October 2024 - Linux-kselftest-mirror

[PATCH v2] selftests: net/rds: add module not found

by Alessandro Zanni

This fix solves this error, when calling kselftest with targets "net/rds": The error was found by running tests manually with the command: make kselftest TARGETS="net/rds" The patch also specifies to import ip() function from the utils module. Signed-off-by: Alessandro Zanni <alessandro.zanni87(a)gmail.com> --- Notes: v2: modified the way the parent path is added added test to reproduce the error tools/testing/selftests/net/rds/test.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/rds/test.py b/tools/testing/selftests/net/rds/test.py index e6bb109bcead..4a7178d11193 100755 --- a/tools/testing/selftests/net/rds/test.py +++ b/tools/testing/selftests/net/rds/test.py @@ -14,8 +14,11 @@ import sys import atexit from pwd import getpwuid from os import stat -from lib.py import ip +# Allow utils module to be imported from different directory +this_dir = os.path.dirname(os.path.realpath(__file__)) +sys.path.append(os.path.join(this_dir, "../")) +from lib.py.utils import ip libc = ctypes.cdll.LoadLibrary('libc.so.6') setns = libc.setns -- 2.43.0

8 months, 3 weeks

3
2
0 0

[PATCH v2] selftests: drivers: net: fix name not defined

by Alessandro Zanni

This fix solves this error, when calling kselftest with targets "drivers/net": File "tools/testing/selftests/net/lib/py/nsim.py", line 64, in __init__ if e.errno == errno.ENOSPC: NameError: name 'errno' is not defined The error was found by running tests manually with the command: make kselftest TARGETS="drivers/net" The module errno makes available standard error system symbols. Reviewed-by: Petr Machata <petrm(a)nvidia.com> Signed-off-by: Alessandro Zanni <alessandro.zanni87(a)gmail.com> --- Notes: v2: added how to run the test tools/testing/selftests/net/lib/py/nsim.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/net/lib/py/nsim.py b/tools/testing/selftests/net/lib/py/nsim.py index f571a8b3139b..1a8cbe9acc48 100644 --- a/tools/testing/selftests/net/lib/py/nsim.py +++ b/tools/testing/selftests/net/lib/py/nsim.py @@ -1,5 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 +import errno import json import os import random -- 2.43.0

8 months, 3 weeks

2
1
0 0

[PATCH v3 0/2] selftests/futex: Create test for robust list

by André Almeida

This patchset creates a selftest for the robust list interface, to track regressions and assure that the interface keeps working as expected. In this version I removed the kselftest_harness include, but I expanded the current futex selftest API a little bit with basic ASSERT_ macros to make the test easier to write and read. In the future, hopefully we can move all futex selftests to the kselftest_harness API anyway. Changes from v2: - Create ASSERT_ macros for futex selftests - Dropped kselftest_harness include, using just futex test API - This is the expected output: TAP version 13 1..6 ok 1 test_robustness ok 2 test_set_robust_list_invalid_size ok 3 test_get_robust_list_self ok 4 test_get_robust_list_child ok 5 test_set_list_op_pending ok 6 test_robust_list_multiple_elements # Totals: pass:6 fail:0 xfail:0 xpass:0 skip:0 error:0 https://lore.kernel.org/lkml/20240903134033.816500-1-andrealmeid@igalia.com André Almeida (2): selftests/futex: Add ASSERT_ macros selftests/futex: Create test for robust list .../selftests/futex/functional/.gitignore | 1 + .../selftests/futex/functional/Makefile | 3 +- .../selftests/futex/functional/robust_list.c | 512 ++++++++++++++++++ .../testing/selftests/futex/include/logging.h | 28 + 4 files changed, 543 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/futex/functional/robust_list.c -- 2.46.0

8 months, 3 weeks

2
3
0 0

[PATCH] selftests/ftrace: Fix check of return value in fgraph-retval.tc test

by Steven Rostedt

From: Steven Rostedt <rostedt(a)goodmis.org> The addition of recording both the function name and return address to the function graph tracer updated the selftest to check for "=-5" from "= -5". But this causes the test to fail on certain configs, as "= -5" is still a value that can be returned if function addresses are not enabled (older kernels). Check for both "=-5" and " -5" as a success value. Fixes: 21e92806d39c6 ("function_graph: Support recording and printing the function return address") Signed-off-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> --- Shuah, this update is only for changes in my tree, so you do not need to add it. tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc b/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc index e8e46378b88d..4307d4eef417 100644 --- a/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc +++ b/tools/testing/selftests/ftrace/test.d/ftrace/fgraph-retval.tc @@ -29,7 +29,7 @@ set -e : "Test printing the error code in signed decimal format" echo 0 > options/funcgraph-retval-hex -count=`cat trace | grep 'proc_reg_write' | grep '=-5' | wc -l` +count=`cat trace | grep 'proc_reg_write' | grep -e '=-5 ' -e '= -5 ' | wc -l` if [ $count -eq 0 ]; then fail "Return value can not be printed in signed decimal format" fi -- 2.45.2

8 months, 3 weeks

2
1
0 0

[PATCH net-next v25 00/13] Device Memory TCP

by Mina Almasry

v25: https://patchwork.kernel.org/project/netdevbpf/list/?series=885396&state=* === Major changes: - Moved devmem.h and mp_dmabuf_devmem.h to internal header files. - Changed the page_pool_params to take in a queue_idx rather than a struct netdev_rx_queue. - Added WARN_ON_ONCE around __skb_checksum readability check and added check to skb_checksum_help(). Other more minor feedback addressed as well. v24: https://patchwork.kernel.org/project/netdevbpf/list/?series=884556&state=* ==== No major changes. Mostly addressing issues in the error paths of dmabuf binding, and code cleanups/improvements from reviewers: Changes: - Fix failing ynl regen error. - Error path fixes & extack error messages in dmabuf binding. - Code cleanup in introspection. - gitignore ynl.d generated file. Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v24/ v23: https://patchwork.kernel.org/project/netdevbpf/list/?series=882978&state=* ==== Fixing relatively minor issues called out in v22. (thanks again!) Mostly code cleanups, extack error messages, and minor reworks. Nothing major really changed, so the exact changes per commit is called in the commit messages. Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v23/ v22: https://patchwork.kernel.org/project/netdevbpf/list/?series=881158&state=* ==== v22 aims to resolve the pending issue pointed to in v21, which is the interaction with xdp. In this series I rebase on top of the minor refactor which refactors propagating xdp configuration to slave devices: https://patchwork.kernel.org/project/netdevbpf/list/?series=881994&state=* I then disable setting xdp on devices using memory providers, and propagating xdp configuration to devices using memory providers. Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v22/ v21: https://patchwork.kernel.org/project/netdevbpf/list/?series=880735&state=* ==== v20 addressed some comments and resolved a test failure, but introduced an unfortunate build error with a config edge case I wasn't testing. v21 simply resolves that error. Major Changes: - Resolve build error with CONFIG_PAGE_POOL=n && CONFIG_NET=y Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v21/ v20: https://patchwork.kernel.org/project/netdevbpf/list/?series=879373&state=* ==== v20 aims to resolve a couple of bug reports against v19, and addresses some review comments around the page_pool_check_memory_provider mechanism. Major changes: - Test edge cases such as header split disabled in selftest. - Change `offset = 0` back to `offset = offset - start` to resolve issue found in RX path by Taehee (thanks!) - Address a few comments around page_pool_check_memory_provider() from Pavel & Jakub. - Removed some unnecessary includes across various patches in the series. - Removed unnecessary EXPORT_SYMBOL(page_pool_mem_providers) (Jakub). - Fix regression caused by incorrect dev_get_max_mp_channel check, along with rename (Jakub). Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v20/ v19: https://patchwork.kernel.org/project/netdevbpf/list/?series=876852&state=* ==== v18 got a thorough review (thanks!), and this iteration addresses the feedback. Major changes: - Prevent deactivating mp bound queues. - Prevent installing xdp on mp bound netdevs, or installing mps on xdp installed netdevs. - Fix corner cases in netlink API vis-a-vis missing attributes. - Iron out the unreadable netmem driver support story. To be honest, the conversation with Jakub & Pavel got a bit confusing for me. I've implemented an approach in this set that makes sense to me, and AFAICT, addresses the requirements. It may be good as-is, or it may be a conversation starter/continuer. To be honest IMO there are many ways to skin this cat and I don't see an extremely strong reason to go for one approach over another. Here is one approach you may like. - Don't reset niov dma_addr on allocation & free. - Add some tests to the selftest that catches some of the issues around missing netlink attributes or deactivating mp-bound queues. Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v19/ v18: https://patchwork.kernel.org/project/netdevbpf/list/?series=874848&state=* ==== v17 got minor feedback: (a) to beef up the description on patch 1 and (b) to remove the leading underscores in the header definition. I applied (a). (b) seems to be against current conventions so I did not apply before further discussion. Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v17/ v17: https://patchwork.kernel.org/project/netdevbpf/list/?series=869900&state=* ==== v16 also got a very thorough review and some testing (thanks again!). Thes version addresses all the concerns reported on v15, in terms of feedback and issues reported. Major changes: - Use ASSERT_RTNL. - Moved around some of the page_pool helpers definitions so I can hide some netmem helpers in private files as Jakub suggested. - Don't make every net_iov hold a ref on the binding as Jakub suggested. - Fix issue reported by Taehee where we access queues after they have been freed. Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v17/ v16: https://patchwork.kernel.org/project/netdevbpf/list/?series=866353&state=* ==== v15 got a thorough review and some testing, and this version addresses almost all the feedback. Some more minor comments where the authors said it could be done later, I left out. Major changes: - Addition of dma-buf introspection to page-pool-get and queue-get. - Fixes to selftests suggested by Taehee. - Fixes to documentation suggested by Donald. - A couple of suggestions and fixes to TCP patches by Eric and David. - Fixes to number assignements suggested by Arnd. - Use rtnl_lock()ing to guard against queue reconfiguration while the page_pool initialization is happening. (Jakub). - Fixes to a few warnings reproduced by Taehee. - Fixes to dma-buf binding suggested by Taehee and Jakub. - Fixes to netlink UAPI suggested by Jakub - Applied a number of Reviewed-bys and Acked-bys (including ones I lost from v13+). Full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v16/ One caveat: Taehee reproduced a KASAN warning and reported it here: https://lore.kernel.org/netdev/CAMArcTUdCxOBYGF3vpbq=eBvqZfnc44KBaQTN7H-wqd… I estimate the issue to be minor and easily fixable: https://lore.kernel.org/netdev/CAHS8izNgaqC--GGE2xd85QB=utUnOHmioCsDd1TNxJW… I hope to be able to follow up with a fix to net tree as net-next closes imminently, but if this iteration doesn't make it in, I will repost with a fix squashed after net-next reopens, no problem. v15: https://patchwork.kernel.org/project/netdevbpf/list/?series=865481&state=* ==== No material changes in this version, only a fix to linking against libynl.a from the last version. Per Jakub's instructions I've pulled one of his patches into this series, and now use the new libynl.a correctly, I hope. As usual, the full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v15/ 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. 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> 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> Cc: Taehee Yoo <ap420073(a)gmail.com> Cc: Donald Hunter <donald.hunter(a)gmail.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: 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 netdev: add dmabuf introspection Documentation/netlink/specs/netdev.yaml | 61 +++ Documentation/networking/devmem.rst | 269 +++++++++++ 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/netdevice.h | 2 + include/linux/skbuff.h | 61 ++- include/linux/skbuff_ref.h | 9 +- include/linux/socket.h | 1 + include/net/netdev_rx_queue.h | 5 + include/net/netmem.h | 132 +++++- include/net/page_pool/helpers.h | 39 +- include/net/page_pool/types.h | 23 +- include/net/sock.h | 2 + include/net/tcp.h | 3 +- include/trace/events/page_pool.h | 12 +- include/uapi/asm-generic/socket.h | 6 + include/uapi/linux/netdev.h | 13 + include/uapi/linux/uio.h | 17 + net/Kconfig | 5 + net/core/Makefile | 2 + net/core/datagram.c | 6 + net/core/dev.c | 33 +- net/core/devmem.c | 389 ++++++++++++++++ net/core/devmem.h | 180 ++++++++ net/core/gro.c | 3 +- net/core/mp_dmabuf_devmem.h | 44 ++ net/core/netdev-genl-gen.c | 23 + net/core/netdev-genl-gen.h | 6 + net/core/netdev-genl.c | 139 +++++- net/core/netdev_rx_queue.c | 81 ++++ net/core/netmem_priv.h | 31 ++ net/core/page_pool.c | 120 +++-- net/core/page_pool_priv.h | 46 ++ net/core/page_pool_user.c | 32 +- net/core/skbuff.c | 77 +++- net/core/sock.c | 68 +++ net/ethtool/common.c | 8 + net/ipv4/esp4.c | 3 +- net/ipv4/tcp.c | 263 ++++++++++- 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 +- net/xdp/xsk_buff_pool.c | 5 + tools/include/uapi/linux/netdev.h | 13 + tools/net/ynl/lib/.gitignore | 1 + tools/testing/selftests/net/.gitignore | 1 + tools/testing/selftests/net/Makefile | 9 + tools/testing/selftests/net/ncdevmem.c | 570 ++++++++++++++++++++++++ 54 files changed, 2757 insertions(+), 124 deletions(-) create mode 100644 Documentation/networking/devmem.rst create mode 100644 net/core/devmem.c create mode 100644 net/core/devmem.h create mode 100644 net/core/mp_dmabuf_devmem.h create mode 100644 net/core/netdev_rx_queue.c create mode 100644 net/core/netmem_priv.h create mode 100644 tools/testing/selftests/net/ncdevmem.c -- 2.46.0.469.g59c65b2a67-goog

8 months, 3 weeks

5
27
0 0

[PATCH net-next v02 0/2] net: af_packet: allow joining a fanout when link is down

by Gur Stavi

PACKET socket can retain its fanout membership through link down and up and leave a fanout while closed regardless of link state. However, socket was forbidden from joining a fanout while it was not RUNNING. This patch allows PACKET socket to join a fanout while not RUNNING. Selftest psock_fanout is extended to test this scenario. This is the only test that was performed. This scenario was identified while studying DPDK pmd_af_packet_drv. Since sockets are only created during initialization, there is no reason to fail the initialization if a single link is temporarily down. I hope it is not considered as breaking user space and that applications are not designed to expect this failure. Changes: V02: * psock_fanout: use explicit loopback up/down instead of toggle. * psock_fanout: don't try to restore loopback state on failure. * Rephrase commit message about "leaving a fanout". V01: https://lore.kernel.org/netdev/cover.1728303615.git.gur.stavi@huawei.com/ Gur Stavi (2): af_packet: allow fanout_add when socket is not RUNNING selftests: net/psock_fanout: socket joins fanout when link is down net/packet/af_packet.c | 10 +++--- tools/testing/selftests/net/psock_fanout.c | 42 ++++++++++++++++++++-- 2 files changed, 44 insertions(+), 8 deletions(-) base-commit: f95b4725e796b12e5f347a0d161e1d3843142aa8 -- 2.45.2

8 months, 3 weeks

2
18
0 0

[PATCH net-next 2/2] selftests: drv-net: rss_ctx: add rss ctx busy testcase

by Daniel Zahka

It should be invalid to delete an rss context while it is being referenced from an ntuple filter. ethtool core should prevent this from happening. This patch adds a testcase to verify this behavior. Signed-off-by: Daniel Zahka <daniel.zahka(a)gmail.com> --- .../selftests/drivers/net/hw/rss_ctx.py | 32 +++++++++++++++++-- 1 file changed, 30 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/drivers/net/hw/rss_ctx.py b/tools/testing/selftests/drivers/net/hw/rss_ctx.py index 9d7adb3cf33b..29995586993c 100755 --- a/tools/testing/selftests/drivers/net/hw/rss_ctx.py +++ b/tools/testing/selftests/drivers/net/hw/rss_ctx.py @@ -6,7 +6,7 @@ import random from lib.py import ksft_run, ksft_pr, ksft_exit, ksft_eq, ksft_ne, ksft_ge, ksft_lt from lib.py import NetDrvEpEnv from lib.py import EthtoolFamily, NetdevFamily -from lib.py import KsftSkipEx +from lib.py import KsftSkipEx, KsftFailEx from lib.py import rand_port from lib.py import ethtool, ip, defer, GenerateTraffic, CmdExitFailure @@ -606,6 +606,33 @@ def test_rss_context_overlap2(cfg): test_rss_context_overlap(cfg, True) +def test_delete_rss_context_busy(cfg): + """ + Test that deletion returns -EBUSY when an rss context is being used + by an ntuple filter. + """ + + require_ntuple(cfg) + + # create additional rss context + ctx_id = ethtool_create(cfg, "-X", "context new") + ctx_deleter = defer(ethtool, f"-X {cfg.ifname} context {ctx_id} delete") + + # utilize context from ntuple filter + port = rand_port() + flow = f"flow-type tcp{cfg.addr_ipver} dst-port {port} context {ctx_id}" + ntuple_id = ethtool_create(cfg, "-N", flow) + defer(ethtool, f"-N {cfg.ifname} delete {ntuple_id}") + + # attempt to delete in-use context + try: + ctx_deleter.exec_only() + ctx_deleter.cancel() + raise KsftFailEx(f"deleted context {ctx_id} used by rule {ntuple_id}") + except CmdExitFailure: + pass + + def main() -> None: with NetDrvEpEnv(__file__, nsim_test=False) as cfg: cfg.ethnl = EthtoolFamily() @@ -616,7 +643,8 @@ def main() -> None: test_rss_context, test_rss_context4, test_rss_context32, test_rss_context_dump, test_rss_context_queue_reconfigure, test_rss_context_overlap, test_rss_context_overlap2, - test_rss_context_out_of_order, test_rss_context4_create_with_cfg], + test_rss_context_out_of_order, test_rss_context4_create_with_cfg, + test_delete_rss_context_busy], args=(cfg, )) ksft_exit() -- 2.43.5

8 months, 3 weeks

1
0
0 0

[PATCH,bpf-next v3 0/4] selftests/bpf: migrate and remove cgroup/tracing related tests

by Daniel T. Lee

The BPF testing framework has evolved significantly over time. However, some legacy tests in the samples/bpf directory have not kept up with these changes. These outdated tests can cause confusion and increase maintenance efforts. This patchset focuses on migrating outdated cgroup and tracing-related tests from samples/bpf to selftests/bpf, ensuring the BPF test suite remains current and efficient. Tests that are already covered by selftests/bpf are removed, while those not yet covered are migrated. This includes cgroup sock create tests for setting socket attributes and blocking socket creation, as well as the removal of redundant cgroup and tracing tests that have been replaced by newer tests. This patchset covers the following cgroup/tracing tests: - test_overhead: tests the overhead of BPF programs with task_rename, now covered by selftests and benchmark tests (rename-*). [1] - test_override_return: tests the return override functionality, now handled by kprobe_multi_override in selftests. - test_probe_write_user: tests the probe_write_user functionality, now replaced by the probe_user test in selftests. - test_cgrp2_sock: tests cgroup BPF's ability to set sk_bound_dev_if, mark, and priority during socket creation. Migrated to selftests as 'sock_create' since no existing tests fully cover this. - test_cgrp2_sock2: tests blocking socket creation for specific types (AF_INET{6}, SOCK_DGRAM, IPPROTO_ICMP{V6}). Migrated to selftests in 'sock_create' test for coverage. - test_current_task_under_cgroup: tests bpf_current_task_under_cgroup() to check if a task belongs to a cgroup. Already covered by task_under_cgroup at selftest and other cgroup ID tests. - test_cgrp2_tc: tests bpf_skb_under_cgroup() to filter packets based on cgroup. This behavior is now validated by cgroup_skb_sk_lookup, which uses bpf_skb_cgroup_id, making this test redundant. [1]: https://patchwork.kernel.org/cover/13759916 --- Changes in v2: - commit message fix Changes in v3: - Makefile fix Daniel T. Lee (4): selftests/bpf: migrate cgroup sock create test for setting iface/mark/prio selftests/bpf: migrate cgroup sock create test for prohibiting sockets samples/bpf: remove obsolete cgroup related tests samples/bpf: remove obsolete tracing related tests samples/bpf/Makefile | 25 -- samples/bpf/sock_flags.bpf.c | 47 --- samples/bpf/test_cgrp2_array_pin.c | 106 ------ samples/bpf/test_cgrp2_attach.c | 177 ---------- samples/bpf/test_cgrp2_sock.c | 296 ---------------- samples/bpf/test_cgrp2_sock.sh | 137 ------- samples/bpf/test_cgrp2_sock2.c | 95 ----- samples/bpf/test_cgrp2_sock2.sh | 103 ------ samples/bpf/test_cgrp2_tc.bpf.c | 56 --- samples/bpf/test_cgrp2_tc.sh | 187 ---------- .../bpf/test_current_task_under_cgroup.bpf.c | 43 --- .../bpf/test_current_task_under_cgroup_user.c | 115 ------ samples/bpf/test_overhead_kprobe.bpf.c | 41 --- samples/bpf/test_overhead_raw_tp.bpf.c | 17 - samples/bpf/test_overhead_tp.bpf.c | 23 -- samples/bpf/test_overhead_user.c | 225 ------------ samples/bpf/test_override_return.sh | 16 - samples/bpf/test_probe_write_user.bpf.c | 52 --- samples/bpf/test_probe_write_user_user.c | 108 ------ samples/bpf/tracex7.bpf.c | 15 - samples/bpf/tracex7_user.c | 56 --- .../selftests/bpf/prog_tests/sock_create.c | 333 ++++++++++++++++++ 22 files changed, 333 insertions(+), 1940 deletions(-) delete mode 100644 samples/bpf/sock_flags.bpf.c delete mode 100644 samples/bpf/test_cgrp2_array_pin.c delete mode 100644 samples/bpf/test_cgrp2_attach.c delete mode 100644 samples/bpf/test_cgrp2_sock.c delete mode 100755 samples/bpf/test_cgrp2_sock.sh delete mode 100644 samples/bpf/test_cgrp2_sock2.c delete mode 100755 samples/bpf/test_cgrp2_sock2.sh delete mode 100644 samples/bpf/test_cgrp2_tc.bpf.c delete mode 100755 samples/bpf/test_cgrp2_tc.sh delete mode 100644 samples/bpf/test_current_task_under_cgroup.bpf.c delete mode 100644 samples/bpf/test_current_task_under_cgroup_user.c delete mode 100644 samples/bpf/test_overhead_kprobe.bpf.c delete mode 100644 samples/bpf/test_overhead_raw_tp.bpf.c delete mode 100644 samples/bpf/test_overhead_tp.bpf.c delete mode 100644 samples/bpf/test_overhead_user.c delete mode 100755 samples/bpf/test_override_return.sh delete mode 100644 samples/bpf/test_probe_write_user.bpf.c delete mode 100644 samples/bpf/test_probe_write_user_user.c delete mode 100644 samples/bpf/tracex7.bpf.c delete mode 100644 samples/bpf/tracex7_user.c create mode 100644 tools/testing/selftests/bpf/prog_tests/sock_create.c -- 2.43.0

8 months, 3 weeks

2
5
0 0

[PATCH bpf-next v3 0/3] selftests/bpf: add coverage for xdp_features in test_progs

by Alexis Lothoré (eBPF Foundation)

Hello, this small series aims to increase coverage of xdp features in test_progs. The initial versions proposed to rework test_xdp_features.sh to make it fit in test_progs, but some discussions in v1 and v2 showed that the script is still needed as a standalone tool. So this new revision lets test_xdp_features.sh as-is, and rather adds missing coverage in existing test (cpu map). The new revision is now also a follow-up to the update performed by Florian Kauer in [1] for devmap programs testing. [1] https://lore.kernel.org/bpf/20240911-devel-koalo-fix-ingress-ifindex-v4-2-5… --- Changes in v3: - Drop xdp_features rework commit - update xdp_cpumap_attach to extend its coverage - Link to v2: https://lore.kernel.org/r/20240910-convert_xdp_tests-v2-1-a46367c9d038@boot… Changes in v2: - fix endianness management in userspace packet parsing (call htonl on constant rather than packet part) The new test has been run in a local x86 environment and in CI: #560/1 xdp_cpumap_attach/CPUMAP with programs in entries:OK #560/2 xdp_cpumap_attach/CPUMAP with frags programs in entries:OK #560/3 xdp_cpumap_attach/CPUMAP attach with programs in entries on veth:OK #560 xdp_cpumap_attach:OK Summary: 1/3 PASSED, 0 SKIPPED, 0 FAILED --- Alexis Lothoré (eBPF Foundation) (3): selftests/bpf: fix bpf_map_redirect call for cpu map test selftests/bpf: make xdp_cpumap_attach keep redirect prog attached selftests/bpf: check program redirect in xdp_cpumap_attach .../selftests/bpf/prog_tests/xdp_cpumap_attach.c | 130 +++++++++++++++++++-- .../bpf/progs/test_xdp_with_cpumap_helpers.c | 7 +- 2 files changed, 129 insertions(+), 8 deletions(-) --- base-commit: 058d7c3d1691e2e4a4963716ec6c047dff778637 change-id: 20240730-convert_xdp_tests-ccd66bfe33db Best regards, -- Alexis Lothoré, Bootlin Embedded Linux and Kernel engineering https://bootlin.com

8 months, 3 weeks

4
6
0 0

[RFC PATCH 0/3] Allow sk_lookup UDP return traffic to egress.

by Tiago Lam

Currently, sk_lookup allows an ebpf program to run on the ingress socket lookup path, and accept traffic not only on a range of addresses, but also on a range of ports. At Cloudflare we use sk_lookup for two main cases: 1. Sharing a single port between multiple services - i.e. two services (or more) use disjoint IP ranges but share the same port; 2. Receiving traffic on all ports - i.e. a service which accepts traffic on specific IP ranges but any port [1]. However, one main challenge we face while using sk_lookup for these use cases is how to source return UDP traffic: - On point 1. above, sometimes this range of addresses are not local (i.e. there's no local routes for these in the server), which means we need IP_TRANSPARENT set to be able to egress traffic from addresses we've received traffic on (or simply IP_FREEBIND in the case of IPv6); - And on point 2. above, allowing traffic to a range of ports means a service could get traffic on multiple ports, but currently there's no way to set the source UDP port egress traffic should be sourced from - it's possible to receive the original destination port using the IP_ORIGDSTADDR ancilliary message in recvmsg, but not set it in sendmsg. Both of these limitations can be worked around, but in a sub-optimal way. Using IP_TRANSPARENT, for instance, requires special privileges. And while one could use UDP connected sockets to send return traffic, creating a connected socket for each different address a UDP traffic is received on does have performance implications. Given sk_lookup allows services to accept traffic on a range of addresses or ports, it seems sensible to also allow return traffic to proceed through as well, without needing extra configurations / set ups. This patch set allows to do exactly this by performing a reverse socket lookup on the egress path - where it looks to see if the egress socket matches a socket in the attached sk_lookup ebpf program for the traffic that's being sent. If it does, traffic is allowed to proceed. The downsides to this is that this runs on the egress hot path, although this work tries to minimise its impact by only performing the reverse socket lookup when necessary. Further performance measurements are to be taken, but we're reaching out early for feedback to see what the technical concerns are and if we can address them. [1] https://blog.cloudflare.com/how-we-built-spectrum/ Suggested-by: Jakub Sitnicki <jakub(a)cloudflare.com> Signed-off-by: Tiago Lam <tiagolam(a)cloudflare.com> --- Tiago Lam (3): ipv4: Run a reverse sk_lookup on sendmsg. ipv6: Run a reverse sk_lookup on sendmsg. bpf: Add sk_lookup test to use ORIGDSTADDR cmsg. include/net/ip.h | 1 + net/ipv4/ip_sockglue.c | 11 ++++ net/ipv4/udp.c | 33 +++++++++- net/ipv6/datagram.c | 76 ++++++++++++++++++++++ net/ipv6/udp.c | 8 ++- tools/testing/selftests/bpf/prog_tests/sk_lookup.c | 70 +++++++++++++------- 6 files changed, 174 insertions(+), 25 deletions(-) --- base-commit: da3ea35007d0af457a0afc87e84fddaebc4e0b63 change-id: 20240909-reverse-sk-lookup-f7bf36292bc4 Best regards, -- Tiago Lam <tiagolam(a)cloudflare.com>

8 months, 3 weeks

6
15
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror October 2024