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
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
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
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
Hello,
This v2 addresses some issues observed when running the ACPI probe
kselftest proposed in v1[1] across various devices and improves the overall
reliability of the test.
The acpi-extract-ids script has been improved to:
- Parse both .c and .h files
- Add an option to print only IDs matched by a driver (i.e. defined in an
ACPI match tables or in lists of IDs provided by the drivers)
The test_unprobed_devices.sh script relies on sysfs information to
determine if a device was successfully bound to a driver. Not all devices
listed in /sys/devices are expected to have a driver folder, so the script
has been adjusted to handle these cases and avoid generating false
negatives.
The test_unprobed_devices.sh test script logic has been modified to:
- Check the status attribute (when available) to exclusively test hardware
devices that are physically present, enabled and operational
- Traverse only ACPI objects with a physical_node* link, to ensure testing
of correctly enumerated devices
- Skip devices whose HID or CID are not matched by any driver, as
determined by the list generated through the acpi-extract-ids script
- Skip devices with HID or CID listed in the ignored IDs list. This list
has been added to contain IDs of devices that don't require a driver or
cannot be represented as platform devices (e.g. ACPI container and module
devices).
- Skip devices that are natively enumerated and don't need a driver, such
as certain PCI bridges
- Skip devices unassigned to any subsystem, devices linked to other devices
and class devices
Some of the heuristics used by the script are suboptimal and might require
adjustments over time. This kind of tests would greatly benefit from a
dedicated interface that exposes information about devices expected to be
matched by drivers and their probe status. Discussion regarding this matter
was initiated in v1.
As of now, I have not identified a suitable method for exposing this
information; I plan on submitting a separate RFC to propose some options
and engage in discussion. Meanwhile, this v2 focuses on utilizing already
available information to provide an ACPI equivalent of the existing DT
kselftest [2].
Adding in CC the people involved in the discussion at Plumbers [3], feel
free to add anyone that might be interested in this.
This series depends on:
- https://lore.kernel.org/all/20240102141528.169947-1-laura.nao@collabora.com…
- https://lore.kernel.org/all/20240131-ktap-sh-helpers-extend-v1-0-98ffb46871…
Thanks,
Laura
[1] https://lore.kernel.org/all/20230925155806.1812249-2-laura.nao@collabora.co…
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/too…
[3] https://www.youtube.com/watch?v=oE73eVSyFXQ&t=9377s
Original cover letter:
Regressions that prevent a driver from probing a device can significantly
affect the functionality of a platform.
A kselftest to verify if devices on a DT-based platform are probed
correctly was recently introduced [4], but no such generic test is
available for ACPI platforms yet. bootrr [5] provides device probe
testing, but relies on a pre-defined list of the peripherals present on
each DUT.
On ACPI based hardware, a complete description of the platform is
provided to the OS by the system firmware. ACPI namespace objects are
mapped by the Linux ACPI subsystem into a device tree in
/sys/devices/LNXSYSTEM:00; the information in this subtree can be parsed
to build a list of the hw peripherals present on the DUT dynamically.
This series adds a test to verify if the devices declared in the ACPI
namespace and supported by the kernel are probed correctly.
This work follows a similar approach to [4], adapted for the ACPI use
case.
The first patch introduces a script that builds a list of all ACPI device
IDs supported by the kernel, by inspecting the acpi_device_id structs in
the sources. This list can be used to avoid testing ACPI-enumerated
devices that don't have a matching driver in the kernel. This script was
highly inspired by the dt-extract-compatibles script [6].
In the second patch, a new kselftest is added. It parses the
/sys/devices/LNXSYSTEM:00 tree to obtain a list of all platform
peripherals and verifies which of those, if supported, are correctly
bound to a driver.
Feedback is much appreciated,
Thank you,
Laura
[4] https://lore.kernel.org/all/20230828211424.2964562-1-nfraprado@collabora.co…
[5] https://github.com/kernelci/bootr
[6] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/scr…
Laura Nao (2):
acpi: Add script to extract ACPI device ids in the kernel
kselftest: Add test to detect unprobed devices on ACPI platforms
MAINTAINERS | 2 +
scripts/acpi/acpi-extract-ids | 99 +++++++++++++
tools/testing/selftests/Makefile | 1 +
tools/testing/selftests/acpi/.gitignore | 1 +
tools/testing/selftests/acpi/Makefile | 21 +++
tools/testing/selftests/acpi/id_ignore_list | 3 +
.../selftests/acpi/test_unprobed_devices.sh | 138 ++++++++++++++++++
7 files changed, 265 insertions(+)
create mode 100755 scripts/acpi/acpi-extract-ids
create mode 100644 tools/testing/selftests/acpi/.gitignore
create mode 100644 tools/testing/selftests/acpi/Makefile
create mode 100644 tools/testing/selftests/acpi/id_ignore_list
create mode 100755 tools/testing/selftests/acpi/test_unprobed_devices.sh
--
2.30.2
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 HugeTLB hugepages.
Soft-offline dissolves a hugepage, 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 addition, discarding the entire 1G memory 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 soft-offlining
HugeTLB pages: kernel only soft offlines hugepage if userspace has
opt-ed in for that specific hugepage size, and exposed to userspace
by a new sysfs entry called softoffline_corrected_errors under
/sys/kernel/mm/hugepages/hugepages-${size}kB directory:
* When softoffline_corrected_errors=0, skip soft offlining for all
hugepages of size ${size}kB.
* When softoffline_corrected_errors=1, soft offline as before this
patch series.
By default softoffline_corrected_errors is 1.
This patch set is based at
commit a52b4f11a2e1 ("selftest mm/mseal read-only elf memory segment").
Jiaqi Yan (3):
mm/memory-failure: userspace controls soft-offlining hugetlb pages
selftest/mm: test softoffline_corrected_errors behaviors
docs: hugetlbpage.rst: add softoffline_corrected_errors
Documentation/admin-guide/mm/hugetlbpage.rst | 15 +-
include/linux/hugetlb.h | 17 ++
mm/hugetlb.c | 34 +++
mm/memory-failure.c | 7 +
tools/testing/selftests/mm/.gitignore | 1 +
tools/testing/selftests/mm/Makefile | 1 +
.../selftests/mm/hugetlb-soft-offline.c | 262 ++++++++++++++++++
tools/testing/selftests/mm/run_vmtests.sh | 4 +
8 files changed, 340 insertions(+), 1 deletion(-)
create mode 100644 tools/testing/selftests/mm/hugetlb-soft-offline.c
--
2.45.1.288.g0e0cd299f1-goog