9:
===
Major Changes:
--------------
GVE queue API has been merged. Submitting this version as non-RFC after
rebasing on top of the merged API, and dropped the out of tree queue API
I was carrying on github. Addressed the little feedback v8 has received.
Detailed changelog:
------------------
- Added new patch from David Wei to this series for
netdev_rx_queue_restart()
- Fixed sparse error.
- Removed CONFIG_ checks in netmem_is_net_iov()
- Flipped skb->readable to skb->unreadable
- Minor fixes to selftests & docs.
RFC v8:
=======
Major Changes:
--------------
- Fixed build error generated by patch-by-patch build.
- Applied docs suggestions from Randy.
RFC v7:
=======
Major Changes:
--------------
This revision largely rebases on top of net-next and addresses the feedback
RFCv6 received from folks, namely Jakub, Yunsheng, Arnd, David, & Pavel.
The series remains in RFC because the queue-API ndos defined in this
series are not yet implemented. I have a GVE implementation I carry out
of tree for my testing. A upstreamable GVE implementation is in the
works. Aside from that, in my estimation all the patches are ready for
review/merge. Please do take a look.
As usual the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v7/
Detailed changelog:
- Use admin-perm in netlink API.
- Addressed feedback from Jakub with regards to netlink API
implementation.
- Renamed devmem.c functions to something more appropriate for that
file.
- Improve the performance seen through the page_pool benchmark.
- Fix the value definition of all the SO_DEVMEM_* uapi.
- Various fixes to documentation.
Perf - page-pool benchmark:
---------------------------
Improved performance of bench_page_pool_simple.ko tests compared to v6:
https://pastebin.com/raw/v5dYRg8L
net-next base: 8 cycle fast path.
RFC v6: 10 cycle fast path.
RFC v7: 9 cycle fast path.
RFC v7 with CONFIG_DMA_SHARED_BUFFER disabled: 8 cycle fast path,
same as baseline.
Perf - Devmem TCP benchmark:
---------------------
Perf is about the same regardless of the changes in v7, namely the
removal of the static_branch_unlikely to improve the page_pool benchmark
performance:
189/200gbps bi-directional throughput with RX devmem TCP and regular TCP
TX i.e. ~95% line rate.
RFC v6:
=======
Major Changes:
--------------
This revision largely rebases on top of net-next and addresses the little
feedback RFCv5 received.
The series remains in RFC because the queue-API ndos defined in this
series are not yet implemented. I have a GVE implementation I carry out
of tree for my testing. A upstreamable GVE implementation is in the
works. Aside from that, in my estimation all the patches are ready for
review/merge. Please do take a look.
As usual the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v6/
This version also comes with some performance data recorded in the cover
letter (see below changelog).
Detailed changelog:
- Rebased on top of the merged netmem_ref changes.
- Converted skb->dmabuf to skb->readable (Pavel). Pavel's original
suggestion was to remove the skb->dmabuf flag entirely, but when I
looked into it closely, I found the issue that if we remove the flag
we have to dereference the shinfo(skb) pointer to obtain the first
frag to tell whether an skb is readable or not. This can cause a
performance regression if it dirties the cache line when the
shinfo(skb) was not really needed. Instead, I converted the skb->dmabuf
flag into a generic skb->readable flag which can be re-used by io_uring
0-copy RX.
- Squashed a few locking optimizations from Eric Dumazet in the RX path
and the DEVMEM_DONTNEED setsockopt.
- Expanded the tests a bit. Added validation for invalid scenarios and
added some more coverage.
Perf - page-pool benchmark:
---------------------------
bench_page_pool_simple.ko tests with and without these changes:
https://pastebin.com/raw/ncHDwAbn
AFAIK the number that really matters in the perf tests is the
'tasklet_page_pool01_fast_path Per elem'. This one measures at about 8
cycles without the changes but there is some 1 cycle noise in some
results.
With the patches this regresses to 9 cycles with the changes but there
is 1 cycle noise occasionally running this test repeatedly.
Lastly I tried disable the static_branch_unlikely() in
netmem_is_net_iov() check. To my surprise disabling the
static_branch_unlikely() check reduces the fast path back to 8 cycles,
but the 1 cycle noise remains.
Perf - Devmem TCP benchmark:
---------------------
189/200gbps bi-directional throughput with RX devmem TCP and regular TCP
TX i.e. ~95% line rate.
Major changes in RFC v5:
========================
1. Rebased on top of 'Abstract page from net stack' series and used the
new netmem type to refer to LSB set pointers instead of re-using
struct page.
2. Downgraded this series back to RFC and called it RFC v5. This is
because this series is now dependent on 'Abstract page from net
stack'[1] and the queue API. Both are removed from the series to
reduce the patch # and those bits are fairly independent or
pre-requisite work.
3. Reworked the page_pool devmem support to use netmem and for some
more unified handling.
4. Reworked the reference counting of net_iov (renamed from
page_pool_iov) to use pp_ref_count for refcounting.
The full changes including the dependent series and GVE page pool
support is here:
https://github.com/mina/linux/commits/tcpdevmem-rfcv5/
[1] https://patchwork.kernel.org/project/netdevbpf/list/?series=810774
Major changes in v1:
====================
1. Implemented MVP queue API ndos to remove the userspace-visible
driver reset.
2. Fixed issues in the napi_pp_put_page() devmem frag unref path.
3. Removed RFC tag.
Many smaller addressed comments across all the patches (patches have
individual change log).
Full tree including the rest of the GVE driver changes:
https://github.com/mina/linux/commits/tcpdevmem-v1
Changes in RFC v3:
==================
1. Pulled in the memory-provider dependency from Jakub's RFC[1] to make the
series reviewable and mergeable.
2. Implemented multi-rx-queue binding which was a todo in v2.
3. Fix to cmsg handling.
The sticking point in RFC v2[2] was the device reset required to refill
the device rx-queues after the dmabuf bind/unbind. The solution
suggested as I understand is a subset of the per-queue management ops
Jakub suggested or similar:
https://lore.kernel.org/netdev/20230815171638.4c057dcd@kernel.org/
This is not addressed in this revision, because:
1. This point was discussed at netconf & netdev and there is openness to
using the current approach of requiring a device reset.
2. Implementing individual queue resetting seems to be difficult for my
test bed with GVE. My prototype to test this ran into issues with the
rx-queues not coming back up properly if reset individually. At the
moment I'm unsure if it's a mistake in the POC or a genuine issue in
the virtualization stack behind GVE, which currently doesn't test
individual rx-queue restart.
3. Our usecases are not bothered by requiring a device reset to refill
the buffer queues, and we'd like to support NICs that run into this
limitation with resetting individual queues.
My thought is that drivers that have trouble with per-queue configs can
use the support in this series, while drivers that support new netdev
ops to reset individual queues can automatically reset the queue as
part of the dma-buf bind/unbind.
The same approach with device resets is presented again for consideration
with other sticking points addressed.
This proposal includes the rx devmem path only proposed for merge. For a
snapshot of my entire tree which includes the GVE POC page pool support &
device memory support:
https://github.com/torvalds/linux/compare/master...mina:linux:tcpdevmem-v3
[1] https://lore.kernel.org/netdev/f8270765-a27b-6ccf-33ea-cda097168d79@redhat.…
[2] https://lore.kernel.org/netdev/CAHS8izOVJGJH5WF68OsRWFKJid1_huzzUK+hpKbLcL4…
Changes in RFC v2:
==================
The sticking point in RFC v1[1] was the dma-buf pages approach we used to
deliver the device memory to the TCP stack. RFC v2 is a proof-of-concept
that attempts to resolve this by implementing scatterlist support in the
networking stack, such that we can import the dma-buf scatterlist
directly. This is the approach proposed at a high level here[2].
Detailed changes:
1. Replaced dma-buf pages approach with importing scatterlist into the
page pool.
2. Replace the dma-buf pages centric API with a netlink API.
3. Removed the TX path implementation - there is no issue with
implementing the TX path with scatterlist approach, but leaving
out the TX path makes it easier to review.
4. Functionality is tested with this proposal, but I have not conducted
perf testing yet. I'm not sure there are regressions, but I removed
perf claims from the cover letter until they can be re-confirmed.
5. Added Signed-off-by: contributors to the implementation.
6. Fixed some bugs with the RX path since RFC v1.
Any feedback welcome, but specifically the biggest pending questions
needing feedback IMO are:
1. Feedback on the scatterlist-based approach in general.
2. Netlink API (Patch 1 & 2).
3. Approach to handle all the drivers that expect to receive pages from
the page pool (Patch 6).
[1] https://lore.kernel.org/netdev/dfe4bae7-13a0-3c5d-d671-f61b375cb0b4@gmail.c…
[2] https://lore.kernel.org/netdev/CAHS8izPm6XRS54LdCDZVd0C75tA1zHSu6jLVO8nzTLX…
==================
* TL;DR:
Device memory TCP (devmem TCP) is a proposal for transferring data to and/or
from device memory efficiently, without bouncing the data to a host memory
buffer.
* Problem:
A large amount of data transfers have device memory as the source and/or
destination. Accelerators drastically increased the volume of such transfers.
Some examples include:
- ML accelerators transferring large amounts of training data from storage into
GPU/TPU memory. In some cases ML training setup time can be as long as 50% of
TPU compute time, improving data transfer throughput & efficiency can help
improving GPU/TPU utilization.
- Distributed training, where ML accelerators, such as GPUs on different hosts,
exchange data among them.
- Distributed raw block storage applications transfer large amounts of data with
remote SSDs, much of this data does not require host processing.
Today, the majority of the Device-to-Device data transfers the network are
implemented as the following low level operations: Device-to-Host copy,
Host-to-Host network transfer, and Host-to-Device copy.
The implementation is suboptimal, especially for bulk data transfers, and can
put significant strains on system resources, such as host memory bandwidth,
PCIe bandwidth, etc. One important reason behind the current state is the
kernel’s lack of semantics to express device to network transfers.
* Proposal:
In this patch series we attempt to optimize this use case by implementing
socket APIs that enable the user to:
1. send device memory across the network directly, and
2. receive incoming network packets directly into device memory.
Packet _payloads_ go directly from the NIC to device memory for receive and from
device memory to NIC for transmit.
Packet _headers_ go to/from host memory and are processed by the TCP/IP stack
normally. The NIC _must_ support header split to achieve this.
Advantages:
- Alleviate host memory bandwidth pressure, compared to existing
network-transfer + device-copy semantics.
- Alleviate PCIe BW pressure, by limiting data transfer to the lowest level
of the PCIe tree, compared to traditional path which sends data through the
root complex.
* Patch overview:
** Part 1: netlink API
Gives user ability to bind dma-buf to an RX queue.
** Part 2: scatterlist support
Currently the standard for device memory sharing is DMABUF, which doesn't
generate struct pages. On the other hand, networking stack (skbs, drivers, and
page pool) operate on pages. We have 2 options:
1. Generate struct pages for dmabuf device memory, or,
2. Modify the networking stack to process scatterlist.
Approach #1 was attempted in RFC v1. RFC v2 implements approach #2.
** part 3: page pool support
We piggy back on page pool memory providers proposal:
https://github.com/kuba-moo/linux/tree/pp-providers
It allows the page pool to define a memory provider that provides the
page allocation and freeing. It helps abstract most of the device memory
TCP changes from the driver.
** part 4: support for unreadable skb frags
Page pool iovs are not accessible by the host; we implement changes
throughput the networking stack to correctly handle skbs with unreadable
frags.
** Part 5: recvmsg() APIs
We define user APIs for the user to send and receive device memory.
Not included with this series is the GVE devmem TCP support, just to
simplify the review. Code available here if desired:
https://github.com/mina/linux/tree/tcpdevmem
This series is built on top of net-next with Jakub's pp-providers changes
cherry-picked.
* NIC dependencies:
1. (strict) Devmem TCP require the NIC to support header split, i.e. the
capability to split incoming packets into a header + payload and to put
each into a separate buffer. Devmem TCP works by using device memory
for the packet payload, and host memory for the packet headers.
2. (optional) Devmem TCP works better with flow steering support & RSS support,
i.e. the NIC's ability to steer flows into certain rx queues. This allows the
sysadmin to enable devmem TCP on a subset of the rx queues, and steer
devmem TCP traffic onto these queues and non devmem TCP elsewhere.
The NIC I have access to with these properties is the GVE with DQO support
running in Google Cloud, but any NIC that supports these features would suffice.
I may be able to help reviewers bring up devmem TCP on their NICs.
* Testing:
The series includes a udmabuf kselftest that show a simple use case of
devmem TCP and validates the entire data path end to end without
a dependency on a specific dmabuf provider.
** Test Setup
Kernel: net-next with this series and memory provider API cherry-picked
locally.
Hardware: Google Cloud A3 VMs.
NIC: GVE with header split & RSS & flow steering support.
Cc: Pavel Begunkov <asml.silence(a)gmail.com>
Cc: David Wei <dw(a)davidwei.uk>
Cc: Jason Gunthorpe <jgg(a)ziepe.ca>
Cc: Yunsheng Lin <linyunsheng(a)huawei.com>
Cc: Shailend Chand <shailend(a)google.com>
Cc: Harshitha Ramamurthy <hramamurthy(a)google.com>
Cc: Shakeel Butt <shakeel.butt(a)linux.dev>
Cc: Jeroen de Borst <jeroendb(a)google.com>
Cc: Praveen Kaligineedi <pkaligineedi(a)google.com>
Mina Almasry (13):
netdev: add netdev_rx_queue_restart()
net: netdev netlink api to bind dma-buf to a net device
netdev: support binding dma-buf to netdevice
netdev: netdevice devmem allocator
page_pool: convert to use netmem
page_pool: devmem support
memory-provider: dmabuf devmem memory provider
net: support non paged skb frags
net: add support for skbs with unreadable frags
tcp: RX path for devmem TCP
net: add SO_DEVMEM_DONTNEED setsockopt to release RX frags
net: add devmem TCP documentation
selftests: add ncdevmem, netcat for devmem TCP
Documentation/netlink/specs/netdev.yaml | 57 +++
Documentation/networking/devmem.rst | 258 +++++++++++
Documentation/networking/index.rst | 1 +
arch/alpha/include/uapi/asm/socket.h | 6 +
arch/mips/include/uapi/asm/socket.h | 6 +
arch/parisc/include/uapi/asm/socket.h | 6 +
arch/sparc/include/uapi/asm/socket.h | 6 +
drivers/net/ethernet/renesas/ravb_main.c | 5 +-
include/linux/skbuff.h | 61 ++-
include/linux/skbuff_ref.h | 11 +-
include/linux/socket.h | 1 +
include/net/devmem.h | 124 ++++++
include/net/mp_dmabuf_devmem.h | 44 ++
include/net/netdev_rx_queue.h | 5 +
include/net/netmem.h | 208 ++++++++-
include/net/page_pool/helpers.h | 153 +++++--
include/net/page_pool/types.h | 22 +-
include/net/sock.h | 2 +
include/net/tcp.h | 5 +-
include/trace/events/page_pool.h | 30 +-
include/uapi/asm-generic/socket.h | 6 +
include/uapi/linux/netdev.h | 19 +
include/uapi/linux/uio.h | 17 +
net/bpf/test_run.c | 5 +-
net/core/Makefile | 3 +-
net/core/datagram.c | 6 +
net/core/dev.c | 6 +-
net/core/devmem.c | 373 ++++++++++++++++
net/core/gro.c | 3 +-
net/core/netdev-genl-gen.c | 23 +
net/core/netdev-genl-gen.h | 6 +
net/core/netdev-genl.c | 103 +++++
net/core/netdev_rx_queue.c | 74 ++++
net/core/page_pool.c | 360 ++++++++-------
net/core/skbuff.c | 83 +++-
net/core/sock.c | 61 +++
net/ipv4/esp4.c | 3 +-
net/ipv4/tcp.c | 261 ++++++++++-
net/ipv4/tcp_input.c | 13 +-
net/ipv4/tcp_ipv4.c | 10 +
net/ipv4/tcp_minisocks.c | 2 +
net/ipv4/tcp_output.c | 5 +-
net/ipv6/esp6.c | 3 +-
net/packet/af_packet.c | 4 +-
tools/include/uapi/linux/netdev.h | 19 +
tools/testing/selftests/net/.gitignore | 1 +
tools/testing/selftests/net/Makefile | 5 +
tools/testing/selftests/net/ncdevmem.c | 542 +++++++++++++++++++++++
48 files changed, 2759 insertions(+), 268 deletions(-)
create mode 100644 Documentation/networking/devmem.rst
create mode 100644 include/net/devmem.h
create mode 100644 include/net/mp_dmabuf_devmem.h
create mode 100644 net/core/devmem.c
create mode 100644 net/core/netdev_rx_queue.c
create mode 100644 tools/testing/selftests/net/ncdevmem.c
--
2.45.2.505.gda0bf45e8d-goog
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v3:
- a new patch to add backlog for network_helper_opts.
- use start_server_str in sockmap_ktls now, not start_server.
v2:
- address Eduard's comments in v1. (thanks)
- fix errors reported by CI.
This patch set uses network helpers in sockmap_ktls and sk_lookup, and drop
three local helpers tcp_server(), inetaddr_len() and make_socket() in them.
Geliang Tang (8):
selftests/bpf: Add backlog for network_helper_opts
selftests/bpf: Use start_server_str in sockmap_ktls
selftests/bpf: Use connect_to_fd in sockmap_ktls
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Use connect_to_fd in sk_lookup
selftests/bpf: Use connect_to_addr in sk_lookup
selftests/bpf: Drop inetaddr_len in sk_lookup
selftests/bpf: Drop make_socket in sk_lookup
tools/testing/selftests/bpf/network_helpers.c | 2 +-
tools/testing/selftests/bpf/network_helpers.h | 1 +
.../selftests/bpf/prog_tests/sk_lookup.c | 138 ++++++------------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 35 +----
4 files changed, 57 insertions(+), 119 deletions(-)
--
2.43.0
'==' is bashism, '=' needs to be used for comparison.
With this fix script can work on systems where the default shell is
dash, busybox ash or any other strictly POSIX compatible shell.
While at it, also improve:
* remove "x" in the comparison (not needed for decades)
* use $# for checking number of arguments
* cleanup whitespace
Fixes: 4ab5a5d2a4a22 ("tools: add a kernel-chktaint to tools/debugging")
Signed-off-by: Petr Vorel <pvorel(a)suse.cz>
---
tools/debugging/kernel-chktaint | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/tools/debugging/kernel-chktaint b/tools/debugging/kernel-chktaint
index 279be06332be9..adbb1d621ccd4 100755
--- a/tools/debugging/kernel-chktaint
+++ b/tools/debugging/kernel-chktaint
@@ -18,11 +18,11 @@ retrieved from /proc/sys/kernel/tainted on another system.
EOF
}
-if [ "$1"x != "x" ]; then
- if [ "$1"x == "--helpx" ] || [ "$1"x == "-hx" ] ; then
+if [ $# -gt 0 ]; then
+ if [ "$1" = "--help" ] || [ "$1" = "-h" ]; then
usage
exit 1
- elif [ $1 -ge 0 ] 2>/dev/null ; then
+ elif [ $1 -ge 0 ] 2>/dev/null; then
taint=$1
else
echo "Error: Parameter '$1' not a positive integer. Aborting." >&2
--
2.45.1
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v2:
- address Eduard's comments in v1. (thanks)
- fix errors reported by CI.
This patch set uses network helpers in sockmap_ktls and sk_lookup, and drop
three local helpers tcp_server(), inetaddr_len() and make_socket() in them.
Geliang Tang (7):
selftests/bpf: Use start_server in sockmap_ktls
selftests/bpf: Use connect_to_fd in sockmap_ktls
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Use connect_to_fd in sk_lookup
selftests/bpf: Use connect_to_addr in sk_lookup
selftests/bpf: Drop inetaddr_len in sk_lookup
selftests/bpf: Drop make_socket in sk_lookup
.../selftests/bpf/prog_tests/sk_lookup.c | 138 ++++++------------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 32 +---
2 files changed, 52 insertions(+), 118 deletions(-)
--
2.43.0
Hi all,
This series does a number of cleanups into resctrl_val() and
generalizes it by removing test name specific handling from the
function.
v7:
- Truly use "bound to", not bounded to.
- Fix separator to use 3 dashes
v6:
- Adjust closing/rollback of the IMC perf
- Move the comment in measure_vals() to function level
- Capitalize MBM
- binded to -> bound to
- Language tweak into kerneldoc
- Removed stale paragraph from commit message
v5:
- Open mem bw file only once and use rewind().
- Add \n to mem bw file read to allow reading fresh values from the file.
- Return 0 if create_grp() is given NULL grp_name (matches the original
behavior). Mention this in function's kerneldoc.
- Cast pid_t to int before printing with %d.
- Caps/typo fixes to kerneldoc and commit messages.
- Use imperative tone in commit messages and improve them based on points
that came up during review.
v4:
- Merged close fix into IMC READ+WRITE rework patch
- Add loop to reset imc_counters_config fds to -1 to be able know which
need closing
- Introduce perf_close_imc_mem_bw() to close fds
- Open resctrl mem bw file (twice) beforehand to avoid opening it during
the test
- Remove MBM .mongrp setup
- Remove mongrp from CMT test
v3:
- Rename init functions to <testname>_init()
- Replace for loops with READ+WRITE statements for clarity
- Don't drop Return: entry from perf_open_imc_mem_bw() func comment
- New patch: Fix closing of IMC fds in case of error
- New patch: Make "bandwidth" consistent in comments & prints
- New patch: Simplify mem bandwidth file code
- Remove wrong comment
- Changed grp_name check to return -1 on fail (internal sanity check)
v2:
- Resolved conflicts with kselftest/next
- Spaces -> tabs correction
Ilpo Järvinen (16):
selftests/resctrl: Fix closing IMC fds on error and open-code R+W
instead of loops
selftests/resctrl: Calculate resctrl FS derived mem bw over sleep(1)
only
selftests/resctrl: Make "bandwidth" consistent in comments & prints
selftests/resctrl: Consolidate get_domain_id() into resctrl_val()
selftests/resctrl: Use correct type for pids
selftests/resctrl: Cleanup bm_pid and ppid usage & limit scope
selftests/resctrl: Rename measure_vals() to measure_mem_bw_vals() &
document
selftests/resctrl: Simplify mem bandwidth file code for MBA & MBM
tests
selftests/resctrl: Add ->measure() callback to resctrl_val_param
selftests/resctrl: Add ->init() callback into resctrl_val_param
selftests/resctrl: Simplify bandwidth report type handling
selftests/resctrl: Make some strings passed to resctrlfs functions
const
selftests/resctrl: Convert ctrlgrp & mongrp to pointers
selftests/resctrl: Remove mongrp from MBA test
selftests/resctrl: Remove mongrp from CMT test
selftests/resctrl: Remove test name comparing from
write_bm_pid_to_resctrl()
tools/testing/selftests/resctrl/cache.c | 10 +-
tools/testing/selftests/resctrl/cat_test.c | 5 +-
tools/testing/selftests/resctrl/cmt_test.c | 22 +-
tools/testing/selftests/resctrl/mba_test.c | 26 +-
tools/testing/selftests/resctrl/mbm_test.c | 26 +-
tools/testing/selftests/resctrl/resctrl.h | 49 ++-
tools/testing/selftests/resctrl/resctrl_val.c | 371 ++++++++----------
tools/testing/selftests/resctrl/resctrlfs.c | 67 ++--
8 files changed, 291 insertions(+), 285 deletions(-)
--
2.39.2
From: Geliang Tang <tanggeliang(a)kylinos.cn>
This patch set uses network helpers in sockmap_ktls and sk_lookup, and drop
three local helpers tcp_server(), inetaddr_len() and make_socket() in them.
Geliang Tang (6):
selftests/bpf: Use start_server in sockmap_ktls
selftests/bpf: Use connect_to_fd in sockmap_ktls
selftests/bpf: Use start_server_str in sk_lookup
selftests/bpf: Drop inetaddr_len in sk_lookup
selftests/bpf: Use connect_to_fd in sk_lookup
selftests/bpf: Drop make_socket in sk_lookup
.../selftests/bpf/prog_tests/sk_lookup.c | 144 ++++++------------
.../selftests/bpf/prog_tests/sockmap_ktls.c | 32 +---
2 files changed, 53 insertions(+), 123 deletions(-)
--
2.43.0
** Background **
Currently, OVS supports several packet sampling mechanisms (sFlow,
per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
userspace action that needs to be handled by ovs-vswitchd's handler
threads only to be forwarded to some third party application that
will somehow process the sample and provide observability on the
datapath.
A particularly interesting use-case is controller-driven
per-flow IPFIX sampling where the OpenFlow controller can add metadata
to samples (via two 32bit integers) and this metadata is then available
to the sample-collecting system for correlation.
** Problem **
The fact that sampled traffic share netlink sockets and handler thread
time with upcalls, apart from being a performance bottleneck in the
sample extraction itself, can severely compromise the datapath,
yielding this solution unfit for highly loaded production systems.
Users are left with little options other than guessing what sampling
rate will be OK for their traffic pattern and system load and dealing
with the lost accuracy.
Looking at available infrastructure, an obvious candidated would be
to use psample. However, it's current state does not help with the
use-case at stake because sampled packets do not contain user-defined
metadata.
** Proposal **
This series is an attempt to fix this situation by extending the
existing psample infrastructure to carry a variable length
user-defined cookie.
The main existing user of psample is tc's act_sample. It is also
extended to forward the action's cookie to psample.
Finally, a new OVS action (OVS_SAMPLE_ATTR_EMIT_SAMPLE) is created.
It accepts a group and an optional cookie and uses psample to
multicast the packet and the metadata.
--
v3 -> v4:
- Rebased.
- Addressed Jakub's comment on private and unused nla attributes.
v2 -> v3:
- Addressed comments from Simon, Aaron and Ilya.
- Dropped probability propagation in nested sample actions.
- Dropped patch v2's 7/9 in favor of a userspace implementation and
consume skb if emit_sample is the last action, same as we do with
userspace.
- Split ovs-dpctl.py features in independent patches.
v1 -> v2:
- Create a new action ("emit_sample") rather than reuse existing
"sample" one.
- Add probability semantics to psample's sampling rate.
- Store sampling probability in skb's cb area and use it in emit_sample.
- Test combining "emit_sample" with "trunc"
- Drop group_id filtering and tracepoint in psample.
rfc_v2 -> v1:
- Accomodate Ilya's comments.
- Split OVS's attribute in two attributes and simplify internal
handling of psample arguments.
- Extend psample and tc with a user-defined cookie.
- Add a tracepoint to psample to facilitate troubleshooting.
rfc_v1 -> rfc_v2:
- Use psample instead of a new OVS-only multicast group.
- Extend psample and tc with a user-defined cookie.
Adrian Moreno (10):
net: psample: add user cookie
net: sched: act_sample: add action cookie to sample
net: psample: skip packet copy if no listeners
net: psample: allow using rate as probability
net: openvswitch: add emit_sample action
net: openvswitch: store sampling probability in cb.
selftests: openvswitch: add emit_sample action
selftests: openvswitch: add userspace parsing
selftests: openvswitch: parse trunc action
selftests: openvswitch: add emit_sample test
Documentation/netlink/specs/ovs_flow.yaml | 17 ++
include/net/psample.h | 5 +-
include/uapi/linux/openvswitch.h | 31 +-
include/uapi/linux/psample.h | 11 +-
include/uapi/linux/tc_act/tc_sample.h | 1 +
net/openvswitch/Kconfig | 1 +
net/openvswitch/actions.c | 63 +++-
net/openvswitch/datapath.h | 3 +
net/openvswitch/flow_netlink.c | 33 ++-
net/openvswitch/vport.c | 1 +
net/psample/psample.c | 16 +-
net/sched/act_sample.c | 12 +
.../selftests/net/openvswitch/openvswitch.sh | 110 ++++++-
.../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
14 files changed, 560 insertions(+), 16 deletions(-)
--
2.45.1
From: Roberto Sassu <roberto.sassu(a)huawei.com>
Integrity detection and protection has long been a desirable feature, to
reach a large user base and mitigate the risk of flaws in the software
and attacks.
However, while solutions exist, they struggle to reach the large user
base, due to requiring higher than desired constraints on performance,
flexibility and configurability, that only security conscious people are
willing to accept.
This is where the new digest_cache LSM comes into play, it offers
additional support for new and existing integrity solutions, to make
them faster and easier to deploy.
The full documentation with the motivation and the solution details can be
found in patch 14.
The IMA integration patch set will be introduced separately. Also a PoC
based on the current version of IPE can be provided.
v3:
- Rewrite documentation, and remove the installation instructions since
they are now included in the README of digest-cache-tools
- Add digest cache event notifier
- Drop digest_cache_was_reset(), and send instead to asynchronous
notifications
- Fix digest_cache LSM Kconfig style issues (suggested by Randy Dunlap)
- Propagate digest cache reset to directory entries
- Destroy per directory entry mutex
- Introduce RESET_USER bit, to clear the dig_user pointer on
set/removexattr
- Replace 'file content' with 'file data' (suggested by Mimi)
- Introduce per digest cache mutex and replace verif_data_lock spinlock
- Track changes of security.digest_list xattr
- Stop tracking file_open and use file_release instead also for file writes
- Add error messages in digest_cache_create()
- Load/unload testing kernel module automatically during execution of test
- Add tests for digest cache event notifier
- Add test for ftruncate()
- Remove DIGEST_CACHE_RESET_PREFETCH_BUF command in test and clear the
buffer on read instead
v2:
- Include the TLV parser in this patch set (from user asymmetric keys and
signatures)
- Move from IMA and make an independent LSM
- Remove IMA-specific stuff from this patch set
- Add per algorithm hash table
- Expect all digest lists to be in the same directory and allow changing
the default directory
- Support digest lookup on directories, when there is no
security.digest_list xattr
- Add seq num to digest list file name, to impose ordering on directory
iteration
- Add a new data type DIGEST_LIST_ENTRY_DATA for the nested data in the
tlv digest list format
- Add the concept of verification data attached to digest caches
- Add the reset mechanism to track changes on digest lists and directory
containing the digest lists
- Add kernel selftests
v1:
- Add documentation in Documentation/security/integrity-digest-cache.rst
- Pass the mask of IMA actions to digest_cache_alloc()
- Add a reference count to the digest cache
- Remove the path parameter from digest_cache_get(), and rely on the
reference count to avoid the digest cache disappearing while being used
- Rename the dentry_to_check parameter of digest_cache_get() to dentry
- Rename digest_cache_get() to digest_cache_new() and add
digest_cache_get() to set the digest cache in the iint of the inode for
which the digest cache was requested
- Add dig_owner and dig_user to the iint, to distinguish from which inode
the digest cache was created from, and which is using it; consequently it
makes the digest cache usable to measure/appraise other digest caches
(support not yet enabled)
- Add dig_owner_mutex and dig_user_mutex to serialize accesses to dig_owner
and dig_user until they are initialized
- Enforce strong synchronization and make the contenders wait until
dig_owner and dig_user are assigned to the iint the first time
- Move checking IMA actions on the digest list earlier, and fail if no
action were performed (digest cache not usable)
- Remove digest_cache_put(), not needed anymore with the introduction of
the reference count
- Fail immediately in digest_cache_lookup() if the digest algorithm is
not set in the digest cache
- Use 64 bit mask for IMA actions on the digest list instead of 8 bit
- Return NULL in the inline version of digest_cache_get()
- Use list_add_tail() instead of list_add() in the iterator
- Copy the digest list path to a separate buffer in digest_cache_iter_dir()
- Use digest list parsers verified with Frama-C
- Explicitly disable (for now) the possibility in the IMA policy to use the
digest cache to measure/appraise other digest lists
- Replace exit(<value>) with return <value> in manage_digest_lists.c
Roberto Sassu (14):
lib: Add TLV parser
security: Introduce the digest_cache LSM
digest_cache: Add securityfs interface
digest_cache: Add hash tables and operations
digest_cache: Populate the digest cache from a digest list
digest_cache: Parse tlv digest lists
digest_cache: Parse rpm digest lists
digest_cache: Add management of verification data
digest_cache: Add support for directories
digest cache: Prefetch digest lists if requested
digest_cache: Reset digest cache on file/directory change
digest_cache: Notify digest cache events
selftests/digest_cache: Add selftests for digest_cache LSM
docs: Add documentation of the digest_cache LSM
Documentation/security/digest_cache.rst | 763 ++++++++++++++++
Documentation/security/index.rst | 1 +
MAINTAINERS | 16 +
include/linux/digest_cache.h | 117 +++
include/linux/kernel_read_file.h | 1 +
include/linux/tlv_parser.h | 28 +
include/uapi/linux/lsm.h | 1 +
include/uapi/linux/tlv_digest_list.h | 72 ++
include/uapi/linux/tlv_parser.h | 59 ++
include/uapi/linux/xattr.h | 6 +
lib/Kconfig | 3 +
lib/Makefile | 3 +
lib/tlv_parser.c | 214 +++++
lib/tlv_parser.h | 17 +
security/Kconfig | 11 +-
security/Makefile | 1 +
security/digest_cache/Kconfig | 33 +
security/digest_cache/Makefile | 11 +
security/digest_cache/dir.c | 252 ++++++
security/digest_cache/htable.c | 268 ++++++
security/digest_cache/internal.h | 290 +++++++
security/digest_cache/main.c | 570 ++++++++++++
security/digest_cache/modsig.c | 66 ++
security/digest_cache/notifier.c | 135 +++
security/digest_cache/parsers/parsers.h | 15 +
security/digest_cache/parsers/rpm.c | 223 +++++
security/digest_cache/parsers/tlv.c | 299 +++++++
security/digest_cache/populate.c | 163 ++++
security/digest_cache/reset.c | 235 +++++
security/digest_cache/secfs.c | 87 ++
security/digest_cache/verif.c | 119 +++
security/security.c | 3 +-
tools/testing/selftests/Makefile | 1 +
.../testing/selftests/digest_cache/.gitignore | 3 +
tools/testing/selftests/digest_cache/Makefile | 24 +
.../testing/selftests/digest_cache/all_test.c | 815 ++++++++++++++++++
tools/testing/selftests/digest_cache/common.c | 78 ++
tools/testing/selftests/digest_cache/common.h | 135 +++
.../selftests/digest_cache/common_user.c | 47 +
.../selftests/digest_cache/common_user.h | 17 +
tools/testing/selftests/digest_cache/config | 1 +
.../selftests/digest_cache/generators.c | 248 ++++++
.../selftests/digest_cache/generators.h | 19 +
.../selftests/digest_cache/testmod/Makefile | 16 +
.../selftests/digest_cache/testmod/kern.c | 564 ++++++++++++
.../selftests/lsm/lsm_list_modules_test.c | 3 +
46 files changed, 6047 insertions(+), 6 deletions(-)
create mode 100644 Documentation/security/digest_cache.rst
create mode 100644 include/linux/digest_cache.h
create mode 100644 include/linux/tlv_parser.h
create mode 100644 include/uapi/linux/tlv_digest_list.h
create mode 100644 include/uapi/linux/tlv_parser.h
create mode 100644 lib/tlv_parser.c
create mode 100644 lib/tlv_parser.h
create mode 100644 security/digest_cache/Kconfig
create mode 100644 security/digest_cache/Makefile
create mode 100644 security/digest_cache/dir.c
create mode 100644 security/digest_cache/htable.c
create mode 100644 security/digest_cache/internal.h
create mode 100644 security/digest_cache/main.c
create mode 100644 security/digest_cache/modsig.c
create mode 100644 security/digest_cache/notifier.c
create mode 100644 security/digest_cache/parsers/parsers.h
create mode 100644 security/digest_cache/parsers/rpm.c
create mode 100644 security/digest_cache/parsers/tlv.c
create mode 100644 security/digest_cache/populate.c
create mode 100644 security/digest_cache/reset.c
create mode 100644 security/digest_cache/secfs.c
create mode 100644 security/digest_cache/verif.c
create mode 100644 tools/testing/selftests/digest_cache/.gitignore
create mode 100644 tools/testing/selftests/digest_cache/Makefile
create mode 100644 tools/testing/selftests/digest_cache/all_test.c
create mode 100644 tools/testing/selftests/digest_cache/common.c
create mode 100644 tools/testing/selftests/digest_cache/common.h
create mode 100644 tools/testing/selftests/digest_cache/common_user.c
create mode 100644 tools/testing/selftests/digest_cache/common_user.h
create mode 100644 tools/testing/selftests/digest_cache/config
create mode 100644 tools/testing/selftests/digest_cache/generators.c
create mode 100644 tools/testing/selftests/digest_cache/generators.h
create mode 100644 tools/testing/selftests/digest_cache/testmod/Makefile
create mode 100644 tools/testing/selftests/digest_cache/testmod/kern.c
--
2.34.1
Like fbf4dec70277 ("selftests/futex: Order calls to futex_lock_pi"),
which fixed a flake in futex_lock_pi due to racing between the parent
and child threads.
The same issue can occur in the futex_requeue test, because it expects
waiterfn to make progress to futex_wait before the parent starts to
requeue. This is mitigated by the parent sleeping for WAKE_WAIT_US, but
it still fails occasionally. This can be reproduced by adding a sleep in
the waiterfn before futex_wait:
TAP version 13
1..2
not ok 1 futex_requeue simple returned: 0
not ok 2 futex_requeue simple returned: 0
not ok 3 futex_requeue many returned: 0
not ok 4 futex_requeue many returned: 0
Instead, replace the sleep with barriers to make the sequencing
explicit.
Fixes: 7cb5dd8e2c8c ("selftests: futex: Add futex compare requeue test")
Signed-off-by: Edward Liaw <edliaw(a)google.com>
---
.../selftests/futex/functional/futex_requeue.c | 12 +++++++++---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/tools/testing/selftests/futex/functional/futex_requeue.c b/tools/testing/selftests/futex/functional/futex_requeue.c
index 51485be6eb2f..8f7d3e8bf32a 100644
--- a/tools/testing/selftests/futex/functional/futex_requeue.c
+++ b/tools/testing/selftests/futex/functional/futex_requeue.c
@@ -12,9 +12,9 @@
#define TEST_NAME "futex-requeue"
#define timeout_ns 30000000
-#define WAKE_WAIT_US 10000
volatile futex_t *f1;
+static pthread_barrier_t barrier;
void usage(char *prog)
{
@@ -32,6 +32,8 @@ void *waiterfn(void *arg)
to.tv_sec = 0;
to.tv_nsec = timeout_ns;
+ pthread_barrier_wait(&barrier);
+
if (futex_wait(f1, *f1, &to, 0))
printf("waiter failed errno %d\n", errno);
@@ -70,13 +72,15 @@ int main(int argc, char *argv[])
ksft_print_msg("%s: Test futex_requeue\n",
basename(argv[0]));
+ pthread_barrier_init(&barrier, NULL, 2);
/*
* Requeue a waiter from f1 to f2, and wake f2.
*/
if (pthread_create(&waiter[0], NULL, waiterfn, NULL))
error("pthread_create failed\n", errno);
- usleep(WAKE_WAIT_US);
+ pthread_barrier_wait(&barrier);
+ pthread_barrier_destroy(&barrier);
info("Requeuing 1 futex from f1 to f2\n");
res = futex_cmp_requeue(f1, 0, &f2, 0, 1, 0);
@@ -99,6 +103,7 @@ int main(int argc, char *argv[])
ksft_test_result_pass("futex_requeue simple succeeds\n");
}
+ pthread_barrier_init(&barrier, NULL, 11);
/*
* Create 10 waiters at f1. At futex_requeue, wake 3 and requeue 7.
@@ -109,7 +114,8 @@ int main(int argc, char *argv[])
error("pthread_create failed\n", errno);
}
- usleep(WAKE_WAIT_US);
+ pthread_barrier_wait(&barrier);
+ pthread_barrier_destroy(&barrier);
info("Waking 3 futexes at f1 and requeuing 7 futexes from f1 to f2\n");
res = futex_cmp_requeue(f1, 0, &f2, 3, 7, 0);
--
2.45.2.627.g7a2c4fd464-goog
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v6:
- update ASSERT strings in patch 4 as Eduard suggested. (thanks)
v5:
- update patch 1, add getsockopt(SO_PROTOCOL) in connect_to_fd() to
fix errors reported by CI.
v4:
- fix errors reported by CI.
v3:
- rename start_client to client_socket
- Use connect_to_addr in connect_to_fd_opt
v2:
- update patch 2, extract a new helper start_client.
- drop patch 3, keep must_fail in network_helper_opts.
Drop type and noconnect from network_helper_opts. And use start_server_str
in mptcp and test_tcp_check_syncookie_user.
Patches 1-4 address Martin's comments in the previous series.
Geliang Tang (6):
selftests/bpf: Drop type from network_helper_opts
selftests/bpf: Use connect_to_addr in connect_to_fd_opt
selftests/bpf: Add client_socket helper
selftests/bpf: Drop noconnect from network_helper_opts
selftests/bpf: Use start_server_str in mptcp
selftests/bpf: Use start_server_str in test_tcp_check_syncookie_user
tools/testing/selftests/bpf/network_helpers.c | 100 +++++++++---------
tools/testing/selftests/bpf/network_helpers.h | 6 +-
.../selftests/bpf/prog_tests/bpf_tcp_ca.c | 2 +-
.../selftests/bpf/prog_tests/cgroup_v1v2.c | 4 +-
.../bpf/prog_tests/ip_check_defrag.c | 14 ++-
.../testing/selftests/bpf/prog_tests/mptcp.c | 7 +-
.../bpf/test_tcp_check_syncookie_user.c | 29 +----
7 files changed, 64 insertions(+), 98 deletions(-)
--
2.43.0
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v5:
- update patch 1, add getsockopt(SO_PROTOCOL) in connect_to_fd() to
fix errors reported by CI.
v4:
- fix errors reported by CI.
v3:
- rename start_client to client_socket
- Use connect_to_addr in connect_to_fd_opt
v2:
- update patch 2, extract a new helper start_client.
- drop patch 3, keep must_fail in network_helper_opts.
Drop type and noconnect from network_helper_opts. And use start_server_str
in mptcp and test_tcp_check_syncookie_user.
Patches 1-4 address Martin's comments in the previous series.
Geliang Tang (6):
selftests/bpf: Drop type from network_helper_opts
selftests/bpf: Use connect_to_addr in connect_to_fd_opt
selftests/bpf: Add client_socket helper
selftests/bpf: Drop noconnect from network_helper_opts
selftests/bpf: Use start_server_str in mptcp
selftests/bpf: Use start_server_str in test_tcp_check_syncookie_user
tools/testing/selftests/bpf/network_helpers.c | 100 +++++++++---------
tools/testing/selftests/bpf/network_helpers.h | 6 +-
.../selftests/bpf/prog_tests/bpf_tcp_ca.c | 2 +-
.../selftests/bpf/prog_tests/cgroup_v1v2.c | 4 +-
.../bpf/prog_tests/ip_check_defrag.c | 10 +-
.../testing/selftests/bpf/prog_tests/mptcp.c | 7 +-
.../bpf/test_tcp_check_syncookie_user.c | 29 +----
7 files changed, 62 insertions(+), 96 deletions(-)
--
2.43.0
The test is inspired by the pmu_event_filter_test which implemented by x86. On
the arm64 platform, there is the same ability to set the pmu_event_filter
through the KVM_ARM_VCPU_PMU_V3_FILTER attribute. So add the test for arm64.
The series first create the helper function which can be used
for the vpmu related tests. Then, it implement the test.
Changelog:
----------
v9->v10:
- Remove the first_filter checking in the prepare_expected_pmce function.
- Add a new macro EVENT_[ALLOW|DENY] to make the definition of filter more
readable.
- Some small improvements.
v8->v9:
- Rebased to latest kvm-arm/next.
v7->v8:
- Rebased to kvm-arm/next.
- Deleted the GIC layout related staff.
- Fixed the checking logic in the kvm_pmu_support_events.
v6->v7:
- Rebased to v6.9-rc3.
v5->v6:
- Rebased to v6.9-rc1.
- Collect RB.
- Add multiple filter test.
v4->v5:
- Rebased to v6.8-rc6.
- Refactor the helper function, make it fine-grained and easy to be used.
- Namimg improvements.
- Use the kvm_device_attr_set() helper.
- Make the test descriptor array readable and clean.
- Delete the patch which moves the pmu related helper to vpmu.h.
- Remove the kvm_supports_pmu_event_filter() function since nobody will run
this on a old kernel.
v3->v4:
- Rebased to the v6.8-rc2.
v2->v3:
- Check the pmceid in guest code instead of pmu event count since different
hardware may have different event count result, check pmceid makes it stable
on different platform. [Eric]
- Some typo fixed and commit message improved.
v1->v2:
- Improve the commit message. [Eric]
- Fix the bug in [enable|disable]_counter. [Raghavendra & Marc]
- Add the check if kvm has attr KVM_ARM_VCPU_PMU_V3_FILTER.
- Add if host pmu support the test event throught pmceid0.
- Split the test_invalid_filter() to another patch. [Eric]
Shaoqin Huang (3):
KVM: selftests: aarch64: Add helper function for the vpmu vcpu
creation
KVM: selftests: aarch64: Introduce pmu_event_filter_test
KVM: selftests: aarch64: Add invalid filter test in
pmu_event_filter_test
tools/testing/selftests/kvm/Makefile | 1 +
.../kvm/aarch64/pmu_event_filter_test.c | 352 ++++++++++++++++++
.../kvm/aarch64/vpmu_counter_access.c | 32 +-
.../selftests/kvm/include/aarch64/vpmu.h | 28 ++
4 files changed, 387 insertions(+), 26 deletions(-)
create mode 100644 tools/testing/selftests/kvm/aarch64/pmu_event_filter_test.c
create mode 100644 tools/testing/selftests/kvm/include/aarch64/vpmu.h
--
2.40.1
The MBM (Memory Bandwidth Monitoring) and MBA (Memory Bandwidth Allocation)
features are not enabled for AMD systems. The reason was lack of perf
counters to compare the resctrl test results.
Starting with the commit
25e56847821f ("perf/x86/amd/uncore: Add memory controller support"), AMD
now supports the UMC (Unified Memory Controller) perf events. These events
can be used to compare the test results.
This series adds the support to detect the UMC events and enable MBM/MBA
tests for AMD systems.
v3:
Note: Based the series on top of latest kselftests/master
1613e604df0cd359cf2a7fbd9be7a0bcfacfabd0 (tag: v6.10-rc1).
Also applied the patches from the series
https://lore.kernel.org/lkml/20240531131142.1716-1-ilpo.jarvinen@linux.inte…
Separated the fix patch.
Renamed the imc to just mc to make it generic.
Changed the search string "uncore_imc_" and "amd_umc_"
Changes related rebase to latest kselftest tree.
v2: Changes.
a. Rebased on top of tip/master (Apr 25, 2024)
b. Addressed Ilpo comments except the one about close call.
It seems more clear to keep READ and WRITE separate.
https://lore.kernel.org/lkml/8e4badb7-6cc5-61f1-e041-d902209a90d5@linux.int…
c. Used ksft_perror call when applicable.
d. Added vendor check for non contiguous CBM check.
v1: https://lore.kernel.org/lkml/cover.1708637563.git.babu.moger@amd.com/
Babu Moger (4):
selftests/resctrl: Rename variables and functions to generic names
selftests/resctrl: Pass sysfs controller name of the vendor
selftests/resctrl: Add support for MBM and MBA tests on AMD
selftests/resctrl: Enable MBA/MBA tests on AMD
tools/testing/selftests/resctrl/mba_test.c | 25 +-
tools/testing/selftests/resctrl/mbm_test.c | 23 +-
tools/testing/selftests/resctrl/resctrl.h | 2 +-
tools/testing/selftests/resctrl/resctrl_val.c | 305 ++++++++++--------
tools/testing/selftests/resctrl/resctrlfs.c | 2 +-
5 files changed, 191 insertions(+), 166 deletions(-)
--
2.34.1
Correctable memory errors are very common on servers with large
amount of memory, and are corrected by ECC, but with two
pain points to users:
1. Correction usually happens on the fly and adds latency overhead
2. Not-fully-proved theory states excessive correctable memory
errors can develop into uncorrectable memory error.
Soft offline is kernel's additional solution for memory pages
having (excessive) corrected memory errors. Impacted page is migrated
to healthy page if it is in use, then the original page is discarded
for any future use.
The actual policy on whether (and when) to soft offline should be
maintained by userspace, especially in case of an 1G HugeTLB page.
Soft-offline dissolves the HugeTLB page, either in-use or free, into
chunks of 4K pages, reducing HugeTLB pool capacity by 1 hugepage.
If userspace has not acknowledged such behavior, it may be surprised
when later mmap hugepages MAP_FAILED due to lack of hugepages.
In case of a transparent hugepage, it will be split into 4K pages
as well; userspace will stop enjoying the transparent performance.
In addition, discarding the entire 1G HugeTLB page only because of
corrected memory errors sounds very costly and kernel better not
doing under the hood. But today there are at least 2 such cases:
1. GHES driver sees both GHES_SEV_CORRECTED and
CPER_SEC_ERROR_THRESHOLD_EXCEEDED after parsing CPER.
2. RAS Correctable Errors Collector counts correctable errors per
PFN and when the counter for a PFN reaches threshold
In both cases, userspace has no control of the soft offline performed
by kernel's memory failure recovery.
This patch series give userspace the control of softofflining any page:
kernel only soft offlines raw page / transparent hugepage / HugeTLB
hugepage if userspace has agreed to. The interface to userspace is a
new sysctl called enable_soft_offline under /proc/sys/vm. By default
enable_soft_line is 1 to preserve existing behavior in kernel.
Changelog
v2 => v3:
* incorporate feedbacks from Miaohe Lin <linmiaohe(a)huawei.come>,
Lance Yang <ioworker0(a)gmail.com>, Oscar Salvador <osalvador(a)suse.de>,
and David Rientjes <rientjes(a)google.com>.
* release potential refcount if enable_soft_offline is 0.
* soft_offline_page() returns EOPNOTSUPP if enable_soft_offline is 0.
* refactor hugetlb-soft-offline.c, for example, introduce
test_soft_offline_common to reduce repeated code.
* rewrite enable_soft_offline's documentation, adds more details about
the cost of soft-offline for transparent and hugetlb hugepages, and
components that are impacted when enable_soft_offline becomes 0.
* fix typos in commit messages.
* v3 is still based on commit 83a7eefedc9b ("Linux 6.10-rc3").
v1 => v2:
* incorporate feedbacks from both Miaohe Lin <linmiaohe(a)huawei.com> and
Jane Chu <jane.chu(a)oracle.com>.
* make the switch to control all pages, instead of HugeTLB specific.
* change the API from
/sys/kernel/mm/hugepages/hugepages-${size}kB/softoffline_corrected_errors
to /proc/sys/vm/enable_soft_offline.
* minor update to test code.
* update documentation of the user control API.
* v2 is based on commit 83a7eefedc9b ("Linux 6.10-rc3").
Jiaqi Yan (3):
mm/memory-failure: userspace controls soft-offlining pages
selftest/mm: test enable_soft_offline behaviors
docs: mm: add enable_soft_offline sysctl
Documentation/admin-guide/sysctl/vm.rst | 33 +++
mm/memory-failure.c | 22 +-
tools/testing/selftests/mm/.gitignore | 1 +
tools/testing/selftests/mm/Makefile | 1 +
.../selftests/mm/hugetlb-soft-offline.c | 229 ++++++++++++++++++
tools/testing/selftests/mm/run_vmtests.sh | 4 +
6 files changed, 288 insertions(+), 2 deletions(-)
create mode 100644 tools/testing/selftests/mm/hugetlb-soft-offline.c
--
2.45.2.627.g7a2c4fd464-goog
Filter out nodes that have one of its ancestors disabled as they aren't
expected to probe.
This removes the following false-positive failures on the
sc7180-trogdor-lazor-limozeen-nots-r5 platform:
/soc@0/geniqup@8c0000/i2c@894000/proximity@28
/soc@0/geniqup@ac0000/spi@a90000/ec@0
/soc@0/remoteproc@62400000/glink-edge/apr
/soc@0/remoteproc@62400000/glink-edge/apr/service@3
/soc@0/remoteproc@62400000/glink-edge/apr/service@4
/soc@0/remoteproc@62400000/glink-edge/apr/service@4/clock-controller
/soc@0/remoteproc@62400000/glink-edge/apr/service@4/dais
/soc@0/remoteproc@62400000/glink-edge/apr/service@7
/soc@0/remoteproc@62400000/glink-edge/apr/service@7/dais
/soc@0/remoteproc@62400000/glink-edge/apr/service@8
/soc@0/remoteproc@62400000/glink-edge/apr/service@8/routing
/soc@0/remoteproc@62400000/glink-edge/fastrpc
/soc@0/remoteproc@62400000/glink-edge/fastrpc/compute-cb@3
/soc@0/remoteproc@62400000/glink-edge/fastrpc/compute-cb@4
/soc@0/remoteproc@62400000/glink-edge/fastrpc/compute-cb@5
/soc@0/spmi@c440000/pmic@0/pon@800/pwrkey
Fixes: 14571ab1ad21 ("kselftest: Add new test for detecting unprobed Devicetree devices")
Signed-off-by: Nícolas F. R. A. Prado <nfraprado(a)collabora.com>
---
tools/testing/selftests/dt/test_unprobed_devices.sh | 15 ++++++++++++++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/dt/test_unprobed_devices.sh b/tools/testing/selftests/dt/test_unprobed_devices.sh
index 2d7e70c5ad2d..5e3f42ef249e 100755
--- a/tools/testing/selftests/dt/test_unprobed_devices.sh
+++ b/tools/testing/selftests/dt/test_unprobed_devices.sh
@@ -34,8 +34,21 @@ nodes_compatible=$(
# Check if node is available
if [[ -e "${node}"/status ]]; then
status=$(tr -d '\000' < "${node}"/status)
- [[ "${status}" != "okay" && "${status}" != "ok" ]] && continue
+ if [[ "${status}" != "okay" && "${status}" != "ok" ]]; then
+ if [ -n "${disabled_nodes_regex}" ]; then
+ disabled_nodes_regex="${disabled_nodes_regex}|${node}"
+ else
+ disabled_nodes_regex="${node}"
+ fi
+ continue
+ fi
fi
+
+ # Ignore this node if one of its ancestors was disabled
+ if [ -n "${disabled_nodes_regex}" ]; then
+ echo "${node}" | grep -q -E "${disabled_nodes_regex}" && continue
+ fi
+
echo "${node}" | sed -e 's|\/proc\/device-tree||'
done | sort
)
---
base-commit: 6906a84c482f098d31486df8dc98cead21cce2d0
change-id: 20240619-dt-kselftest-parent-disabled-2282a7223d26
Best regards,
--
Nícolas F. R. A. Prado <nfraprado(a)collabora.com>
** Background **
Currently, OVS supports several packet sampling mechanisms (sFlow,
per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
userspace action that needs to be handled by ovs-vswitchd's handler
threads only to be forwarded to some third party application that
will somehow process the sample and provide observability on the
datapath.
A particularly interesting use-case is controller-driven
per-flow IPFIX sampling where the OpenFlow controller can add metadata
to samples (via two 32bit integers) and this metadata is then available
to the sample-collecting system for correlation.
** Problem **
The fact that sampled traffic share netlink sockets and handler thread
time with upcalls, apart from being a performance bottleneck in the
sample extraction itself, can severely compromise the datapath,
yielding this solution unfit for highly loaded production systems.
Users are left with little options other than guessing what sampling
rate will be OK for their traffic pattern and system load and dealing
with the lost accuracy.
Looking at available infrastructure, an obvious candidated would be
to use psample. However, it's current state does not help with the
use-case at stake because sampled packets do not contain user-defined
metadata.
** Proposal **
This series is an attempt to fix this situation by extending the
existing psample infrastructure to carry a variable length
user-defined cookie.
The main existing user of psample is tc's act_sample. It is also
extended to forward the action's cookie to psample.
Finally, a new OVS action (OVS_SAMPLE_ATTR_EMIT_SAMPLE) is created.
It accepts a group and an optional cookie and uses psample to
multicast the packet and the metadata.
--
v2 -> v3:
- Addressed comments from Simon, Aaron and Ilya.
- Dropped probability propagation in nested sample actions.
- Dropped patch v2's 7/9 in favor of a userspace implementation and
consume skb if emit_sample is the last action, same as we do with
userspace.
- Split ovs-dpctl.py features in independent patches.
v1 -> v2:
- Create a new action ("emit_sample") rather than reuse existing
"sample" one.
- Add probability semantics to psample's sampling rate.
- Store sampling probability in skb's cb area and use it in emit_sample.
- Test combining "emit_sample" with "trunc"
- Drop group_id filtering and tracepoint in psample.
rfc_v2 -> v1:
- Accomodate Ilya's comments.
- Split OVS's attribute in two attributes and simplify internal
handling of psample arguments.
- Extend psample and tc with a user-defined cookie.
- Add a tracepoint to psample to facilitate troubleshooting.
rfc_v1 -> rfc_v2:
- Use psample instead of a new OVS-only multicast group.
- Extend psample and tc with a user-defined cookie.
Adrian Moreno (10):
net: psample: add user cookie
net: sched: act_sample: add action cookie to sample
net: psample: skip packet copy if no listeners
net: psample: allow using rate as probability
net: openvswitch: add emit_sample action
net: openvswitch: store sampling probability in cb.
selftests: openvswitch: add emit_sample action
selftests: openvswitch: add userspace parsing
selftests: openvswitch: parse trunc action
selftests: openvswitch: add emit_sample test
Documentation/netlink/specs/ovs_flow.yaml | 17 ++
include/net/psample.h | 5 +-
include/uapi/linux/openvswitch.h | 30 +-
include/uapi/linux/psample.h | 11 +-
include/uapi/linux/tc_act/tc_sample.h | 1 +
net/openvswitch/Kconfig | 1 +
net/openvswitch/actions.c | 63 +++-
net/openvswitch/datapath.h | 3 +
net/openvswitch/flow_netlink.c | 33 ++-
net/openvswitch/vport.c | 1 +
net/psample/psample.c | 16 +-
net/sched/act_sample.c | 12 +
.../selftests/net/openvswitch/openvswitch.sh | 110 ++++++-
.../selftests/net/openvswitch/ovs-dpctl.py | 272 +++++++++++++++++-
14 files changed, 559 insertions(+), 16 deletions(-)
--
2.45.1
Hello,
KernelCI is hosting a bi-weekly call on Thursday to discuss improvements
to existing upstream tests, the development of new tests to increase
kernel testing coverage, and the enablement of these tests in KernelCI.
In recent months, we at Collabora have focused on various kernel areas,
assessing the tests already available upstream and contributing patches
to make them easily runnable in CIs.
Below is a list of the tests we've been working on and their latest
status updates, as discussed in the last meeting held on 2024-06-13:
*ACPI probe kselftest*
- Proposing new kselftest to detect unprobed devices on ACPI platforms:
https://lore.kernel.org/all/20240308144933.337107-1-laura.nao@collabora.com
- Regression on acpi_fan driver detected and fixed upstream after
preliminary testing in KernelCI. Sent follow-up on RFCv2.
*USB/PCI devices kselftest*
- Upstream test to detect unprobed devices on discoverable buses:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?…
- Preparing to enable the test in KernelCI on MediaTek Chromebooks in the
Collabora LAVA lab - see: https://github.com/kernelci/kernelci-pipeline/issues/579
*Watchdog kselftest*
- Adding KTAP output and limiting the ping loop to run the test
individually in CIs: https://lore.kernel.org/all/20240506111359.224579-1-laura.nao@collabora.com
- Received feedback on series, discussion ongoing
*Watchdog reset test*
- Implementing out-of-tree test in KernelCI to detect device reset after
watchdog timeout expiration:
https://github.com/kernelci/kernelci-pipeline/issues/608
- Enabled test in KernelCI for all ARM64 and AMD Chromebooks, currently
monitoring the results
*Error log test*
- Proposing new kselftest to report device log errors:
https://lore.kernel.org/all/20240423-dev-err-log-selftest-v1-0-690c1741d68b…
- Standardizing error log format suggested for easier parsing, including
for kernel panic/oops/warning messages.
*Suspend/resume in cpufreq kselftest*
- Enabling suspend/resume test within the cpufreq kselftest in KernelCI
- Adding parameter support to kselftest script in KernelCI:
https://github.com/Linaro/test-definitions/pull/511
- Planning to add support for rtcwake in the test, to enable automated
resume
- Considering measuring latency for suspend and resume processes
*TAP conformance in kselftests*
- Ongoing improvements to KVM selftests, discussion ongoing upstream
https://lore.kernel.org/all/20240422170842.2073979-1-usama.anjum@collabora.…https://lore.kernel.org/all/20240612104500.425012-1-usama.anjum@collabora.c…
- Enabling more kselftests in KernelCI (dmabuf-heaps, exec, and iommu):
https://github.com/kernelci/kernelci-pipeline/pull/640/
*Boot time test*
- Investigating possibility of adding new test upstream to measure the
kernel boot time and detect regressions
- Need to investigate available interfaces for reporting boot/probe times
- Multiple measurement points may be necessary, need to establish a clear
definition of "boot" first
- Influenced by defconfig; testing with various kernel configurations
recommended
*Possible improvements to kselftest documentation*
- Document tests excluded from default kselftest run and parameter
support
*Test configuration reporting in KernelCI*
- Discussing reporting more details on the test configuration used:
https://github.com/kernelci/kernelci-pipeline/issues/636
Please reply to this thread if you'd like to join the call or discuss
any of the topics further. We look forward to collaborating with the
community to improve upstream tests and expand coverage to more areas
of interest within the kernel.
Best regards,
Laura Nao
Netlink flags, although they don't have payload at the netlink level,
are represented as having "True" as value in pyroute2.
Without it, trying to add a flow with a flag-type action (e.g: pop_vlan)
fails with the following traceback:
Traceback (most recent call last):
File "[...]/ovs-dpctl.py", line 2498, in <module>
sys.exit(main(sys.argv))
^^^^^^^^^^^^^^
File "[...]/ovs-dpctl.py", line 2487, in main
ovsflow.add_flow(rep["dpifindex"], flow)
File "[...]/ovs-dpctl.py", line 2136, in add_flow
reply = self.nlm_request(
^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/nlsocket.py", line 822, in nlm_request
return tuple(self._genlm_request(*argv, **kwarg))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/generic/__init__.py", line 126, in
nlm_request
return tuple(super().nlm_request(*argv, **kwarg))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/nlsocket.py", line 1124, in nlm_request
self.put(msg, msg_type, msg_flags, msg_seq=msg_seq)
File "[...]/pyroute2/netlink/nlsocket.py", line 389, in put
self.sendto_gate(msg, addr)
File "[...]/pyroute2/netlink/nlsocket.py", line 1056, in sendto_gate
msg.encode()
File "[...]/pyroute2/netlink/__init__.py", line 1245, in encode
offset = self.encode_nlas(offset)
^^^^^^^^^^^^^^^^^^^^^^^^
File "[...]/pyroute2/netlink/__init__.py", line 1560, in encode_nlas
nla_instance.setvalue(cell[1])
File "[...]/pyroute2/netlink/__init__.py", line 1265, in setvalue
nlv.setvalue(nla_tuple[1])
~~~~~~~~~^^^
IndexError: list index out of range
Signed-off-by: Adrian Moreno <amorenoz(a)redhat.com>
---
tools/testing/selftests/net/openvswitch/ovs-dpctl.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
index 1dd057afd3fb..9f8dec2f6539 100644
--- a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
+++ b/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
@@ -531,7 +531,7 @@ class ovsactions(nla):
for flat_act in parse_flat_map:
if parse_starts_block(actstr, flat_act[0], False):
actstr = actstr[len(flat_act[0]):]
- self["attrs"].append([flat_act[1]])
+ self["attrs"].append([flat_act[1], True])
actstr = actstr[strspn(actstr, ", ") :]
parsed = True
--
2.45.1
openvswitch.sh makes use of substitutions of the form ${ns:0:1}, to
obtain the first character of $ns. Empirically, this is works with bash
but not dash. When run with dash these evaluate to an empty string and
printing an error to stdout.
# dash -c 'ns=client; echo "${ns:0:1}"' 2>error
# cat error
dash: 1: Bad substitution
# bash -c 'ns=client; echo "${ns:0:1}"' 2>error
c
# cat error
This leads to tests that neither pass nor fail.
F.e.
TEST: arp_ping [START]
adding sandbox 'test_arp_ping'
Adding DP/Bridge IF: sbx:test_arp_ping dp:arpping {, , }
create namespaces
./openvswitch.sh: 282: eval: Bad substitution
TEST: ct_connect_v4 [START]
adding sandbox 'test_ct_connect_v4'
Adding DP/Bridge IF: sbx:test_ct_connect_v4 dp:ct4 {, , }
./openvswitch.sh: 322: eval: Bad substitution
create namespaces
Resolve this by making openvswitch.sh a bash script.
Fixes: 918423fda910 ("selftests: openvswitch: add an initial flow programming case")
Signed-off-by: Simon Horman <horms(a)kernel.org>
---
tools/testing/selftests/net/openvswitch/openvswitch.sh | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/openvswitch/openvswitch.sh b/tools/testing/selftests/net/openvswitch/openvswitch.sh
index 5cae53543849..15bca0708717 100755
--- a/tools/testing/selftests/net/openvswitch/openvswitch.sh
+++ b/tools/testing/selftests/net/openvswitch/openvswitch.sh
@@ -1,4 +1,4 @@
-#!/bin/sh
+#!/bin/bash
# SPDX-License-Identifier: GPL-2.0
#
# OVS kernel module self tests
Although "TAP" word is being used already in documentation, but it hasn't
been defined in informative way for developers that how to write TAP
conformant tests and what are the benefits. Write a short brief about it.
Signed-off-by: Muhammad Usama Anjum <usama.anjum(a)collabora.com>
---
Changes since v1:
- Update documentation by modifying and removing sentences
---
Documentation/dev-tools/kselftest.rst | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/Documentation/dev-tools/kselftest.rst b/Documentation/dev-tools/kselftest.rst
index dcf634e411bd9..f3766e326d1e3 100644
--- a/Documentation/dev-tools/kselftest.rst
+++ b/Documentation/dev-tools/kselftest.rst
@@ -228,6 +228,13 @@ In general, the rules for selftests are
* Don't cause the top-level "make run_tests" to fail if your feature is
unconfigured.
+ * The output of tests must conform to the TAP standard to ensure high
+ testing quality and to capture failures/errors with specific details.
+ The kselftest.h and kselftest_harness.h headers provide wrappers for
+ outputting test results. These wrappers should be used for pass,
+ fail, exit, and skip messages. CI systems can easily parse TAP output
+ messages to detect test results.
+
Contributing new tests (details)
================================
--
2.39.2
Changes since v2:
a) After some disussion with David Hildenbrand, simplified the "fix
missing __NR_mseal" patch down to just two lines of diff, after all.
b) Improved the "kvm, mdwe fixes to avoid requiring "make headers""
patch by taking a snapshot of the prctl.h, instead of manually creating
defines. Thanks also to David Hildenbrand for that.
c) Fixed up the final patch to use asm-generic/unistd.h.
d) Added acks.
e) Rebased onto 6.10-rc4+ (today's top of tree).
Changes since v1:
Jeff Xu, I apologize for this churn: I was forced to drop your
Reviewed-by and Tested-by tags from 2 of the 3 mseal patches, because
the __NR_mseal fix is completely different now.
a) Reworked the mseal fix to use the kernel's in-tree unistd*.h files,
instead of hacking in a __NR_mseal definition directly. (Thanks to David
Hildenbrand for pointing out that this needed to be done.)
b) Fixed the subject line of the kvm and mdwe patch.
c) Reordered the patches so as to group the mseal changes together.
d) ADDED an additional patch, 6/6, to remove various __NR_xx items and
checks from the mm selftests.
Cover letter, updated for v2:
Eventually, once the build succeeds on a sufficiently old distro, the
idea is to delete $(KHDR_INCLUDES) from the selftests/mm build, and then
after that, from selftests/lib.mk and all of the other selftest builds.
For now, this series merely achieves a clean build of selftests/mm on a
not-so-old distro: Ubuntu 23.04. In other words, after this series is
applied, it is possible to delete $(KHDR_INCLUDES) from
selftests/mm/Makefile and the build will still succeed.
1. Add tools/uapi/asm/unistd_[32|x32|64].h files, which include
definitions of __NR_mseal, and include them (indirectly) from the files
that use __NR_mseal. The new files are copied from ./usr/include/asm,
which is how we have agreed to do this sort of thing, see [1].
2. Add fs.h, similarly created: it was copied directly from a snapshot
of ./usr/include/linux/fs.h after running "make headers".
3. Add a few selected prctl.h values that the ksm and mdwe tests require.
4. Factor out some common code from mseal_test.c and seal_elf.c, into a
new mseal_helpers.h file.
5. Remove local __NR_* definitions and checks.
[1] commit e076eaca5906 ("selftests: break the dependency upon local
header files")
John Hubbard (6):
selftests/mm: mseal, self_elf: fix missing __NR_mseal
selftests/mm: mseal, self_elf: factor out test macros and other
duplicated items
selftests/mm: mseal, self_elf: rename TEST_END_CHECK to
REPORT_TEST_PASS
selftests/mm: fix vm_util.c build failures: add snapshot of fs.h
selftests/mm: kvm, mdwe fixes to avoid requiring "make headers"
selftests/mm: remove local __NR_* definitions
tools/include/uapi/linux/fs.h | 392 ++++++++++++++++++
tools/include/uapi/linux/prctl.h | 331 +++++++++++++++
tools/testing/selftests/mm/hugepage-mremap.c | 2 +-
.../selftests/mm/ksm_functional_tests.c | 8 +-
tools/testing/selftests/mm/memfd_secret.c | 14 +-
tools/testing/selftests/mm/mkdirty.c | 8 +-
tools/testing/selftests/mm/mlock2.h | 1 +
tools/testing/selftests/mm/mseal_helpers.h | 41 ++
tools/testing/selftests/mm/mseal_test.c | 143 +++----
tools/testing/selftests/mm/pagemap_ioctl.c | 2 +-
tools/testing/selftests/mm/protection_keys.c | 2 +-
tools/testing/selftests/mm/seal_elf.c | 37 +-
tools/testing/selftests/mm/uffd-common.c | 4 -
tools/testing/selftests/mm/uffd-stress.c | 16 +-
tools/testing/selftests/mm/uffd-unit-tests.c | 14 +-
15 files changed, 824 insertions(+), 191 deletions(-)
create mode 100644 tools/include/uapi/linux/fs.h
create mode 100644 tools/include/uapi/linux/prctl.h
create mode 100644 tools/testing/selftests/mm/mseal_helpers.h
base-commit: 14d7c92f8df9c0964ae6f8b813c1b3ac38120825
--
2.45.2
Hi Linus,
Please pull this kselftest fixes update for Linux 6.10-rc5.
This kselftest fixes update consists of 4 fixes to the following
build warnings:
- filesystems: warn_unused_result warnings
- seccomp: format-zero-length warnings
- fchmodat2: clang build warnings due to-static-libasan
- openat2: clang build warnings due to static-libasan, LOCAL_HDRS
diff for pull request is attached.
thanks,
-- Shuah
----------------------------------------------------------------
The following changes since commit 4bf15b1c657d22d1d70173e43264e4606dfe75ff:
selftests/futex: don't pass a const char* to asprintf(3) (2024-05-31 14:37:10 -0600)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest tags/linux_kselftest-fixes-6.10-rc5
for you to fetch changes up to ed3994ac847e0d6605f248e7f6776b1d4f445f4b:
selftests/fchmodat2: fix clang build failure due to -static-libasan (2024-06-11 15:05:05 -0600)
----------------------------------------------------------------
linux_kselftest-fixes-6.10-rc5
This kselftest fixes update consists of 4 fixes to the following
build warnings:
- filesystems: warn_unused_result warnings
- seccomp: format-zero-length warnings
- fchmodat2: clang build warnings due to-static-libasan
- openat2: clang build warnings due to static-libasan, LOCAL_HDRS
----------------------------------------------------------------
Amer Al Shanawany (2):
selftests: filesystems: fix warn_unused_result build warnings
selftests: seccomp: fix format-zero-length warnings
John Hubbard (2):
selftests/openat2: fix clang build failures: -static-libasan, LOCAL_HDRS
selftests/fchmodat2: fix clang build failure due to -static-libasan
tools/testing/selftests/fchmodat2/Makefile | 11 ++++++++++-
.../selftests/filesystems/statmount/statmount_test.c | 12 ++++++++++--
tools/testing/selftests/openat2/Makefile | 14 ++++++++++++--
tools/testing/selftests/seccomp/seccomp_benchmark.c | 6 +++---
4 files changed, 35 insertions(+), 8 deletions(-)
----------------------------------------------------------------
This patchset enables both detecting as well as dumping compilable
prototypes for kfuncs.
The first commit instructs pahole to DECL_TAG kfuncs when available.
This requires v1.27 or newer. v1.27 is nearing release at time of
writing. Following this, users will be able to look at BTF inside
vmlinux (or modules) and check if the kfunc they want is available.
The final commit teaches bpftool how to dump kfunc prototypes. This
is done for developer convenience.
The rest of the commits are fixups to enable selftests to use the
newly dumped kfunc prototypes. With these, selftests will regularly
exercise the newly added codepaths.
=== Changelog ===
From v3:
* Teach selftests to use dumped prototypes
From v2:
* Update Makefile.btf with pahole flag
* More error checking
* Output formatting changes
* Drop already-merged commit
From v1:
* Add __weak annotation
* Use btf_dump for kfunc prototypes
* Update kernel bpf_rdonly_cast() signature
Daniel Xu (12):
kbuild: bpf: Tell pahole to DECL_TAG kfuncs
bpf: selftests: Fix bpf_iter_task_vma_new() prototype
bpf: selftests: Fix fentry test kfunc prototypes
bpf: selftests: Fix bpf_cpumask_first_zero() kfunc prototype
bpf: selftests: Fix bpf_map_sum_elem_count() kfunc prototype
bpf: selftests: Fix bpf_session_cookie() kfunc prototype
bpf: selftests: Namespace struct_opt callbacks in bpf_dctcp
bpf: verifier: Relax caller requirements for kfunc projection type
args
bpf: treewide: Align kfunc signatures to prog point-of-view
bpf: selftests: nf: Opt out of using generated kfunc prototypes
bpf: selftests: xfrm: Opt out of using generated kfunc prototypes
bpftool: Support dumping kfunc prototypes from BTF
fs/verity/measure.c | 3 +-
include/linux/bpf.h | 8 +--
kernel/bpf/crypto.c | 24 +++++---
kernel/bpf/helpers.c | 39 +++++++++----
kernel/bpf/verifier.c | 12 +++-
kernel/trace/bpf_trace.c | 9 ++-
net/core/filter.c | 32 +++++++----
scripts/Makefile.btf | 2 +-
tools/bpf/bpftool/btf.c | 55 +++++++++++++++++++
.../testing/selftests/bpf/bpf_experimental.h | 2 +-
tools/testing/selftests/bpf/bpf_kfuncs.h | 2 +-
tools/testing/selftests/bpf/progs/bpf_dctcp.c | 36 ++++++------
.../selftests/bpf/progs/get_func_ip_test.c | 14 ++---
.../selftests/bpf/progs/ip_check_defrag.c | 10 ++--
.../bpf/progs/kprobe_multi_session_cookie.c | 2 +-
.../selftests/bpf/progs/map_percpu_stats.c | 2 +-
.../selftests/bpf/progs/nested_trust_common.h | 2 +-
.../testing/selftests/bpf/progs/test_bpf_nf.c | 1 +
.../selftests/bpf/progs/test_bpf_nf_fail.c | 1 +
.../bpf/progs/verifier_netfilter_ctx.c | 6 +-
.../selftests/bpf/progs/xdp_synproxy_kern.c | 1 +
tools/testing/selftests/bpf/progs/xfrm_info.c | 1 +
22 files changed, 183 insertions(+), 81 deletions(-)
--
2.44.0
Currently, if a user wants to run pmtu.sh and cover all the provided test
cases, they need to install the Open vSwitch userspace utilities. This
dependency is difficult for users as well as CI environments, because the
userspace build and setup may require lots of support and devel packages
to be installed, system setup to be correct, and things like permissions
and selinux policies to be properly configured.
The kernel selftest suite includes an ovs-dpctl.py utility which can
interact with the openvswitch module directly. This lets developers and
CI environments run without needing too many extra dependencies - just
the pyroute2 python package.
This series enhances the ovs-dpctl utility to provide support for set()
and tunnel() flow specifiers, better ipv6 handling support, and the
ability to add tunnel vports, and LWT interfaces. Finally, it modifies
the pmtu.sh script to call the ovs-dpctl.py utility rather than the
typical OVS userspace utilities.
Aaron Conole (7):
selftests: openvswitch: Support explicit tunnel port creation.
selftests: openvswitch: Refactor actions parsing.
selftests: openvswitch: Add set() and set_masked() support.
selftests: openvswitch: Add support for tunnel() key.
selftests: openvswitch: Support implicit ipv6 arguments.
selftests: net: Use the provided dpctl rather than the vswitchd for
tests.
selftests: net: add config for openvswitch
tools/testing/selftests/net/config | 5 +
.../selftests/net/openvswitch/ovs-dpctl.py | 372 +++++++++++++++---
tools/testing/selftests/net/pmtu.sh | 145 +++++--
3 files changed, 453 insertions(+), 69 deletions(-)
--
2.45.1
From: John Hubbard <jhubbard(a)nvidia.com>
[ Upstream commit cb708ab9f584f159798b60853edcf0c8b67ce295 ]
It's slightly better to set _GNU_SOURCE in the source code, but if one
must do it via the compiler invocation, then the best way to do so is
this:
$(CC) -D_GNU_SOURCE=
...because otherwise, if this form is used:
$(CC) -D_GNU_SOURCE
...then that leads the compiler to set a value, as if you had passed in:
$(CC) -D_GNU_SOURCE=1
That, in turn, leads to warnings under both gcc and clang, like this:
futex_requeue_pi.c:20: warning: "_GNU_SOURCE" redefined
Fix this by using the "-D_GNU_SOURCE=" form.
Reviewed-by: Edward Liaw <edliaw(a)google.com>
Reviewed-by: Davidlohr Bueso <dave(a)stgolabs.net>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
Signed-off-by: Shuah Khan <skhan(a)linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/futex/functional/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/futex/functional/Makefile b/tools/testing/selftests/futex/functional/Makefile
index a392d0917b4e5..994fa3468f170 100644
--- a/tools/testing/selftests/futex/functional/Makefile
+++ b/tools/testing/selftests/futex/functional/Makefile
@@ -1,6 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
INCLUDES := -I../include -I../../ $(KHDR_INCLUDES)
-CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE -pthread $(INCLUDES) $(KHDR_INCLUDES)
+CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE= -pthread $(INCLUDES) $(KHDR_INCLUDES)
LDLIBS := -lpthread -lrt
LOCAL_HDRS := \
--
2.43.0
From: John Hubbard <jhubbard(a)nvidia.com>
[ Upstream commit cb708ab9f584f159798b60853edcf0c8b67ce295 ]
It's slightly better to set _GNU_SOURCE in the source code, but if one
must do it via the compiler invocation, then the best way to do so is
this:
$(CC) -D_GNU_SOURCE=
...because otherwise, if this form is used:
$(CC) -D_GNU_SOURCE
...then that leads the compiler to set a value, as if you had passed in:
$(CC) -D_GNU_SOURCE=1
That, in turn, leads to warnings under both gcc and clang, like this:
futex_requeue_pi.c:20: warning: "_GNU_SOURCE" redefined
Fix this by using the "-D_GNU_SOURCE=" form.
Reviewed-by: Edward Liaw <edliaw(a)google.com>
Reviewed-by: Davidlohr Bueso <dave(a)stgolabs.net>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
Signed-off-by: Shuah Khan <skhan(a)linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/futex/functional/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/futex/functional/Makefile b/tools/testing/selftests/futex/functional/Makefile
index a392d0917b4e5..994fa3468f170 100644
--- a/tools/testing/selftests/futex/functional/Makefile
+++ b/tools/testing/selftests/futex/functional/Makefile
@@ -1,6 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
INCLUDES := -I../include -I../../ $(KHDR_INCLUDES)
-CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE -pthread $(INCLUDES) $(KHDR_INCLUDES)
+CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE= -pthread $(INCLUDES) $(KHDR_INCLUDES)
LDLIBS := -lpthread -lrt
LOCAL_HDRS := \
--
2.43.0
From: John Hubbard <jhubbard(a)nvidia.com>
[ Upstream commit cb708ab9f584f159798b60853edcf0c8b67ce295 ]
It's slightly better to set _GNU_SOURCE in the source code, but if one
must do it via the compiler invocation, then the best way to do so is
this:
$(CC) -D_GNU_SOURCE=
...because otherwise, if this form is used:
$(CC) -D_GNU_SOURCE
...then that leads the compiler to set a value, as if you had passed in:
$(CC) -D_GNU_SOURCE=1
That, in turn, leads to warnings under both gcc and clang, like this:
futex_requeue_pi.c:20: warning: "_GNU_SOURCE" redefined
Fix this by using the "-D_GNU_SOURCE=" form.
Reviewed-by: Edward Liaw <edliaw(a)google.com>
Reviewed-by: Davidlohr Bueso <dave(a)stgolabs.net>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
Signed-off-by: Shuah Khan <skhan(a)linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/futex/functional/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/futex/functional/Makefile b/tools/testing/selftests/futex/functional/Makefile
index a392d0917b4e5..994fa3468f170 100644
--- a/tools/testing/selftests/futex/functional/Makefile
+++ b/tools/testing/selftests/futex/functional/Makefile
@@ -1,6 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
INCLUDES := -I../include -I../../ $(KHDR_INCLUDES)
-CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE -pthread $(INCLUDES) $(KHDR_INCLUDES)
+CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE= -pthread $(INCLUDES) $(KHDR_INCLUDES)
LDLIBS := -lpthread -lrt
LOCAL_HDRS := \
--
2.43.0
From: Michael Ellerman <mpe(a)ellerman.id.au>
[ Upstream commit e8b8c5264d4ebd248f60a5cef077fe615806e7a0 ]
Fix build error on ppc64:
dev_in_maps.c: In function ‘get_file_dev_and_inode’:
dev_in_maps.c:60:59: error: format ‘%llu’ expects argument of type
‘long long unsigned int *’, but argument 7 has type ‘__u64 *’ {aka ‘long
unsigned int *’} [-Werror=format=]
By switching to unsigned long long for u64 for ppc64 builds.
Signed-off-by: Michael Ellerman <mpe(a)ellerman.id.au>
Signed-off-by: Shuah Khan <skhan(a)linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c b/tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c
index 759f86e7d263e..2862aae58b79a 100644
--- a/tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c
+++ b/tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c
@@ -1,5 +1,6 @@
// SPDX-License-Identifier: GPL-2.0
#define _GNU_SOURCE
+#define __SANE_USERSPACE_TYPES__ // Use ll64
#include <inttypes.h>
#include <unistd.h>
--
2.43.0
v10: https://patchwork.kernel.org/project/netdevbpf/list/?series=852422&state=*
====
Major Changes:
--------------
v9 was sent right before the merge window closed (sorry!). v10 is almost
a re-send of the series now that the merge window re-opened. Only
rebased to latest net-next and addressed some minor iterative comments
received on v9.
As usual, the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v10/
Detailed changelog:
-------------------
- Fixed tokens leaking in DONTNEED setsockopt (Nikolay).
- Moved net_iov_dma_addr() to devmem.c and made it a devmem specific
helpers (David).
- Rename hook alloc_pages to alloc_netmems as alloc_pages is now
preprocessor macro defined and causes a build error.
v9:
===
Major Changes:
--------------
GVE queue API has been merged. Submitting this version as non-RFC after
rebasing on top of the merged API, and dropped the out of tree queue API
I was carrying on github. Addressed the little feedback v8 has received.
Detailed changelog:
------------------
- Added new patch from David Wei to this series for
netdev_rx_queue_restart()
- Fixed sparse error.
- Removed CONFIG_ checks in netmem_is_net_iov()
- Flipped skb->readable to skb->unreadable
- Minor fixes to selftests & docs.
RFC v8:
=======
Major Changes:
--------------
- Fixed build error generated by patch-by-patch build.
- Applied docs suggestions from Randy.
RFC v7:
=======
Major Changes:
--------------
This revision largely rebases on top of net-next and addresses the feedback
RFCv6 received from folks, namely Jakub, Yunsheng, Arnd, David, & Pavel.
The series remains in RFC because the queue-API ndos defined in this
series are not yet implemented. I have a GVE implementation I carry out
of tree for my testing. A upstreamable GVE implementation is in the
works. Aside from that, in my estimation all the patches are ready for
review/merge. Please do take a look.
As usual the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v7/
Detailed changelog:
- Use admin-perm in netlink API.
- Addressed feedback from Jakub with regards to netlink API
implementation.
- Renamed devmem.c functions to something more appropriate for that
file.
- Improve the performance seen through the page_pool benchmark.
- Fix the value definition of all the SO_DEVMEM_* uapi.
- Various fixes to documentation.
Perf - page-pool benchmark:
---------------------------
Improved performance of bench_page_pool_simple.ko tests compared to v6:
https://pastebin.com/raw/v5dYRg8L
net-next base: 8 cycle fast path.
RFC v6: 10 cycle fast path.
RFC v7: 9 cycle fast path.
RFC v7 with CONFIG_DMA_SHARED_BUFFER disabled: 8 cycle fast path,
same as baseline.
Perf - Devmem TCP benchmark:
---------------------
Perf is about the same regardless of the changes in v7, namely the
removal of the static_branch_unlikely to improve the page_pool benchmark
performance:
189/200gbps bi-directional throughput with RX devmem TCP and regular TCP
TX i.e. ~95% line rate.
RFC v6:
=======
Major Changes:
--------------
This revision largely rebases on top of net-next and addresses the little
feedback RFCv5 received.
The series remains in RFC because the queue-API ndos defined in this
series are not yet implemented. I have a GVE implementation I carry out
of tree for my testing. A upstreamable GVE implementation is in the
works. Aside from that, in my estimation all the patches are ready for
review/merge. Please do take a look.
As usual the full devmem TCP changes including the full GVE driver
implementation is here:
https://github.com/mina/linux/commits/tcpdevmem-v6/
This version also comes with some performance data recorded in the cover
letter (see below changelog).
Detailed changelog:
- Rebased on top of the merged netmem_ref changes.
- Converted skb->dmabuf to skb->readable (Pavel). Pavel's original
suggestion was to remove the skb->dmabuf flag entirely, but when I
looked into it closely, I found the issue that if we remove the flag
we have to dereference the shinfo(skb) pointer to obtain the first
frag to tell whether an skb is readable or not. This can cause a
performance regression if it dirties the cache line when the
shinfo(skb) was not really needed. Instead, I converted the skb->dmabuf
flag into a generic skb->readable flag which can be re-used by io_uring
0-copy RX.
- Squashed a few locking optimizations from Eric Dumazet in the RX path
and the DEVMEM_DONTNEED setsockopt.
- Expanded the tests a bit. Added validation for invalid scenarios and
added some more coverage.
Perf - page-pool benchmark:
---------------------------
bench_page_pool_simple.ko tests with and without these changes:
https://pastebin.com/raw/ncHDwAbn
AFAIK the number that really matters in the perf tests is the
'tasklet_page_pool01_fast_path Per elem'. This one measures at about 8
cycles without the changes but there is some 1 cycle noise in some
results.
With the patches this regresses to 9 cycles with the changes but there
is 1 cycle noise occasionally running this test repeatedly.
Lastly I tried disable the static_branch_unlikely() in
netmem_is_net_iov() check. To my surprise disabling the
static_branch_unlikely() check reduces the fast path back to 8 cycles,
but the 1 cycle noise remains.
Perf - Devmem TCP benchmark:
---------------------
189/200gbps bi-directional throughput with RX devmem TCP and regular TCP
TX i.e. ~95% line rate.
Major changes in RFC v5:
========================
1. Rebased on top of 'Abstract page from net stack' series and used the
new netmem type to refer to LSB set pointers instead of re-using
struct page.
2. Downgraded this series back to RFC and called it RFC v5. This is
because this series is now dependent on 'Abstract page from net
stack'[1] and the queue API. Both are removed from the series to
reduce the patch # and those bits are fairly independent or
pre-requisite work.
3. Reworked the page_pool devmem support to use netmem and for some
more unified handling.
4. Reworked the reference counting of net_iov (renamed from
page_pool_iov) to use pp_ref_count for refcounting.
The full changes including the dependent series and GVE page pool
support is here:
https://github.com/mina/linux/commits/tcpdevmem-rfcv5/
[1] https://patchwork.kernel.org/project/netdevbpf/list/?series=810774
Major changes in v1:
====================
1. Implemented MVP queue API ndos to remove the userspace-visible
driver reset.
2. Fixed issues in the napi_pp_put_page() devmem frag unref path.
3. Removed RFC tag.
Many smaller addressed comments across all the patches (patches have
individual change log).
Full tree including the rest of the GVE driver changes:
https://github.com/mina/linux/commits/tcpdevmem-v1
Changes in RFC v3:
==================
1. Pulled in the memory-provider dependency from Jakub's RFC[1] to make the
series reviewable and mergeable.
2. Implemented multi-rx-queue binding which was a todo in v2.
3. Fix to cmsg handling.
The sticking point in RFC v2[2] was the device reset required to refill
the device rx-queues after the dmabuf bind/unbind. The solution
suggested as I understand is a subset of the per-queue management ops
Jakub suggested or similar:
https://lore.kernel.org/netdev/20230815171638.4c057dcd@kernel.org/
This is not addressed in this revision, because:
1. This point was discussed at netconf & netdev and there is openness to
using the current approach of requiring a device reset.
2. Implementing individual queue resetting seems to be difficult for my
test bed with GVE. My prototype to test this ran into issues with the
rx-queues not coming back up properly if reset individually. At the
moment I'm unsure if it's a mistake in the POC or a genuine issue in
the virtualization stack behind GVE, which currently doesn't test
individual rx-queue restart.
3. Our usecases are not bothered by requiring a device reset to refill
the buffer queues, and we'd like to support NICs that run into this
limitation with resetting individual queues.
My thought is that drivers that have trouble with per-queue configs can
use the support in this series, while drivers that support new netdev
ops to reset individual queues can automatically reset the queue as
part of the dma-buf bind/unbind.
The same approach with device resets is presented again for consideration
with other sticking points addressed.
This proposal includes the rx devmem path only proposed for merge. For a
snapshot of my entire tree which includes the GVE POC page pool support &
device memory support:
https://github.com/torvalds/linux/compare/master...mina:linux:tcpdevmem-v3
[1] https://lore.kernel.org/netdev/f8270765-a27b-6ccf-33ea-cda097168d79@redhat.…
[2] https://lore.kernel.org/netdev/CAHS8izOVJGJH5WF68OsRWFKJid1_huzzUK+hpKbLcL4…
Changes in RFC v2:
==================
The sticking point in RFC v1[1] was the dma-buf pages approach we used to
deliver the device memory to the TCP stack. RFC v2 is a proof-of-concept
that attempts to resolve this by implementing scatterlist support in the
networking stack, such that we can import the dma-buf scatterlist
directly. This is the approach proposed at a high level here[2].
Detailed changes:
1. Replaced dma-buf pages approach with importing scatterlist into the
page pool.
2. Replace the dma-buf pages centric API with a netlink API.
3. Removed the TX path implementation - there is no issue with
implementing the TX path with scatterlist approach, but leaving
out the TX path makes it easier to review.
4. Functionality is tested with this proposal, but I have not conducted
perf testing yet. I'm not sure there are regressions, but I removed
perf claims from the cover letter until they can be re-confirmed.
5. Added Signed-off-by: contributors to the implementation.
6. Fixed some bugs with the RX path since RFC v1.
Any feedback welcome, but specifically the biggest pending questions
needing feedback IMO are:
1. Feedback on the scatterlist-based approach in general.
2. Netlink API (Patch 1 & 2).
3. Approach to handle all the drivers that expect to receive pages from
the page pool (Patch 6).
[1] https://lore.kernel.org/netdev/dfe4bae7-13a0-3c5d-d671-f61b375cb0b4@gmail.c…
[2] https://lore.kernel.org/netdev/CAHS8izPm6XRS54LdCDZVd0C75tA1zHSu6jLVO8nzTLX…
==================
* TL;DR:
Device memory TCP (devmem TCP) is a proposal for transferring data to and/or
from device memory efficiently, without bouncing the data to a host memory
buffer.
* Problem:
A large amount of data transfers have device memory as the source and/or
destination. Accelerators drastically increased the volume of such transfers.
Some examples include:
- ML accelerators transferring large amounts of training data from storage into
GPU/TPU memory. In some cases ML training setup time can be as long as 50% of
TPU compute time, improving data transfer throughput & efficiency can help
improving GPU/TPU utilization.
- Distributed training, where ML accelerators, such as GPUs on different hosts,
exchange data among them.
- Distributed raw block storage applications transfer large amounts of data with
remote SSDs, much of this data does not require host processing.
Today, the majority of the Device-to-Device data transfers the network are
implemented as the following low level operations: Device-to-Host copy,
Host-to-Host network transfer, and Host-to-Device copy.
The implementation is suboptimal, especially for bulk data transfers, and can
put significant strains on system resources, such as host memory bandwidth,
PCIe bandwidth, etc. One important reason behind the current state is the
kernel’s lack of semantics to express device to network transfers.
* Proposal:
In this patch series we attempt to optimize this use case by implementing
socket APIs that enable the user to:
1. send device memory across the network directly, and
2. receive incoming network packets directly into device memory.
Packet _payloads_ go directly from the NIC to device memory for receive and from
device memory to NIC for transmit.
Packet _headers_ go to/from host memory and are processed by the TCP/IP stack
normally. The NIC _must_ support header split to achieve this.
Advantages:
- Alleviate host memory bandwidth pressure, compared to existing
network-transfer + device-copy semantics.
- Alleviate PCIe BW pressure, by limiting data transfer to the lowest level
of the PCIe tree, compared to traditional path which sends data through the
root complex.
* Patch overview:
** Part 1: netlink API
Gives user ability to bind dma-buf to an RX queue.
** Part 2: scatterlist support
Currently the standard for device memory sharing is DMABUF, which doesn't
generate struct pages. On the other hand, networking stack (skbs, drivers, and
page pool) operate on pages. We have 2 options:
1. Generate struct pages for dmabuf device memory, or,
2. Modify the networking stack to process scatterlist.
Approach #1 was attempted in RFC v1. RFC v2 implements approach #2.
** part 3: page pool support
We piggy back on page pool memory providers proposal:
https://github.com/kuba-moo/linux/tree/pp-providers
It allows the page pool to define a memory provider that provides the
page allocation and freeing. It helps abstract most of the device memory
TCP changes from the driver.
** part 4: support for unreadable skb frags
Page pool iovs are not accessible by the host; we implement changes
throughput the networking stack to correctly handle skbs with unreadable
frags.
** Part 5: recvmsg() APIs
We define user APIs for the user to send and receive device memory.
Not included with this series is the GVE devmem TCP support, just to
simplify the review. Code available here if desired:
https://github.com/mina/linux/tree/tcpdevmem
This series is built on top of net-next with Jakub's pp-providers changes
cherry-picked.
* NIC dependencies:
1. (strict) Devmem TCP require the NIC to support header split, i.e. the
capability to split incoming packets into a header + payload and to put
each into a separate buffer. Devmem TCP works by using device memory
for the packet payload, and host memory for the packet headers.
2. (optional) Devmem TCP works better with flow steering support & RSS support,
i.e. the NIC's ability to steer flows into certain rx queues. This allows the
sysadmin to enable devmem TCP on a subset of the rx queues, and steer
devmem TCP traffic onto these queues and non devmem TCP elsewhere.
The NIC I have access to with these properties is the GVE with DQO support
running in Google Cloud, but any NIC that supports these features would suffice.
I may be able to help reviewers bring up devmem TCP on their NICs.
* Testing:
The series includes a udmabuf kselftest that show a simple use case of
devmem TCP and validates the entire data path end to end without
a dependency on a specific dmabuf provider.
** Test Setup
Kernel: net-next with this series and memory provider API cherry-picked
locally.
Hardware: Google Cloud A3 VMs.
NIC: GVE with header split & RSS & flow steering support.
Cc: Pavel Begunkov <asml.silence(a)gmail.com>
Cc: David Wei <dw(a)davidwei.uk>
Cc: Jason Gunthorpe <jgg(a)ziepe.ca>
Cc: Yunsheng Lin <linyunsheng(a)huawei.com>
Cc: Shailend Chand <shailend(a)google.com>
Cc: Harshitha Ramamurthy <hramamurthy(a)google.com>
Cc: Shakeel Butt <shakeel.butt(a)linux.dev>
Cc: Jeroen de Borst <jeroendb(a)google.com>
Cc: Praveen Kaligineedi <pkaligineedi(a)google.com>
Jakub Kicinski (1):
net: page_pool: create hooks for custom page providers
Mina Almasry (13):
netdev: add netdev_rx_queue_restart()
net: netdev netlink api to bind dma-buf to a net device
netdev: support binding dma-buf to netdevice
netdev: netdevice devmem allocator
page_pool: convert to use netmem
page_pool: devmem support
memory-provider: dmabuf devmem memory provider
net: support non paged skb frags
net: add support for skbs with unreadable frags
tcp: RX path for devmem TCP
net: add SO_DEVMEM_DONTNEED setsockopt to release RX frags
net: add devmem TCP documentation
selftests: add ncdevmem, netcat for devmem TCP
Documentation/netlink/specs/netdev.yaml | 57 +++
Documentation/networking/devmem.rst | 258 +++++++++++
Documentation/networking/index.rst | 1 +
arch/alpha/include/uapi/asm/socket.h | 6 +
arch/mips/include/uapi/asm/socket.h | 6 +
arch/parisc/include/uapi/asm/socket.h | 6 +
arch/sparc/include/uapi/asm/socket.h | 6 +
include/linux/skbuff.h | 61 ++-
include/linux/skbuff_ref.h | 11 +-
include/linux/socket.h | 1 +
include/net/devmem.h | 124 ++++++
include/net/netdev_rx_queue.h | 5 +
include/net/netmem.h | 208 ++++++++-
include/net/page_pool/helpers.h | 153 +++++--
include/net/page_pool/types.h | 33 +-
include/net/sock.h | 2 +
include/net/tcp.h | 5 +-
include/trace/events/page_pool.h | 29 +-
include/uapi/asm-generic/socket.h | 6 +
include/uapi/linux/netdev.h | 19 +
include/uapi/linux/uio.h | 17 +
net/bpf/test_run.c | 5 +-
net/core/Makefile | 3 +-
net/core/datagram.c | 6 +
net/core/dev.c | 6 +-
net/core/devmem.c | 384 +++++++++++++++++
net/core/gro.c | 8 +-
net/core/netdev-genl-gen.c | 23 +
net/core/netdev-genl-gen.h | 6 +
net/core/netdev-genl.c | 103 +++++
net/core/netdev_rx_queue.c | 74 ++++
net/core/page_pool.c | 368 +++++++++-------
net/core/skbuff.c | 83 +++-
net/core/sock.c | 61 +++
net/ipv4/esp4.c | 3 +-
net/ipv4/tcp.c | 254 ++++++++++-
net/ipv4/tcp_input.c | 13 +-
net/ipv4/tcp_ipv4.c | 10 +
net/ipv4/tcp_minisocks.c | 2 +
net/ipv4/tcp_output.c | 5 +-
net/ipv6/esp6.c | 3 +-
net/packet/af_packet.c | 4 +-
tools/include/uapi/linux/netdev.h | 19 +
tools/testing/selftests/net/.gitignore | 1 +
tools/testing/selftests/net/Makefile | 5 +
tools/testing/selftests/net/ncdevmem.c | 542 ++++++++++++++++++++++++
46 files changed, 2738 insertions(+), 267 deletions(-)
create mode 100644 Documentation/networking/devmem.rst
create mode 100644 include/net/devmem.h
create mode 100644 net/core/devmem.c
create mode 100644 net/core/netdev_rx_queue.c
create mode 100644 tools/testing/selftests/net/ncdevmem.c
--
2.45.1.288.g0e0cd299f1-goog
The test is inspired by the pmu_event_filter_test which implemented by x86. On
the arm64 platform, there is the same ability to set the pmu_event_filter
through the KVM_ARM_VCPU_PMU_V3_FILTER attribute. So add the test for arm64.
The series first create the helper function which can be used
for the vpmu related tests. Then, it implement the test.
Changelog:
----------
v8->v9:
- Rebased to latest kvm-arm/next.
v7->v8:
- Rebased to kvm-arm/next.
- Deleted the GIC layout related staff.
- Fixed the checking logic in the kvm_pmu_support_events.
v6->v7:
- Rebased to v6.9-rc3.
v5->v6:
- Rebased to v6.9-rc1.
- Collect RB.
- Add multiple filter test.
v4->v5:
- Rebased to v6.8-rc6.
- Refactor the helper function, make it fine-grained and easy to be used.
- Namimg improvements.
- Use the kvm_device_attr_set() helper.
- Make the test descriptor array readable and clean.
- Delete the patch which moves the pmu related helper to vpmu.h.
- Remove the kvm_supports_pmu_event_filter() function since nobody will run
this on a old kernel.
v3->v4:
- Rebased to the v6.8-rc2.
v2->v3:
- Check the pmceid in guest code instead of pmu event count since different
hardware may have different event count result, check pmceid makes it stable
on different platform. [Eric]
- Some typo fixed and commit message improved.
v1->v2:
- Improve the commit message. [Eric]
- Fix the bug in [enable|disable]_counter. [Raghavendra & Marc]
- Add the check if kvm has attr KVM_ARM_VCPU_PMU_V3_FILTER.
- Add if host pmu support the test event throught pmceid0.
- Split the test_invalid_filter() to another patch. [Eric]
Shaoqin Huang (3):
KVM: selftests: aarch64: Add helper function for the vpmu vcpu
creation
KVM: selftests: aarch64: Introduce pmu_event_filter_test
KVM: selftests: aarch64: Add invalid filter test in
pmu_event_filter_test
tools/testing/selftests/kvm/Makefile | 1 +
.../kvm/aarch64/pmu_event_filter_test.c | 340 ++++++++++++++++++
.../kvm/aarch64/vpmu_counter_access.c | 32 +-
.../selftests/kvm/include/aarch64/vpmu.h | 28 ++
4 files changed, 375 insertions(+), 26 deletions(-)
create mode 100644 tools/testing/selftests/kvm/aarch64/pmu_event_filter_test.c
create mode 100644 tools/testing/selftests/kvm/include/aarch64/vpmu.h
--
2.40.1
Allow userspace to change the guest-visible value of the register with
some severe limitation:
- No changes to features not virtualized by KVM (MPAM_frac, RAS_frac)
Also add the selftest for it.
Changelog:
----------
v1 -> v2:
* Tackling the full register instead of single field.
* Changing the patch title and commit message.
RFCv1 -> v1:
* Fix the compilation error.
* Delete the machine specific information and make the description more
generable.
RFCv1: https://lore.kernel.org/all/20240612023553.127813-1-shahuang@redhat.com/
v1: https://lore.kernel.org/all/20240617075131.1006173-1-shahuang@redhat.com/
Shaoqin Huang (2):
KVM: arm64: Allow userspace to change ID_AA64PFR1_EL1
KVM: selftests: aarch64: Add writable test for ID_AA64PFR1_EL1
arch/arm64/kvm/sys_regs.c | 3 ++-
tools/testing/selftests/kvm/aarch64/set_id_regs.c | 9 +++++++++
2 files changed, 11 insertions(+), 1 deletion(-)
--
2.40.1
In this patch series, we try to make more register fields writable like
ID_AA64PFR1_EL1.BT since this can benifit the migration between some of the
machines which have different BT values.
Changelog:
----------
RFCv1 -> v1:
* Fix the compilation error.
* Delete the machine specific information and make the description more
generable.
RFCv1: https://lore.kernel.org/all/20240612023553.127813-1-shahuang@redhat.com/
Shaoqin Huang (2):
KVM: arm64: Allow BT field in ID_AA64PFR1_EL1 writable
KVM: selftests: aarch64: Add writable test for ID_AA64PFR1_EL1
arch/arm64/kvm/sys_regs.c | 2 +-
tools/testing/selftests/kvm/aarch64/set_id_regs.c | 6 ++++++
2 files changed, 7 insertions(+), 1 deletion(-)
--
2.40.1
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v4:
- fix errors reported by CI.
v3:
- rename start_client to client_socket
- Use connect_to_addr in connect_to_fd_opt
v2:
- update patch 2, extract a new helper start_client.
- drop patch 3, keep must_fail in network_helper_opts.
Drop type and noconnect from network_helper_opts. And use start_server_str
in mptcp and test_tcp_check_syncookie_user.
Patches 1-4 address Martin's comments in the previous series.
Geliang Tang (6):
selftests/bpf: Drop type from network_helper_opts
selftests/bpf: Use connect_to_addr in connect_to_fd_opt
selftests/bpf: Add client_socket helper
selftests/bpf: Drop noconnect from network_helper_opts
selftests/bpf: Use start_server_str in mptcp
selftests/bpf: Use start_server_str in test_tcp_check_syncookie_user
tools/testing/selftests/bpf/network_helpers.c | 94 +++++++++----------
tools/testing/selftests/bpf/network_helpers.h | 6 +-
.../selftests/bpf/prog_tests/bpf_tcp_ca.c | 2 +-
.../selftests/bpf/prog_tests/cgroup_v1v2.c | 4 +-
.../bpf/prog_tests/ip_check_defrag.c | 10 +-
.../testing/selftests/bpf/prog_tests/mptcp.c | 7 +-
.../bpf/test_tcp_check_syncookie_user.c | 29 +-----
7 files changed, 56 insertions(+), 96 deletions(-)
--
2.43.0
Currently, if a user wants to run pmtu.sh and cover all the provided test
cases, they need to install the Open vSwitch userspace utilities. This
dependency is difficult for users as well as CI environments, because the
userspace build and setup may require lots of support and devel packages
to be installed, system setup to be correct, and things like permissions
and selinux policies to be properly configured.
The kernel selftest suite includes an ovs-dpctl.py utility which can
interact with the openvswitch module directly. This lets developers and
CI environments run without needing too many extra dependencies - just
the pyroute2 python package.
This series enhances the ovs-dpctl utility to provide support for set()
and tunnel() flow specifiers, better ipv6 handling support, and the
ability to add tunnel vports, and LWT interfaces. Finally, it modifies
the pmtu.sh script to call the ovs-dpctl.py utility rather than the
typical OVS userspace utilities.
NOTE: This could also be applied as-is. I'm trying to get the vng test
working in my environment, so I submitted as RFC because I didn't
get to test with the config change in 7/7.
Aaron Conole (6):
selftests: openvswitch: Support explicit tunnel port creation.
selftests: openvswitch: Refactor actions parsing.
selftests: openvswitch: Add set() and set_masked() support.
selftests: openvswitch: Add support for tunnel() key.
selftests: openvswitch: Support implicit ipv6 arguments.
selftests: net: Use the provided dpctl rather than the vswitchd for
tests.
selftests: net: add config for openvswitch
.../selftests/net/openvswitch/ovs-dpctl.py | 370 +++++++++++++++---
tools/testing/selftests/net/config | 5 ++++
tools/testing/selftests/net/pmtu.sh | 87 +++-
3 files changed, 394 insertions(+), 68 deletions(-)
--
2.45.1
On 6/14/24 12:06 PM, Jason A. Donenfeld wrote:
> Hook up the generic vDSO implementation to the x86 vDSO data page. Since
> the existing vDSO infrastructure is heavily based on the timekeeping
> functionality, which works over arrays of bases, a new macro is
> introduced for vvars that are not arrays.
>
> The vDSO function requires a ChaCha20 implementation that does not write
> to the stack, yet can still do an entire ChaCha20 permutation, so
> provide this using SSE2, since this is userland code that must work on
> all x86-64 processors. There's a simple test for this code as well.
>
> Reviewed-by: Samuel Neves <sneves(a)dei.uc.pt> # for vgetrandom-chacha.S
> Signed-off-by: Jason A. Donenfeld <Jason(a)zx2c4.com>
> ---
> arch/x86/Kconfig | 1 +
> arch/x86/entry/vdso/Makefile | 3 +-
> arch/x86/entry/vdso/vdso.lds.S | 2 +
> arch/x86/entry/vdso/vgetrandom-chacha.S | 178 ++++++++++++++++++
> arch/x86/entry/vdso/vgetrandom.c | 17 ++
> arch/x86/include/asm/vdso/getrandom.h | 55 ++++++
> arch/x86/include/asm/vdso/vsyscall.h | 2 +
> arch/x86/include/asm/vvar.h | 16 ++
> tools/testing/selftests/vDSO/.gitignore | 1 +
> tools/testing/selftests/vDSO/Makefile | 13 ++
> .../testing/selftests/vDSO/vdso_test_chacha.c | 43 +++++
Hi Jason,
This is a large patch, so it might be helpful to split out the selftests
into their own patch. In fact, my comments here are only about those.
I'm adding linux-kselftest to Cc for visibility, and I've also Cc'd you
on a related selftests/vDSO series I just now posted [1].
In fact, I think it might work well if you insert patches 2/3 and 3/3
from that series, and build on top of those for the
selftests/vDSO/Makefile. See below for details.
...
> diff --git a/tools/testing/selftests/vDSO/Makefile b/tools/testing/selftests/vDSO/Makefile
> index a33b4d200a32..8b87ebea1630 100644
> --- a/tools/testing/selftests/vDSO/Makefile
> +++ b/tools/testing/selftests/vDSO/Makefile
> @@ -3,6 +3,7 @@ include ../lib.mk
>
> uname_M := $(shell uname -m 2>/dev/null || echo not)
> ARCH ?= $(shell echo $(uname_M) | sed -e s/i.86/x86/ -e s/x86_64/x86/)
> +SODIUM := $(shell pkg-config --libs libsodium 2>/dev/null)
>
> TEST_GEN_PROGS := $(OUTPUT)/vdso_test_gettimeofday $(OUTPUT)/vdso_test_getcpu
> TEST_GEN_PROGS += $(OUTPUT)/vdso_test_abi
> @@ -12,9 +13,15 @@ TEST_GEN_PROGS += $(OUTPUT)/vdso_standalone_test_x86
> endif
> TEST_GEN_PROGS += $(OUTPUT)/vdso_test_correctness
> TEST_GEN_PROGS += $(OUTPUT)/vdso_test_getrandom
> +ifeq ($(uname_M),x86_64)
> +ifneq ($(SODIUM),)
> +TEST_GEN_PROGS += $(OUTPUT)/vdso_test_chacha
Unfortunately, this is "pre-existing wrong". :) That is, it is following
a pre-existing pattern that is broken: the $(OUTPUT) is not supposed to
be part of TEST_GEN_PROGS. Fixing it requires other changes, though, as
I've done in [2].
[1] https://lore.kernel.org/all/20240614233105.265009-1-jhubbard@nvidia.com/
[2] https://lore.kernel.org/all/20240614233105.265009-3-jhubbard@nvidia.com/
[3] https://lore.kernel.org/all/20240614233105.265009-4-jhubbard@nvidia.com/
> +endif
> +endif
>
> CFLAGS := -std=gnu99
> CFLAGS_vdso_standalone_test_x86 := -nostdlib -fno-asynchronous-unwind-tables -fno-stack-protector
> +CFLAGS_vdso_test_chacha := $(SODIUM) -idirafter $(top_srcdir)/include -idirafter $(top_srcdir)/arch/$(ARCH)/include -idirafter include -D__ASSEMBLY__ -DBULID_VDSO -DCONFIG_FUNCTION_ALIGNMENT=0 -Wa,--noexecstack
Line breaks via "\" are allowed in Makefiles. Might need two or three here.
> LDFLAGS_vdso_test_correctness := -ldl
> ifeq ($(CONFIG_X86_32),y)
> LDLIBS += -lgcc_s
> @@ -35,3 +42,9 @@ $(OUTPUT)/vdso_test_correctness: vdso_test_correctness.c
> -o $@ \
> $(LDFLAGS_vdso_test_correctness)
> $(OUTPUT)/vdso_test_getrandom: parse_vdso.c
> +$(OUTPUT)/vdso_test_chacha: CFLAGS += $(CFLAGS_vdso_test_chacha)
This also follows an unfortunate pattern, which I've also fixed just today
in a patch [3]. Please just add to CFLAGS directly, rather than creating
these name-mangled intermediate variables. See [3] for how that looks.
> +$(OUTPUT)/vdso_test_chacha: $(top_srcdir)/arch/$(ARCH)/entry/vdso/vgetrandom-chacha.S
> +$(OUTPUT)/vdso_test_chacha: include/asm/rwonce.h
> +include/asm/rwonce.h:
> + mkdir -p include/asm
> + touch $@
thanks,
--
John Hubbard
NVIDIA
Correctable memory errors are very common on servers with large
amount of memory, and are corrected by ECC, but with two
pain points to users:
1. Correction usually happens on the fly and adds latency overhead
2. Not-fully-proved theory states excessive correctable memory
errors can develop into uncorrectable memory error.
Soft offline is kernel's additional solution for memory pages
having (excessive) corrected memory errors. Impacted page is migrated
to healthy page if it is in use, then the original page is discarded
for any future use.
The actual policy on whether (and when) to soft offline should be
maintained by userspace, especially in case of an 1G HugeTLB page.
Soft-offline dissolves the HugeTLB page, either in-use or free, into
chunks of 4K pages, reducing HugeTLB pool capacity by 1 hugepage.
If userspace has not acknowledged such behavior, it may be surprised
when later mmap hugepages MAP_FAILED due to lack of hugepages.
In case of a transparent hugepage, it will be split into 4K pages
as well; userspace will stop enjoying the transparent performance.
In addition, discarding the entire 1G HugeTLB page only because of
corrected memory errors sounds very costly and kernel better not
doing under the hood. But today there are at least 2 such cases:
1. GHES driver sees both GHES_SEV_CORRECTED and
CPER_SEC_ERROR_THRESHOLD_EXCEEDED after parsing CPER.
2. RAS Correctable Errors Collector counts correctable errors per
PFN and when the counter for a PFN reaches threshold
In both cases, userspace has no control of the soft offline performed
by kernel's memory failure recovery.
This patch series give userspace the control of softofflining any page:
kernel only soft offlines raw page / transparent hugepage / HugeTLB
hugepage if userspace has agreed to. The interface to userspace is a
new sysctl called enable_soft_offline under /proc/sys/vm. By default
enable_soft_line is 1 to preserve existing behavior in kernel.
Changelog
v1 => v2:
* incorporate feedbacks from both Miaohe Lin <linmiaohe(a)huawei.com> and
Jane Chu <jane.chu(a)oracle.com>.
* make the switch to control all pages, instead of HugeTLB specific.
* change the API from
/sys/kernel/mm/hugepages/hugepages-${size}kB/softoffline_corrected_errors
to /proc/sys/vm/enable_soft_offline.
* minor update to test code.
* update documentation of the user control API.
* v2 is based on commit 83a7eefedc9b ("Linux 6.10-rc3").
Jiaqi Yan (3):
mm/memory-failure: userspace controls soft-offlining pages
selftest/mm: test enable_soft_offline behaviors
docs: mm: add enable_soft_offline sysctl
Documentation/admin-guide/sysctl/vm.rst | 15 +
mm/memory-failure.c | 16 ++
tools/testing/selftests/mm/.gitignore | 1 +
tools/testing/selftests/mm/Makefile | 1 +
.../selftests/mm/hugetlb-soft-offline.c | 258 ++++++++++++++++++
tools/testing/selftests/mm/run_vmtests.sh | 4 +
6 files changed, 295 insertions(+)
create mode 100644 tools/testing/selftests/mm/hugetlb-soft-offline.c
--
2.45.2.505.gda0bf45e8d-goog
From: John Hubbard <jhubbard(a)nvidia.com>
[ Upstream commit cb708ab9f584f159798b60853edcf0c8b67ce295 ]
It's slightly better to set _GNU_SOURCE in the source code, but if one
must do it via the compiler invocation, then the best way to do so is
this:
$(CC) -D_GNU_SOURCE=
...because otherwise, if this form is used:
$(CC) -D_GNU_SOURCE
...then that leads the compiler to set a value, as if you had passed in:
$(CC) -D_GNU_SOURCE=1
That, in turn, leads to warnings under both gcc and clang, like this:
futex_requeue_pi.c:20: warning: "_GNU_SOURCE" redefined
Fix this by using the "-D_GNU_SOURCE=" form.
Reviewed-by: Edward Liaw <edliaw(a)google.com>
Reviewed-by: Davidlohr Bueso <dave(a)stgolabs.net>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
Signed-off-by: Shuah Khan <skhan(a)linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/futex/functional/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/futex/functional/Makefile b/tools/testing/selftests/futex/functional/Makefile
index a392d0917b4e5..994fa3468f170 100644
--- a/tools/testing/selftests/futex/functional/Makefile
+++ b/tools/testing/selftests/futex/functional/Makefile
@@ -1,6 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
INCLUDES := -I../include -I../../ $(KHDR_INCLUDES)
-CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE -pthread $(INCLUDES) $(KHDR_INCLUDES)
+CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE= -pthread $(INCLUDES) $(KHDR_INCLUDES)
LDLIBS := -lpthread -lrt
LOCAL_HDRS := \
--
2.43.0
From: John Hubbard <jhubbard(a)nvidia.com>
[ Upstream commit cb708ab9f584f159798b60853edcf0c8b67ce295 ]
It's slightly better to set _GNU_SOURCE in the source code, but if one
must do it via the compiler invocation, then the best way to do so is
this:
$(CC) -D_GNU_SOURCE=
...because otherwise, if this form is used:
$(CC) -D_GNU_SOURCE
...then that leads the compiler to set a value, as if you had passed in:
$(CC) -D_GNU_SOURCE=1
That, in turn, leads to warnings under both gcc and clang, like this:
futex_requeue_pi.c:20: warning: "_GNU_SOURCE" redefined
Fix this by using the "-D_GNU_SOURCE=" form.
Reviewed-by: Edward Liaw <edliaw(a)google.com>
Reviewed-by: Davidlohr Bueso <dave(a)stgolabs.net>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
Signed-off-by: Shuah Khan <skhan(a)linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/futex/functional/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/futex/functional/Makefile b/tools/testing/selftests/futex/functional/Makefile
index a392d0917b4e5..994fa3468f170 100644
--- a/tools/testing/selftests/futex/functional/Makefile
+++ b/tools/testing/selftests/futex/functional/Makefile
@@ -1,6 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
INCLUDES := -I../include -I../../ $(KHDR_INCLUDES)
-CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE -pthread $(INCLUDES) $(KHDR_INCLUDES)
+CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE= -pthread $(INCLUDES) $(KHDR_INCLUDES)
LDLIBS := -lpthread -lrt
LOCAL_HDRS := \
--
2.43.0
From: John Hubbard <jhubbard(a)nvidia.com>
[ Upstream commit cb708ab9f584f159798b60853edcf0c8b67ce295 ]
It's slightly better to set _GNU_SOURCE in the source code, but if one
must do it via the compiler invocation, then the best way to do so is
this:
$(CC) -D_GNU_SOURCE=
...because otherwise, if this form is used:
$(CC) -D_GNU_SOURCE
...then that leads the compiler to set a value, as if you had passed in:
$(CC) -D_GNU_SOURCE=1
That, in turn, leads to warnings under both gcc and clang, like this:
futex_requeue_pi.c:20: warning: "_GNU_SOURCE" redefined
Fix this by using the "-D_GNU_SOURCE=" form.
Reviewed-by: Edward Liaw <edliaw(a)google.com>
Reviewed-by: Davidlohr Bueso <dave(a)stgolabs.net>
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
Signed-off-by: Shuah Khan <skhan(a)linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/futex/functional/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/testing/selftests/futex/functional/Makefile b/tools/testing/selftests/futex/functional/Makefile
index a392d0917b4e5..994fa3468f170 100644
--- a/tools/testing/selftests/futex/functional/Makefile
+++ b/tools/testing/selftests/futex/functional/Makefile
@@ -1,6 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
INCLUDES := -I../include -I../../ $(KHDR_INCLUDES)
-CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE -pthread $(INCLUDES) $(KHDR_INCLUDES)
+CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE= -pthread $(INCLUDES) $(KHDR_INCLUDES)
LDLIBS := -lpthread -lrt
LOCAL_HDRS := \
--
2.43.0
From: Michael Ellerman <mpe(a)ellerman.id.au>
[ Upstream commit e8b8c5264d4ebd248f60a5cef077fe615806e7a0 ]
Fix build error on ppc64:
dev_in_maps.c: In function ‘get_file_dev_and_inode’:
dev_in_maps.c:60:59: error: format ‘%llu’ expects argument of type
‘long long unsigned int *’, but argument 7 has type ‘__u64 *’ {aka ‘long
unsigned int *’} [-Werror=format=]
By switching to unsigned long long for u64 for ppc64 builds.
Signed-off-by: Michael Ellerman <mpe(a)ellerman.id.au>
Signed-off-by: Shuah Khan <skhan(a)linuxfoundation.org>
Signed-off-by: Sasha Levin <sashal(a)kernel.org>
---
tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c b/tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c
index 759f86e7d263e..2862aae58b79a 100644
--- a/tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c
+++ b/tools/testing/selftests/filesystems/overlayfs/dev_in_maps.c
@@ -1,5 +1,6 @@
// SPDX-License-Identifier: GPL-2.0
#define _GNU_SOURCE
+#define __SANE_USERSPACE_TYPES__ // Use ll64
#include <inttypes.h>
#include <unistd.h>
--
2.43.0
Hi guys,
I'm trying to enable migration from MtCollins(Ampere Altra, ARMv8.2+) to
AmpereOne(AmpereOne, ARMv8.6+), the migration always fails when migration from
MtCollins to AmpereOne due to some register fields differing between the
two machines.
In this patch series, we try to make more register fields writable like
ID_AA64PFR1_EL1.BT. This is first step towards making the migration possible.
Some other hurdles need to be overcome. This is not sufficient to make the
migration successful from MtCollins to AmpereOne.
Shaoqin Huang (2):
KVM: arm64: Allow BT field in ID_AA64PFR1_EL1 writable
KVM: selftests: aarch64: Add writable test for ID_AA64PFR1_EL1
arch/arm64/kvm/sys_regs.c | 2 +-
tools/testing/selftests/kvm/aarch64/set_id_regs.c | 6 ++++++
2 files changed, 7 insertions(+), 1 deletion(-)
--
2.40.1
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v3:
- rename start_client to client_socket
- Use connect_to_addr in connect_to_fd_opt
v2:
- update patch 2, extract a new helper start_client.
- drop patch 3, keep must_fail in network_helper_opts.
Drop type and noconnect from network_helper_opts. And use start_server_str
in mptcp and test_tcp_check_syncookie_user.
Patches 1-4 address Martin's comments in the previous series.
Geliang Tang (6):
selftests/bpf: Drop type from network_helper_opts
selftests/bpf: Use connect_to_addr in connect_to_fd_opt
selftests/bpf: Add client_socket helper
selftests/bpf: Drop noconnect from network_helper_opts
selftests/bpf: Use start_server_str in mptcp
selftests/bpf: Use start_server_str in test_tcp_check_syncookie_user
tools/testing/selftests/bpf/network_helpers.c | 100 ++++++++----------
tools/testing/selftests/bpf/network_helpers.h | 6 +-
.../selftests/bpf/prog_tests/bpf_tcp_ca.c | 2 +-
.../selftests/bpf/prog_tests/cgroup_v1v2.c | 4 +-
.../bpf/prog_tests/ip_check_defrag.c | 10 +-
.../testing/selftests/bpf/prog_tests/mptcp.c | 7 +-
.../bpf/test_tcp_check_syncookie_user.c | 29 +----
7 files changed, 56 insertions(+), 102 deletions(-)
--
2.43.0
Hi,
this is a v3 patch set as a follow up of the thread about the errors
reported by kselftest mixer-test. It changes HD-audio and vmaster
control behavior to return -EINVAL for invalid input values.
There is a change in kselftest itself to skip the verification after
write tests for volatile controls, too. It's for the channel map
controls that can't hold the stable values.
v2->v3:
* Replace with Mark's patch for kselftest
* Apply the validation for user controls in put callback instead
v1->v2:
* Skip only verification after write in kselftest
* Add sanity check to HDMI chmap write, too
v2: https://lore.kernel.org/r/20240614153717.30143-1-tiwai@suse.de
v1: https://lore.kernel.org/r/20240614124728.27901-1-tiwai@suse.de
Takashi
===
Mark Brown (1):
kselftest/alsa: Fix validation of writes to volatile controls
Takashi Iwai (5):
ALSA: vmaster: Return error for invalid input values
ALSA: hda: Return -EINVAL for invalid volume/switch inputs
ALSA: control: Apply sanity check of input values for user elements
ALSA: chmap: Mark Channel Map controls as volatile
ALSA: hda: Add input value sanity checks to HDMI channel map controls
sound/core/control.c | 6 ++-
sound/core/pcm_lib.c | 1 +
sound/core/vmaster.c | 8 ++++
sound/hda/hdmi_chmap.c | 18 +++++++++
sound/pci/hda/hda_codec.c | 23 +++++++++---
tools/testing/selftests/alsa/mixer-test.c | 45 +++++++++++++++--------
6 files changed, 79 insertions(+), 22 deletions(-)
--
2.43.0
When validating writes to controls we check that whatever value we wrote
actually appears in the control. For volatile controls we cannot assume
that this will be the case, the value may be changed at any time
including between our write and read. Handle this by moving the check
for volatile controls that we currently do for events to a separate
block and just verifying that whatever value we read is valid for the
control.
Signed-off-by: Mark Brown <broonie(a)kernel.org>
---
tools/testing/selftests/alsa/mixer-test.c | 45 ++++++++++++++++++++-----------
1 file changed, 29 insertions(+), 16 deletions(-)
diff --git a/tools/testing/selftests/alsa/mixer-test.c b/tools/testing/selftests/alsa/mixer-test.c
index 1c04e5f638a0..dd74f8cc7ece 100644
--- a/tools/testing/selftests/alsa/mixer-test.c
+++ b/tools/testing/selftests/alsa/mixer-test.c
@@ -625,6 +625,21 @@ static int write_and_verify(struct ctl_data *ctl,
return err;
}
+ /*
+ * We can't verify any specific value for volatile controls
+ * but we should still check that whatever we read is a valid
+ * vale for the control.
+ */
+ if (snd_ctl_elem_info_is_volatile(ctl->info)) {
+ if (!ctl_value_valid(ctl, read_val)) {
+ ksft_print_msg("Volatile control %s has invalid value\n",
+ ctl->name);
+ return -EINVAL;
+ }
+
+ return 0;
+ }
+
/*
* Check for an event if the value changed, or confirm that
* there was none if it didn't. We rely on the kernel
@@ -632,22 +647,20 @@ static int write_and_verify(struct ctl_data *ctl,
* write, this is currently true, should that ever change this
* will most likely break and need updating.
*/
- if (!snd_ctl_elem_info_is_volatile(ctl->info)) {
- err = wait_for_event(ctl, 0);
- if (snd_ctl_elem_value_compare(initial_val, read_val)) {
- if (err < 1) {
- ksft_print_msg("No event generated for %s\n",
- ctl->name);
- show_values(ctl, initial_val, read_val);
- ctl->event_missing++;
- }
- } else {
- if (err != 0) {
- ksft_print_msg("Spurious event generated for %s\n",
- ctl->name);
- show_values(ctl, initial_val, read_val);
- ctl->event_spurious++;
- }
+ err = wait_for_event(ctl, 0);
+ if (snd_ctl_elem_value_compare(initial_val, read_val)) {
+ if (err < 1) {
+ ksft_print_msg("No event generated for %s\n",
+ ctl->name);
+ show_values(ctl, initial_val, read_val);
+ ctl->event_missing++;
+ }
+ } else {
+ if (err != 0) {
+ ksft_print_msg("Spurious event generated for %s\n",
+ ctl->name);
+ show_values(ctl, initial_val, read_val);
+ ctl->event_spurious++;
}
}
---
base-commit: 83a7eefedc9b56fe7bfeff13b6c7356688ffa670
change-id: 20240614-alsa-selftest-volatile-d6f3e8e28c08
Best regards,
--
Mark Brown <broonie(a)kernel.org>
Hi,
this is a revised patch set as a follow up of the thread about the
errors reported by kselftest mixer-test. It changes HD-audio and
vmaster control behavior to return -EINVAL for invalid input values.
There is a change in kselftest itself to skip the verification after
write tests for volatile controls, too. It's for the channel map
controls that can't hold the stable values.
v1->v2:
* Skip only verification after write in kselftest
* Add sanity check to HDMI chmap write, too
Takashi
===
Takashi Iwai (6):
ALSA: vmaster: Return error for invalid input values
ALSA: hda: Return -EINVAL for invalid volume/switch inputs
ALSA: control: Apply sanity check of input values for user elements
kselftest/alsa: mixer-test: Skip write verification for volatile
controls
ALSA: chmap: Mark Channel Map controls as volatile
ALSA: hda: Add input value sanity checks to HDMI channel map controls
sound/core/control.c | 3 ++-
sound/core/pcm_lib.c | 1 +
sound/core/vmaster.c | 8 ++++++++
sound/hda/hdmi_chmap.c | 18 ++++++++++++++++++
sound/pci/hda/hda_codec.c | 23 ++++++++++++++++++-----
tools/testing/selftests/alsa/mixer-test.c | 4 ++++
6 files changed, 51 insertions(+), 6 deletions(-)
--
2.43.0
The goal of this series is to use helpers from net/lib.sh with MPTCP
selftests.
- Patches 1 to 4 are some clean-ups and preparation in net/lib.sh:
- Patch 1 simplifies the code handling errexit by ignoring possible
errors instead of disabling errexit temporary.
- Patch 2 removes the netns from the list after having cleaned it, not
to try to clean it twice.
- Patch 3 removes the 'readonly' attribute for the netns variable, to
allow using the same name in local variables.
- Patch 4 removes the local 'ns' var, not to conflict with the global
one it needs to setup.
- Patch 5 uses helpers from net/lib.sh to create and delete netns in
MPTCP selftests.
- Patch 6 uses wait_local_port_listen helper from net/net_helper.sh.
Signed-off-by: Matthieu Baerts (NGI0) <matttbe(a)kernel.org>
---
Geliang Tang (3):
selftests: net: lib: remove 'ns' var in setup_ns
selftests: mptcp: lib: use setup/cleanup_ns helpers
selftests: mptcp: lib: use wait_local_port_listen helper
Matthieu Baerts (NGI0) (3):
selftests: net: lib: ignore possible errors
selftests: net: lib: remove ns from list after clean-up
selftests: net: lib: do not set ns var as readonly
tools/testing/selftests/net/lib.sh | 55 +++++++++++++++-----------
tools/testing/selftests/net/mptcp/mptcp_lib.sh | 33 +++++-----------
2 files changed, 42 insertions(+), 46 deletions(-)
---
base-commit: a999973236543f0b8f6daeaa7ecba7488c3a593b
change-id: 20240607-upstream-net-next-20240607-selftests-mptcp-net-lib-365e43e2e1ca
Best regards,
--
Matthieu Baerts (NGI0) <matttbe(a)kernel.org>
This patch addresses the TODO (add non fixed feature on/off check).
I have tested it manually on my system after making changes as suggested
in v1 and v2 linked below for reference.
Patch now restores the features being tested to their initial state.
Signed-off-by: Abhinav Jain <jain.abhinav177(a)gmail.com>
---
PATCH v2:
https://lore.kernel.org/all/20240609132124.51683-1-jain.abhinav177@gmail.co…
Changes since v2:
- Added a check for netdev if it exists.
- If netdev doesn't exist, create a veth pair for testing.
- Restore the feature being tested to its intial state.
PATCH v1:
https://lore.kernel.org/all/20240606212714.27472-1-jain.abhinav177@gmail.co…
Changes since v1:
- Removed the addition of tail command as it was not required after
below change.
- Used read to parse the temp features file rather than using for loop
and took out awk/grep/sed from v1.
---
tools/testing/selftests/net/netdevice.sh | 55 +++++++++++++++++++++++-
1 file changed, 54 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/netdevice.sh b/tools/testing/selftests/net/netdevice.sh
index e3afcb424710..d937d39dda6a 100755
--- a/tools/testing/selftests/net/netdevice.sh
+++ b/tools/testing/selftests/net/netdevice.sh
@@ -104,6 +104,20 @@ kci_netdev_ethtool()
{
netdev=$1
+ #check if netdev is provided as an argument
+ if [ -z "$netdev" ]; then
+ echo "No network device provided, creating a veth pair"
+ ip link add veth0 type veth peer name veth1
+ netdev="veth0"
+ veth_created=1
+ else
+ #check if the provided netdev exists
+ if ! ip link show "$netdev" > /dev/null 2>&1; then
+ echo "Network device $netdev does not exist."
+ return 1
+ fi
+ fi
+
#check presence of ethtool
ethtool --version 2>/dev/null >/dev/null
if [ $? -ne 0 ];then
@@ -124,11 +138,50 @@ kci_netdev_ethtool()
return 1
fi
echo "PASS: $netdev: ethtool list features"
- #TODO for each non fixed features, try to turn them on/off
+
+ while read -r FEATURE VALUE FIXED; do
+ [ "$FEATURE" != "Features" ] || continue # Skip "Features" line
+ [ "$FIXED" != "[fixed]" ] || continue # Skip fixed features
+ feature = "${FEATURE%:*}"
+
+ initial_state=$(ethtool -k "$netdev" | grep "$feature:" | awk '{print $2}')
+ ethtool --offload "$netdev" "$feature" off
+ if [ $? -eq 0 ]; then
+ echo "PASS: $netdev: Turned off feature: $feature"
+ else
+ echo "FAIL: $netdev: Failed to turn off feature: $feature"
+ fi
+
+ ethtool --offload "$netdev" "$feature" on
+ if [ $? -eq 0 ]; then
+ echo "PASS: $netdev: Turned on feature: $feature"
+ else
+ echo "FAIL: $netdev: Failed to turn on feature: $feature"
+ fi
+
+ #restore the feature to its initial state
+ ethtool --offload "$netdev" "$feature" "$initial_state"
+ if [$? -eq 0]; then
+ echo "PASS: $netdev: Restore feature $feature to" \
+ " initial state $initial_state"
+ else
+ echo "FAIL: $netdev: Failed to restore feature " \
+ "$feature to initial state $initial_state"
+ fi
+
+ done < "$TMP_ETHTOOL_FEATURES"
+
rm "$TMP_ETHTOOL_FEATURES"
kci_netdev_ethtool_test 74 'dump' "ethtool -d $netdev"
kci_netdev_ethtool_test 94 'stats' "ethtool -S $netdev"
+
+ #clean up veth interface pair if it was created
+ if ["$veth_created" ]; then
+ ip link delete veth0
+ echo "Removed veth pair"
+ fi
+
return 0
}
--
2.34.1
When building with clang, via:
make LLVM=1 -C tools/testing/selftests
...there are several warnings, and an error. This fixes all of those and
allows these tests to run and pass.
1. Fix linker error (undefined reference to memcpy) by providing a local
version of memcpy.
2. clang complains about using this form:
if (g = h & 0xf0000000)
...so factor out the assignment into a separate step.
3. The code is passing a signed const char* to elf_hash(), which expects
a const unsigned char *. There are several callers, so fix this at
the source by allowing the function to accept a signed argument, and
then converting to unsigned operations, once inside the function.
4. clang doesn't have __attribute__((externally_visible)) and generates
a warning to that effect. Fortunately, gcc 12 and gcc 13 do not seem
to require that attribute in order to build, run and pass tests here,
so remove it.
[1] https://lore.kernel.org/all/20240329-selftests-libmk-llvm-rfc-v1-1-2f9ed7d1…
Signed-off-by: John Hubbard <jhubbard(a)nvidia.com>
---
Changes since the first version:
1) Rebased onto Linux 6.10-rc1
thanks,
John Hubbard
tools/testing/selftests/vDSO/parse_vdso.c | 16 +++++++++++-----
.../selftests/vDSO/vdso_standalone_test_x86.c | 18 ++++++++++++++++--
2 files changed, 27 insertions(+), 7 deletions(-)
diff --git a/tools/testing/selftests/vDSO/parse_vdso.c b/tools/testing/selftests/vDSO/parse_vdso.c
index 413f75620a35..4ae417372e9e 100644
--- a/tools/testing/selftests/vDSO/parse_vdso.c
+++ b/tools/testing/selftests/vDSO/parse_vdso.c
@@ -55,14 +55,20 @@ static struct vdso_info
ELF(Verdef) *verdef;
} vdso_info;
-/* Straight from the ELF specification. */
-static unsigned long elf_hash(const unsigned char *name)
+/*
+ * Straight from the ELF specification...and then tweaked slightly, in order to
+ * avoid a few clang warnings.
+ */
+static unsigned long elf_hash(const char *name)
{
unsigned long h = 0, g;
- while (*name)
+ const unsigned char *uch_name = (const unsigned char *)name;
+
+ while (*uch_name)
{
- h = (h << 4) + *name++;
- if (g = h & 0xf0000000)
+ h = (h << 4) + *uch_name++;
+ g = h & 0xf0000000;
+ if (g)
h ^= g >> 24;
h &= ~g;
}
diff --git a/tools/testing/selftests/vDSO/vdso_standalone_test_x86.c b/tools/testing/selftests/vDSO/vdso_standalone_test_x86.c
index 8a44ff973ee1..27f6fdf11969 100644
--- a/tools/testing/selftests/vDSO/vdso_standalone_test_x86.c
+++ b/tools/testing/selftests/vDSO/vdso_standalone_test_x86.c
@@ -18,7 +18,7 @@
#include "parse_vdso.h"
-/* We need a libc functions... */
+/* We need some libc functions... */
int strcmp(const char *a, const char *b)
{
/* This implementation is buggy: it never returns -1. */
@@ -34,6 +34,20 @@ int strcmp(const char *a, const char *b)
return 0;
}
+/*
+ * The clang build needs this, although gcc does not.
+ * Stolen from lib/string.c.
+ */
+void *memcpy(void *dest, const void *src, size_t count)
+{
+ char *tmp = dest;
+ const char *s = src;
+
+ while (count--)
+ *tmp++ = *s++;
+ return dest;
+}
+
/* ...and two syscalls. This is x86-specific. */
static inline long x86_syscall3(long nr, long a0, long a1, long a2)
{
@@ -70,7 +84,7 @@ void to_base10(char *lastdig, time_t n)
}
}
-__attribute__((externally_visible)) void c_main(void **stack)
+void c_main(void **stack)
{
/* Parse the stack */
long argc = (long)*stack;
base-commit: 2bfcfd584ff5ccc8bb7acde19b42570414bf880b
--
2.45.1
The following selftests: arm64 tests failed on FVP-aemva test and kernel
built with gcc-13 but pass with clang.
arm64_fp-stress_KERNEL-1-0/3-0/4-0/6-0 - gcc-13 - Failed
arm64_fp-stress_KERNEL-1-0/3-0/4-0/6-0 - clang-18 - Pass
Reported-by: Linux Kernel Functional Testing <lkft(a)linaro.org>
Test log:
---------
# timeout set to 45
# selftests: arm64: fp-stress
[ 269.882519] NET: Registered PF_ALG protocol family
# TAP version 13
# 1..96
# # 8 CPUs, 3 SVE VLs, 3 SME VLs, SME2 present
# # Will run for 10s
# # Started FPSIMD-0-0
# # Started KERNEL-0-0
# # Started SVE-VL-64-0
...
# ok 13 FPSIMD-1-0
# # KERNEL-1-0 exited with error code 1
# not ok 14 KERNEL-1-0
# ok 15 SVE-VL-64-1
...
# ok 37 FPSIMD-3-0
# # KERNEL-3-0 exited with error code 1
# not ok 38 KERNEL-3-0
# ok 39 SVE-VL-64-3
...
# ok 49 FPSIMD-4-0
# # KERNEL-4-0 exited with error code 1
# not ok 50 KERNEL-4-0
# ok 51 SVE-VL-64-4
...
# ok 73 FPSIMD-6-0
# # KERNEL-6-0 exited with error code 1
# not ok 74 KERNEL-6-0
# ok 75 SVE-VL-64-6
...
# # Totals: pass:92 fail:4 xfail:0 xpass:0 skip:0 error:0
ok 16 selftests: arm64: fp-stress
metadata:
------
git_describe: next-20240613
git_ref: master
git_repo: https://gitlab.com/Linaro/lkft/mirrors/next/linux-next
git_sha: 6906a84c482f098d31486df8dc98cead21cce2d0
git_short_log: 6906a84c482f ("Add linux-next specific files for 20240613")
arch: arm64
test-environment: fvp-aemva
toolchain: gcc-13
Links:
- https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20240613/te…
- https://qa-reports.linaro.org/lkft/linux-next-master/build/next-20240613/te…
--
Linaro LKFT
https://lkft.linaro.org
Hi,
this is a patch set as a follow up of the thread about the errors
reported by kselftest mixer-test. It changes HD-audio and vmaster
control behavior to return -EINVAL for invalid input values.
There is a change in kselftest itself to skip the write tests for
volatile controls, too. It's for the channel map controls that can't
hold the stable values.
Takashi
===
Takashi Iwai (5):
ALSA: vmaster: Return error for invalid input values
ALSA: hda: Return -EINVAL for invalid volume/switch inputs
ALSA: control: Apply sanity check of input values for user elements
kselftest/alsa: mixer-test: Skip write tests for volatile controls
ALSA: chmap: Mark Channel Map controls as volatile
sound/core/control.c | 3 ++-
sound/core/pcm_lib.c | 1 +
sound/core/vmaster.c | 8 ++++++++
sound/pci/hda/hda_codec.c | 23 ++++++++++++++++++-----
tools/testing/selftests/alsa/mixer-test.c | 21 +++++++++++++++++++++
5 files changed, 50 insertions(+), 6 deletions(-)
--
2.43.0
Define a maximum allowable number of pids that can be livepatched in
test-syscall.sh as with extremely large machines the output from a
large number of processes overflows the dev/kmsg "expect" buffer in
the "check_result" function and causes a false error.
Reported-by: CKI Project <cki-project(a)redhat.com>
Signed-off-by: Ryan Sullivan <rysulliv(a)redhat.com>
---
tools/testing/selftests/livepatch/test-syscall.sh | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/livepatch/test-syscall.sh b/tools/testing/selftests/livepatch/test-syscall.sh
index b76a881d4013..289eb7d4c4b3 100755
--- a/tools/testing/selftests/livepatch/test-syscall.sh
+++ b/tools/testing/selftests/livepatch/test-syscall.sh
@@ -15,7 +15,10 @@ setup_config
start_test "patch getpid syscall while being heavily hammered"
-for i in $(seq 1 $(getconf _NPROCESSORS_ONLN)); do
+NPROC=$(getconf _NPROCESSORS_ONLN)
+MAXPROC=128
+
+for i in $(seq 1 $(($NPROC < $MAXPROC ? $NPROC : $MAXPROC))); do
./test_klp-call_getpid &
pids[$i]="$!"
done
--
2.44.0
Dear Linux folks,
Running the ALSA kselftests with Linux 6.10-rc1, `mixer-test` shows ten
failures:
# Totals: pass:24 fail:0 xfail:0 xpass:0 skip:11 error:0
These are:
not ok 5 write_invalid.0.40
not ok 11 write_valid.0.39
not ok 18 write_valid.0.38
not ok 25 write_valid.0.37
not ok 201 write_invalid.0.12
not ok 208 write_invalid.0.11
not ok 264 write_invalid.0.3
not ok 271 write_invalid.0.2
not ok 278 write_invalid.0.1
not ok 285 write_invalid.0.0
`./pcm-test` finishes with 5 out of 5 passes.
The output `sudo alsa-info` is uploaded to the database [1], and the
full output of `mixer-test` is attached.
Kind regards,
Paul
[1]: https://alsa-project.org/db/?f=cf8e1e0f37775448b5e3361e5c3cd4d4e6037d4f
Eventually, once the build succeeds on a sufficiently old distro, the
idea is to delete $(KHDR_INCLUDES) from the selftests/mm build, and then
after that, from selftests/lib.mk and all of the other selftest builds.
For now, this series merely achieves a clean build of selftests/mm on a
not-so-old distro: Ubuntu 23.04:
1. Add __NR_mseal.
2. Add fs.h, taken as usual from a snapshot of ./usr/include/linux/fs.h
after running "make headers". This is how we have agreed to do this sort
of thing, see [1].
3. Add a few selected prctl.h values that the ksm and mdwe tests require.
[1] commit e076eaca5906 ("selftests: break the dependency upon local
header files")
John Hubbard (5):
selftests/mm: mseal, self_elf: fix missing __NR_mseal
selftests/mm: fix vm_util.c build failures: add snapshot of fs.h
mm/selftests: kvm, mdwe fixes to avoid requiring "make headers"
selftests/mm: mseal, self_elf: factor out test macros and other
duplicated items
selftests/mm: mseal, self_elf: rename TEST_END_CHECK to
REPORT_TEST_PASS
tools/include/uapi/linux/fs.h | 392 +++++++++++++++++++++
tools/testing/selftests/mm/mdwe_test.c | 1 +
tools/testing/selftests/mm/mseal_helpers.h | 45 +++
tools/testing/selftests/mm/mseal_test.c | 141 +++-----
tools/testing/selftests/mm/seal_elf.c | 35 +-
tools/testing/selftests/mm/vm_util.h | 15 +
6 files changed, 502 insertions(+), 127 deletions(-)
create mode 100644 tools/include/uapi/linux/fs.h
create mode 100644 tools/testing/selftests/mm/mseal_helpers.h
base-commit: 8a92980606e3585d72d510a03b59906e96755b8a
--
2.45.2
This patch addresses the TODO (add non fixed feature on/off check).
I have tested it manually on my system and made changes as suggested in v1
Signed-off-by: Abhinav Jain <jain.abhinav177(a)gmail.com>
---
PATCH v1:
https://lore.kernel.org/all/20240606212714.27472-1-jain.abhinav177@gmail.co…
Changes since v1:
- Removed the addition of tail command as it was not required after
below change
- Used read to parse the temp features file rather than using for loop
and took out awk/grep/sed from v1
---
tools/testing/selftests/net/netdevice.sh | 22 +++++++++++++++++++++-
1 file changed, 21 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/netdevice.sh b/tools/testing/selftests/net/netdevice.sh
index e3afcb424710..196b7f985db4 100755
--- a/tools/testing/selftests/net/netdevice.sh
+++ b/tools/testing/selftests/net/netdevice.sh
@@ -124,7 +124,27 @@ kci_netdev_ethtool()
return 1
fi
echo "PASS: $netdev: ethtool list features"
- #TODO for each non fixed features, try to turn them on/off
+
+ while read -r FEATURE VALUE FIXED; do
+ [ "$FEATURE" != "Features" ] || continue # Skip "Features" line
+ [ "$FIXED" != "[fixed]" ] || continue # Skip fixed features
+ feature = "${FEATURE%:*}"
+
+ ethtool --offload "$netdev" "$feature" off
+ if [ $? -eq 0 ]; then
+ echo "PASS: $netdev: Turned off feature: $feature"
+ else
+ echo "FAIL: $netdev: Failed to turn off feature: $feature"
+ fi
+
+ ethtool --offload "$netdev" "$feature" on
+ if [ $? -eq 0 ]; then
+ echo "PASS: $netdev: Turned on feature: $feature"
+ else
+ echo "FAIL: $netdev: Failed to turn on feature: $feature"
+ fi
+ done < "$TMP_ETHTOOL_FEATURES"
+
rm "$TMP_ETHTOOL_FEATURES"
kci_netdev_ethtool_test 74 'dump' "ethtool -d $netdev"
--
2.34.1
The purpose of this series is to rethink how HID-BPF is invoked.
Currently it implies a jmp table, a prog fd bpf_map, a preloaded tracing
bpf program and a lot of manual work for handling the bpf program
lifetime and addition/removal.
OTOH, bpf_struct_ops take care of most of the bpf handling leaving us
with a simple list of ops pointers, and we can directly call the
struct_ops program from the kernel as a regular function.
The net gain right now is in term of code simplicity and lines of code
removal (though is an API breakage), but udev-hid-bpf is able to handle
such breakages.
In the near future, we will be able to extend the HID-BPF struct_ops
with entrypoints for hid_hw_raw_request() and hid_hw_output_report(),
allowing for covering all of the initial use cases:
- firewalling a HID device
- fixing all of the HID device interactions (not just device events as
it is right now).
The matching user-space loader (udev-hid-bpf) MR is at
https://gitlab.freedesktop.org/libevdev/udev-hid-bpf/-/merge_requests/86
I'll put it out of draft once this is merged.
Cheers,
Benjamin
Signed-off-by: Benjamin Tissoires <bentiss(a)kernel.org>
---
Changes in v3:
- took Alexei's review into account
- Link to v2: https://lore.kernel.org/r/20240607-hid_bpf_struct_ops-v2-0-3f95f4d02292@ker…
Changes in v2:
- drop HID_BPF_FLAGS enum and use BPF_F_BEFORE instead
- fix .init_members to not open code member->offset
- allow struct hid_device to be writeable from HID-BPF for its name,
uniq and phys
- Link to v1: https://lore.kernel.org/r/20240528-hid_bpf_struct_ops-v1-0-8c6663df27d8@ker…
---
Benjamin Tissoires (16):
HID: rename struct hid_bpf_ops into hid_ops
HID: bpf: add hid_get/put_device() helpers
HID: bpf: implement HID-BPF through bpf_struct_ops
selftests/hid: convert the hid_bpf selftests with struct_ops
HID: samples: convert the 2 HID-BPF samples into struct_ops
HID: bpf: add defines for HID-BPF SEC in in-tree bpf fixes
HID: bpf: convert in-tree fixes into struct_ops
HID: bpf: remove tracing HID-BPF capability
selftests/hid: add subprog call test
Documentation: HID: amend HID-BPF for struct_ops
Documentation: HID: add a small blurb on udev-hid-bpf
HID: bpf: Artist24: remove unused variable
HID: bpf: error on warnings when compiling bpf objects
bpf: allow bpf helpers to be used into HID-BPF struct_ops
HID: bpf: rework hid_bpf_ops_btf_struct_access
HID: bpf: make part of struct hid_device writable
Documentation/hid/hid-bpf.rst | 173 ++++---
drivers/hid/bpf/Makefile | 2 +-
drivers/hid/bpf/entrypoints/Makefile | 93 ----
drivers/hid/bpf/entrypoints/README | 4 -
drivers/hid/bpf/entrypoints/entrypoints.bpf.c | 25 -
drivers/hid/bpf/entrypoints/entrypoints.lskel.h | 248 ---------
drivers/hid/bpf/hid_bpf_dispatch.c | 266 +++-------
drivers/hid/bpf/hid_bpf_dispatch.h | 12 +-
drivers/hid/bpf/hid_bpf_jmp_table.c | 565 ---------------------
drivers/hid/bpf/hid_bpf_struct_ops.c | 298 +++++++++++
drivers/hid/bpf/progs/FR-TEC__Raptor-Mach-2.bpf.c | 9 +-
drivers/hid/bpf/progs/HP__Elite-Presenter.bpf.c | 6 +-
drivers/hid/bpf/progs/Huion__Kamvas-Pro-19.bpf.c | 9 +-
.../hid/bpf/progs/IOGEAR__Kaliber-MMOmentum.bpf.c | 6 +-
drivers/hid/bpf/progs/Makefile | 2 +-
.../hid/bpf/progs/Microsoft__XBox-Elite-2.bpf.c | 6 +-
drivers/hid/bpf/progs/Wacom__ArtPen.bpf.c | 6 +-
drivers/hid/bpf/progs/XPPen__Artist24.bpf.c | 10 +-
drivers/hid/bpf/progs/XPPen__ArtistPro16Gen2.bpf.c | 24 +-
drivers/hid/bpf/progs/hid_bpf.h | 5 +
drivers/hid/hid-core.c | 6 +-
include/linux/hid_bpf.h | 119 +++--
samples/hid/Makefile | 5 +-
samples/hid/hid_bpf_attach.bpf.c | 18 -
samples/hid/hid_bpf_attach.h | 14 -
samples/hid/hid_mouse.bpf.c | 26 +-
samples/hid/hid_mouse.c | 39 +-
samples/hid/hid_surface_dial.bpf.c | 10 +-
samples/hid/hid_surface_dial.c | 53 +-
tools/testing/selftests/hid/hid_bpf.c | 100 +++-
tools/testing/selftests/hid/progs/hid.c | 100 +++-
.../testing/selftests/hid/progs/hid_bpf_helpers.h | 19 +-
32 files changed, 800 insertions(+), 1478 deletions(-)
---
base-commit: 70ec81c2e2b4005465ad0d042e90b36087c36104
change-id: 20240513-hid_bpf_struct_ops-e3212a224555
Best regards,
--
Benjamin Tissoires <bentiss(a)kernel.org>
From: Geliang Tang <tanggeliang(a)kylinos.cn>
v2:
- update patch 2, extract a new helper start_client.
- drop patch 3, keep must_fail in network_helper_opts.
Drop type and noconnect from network_helper_opts. And use start_server_str
in mptcp and test_tcp_check_syncookie_user.
Patches 1-2 address Martin's comments in the previous series.
Geliang Tang (4):
selftests/bpf: Drop type from network_helper_opts
selftests/bpf: Drop noconnect from network_helper_opts
selftests/bpf: Use start_server_str in mptcp
selftests/bpf: Use start_server_str in test_tcp_check_syncookie_user
tools/testing/selftests/bpf/network_helpers.c | 45 +++++++++++++++----
tools/testing/selftests/bpf/network_helpers.h | 7 +--
.../selftests/bpf/prog_tests/bpf_tcp_ca.c | 2 +-
.../selftests/bpf/prog_tests/cgroup_v1v2.c | 4 +-
.../bpf/prog_tests/ip_check_defrag.c | 7 +--
.../testing/selftests/bpf/prog_tests/mptcp.c | 7 +--
.../bpf/test_tcp_check_syncookie_user.c | 29 ++----------
7 files changed, 49 insertions(+), 52 deletions(-)
--
2.43.0
This series adds two new features required to be able to run the test on
more platforms on KernelCI.
The first patch adds a parameter to allow overriding the directory in
which the board files will be looked for. Since the board files are
hosted in a separate repository [1], this parameter allows overlaying
those files on the filesystem and passing the location to the test.
The second is needed for one platform in particular, MT8195, in which
the usb controllers are instanced from a two-level deep DT node that
doesn't allow unique matching based on the existing properties.
[1] https://github.com/kernelci/platform-test-parameters
Signed-off-by: Nícolas F. R. A. Prado <nfraprado(a)collabora.com>
---
Nícolas F. R. A. Prado (2):
kselftest: devices: Allow specifying boards directory through parameter
kselftest: devices: Add of-fullname-regex property
.../selftests/devices/boards/google,spherion.yaml | 4 +++
.../selftests/devices/test_discoverable_devices.py | 37 +++++++++++++++++++++-
2 files changed, 40 insertions(+), 1 deletion(-)
---
base-commit: d97496ca23a2d4ee80b7302849404859d9058bcd
change-id: 20240612-kselftest-discoverable-probe-mt8195-kci-ca21742776d0
Best regards,
--
Nícolas F. R. A. Prado <nfraprado(a)collabora.com>
xtheadvector is a custom extension that is based upon riscv vector
version 0.7.1 [1]. All of the vector routines have been modified to
support this alternative vector version based upon whether xtheadvector
was determined to be supported at boot.
vlenb is not supported on the existing xtheadvector hardware, so a
devicetree property thead,vlenb is added to provide the vlenb to Linux.
There is a new hwprobe key RISCV_HWPROBE_KEY_VENDOR_EXT_THEAD_0 that is
used to request which thead vendor extensions are supported on the
current platform. This allows future vendors to allocate hwprobe keys
for their vendor.
Support for xtheadvector is also added to the vector kselftests.
Signed-off-by: Charlie Jenkins <charlie(a)rivosinc.com>
[1] https://github.com/T-head-Semi/thead-extension-spec/blob/95358cb2cca9489361…
---
This series is a continuation of a different series that was fragmented
into two other series in an attempt to get part of it merged in the 6.10
merge window. The split-off series did not get merged due to a NAK on
the series that added the generic riscv,vlenb devicetree entry. This
series has converted riscv,vlenb to thead,vlenb to remedy this issue.
The original series is titled "riscv: Support vendor extensions and
xtheadvector" [3].
The series titled "riscv: Extend cpufeature.c to detect vendor
extensions" is still under development and this series is based on that
series! [4]
I have tested this with an Allwinner Nezha board. I ran into issues
booting the board after 6.9-rc1 so I applied these patches to 6.8. There
are a couple of minor merge conflicts that do arrise when doing that, so
please let me know if you have been able to boot this board with a 6.9
kernel. I used SkiffOS [1] to manage building the image, but upgraded
the U-Boot version to Samuel Holland's more up-to-date version [2] and
changed out the device tree used by U-Boot with the device trees that
are present in upstream linux and this series. Thank you Samuel for all
of the work you did to make this task possible.
[1] https://github.com/skiffos/SkiffOS/tree/master/configs/allwinner/nezha
[2] https://github.com/smaeul/u-boot/commit/2e89b706f5c956a70c989cd31665f1429e9…
[3] https://lore.kernel.org/all/20240503-dev-charlie-support_thead_vector_6_9-v…
[4] https://lore.kernel.org/linux-riscv/20240609-support_vendor_extensions-v2-0…
---
Changes in v2:
- Removed extraneous references to "riscv,vlenb" (Jess)
- Moved declaration of "thead,vlenb" into cpus.yaml and added
restriction that it's only applicable to thead cores (Conor)
- Check CONFIG_RISCV_ISA_XTHEADVECTOR instead of CONFIG_RISCV_ISA_V for
thead,vlenb (Jess)
- Fix naming of hwprobe variables (Evan)
- Link to v1: https://lore.kernel.org/r/20240609-xtheadvector-v1-0-3fe591d7f109@rivosinc.…
---
Charlie Jenkins (12):
dt-bindings: riscv: Add xtheadvector ISA extension description
dt-bindings: cpus: add a thead vlen register length property
riscv: dts: allwinner: Add xtheadvector to the D1/D1s devicetree
riscv: Add thead and xtheadvector as a vendor extension
riscv: vector: Use vlenb from DT for thead
riscv: csr: Add CSR encodings for VCSR_VXRM/VCSR_VXSAT
riscv: Add xtheadvector instruction definitions
riscv: vector: Support xtheadvector save/restore
riscv: hwprobe: Add thead vendor extension probing
riscv: hwprobe: Document thead vendor extensions and xtheadvector extension
selftests: riscv: Fix vector tests
selftests: riscv: Support xtheadvector in vector tests
Heiko Stuebner (1):
RISC-V: define the elements of the VCSR vector CSR
Documentation/arch/riscv/hwprobe.rst | 10 +
Documentation/devicetree/bindings/riscv/cpus.yaml | 19 ++
.../devicetree/bindings/riscv/extensions.yaml | 10 +
arch/riscv/Kconfig.vendor | 26 ++
arch/riscv/boot/dts/allwinner/sun20i-d1s.dtsi | 3 +-
arch/riscv/include/asm/cpufeature.h | 2 +
arch/riscv/include/asm/csr.h | 13 +
arch/riscv/include/asm/hwprobe.h | 4 +-
arch/riscv/include/asm/switch_to.h | 2 +-
arch/riscv/include/asm/vector.h | 249 +++++++++++++----
arch/riscv/include/asm/vendor_extensions/thead.h | 42 +++
.../include/asm/vendor_extensions/thead_hwprobe.h | 18 ++
.../include/asm/vendor_extensions/vendor_hwprobe.h | 37 +++
arch/riscv/include/uapi/asm/hwprobe.h | 3 +-
arch/riscv/include/uapi/asm/vendor/thead.h | 3 +
arch/riscv/kernel/cpufeature.c | 51 +++-
arch/riscv/kernel/kernel_mode_vector.c | 8 +-
arch/riscv/kernel/process.c | 4 +-
arch/riscv/kernel/signal.c | 6 +-
arch/riscv/kernel/sys_hwprobe.c | 5 +
arch/riscv/kernel/vector.c | 25 +-
arch/riscv/kernel/vendor_extensions.c | 10 +
arch/riscv/kernel/vendor_extensions/Makefile | 2 +
arch/riscv/kernel/vendor_extensions/thead.c | 18 ++
.../riscv/kernel/vendor_extensions/thead_hwprobe.c | 19 ++
tools/testing/selftests/riscv/vector/.gitignore | 3 +-
tools/testing/selftests/riscv/vector/Makefile | 17 +-
.../selftests/riscv/vector/v_exec_initval_nolibc.c | 93 +++++++
tools/testing/selftests/riscv/vector/v_helpers.c | 67 +++++
tools/testing/selftests/riscv/vector/v_helpers.h | 7 +
tools/testing/selftests/riscv/vector/v_initval.c | 22 ++
.../selftests/riscv/vector/v_initval_nolibc.c | 68 -----
.../selftests/riscv/vector/vstate_exec_nolibc.c | 20 +-
.../testing/selftests/riscv/vector/vstate_prctl.c | 295 ++++++++++++---------
34 files changed, 910 insertions(+), 271 deletions(-)
---
base-commit: 11cc01d4d2af304b7288251aad7e03315db8dffc
change-id: 20240530-xtheadvector-833d3d17b423
--
- Charlie