January 2025 - Linux-kselftest-mirror

[PATCH net-next v9 0/6] Suspend IRQs during application busy periods

by Joe Damato

Greetings: Welcome to v9, see changelog below. This revision addresses feedback Willem gave on the selftests. No functional or code changes to the implementation were made and performance tests were not re-run. This series introduces a new mechanism, IRQ suspension, which allows network applications using epoll to mask IRQs during periods of high traffic while also reducing tail latency (compared to existing mechanisms, see below) during periods of low traffic. In doing so, this balances CPU consumption with network processing efficiency. Martin Karsten (CC'd) and I have been collaborating on this series for several months and have appreciated the feedback from the community on our RFC [1]. We've updated the cover letter and kernel documentation in an attempt to more clearly explain how this mechanism works, how applications can use it, and how it compares to existing mechanisms in the kernel. I briefly mentioned this idea at netdev conf 2024 (for those who were there) and Martin described this idea in an earlier paper presented at Sigmetrics 2024 [2]. ~ The short explanation (TL;DR) We propose adding a new napi config parameter: irq_suspend_timeout to help balance CPU usage and network processing efficiency when using IRQ deferral and napi busy poll. If this parameter is set to a non-zero value *and* a user application has enabled preferred busy poll on a busy poll context (via the EPIOCSPARAMS ioctl introduced in commit 18e2bf0edf4d ("eventpoll: Add epoll ioctl for epoll_params")), then application calls to epoll_wait for that context will cause device IRQs and softirq processing to be suspended as long as epoll_wait successfully retrieves data from the NAPI. Each time data is retrieved, the irq_suspend_timeout is deferred. If/when network traffic subsides and epoll_wait returns no data, IRQ suspension is immediately reverted back to the existing napi_defer_hard_irqs and gro_flush_timeout mechanism which was introduced in commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature")). The irq_suspend_timeout serves as a safety mechanism. If userland takes a long time processing data, irq_suspend_timeout will fire and restart normal NAPI processing. For a more in depth explanation, please continue reading. ~ Comparison with existing mechanisms Interrupt mitigation can be accomplished in napi software, by setting napi_defer_hard_irqs and gro_flush_timeout, or via interrupt coalescing in the NIC. This can be quite efficient, but in both cases, a fixed timeout (or packet count) needs to be configured. However, a fixed timeout cannot effectively support both low- and high-load situations: At low load, an application typically processes a few requests and then waits to receive more input data. In this scenario, a large timeout will cause unnecessary latency. At high load, an application typically processes many requests before being ready to receive more input data. In this case, a small timeout will likely fire prematurely and trigger irq/softirq processing, which interferes with the application's execution. This causes overhead, most likely due to cache contention. While NICs attempt to provide adaptive interrupt coalescing schemes, these cannot properly take into account application-level processing. An alternative packet delivery mechanism is busy-polling, which results in perfect alignment of application processing and network polling. It delivers optimal performance (throughput and latency), but results in 100% cpu utilization and is thus inefficient for below-capacity workloads. We propose to add a new packet delivery mode that properly alternates between busy polling and interrupt-based delivery depending on busy and idle periods of the application. During a busy period, the system operates in busy-polling mode, which avoids interference. During an idle period, the system falls back to interrupt deferral, but with a small timeout to avoid excessive latencies. This delivery mode can also be viewed as an extension of basic interrupt deferral, but alternating between a small and a very large timeout. This delivery mode is efficient, because it avoids softirq execution interfering with application processing during busy periods. It can be used with blocking epoll_wait to conserve cpu cycles during idle periods. The effect of alternating between busy and idle periods is that performance (throughput and latency) is very close to full busy polling, while cpu utilization is lower and very close to interrupt mitigation. ~ Usage details IRQ suspension is introduced via a per-NAPI configuration parameter that controls the maximum time that IRQs can be suspended. Here's how it is intended to work: - The user application (or system administrator) uses the netdev-genl netlink interface to set the pre-existing napi_defer_hard_irqs and gro_flush_timeout NAPI config parameters to enable IRQ deferral. - The user application (or system administrator) sets the proposed irq_suspend_timeout parameter via the netdev-genl netlink interface to a larger value than gro_flush_timeout to enable IRQ suspension. - The user application issues the existing epoll ioctl to set the prefer_busy_poll flag on the epoll context. - The user application then calls epoll_wait to busy poll for network events, as it normally would. - If epoll_wait returns events to userland, IRQs are suspended for the duration of irq_suspend_timeout. - If epoll_wait finds no events and the thread is about to go to sleep, IRQ handling using napi_defer_hard_irqs and gro_flush_timeout is resumed. As long as epoll_wait is retrieving events, IRQs (and softirq processing) for the NAPI being polled remain disabled. When network traffic reduces, eventually a busy poll loop in the kernel will retrieve no data. When this occurs, regular IRQ deferral using gro_flush_timeout for the polled NAPI is re-enabled. Unless IRQ suspension is continued by subsequent calls to epoll_wait, it automatically times out after the irq_suspend_timeout timer expires. Regular deferral is also immediately re-enabled when the epoll context is destroyed. ~ Usage scenario The target scenario for IRQ suspension as packet delivery mode is a system that runs a dominant application with substantial network I/O. The target application can be configured to receive input data up to a certain batch size (via epoll_wait maxevents parameter) and this batch size determines the worst-case latency that application requests might experience. Because packet delivery is suspended during the target application's processing, the batch size also determines the worst-case latency of concurrent applications using the same RX queue(s). gro_flush_timeout should be set as small as possible, but large enough to make sure that a single request is likely not being interfered with. irq_suspend_timeout is largely a safety mechanism against misbehaving applications. It should be set large enough to cover the processing of an entire application batch, i.e., the factor between gro_flush_timeout and irq_suspend_timeout should roughly correspond to the maximum batch size that the target application would process in one go. ~ Important call out in the implementation - Enabling per epoll-context preferred busy poll will now effectively lead to a nonblocking iteration through napi_busy_loop, even when busy_poll_usecs is 0. See patch 4. ~ Benchmark configs & descriptions The changes were benchmarked with memcached [3] using the benchmarking tool mutilate [4]. To facilitate benchmarking, a small patch [5] was applied to memcached 1.6.29 to allow setting per-epoll context preferred busy poll and other settings via environment variables. Another small patch [6] was applied to libevent to enable full busy-polling. Multiple scenarios were benchmarked as described below and the scripts used for producing these results can be found on github [7] (note: all scenarios use NAPI-based traffic splitting via SO_INCOMING_ID by passing -N to memcached): - base: - no other options enabled - deferX: - set defer_hard_irqs to 100 - set gro_flush_timeout to X,000 - napibusy: - set defer_hard_irqs to 100 - set gro_flush_timeout to 200,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 64, busy_poll_budget = 64, prefer_busy_poll = true) - fullbusy: - set defer_hard_irqs to 100 - set gro_flush_timeout to 5,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 1000, busy_poll_budget = 64, prefer_busy_poll = true) - change memcached's nonblocking epoll_wait invocation (via libevent) to using a 1 ms timeout - suspend0: - set defer_hard_irqs to 0 - set gro_flush_timeout to 0 - set irq_suspend_timeout to 20,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 0, busy_poll_budget = 64, prefer_busy_poll = true) - suspendX: - set defer_hard_irqs to 100 - set gro_flush_timeout to X,000 - set irq_suspend_timeout to 20,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 0, busy_poll_budget = 64, prefer_busy_poll = true) ~ Benchmark results Tested on: Single socket AMD EPYC 7662 64-Core Processor Hyperthreading disabled 4 NUMA Zones (NPS=4) 16 CPUs per NUMA zone (64 cores total) 2 x Dual port 100gbps Mellanox Technologies ConnectX-5 Ex EN NIC The test machine is configured such that a single interface has 8 RX queues. The queues' IRQs and memcached are pinned to CPUs that are NUMA-local to the interface which is under test. The NIC's interrupt coalescing configuration is left at boot-time defaults. Results: Results are shown below. The mechanism added by this series is represented by the 'suspend' cases. Data presented shows a summary over nearly 10 runs of each test case [8] using the scripts on github [7]. For latency, the median is shown. For throughput and CPU utilization, the average is shown. The results also include cycles-per-query (cpq) and instruction-per-query (ipq) metrics, following the methodology proposed in [2], to augment the CPU utilization numbers, which could be skewed due to frequency scaling. We find that this does not appear to be the case as CPU utilization and low-level metrics show similar trends. These results were captured using the scripts on github [7] to illustrate how this approach compares with other pre-existing mechanisms. This data is not to be interpreted as scientific data captured in a fully isolated lab setting, but instead as best effort, illustrative information comparing and contrasting tradeoffs. The absolute QPS results shift between submissions, but the relative differences are equivalent. As patches are rebased, several factors likely influence overall performance. Compare: - Throughput (MAX) and latencies of base vs suspend. - CPU usage of napibusy and fullbusy during lower load (200K, 400K for example) vs suspend. - Latency of the defer variants vs suspend as timeout and load increases. - suspend0, which sets defer_hard_irqs and gro_flush_timeout to 0, has nearly the same performance as the base case (this is FAQ item #1). The overall takeaway is that the suspend variants provide a superior combination of high throughput, low latency, and low cpu utilization compared to all other variants. Each of the suspend variants works very well, but some fine-tuning between latency and cpu utilization is still possible by tuning the small timeout (gro_flush_timeout). Note: we've reorganized the results to make comparison among testcases with the same load easier. testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 200K 199946 112 239 416 26 12973 11343 defer10 200K 199971 54 124 142 29 19412 17460 defer20 200K 199986 60 130 153 26 15644 14095 defer50 200K 200025 79 144 182 23 12122 11632 defer200 200K 199999 164 254 309 19 8923 9635 fullbusy 200K 199998 46 118 133 100 43658 23133 napibusy 200K 199983 100 237 277 56 24840 24716 suspend0 200K 200020 105 249 432 30 14264 11796 suspend10 200K 199950 53 123 141 32 19518 16903 suspend20 200K 200037 58 126 151 30 16426 14736 suspend50 200K 199961 73 136 177 26 13310 12633 suspend200 200K 199998 149 251 306 21 9566 10203 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 400K 400014 139 269 707 41 9476 9343 defer10 400K 400016 59 133 166 53 13991 12989 defer20 400K 399952 67 140 172 47 12063 11644 defer50 400K 400007 87 162 198 39 9384 9880 defer200 400K 399979 181 274 330 31 7089 8430 fullbusy 400K 399987 50 123 156 100 21827 16037 napibusy 400K 400014 76 222 272 83 18185 16529 suspend0 400K 400015 127 350 776 47 10699 9603 suspend10 400K 400023 57 129 164 54 13758 13178 suspend20 400K 400043 62 135 169 49 12071 11826 suspend50 400K 400071 76 149 186 42 10011 10301 suspend200 400K 399961 154 269 327 34 7827 8774 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 600K 599951 149 266 574 61 9265 8876 defer10 600K 600006 71 147 203 76 11866 10936 defer20 600K 600123 76 152 203 66 10430 10342 defer50 600K 600162 95 172 217 54 8526 9142 defer200 600K 599942 200 301 357 46 6977 8212 fullbusy 600K 599990 55 127 177 100 14551 13983 napibusy 600K 600035 63 160 250 96 13937 14140 suspend0 600K 599903 127 320 732 68 10166 8963 suspend10 600K 599908 63 137 192 69 10902 11100 suspend20 600K 599961 66 141 194 65 9976 10370 suspend50 600K 599973 80 159 204 57 8678 9381 suspend200 600K 600010 157 277 346 48 7133 8381 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 800K 800039 181 300 536 87 9585 8304 defer10 800K 800038 181 530 939 96 10564 8970 defer20 800K 800029 112 225 329 90 10056 8935 defer50 800K 799999 120 208 296 82 9234 8562 defer200 800K 800066 227 338 401 63 7117 8129 fullbusy 800K 800040 61 134 190 100 10913 12608 napibusy 800K 799944 64 141 214 99 10828 12588 suspend0 800K 799911 126 248 509 85 9346 8498 suspend10 800K 800006 69 143 200 83 9410 9845 suspend20 800K 800120 74 150 207 78 8786 9454 suspend50 800K 799989 87 168 224 71 7946 8833 suspend200 800K 799987 160 292 357 62 6923 8229 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 1000K 906879 4079 5751 6216 98 9496 7904 defer10 1000K 860849 3643 6274 6730 99 10040 8676 defer20 1000K 896063 3298 5840 6349 98 9620 8237 defer50 1000K 919782 2962 5513 5807 97 9284 7951 defer200 1000K 970941 3059 5348 5984 95 8593 7959 fullbusy 1000K 999950 70 150 207 100 8732 10777 napibusy 1000K 999996 78 154 223 100 8722 10656 suspend0 1000K 949706 2666 5770 6660 99 9071 8046 suspend10 1000K 1000024 80 160 220 92 8137 9035 suspend20 1000K 1000059 83 165 226 89 7850 8804 suspend50 1000K 999955 95 180 240 84 7411 8459 suspend200 1000K 999914 163 299 366 77 6833 8078 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base MAX 1037654 4184 5453 5810 100 8411 7938 defer10 MAX 905607 4840 6151 6380 100 9639 8431 defer20 MAX 986463 4455 5594 5796 100 8848 8110 defer50 MAX 1077030 4000 5073 5299 100 8104 7920 defer200 MAX 1040728 4152 5385 5765 100 8379 7849 fullbusy MAX 1247536 3518 3935 3984 100 6998 7930 napibusy MAX 1136310 3799 7756 9964 100 7670 7877 suspend0 MAX 1057509 4132 5724 6185 100 8253 7918 suspend10 MAX 1215147 3580 3957 4041 100 7185 7944 suspend20 MAX 1216469 3576 3953 3988 100 7175 7950 suspend50 MAX 1215871 3577 3961 4075 100 7181 7949 suspend200 MAX 1216882 3556 3951 3988 100 7175 7955 ~ FAQ - Why is a new parameter needed? Does irq_suspend_timeout override gro_flush_timeout? Using the suspend mechanism causes the system to alternate between polling mode and irq-driven packet delivery. During busy periods, irq_suspend_timeout overrides gro_flush_timeout and keeps the system busy polling, but when epoll finds no events, the setting of gro_flush_timeout and napi_defer_hard_irqs determine the next step. There are essentially three possible loops for network processing and packet delivery: 1) hardirq -> softirq -> napi poll; basic interrupt delivery 2) timer -> softirq -> napi poll; deferred irq processing 3) epoll -> busy-poll -> napi poll; busy looping Loop 2 can take control from Loop 1, if gro_flush_timeout and napi_defer_hard_irqs are set. If gro_flush_timeout and napi_defer_hard_irqs are set, Loops 2 and 3 "wrestle" with each other for control. During busy periods, irq_suspend_timeout is used as timer in Loop 2, which essentially tilts this in favour of Loop 3. If gro_flush_timeout and napi_defer_hard_irqs are not set, Loop 3 cannot take control from Loop 1. Therefore, setting gro_flush_timeout and napi_defer_hard_irqs is the recommended usage, because otherwise setting irq_suspend_timeout might not have any discernible effect. This is shown in the results above: compare suspend0 with the base case. Note that the lack of napi_defer_hard_irqs and gro_flush_timeout produce similar results for both, which encourages the use of napi_defer_hard_irqs and gro_flush_timeout in addition to irq_suspend_timeout. - Can the new timeout value be threaded through the new epoll ioctl ? It is possible, but presents challenges for userspace. User applications must ensure that the file descriptors added to epoll contexts have the same NAPI ID to support busy polling. An epoll context is not permanently tied to any particular NAPI ID. So, a user application could decide to clear the file descriptors from the context and add a new set of file descriptors with a different NAPI ID to the context. Busy polling would work as expected, but the meaning of the suspend timeout becomes ambiguous because IRQs are not inherently associated with epoll contexts, but rather with the NAPI. The user program would need to reissue the ioctl to set the irq_suspend_timeout, but the napi_defer_hard_irqs and gro_flush_timeout settings would come from the NAPI's napi_config (which are set either by sysfs or by netlink). Such an interface seems awkard to use from a user perspective. Further, IRQs are related to NAPIs, which is why they are stored in the napi_config space. Putting the irq_suspend_timeout in the epoll context while other IRQ deferral mechanisms remain in the NAPI's napi_config space seems like an odd design choice. We've opted to keep all of the IRQ deferral parameters together and place the irq_suspend_timeout in napi_config. This has nice benefits for userspace: if a user app were to remove all file descriptors from an epoll context and add new file descriptors with a new NAPI ID, the correct suspend timeout for that NAPI ID would be used automatically without the user application needing to do anything (like re-issuing an ioctl, for example). All IRQ deferral related parameters are in one place and can all be set the same way: with netlink. - Can irq suspend be built by combining NIC coalescing and gro_flush_timeout ? No. The problem is that the long timeout must engage if and only if prefer-busy is active. When using NIC coalescing for the short timeout (without napi_defer_hard_irqs/gro_flush_timeout), an interrupt after an idle period will trigger softirq, which will run napi polling. At this point, prefer-busy is not active, so NIC interrupts would be re-enabled. Then it is not possible for the longer timeout to interject to switch control back to polling. In other words, only by using the software timer for the short timeout, it is possible to extend the timeout without having to reprogram the NIC timer or reach down directly and disable interrupts. Using gro_flush_timeout for the long timeout also has problems, for the same underlying reason. In the current napi implementation, gro_flush_timeout is not tied to prefer-busy. We'd either have to change that and in the process modify the existing deferral mechanism, or introduce a state variable to determine whether gro_flush_timeout is used as long timeout for irq suspend or whether it is used for its default purpose. In an earlier version, we did try something similar to the latter and made it work, but it ends up being a lot more convoluted than our current proposal. - Isn't it already possible to combine busy looping with irq deferral? Yes, in fact enabling irq deferral via napi_defer_hard_irqs and gro_flush_timeout is a precondition for prefer_busy_poll to have an effect. If the application also uses a tight busy loop with essentially nonblocking epoll_wait (accomplished with a very short timeout parameter), this is the fullbusy case shown in the results. An application using blocking epoll_wait is shown as the napibusy case in the results. It's a hybrid approach that provides limited latency benefits compared to the base case and plain irq deferral, but not as good as fullbusy or suspend. ~ Special thanks Several people were involved in earlier stages of the development of this mechanism whom we'd like to thank: - Peter Cai (CC'd), for the initial kernel patch and his contributions to the paper. - Mohammadamin Shafie (CC'd), for testing various versions of the kernel patch and providing helpful feedback. Thanks, Martin and Joe [1]: https://lore.kernel.org/netdev/20240812125717.413108-1-jdamato@fastly.com/ [2]: https://doi.org/10.1145/3626780 [3]: https://github.com/memcached/memcached/blob/master/doc/napi_ids.txt [4]: https://github.com/leverich/mutilate [5]: https://raw.githubusercontent.com/martinkarsten/irqsuspend/main/patches/mem… [6]: https://raw.githubusercontent.com/martinkarsten/irqsuspend/main/patches/lib… [7]: https://github.com/martinkarsten/irqsuspend [8]: https://github.com/martinkarsten/irqsuspend/tree/main/results v9: - Addresses Willem's feedback on the selftests in patch 5 by fixing the SPDX-License-Identifier, moving constants into variables in the test script, reducing code duplication, shortening long lines, and renaming variables to be more reader friendly. In the C test file, added a comment explaining the if def blob and changed a few types for strtoul. v8: https://lore.kernel.org/netdev/20241108045337.292905-1-jdamato@fastly.com/ - Update patch 2 to drop the exports, as requested by Jakub. v7: https://lore.kernel.org/netdev/20241108023912.98416-1-jdamato@fastly.com/ - Jakub noted that patch 2 adds unnecessary complexity by checking the suspend timeout in the NAPI loop. This makes the code more complicated and difficult to reason about. He's right; we've dropped patch 2 which simplifies this series. - Updated the cover letter with a full re-run of all test cases. - Updated FAQ #2. v6: https://lore.kernel.org/netdev/20241104215542.215919-1-jdamato@fastly.com/ - Updated the cover letter with a full re-run of all test cases, including a new case suspend0, as requested by Sridhar previously. - Updated the kernel documentation in patch 7 as suggested by Bagas Sanjaya, which improved the htmldoc output. v5: https://lore.kernel.org/netdev/20241103052421.518856-1-jdamato@fastly.com/ - Adjusted patch 5 to only suspend IRQs when ep_send_events returns a positive return value. This issue was pointed out by Hillf Danton. - Updated the commit message of patch 6 which still mentioned netcat, despite the code being updated in v4 to replace it with socat and fixed misspelling of netdevsim. - Fixed a minor typo in patch 7 and removed an unnecessary paragraph. - Added Sridhar Samudrala's Reviewed-by to patch 1-5 and 7. v4: https://lore.kernel.org/netdev/20241102005214.32443-1-jdamato@fastly.com/ - Added a new FAQ item to cover letter. - Updated patch 6 to use socat instead of nc in busy_poll_test.sh and updated busy_poller.c to use netlink directly to configure napi params. - Updated the kernel documentation in patch 7 to include more details. - Dropped Stanislav's Acked-by and Bagas' Reviewed-by from patch 7 since the documentation was updated. v3: https://lore.kernel.org/netdev/20241101004846.32532-1-jdamato@fastly.com/ - Added Stanislav Fomichev's Acked-by to every patch except the newly added selftest. - Added Bagas Sanjaya's Reviewed-by to the documentation patch. - Fixed the commit message of patch 2 to remove a reference to the now non-existent sysfs setting. - Added a self test which tests both "regular" busy poll and busy poll with suspend enabled. This was added as patch 6 as requested by Paolo. netdevsim was chosen instead of veth due to netdevsim's pre-existing support for netdev-genl. See the commit message of patch 6 for more details. v2: https://lore.kernel.org/bpf/20241021015311.95468-1-jdamato@fastly.com/ - Cover letter updated, including a re-run of test data. - Patch 1 rewritten to use netdev-genl instead of sysfs. - Patch 3 updated with a comment added to napi_resume_irqs. - Patch 4 rebased to apply now that commit b9ca079dd6b0 ("eventpoll: Annotate data-race of busy_poll_usecs") has been picked up from VFS. - Patch 6 updated the kernel documentation. rfc -> v1: - Cover letter updated to include more details. - Patch 1 updated to remove the documentation added. This was moved to patch 6 with the rest of the docs (see below). - Patch 5 updated to fix an error uncovered by the kernel build robot. See patch 5's changelog for more details. - Patch 6 added which updates kernel documentation. Joe Damato (2): selftests: net: Add busy_poll_test docs: networking: Describe irq suspension Martin Karsten (4): net: Add napi_struct parameter irq_suspend_timeout net: Add control functions for irq suspension eventpoll: Trigger napi_busy_loop, if prefer_busy_poll is set eventpoll: Control irq suspension for prefer_busy_poll Documentation/netlink/specs/netdev.yaml | 7 + Documentation/networking/napi.rst | 170 ++++++++- fs/eventpoll.c | 36 +- include/linux/netdevice.h | 2 + include/net/busy_poll.h | 3 + include/uapi/linux/netdev.h | 1 + net/core/dev.c | 39 ++ net/core/dev.h | 25 ++ net/core/netdev-genl-gen.c | 5 +- net/core/netdev-genl.c | 12 + tools/include/uapi/linux/netdev.h | 1 + tools/testing/selftests/net/.gitignore | 1 + tools/testing/selftests/net/Makefile | 3 +- tools/testing/selftests/net/busy_poll_test.sh | 165 +++++++++ tools/testing/selftests/net/busy_poller.c | 346 ++++++++++++++++++ 15 files changed, 809 insertions(+), 7 deletions(-) create mode 100755 tools/testing/selftests/net/busy_poll_test.sh create mode 100644 tools/testing/selftests/net/busy_poller.c base-commit: dc7c381bb8649e3701ed64f6c3e55316675904d7 -- 2.25.1

1 day, 15 hours

4
5
0 0

[PATCH RESEND] x86: checksum: Fix unaligned checksums on < i686

by David Gow

The checksum_32 code was originally written to only handle 2-byte aligned buffers, but was later extended to support arbitrary alignment. However, the non-PPro variant doesn't apply the carry before jumping to the 2- or 4-byte aligned versions, which clear CF. This causes the new checksum_kunit test to fail, as it runs with a large number of different possible alignments and both with and without carries. For example: ./tools/testing/kunit/kunit.py run --arch i386 --kconfig_add CONFIG_M486=y checksum Gives: KTAP version 1 # Subtest: checksum 1..3 ok 1 test_csum_fixed_random_inputs # test_csum_all_carry_inputs: ASSERTION FAILED at lib/checksum_kunit.c:267 Expected result == expec, but result == 65281 (0xff01) expec == 65280 (0xff00) not ok 2 test_csum_all_carry_inputs # test_csum_no_carry_inputs: ASSERTION FAILED at lib/checksum_kunit.c:314 Expected result == expec, but result == 65535 (0xffff) expec == 65534 (0xfffe) not ok 3 test_csum_no_carry_inputs With this patch, it passes. KTAP version 1 # Subtest: checksum 1..3 ok 1 test_csum_fixed_random_inputs ok 2 test_csum_all_carry_inputs ok 3 test_csum_no_carry_inputs I also tested it on a real 486DX2, with the same results. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: David Gow <davidgow(a)google.com> --- Re-sending this from [1]. While there's an argument that the whole 32-bit checksum code could do with rewriting, it's: (a) worth fixing before someone takes the time to rewrite it, and (b) worth any future rewrite starting from a point where the tests pass I don't think there should be any downside to this fix: it only affects ancient computers, and adds a single instruction which isn't in a loop. Cheers, -- David [1]: https://lore.kernel.org/lkml/20230704083206.693155-2-davidgow@google.com/ --- arch/x86/lib/checksum_32.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/x86/lib/checksum_32.S b/arch/x86/lib/checksum_32.S index 68f7fa3e1322..a5123b29b403 100644 --- a/arch/x86/lib/checksum_32.S +++ b/arch/x86/lib/checksum_32.S @@ -62,6 +62,7 @@ SYM_FUNC_START(csum_partial) jl 8f movzbl (%esi), %ebx adcl %ebx, %eax + adcl $0, %eax roll $8, %eax inc %esi testl $2, %esi -- 2.45.2.1089.g2a221341d9-goog

1 day, 23 hours

2
2
0 0

[PATCH v12] exec: Fix dead-lock in de_thread with ptrace_attach

by Bernd Edlinger

This introduces signal->exec_bprm, which is used to fix the case when at least one of the sibling threads is traced, and therefore the trace process may dead-lock in ptrace_attach, but de_thread will need to wait for the tracer to continue execution. The solution is to detect this situation and allow ptrace_attach to continue by temporarily releasing the cred_guard_mutex, while de_thread() is still waiting for traced zombies to be eventually released by the tracer. In the case of the thread group leader we only have to wait for the thread to become a zombie, which may also need co-operation from the tracer due to PTRACE_O_TRACEEXIT. When a tracer wants to ptrace_attach a task that already is in execve, we simply retry the ptrace_may_access check while temporarily installing the new credentials and dumpability which are about to be used after execve completes. If the ptrace_attach happens on a thread that is a sibling-thread of the thread doing execve, it is sufficient to check against the old credentials, as this thread will be waited for, before the new credentials are installed. Other threads die quickly since the cred_guard_mutex is released, but a deadly signal is already pending. In case the mutex_lock_killable misses the signal, the non-zero current->signal->exec_bprm makes sure they release the mutex immediately and return with -ERESTARTNOINTR. This means there is no API change, unlike the previous version of this patch which was discussed here: https://lore.kernel.org/lkml/b6537ae6-31b1-5c50-f32b-8b8332ace882@hotmail.d… See tools/testing/selftests/ptrace/vmaccess.c for a test case that gets fixed by this change. Note that since the test case was originally designed to test the ptrace_attach returning an error in this situation, the test expectation needed to be adjusted, to allow the API to succeed at the first attempt. Signed-off-by: Bernd Edlinger <bernd.edlinger(a)hotmail.de> --- fs/exec.c | 69 ++++++++++++++++------- fs/proc/base.c | 6 ++ include/linux/cred.h | 1 + include/linux/sched/signal.h | 18 ++++++ kernel/cred.c | 28 +++++++-- kernel/ptrace.c | 32 +++++++++++ kernel/seccomp.c | 12 +++- tools/testing/selftests/ptrace/vmaccess.c | 23 +++++--- 8 files changed, 155 insertions(+), 34 deletions(-) v10: Changes to previous version, make the PTRACE_ATTACH retun -EAGAIN, instead of execve return -ERESTARTSYS. Added some lessions learned to the description. v11: Check old and new credentials in PTRACE_ATTACH again without changing the API. Note: I got actually one response from an automatic checker to the v11 patch, https://lore.kernel.org/lkml/202107121344.wu68hEPF-lkp@intel.com/ which is complaining about: >> kernel/ptrace.c:425:26: sparse: sparse: incorrect type in assignment (different address spaces) @@ expected struct cred const *old_cred @@ got struct cred const [noderef] __rcu *real_cred @@ 417 struct linux_binprm *bprm = task->signal->exec_bprm; 418 const struct cred *old_cred; 419 struct mm_struct *old_mm; 420 421 retval = down_write_killable(&task->signal->exec_update_lock); 422 if (retval) 423 goto unlock_creds; 424 task_lock(task); > 425 old_cred = task->real_cred; v12: Essentially identical to v11. - Fixed a minor merge conflict in linux v5.17, and fixed the above mentioned nit by adding __rcu to the declaration. - re-tested the patch with all linux versions from v5.11 to v6.6 v10 was an alternative approach which did imply an API change. But I would prefer to avoid such an API change. The difficult part is getting the right dumpability flags assigned before de_thread starts, hope you like this version. If not, the v10 is of course also acceptable. Thanks Bernd. diff --git a/fs/exec.c b/fs/exec.c index 2f2b0acec4f0..902d3b230485 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1041,11 +1041,13 @@ static int exec_mmap(struct mm_struct *mm) return 0; } -static int de_thread(struct task_struct *tsk) +static int de_thread(struct task_struct *tsk, struct linux_binprm *bprm) { struct signal_struct *sig = tsk->signal; struct sighand_struct *oldsighand = tsk->sighand; spinlock_t *lock = &oldsighand->siglock; + struct task_struct *t = tsk; + bool unsafe_execve_in_progress = false; if (thread_group_empty(tsk)) goto no_thread_group; @@ -1068,6 +1070,19 @@ static int de_thread(struct task_struct *tsk) if (!thread_group_leader(tsk)) sig->notify_count--; + while_each_thread(tsk, t) { + if (unlikely(t->ptrace) + && (t != tsk->group_leader || !t->exit_state)) + unsafe_execve_in_progress = true; + } + + if (unlikely(unsafe_execve_in_progress)) { + spin_unlock_irq(lock); + sig->exec_bprm = bprm; + mutex_unlock(&sig->cred_guard_mutex); + spin_lock_irq(lock); + } + while (sig->notify_count) { __set_current_state(TASK_KILLABLE); spin_unlock_irq(lock); @@ -1158,6 +1173,11 @@ static int de_thread(struct task_struct *tsk) release_task(leader); } + if (unlikely(unsafe_execve_in_progress)) { + mutex_lock(&sig->cred_guard_mutex); + sig->exec_bprm = NULL; + } + sig->group_exec_task = NULL; sig->notify_count = 0; @@ -1169,6 +1189,11 @@ static int de_thread(struct task_struct *tsk) return 0; killed: + if (unlikely(unsafe_execve_in_progress)) { + mutex_lock(&sig->cred_guard_mutex); + sig->exec_bprm = NULL; + } + /* protects against exit_notify() and __exit_signal() */ read_lock(&tasklist_lock); sig->group_exec_task = NULL; @@ -1253,6 +1278,24 @@ int begin_new_exec(struct linux_binprm * bprm) if (retval) return retval; + /* If the binary is not readable then enforce mm->dumpable=0 */ + would_dump(bprm, bprm->file); + if (bprm->have_execfd) + would_dump(bprm, bprm->executable); + + /* + * Figure out dumpability. Note that this checking only of current + * is wrong, but userspace depends on it. This should be testing + * bprm->secureexec instead. + */ + if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP || + is_dumpability_changed(current_cred(), bprm->cred) || + !(uid_eq(current_euid(), current_uid()) && + gid_eq(current_egid(), current_gid()))) + set_dumpable(bprm->mm, suid_dumpable); + else + set_dumpable(bprm->mm, SUID_DUMP_USER); + /* * Ensure all future errors are fatal. */ @@ -1261,7 +1304,7 @@ int begin_new_exec(struct linux_binprm * bprm) /* * Make this the only thread in the thread group. */ - retval = de_thread(me); + retval = de_thread(me, bprm); if (retval) goto out; @@ -1284,11 +1327,6 @@ int begin_new_exec(struct linux_binprm * bprm) if (retval) goto out; - /* If the binary is not readable then enforce mm->dumpable=0 */ - would_dump(bprm, bprm->file); - if (bprm->have_execfd) - would_dump(bprm, bprm->executable); - /* * Release all of the old mmap stuff */ @@ -1350,18 +1388,6 @@ int begin_new_exec(struct linux_binprm * bprm) me->sas_ss_sp = me->sas_ss_size = 0; - /* - * Figure out dumpability. Note that this checking only of current - * is wrong, but userspace depends on it. This should be testing - * bprm->secureexec instead. - */ - if (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP || - !(uid_eq(current_euid(), current_uid()) && - gid_eq(current_egid(), current_gid()))) - set_dumpable(current->mm, suid_dumpable); - else - set_dumpable(current->mm, SUID_DUMP_USER); - perf_event_exec(); __set_task_comm(me, kbasename(bprm->filename), true); @@ -1480,6 +1506,11 @@ static int prepare_bprm_creds(struct linux_binprm *bprm) if (mutex_lock_interruptible(&current->signal->cred_guard_mutex)) return -ERESTARTNOINTR; + if (unlikely(current->signal->exec_bprm)) { + mutex_unlock(&current->signal->cred_guard_mutex); + return -ERESTARTNOINTR; + } + bprm->cred = prepare_exec_creds(); if (likely(bprm->cred)) return 0; diff --git a/fs/proc/base.c b/fs/proc/base.c index ffd54617c354..0da9adfadb48 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2788,6 +2788,12 @@ static ssize_t proc_pid_attr_write(struct file * file, const char __user * buf, if (rv < 0) goto out_free; + if (unlikely(current->signal->exec_bprm)) { + mutex_unlock(&current->signal->cred_guard_mutex); + rv = -ERESTARTNOINTR; + goto out_free; + } + rv = security_setprocattr(PROC_I(inode)->op.lsm, file->f_path.dentry->d_name.name, page, count); diff --git a/include/linux/cred.h b/include/linux/cred.h index f923528d5cc4..b01e309f5686 100644 --- a/include/linux/cred.h +++ b/include/linux/cred.h @@ -159,6 +159,7 @@ extern const struct cred *get_task_cred(struct task_struct *); extern struct cred *cred_alloc_blank(void); extern struct cred *prepare_creds(void); extern struct cred *prepare_exec_creds(void); +extern bool is_dumpability_changed(const struct cred *, const struct cred *); extern int commit_creds(struct cred *); extern void abort_creds(struct cred *); extern const struct cred *override_creds(const struct cred *); diff --git a/include/linux/sched/signal.h b/include/linux/sched/signal.h index 0014d3adaf84..14df7073a0a8 100644 --- a/include/linux/sched/signal.h +++ b/include/linux/sched/signal.h @@ -234,9 +234,27 @@ struct signal_struct { struct mm_struct *oom_mm; /* recorded mm when the thread group got * killed by the oom killer */ + struct linux_binprm *exec_bprm; /* Used to check ptrace_may_access + * against new credentials while + * de_thread is waiting for other + * traced threads to terminate. + * Set while de_thread is executing. + * The cred_guard_mutex is released + * after de_thread() has called + * zap_other_threads(), therefore + * a fatal signal is guaranteed to be + * already pending in the unlikely + * event, that + * current->signal->exec_bprm happens + * to be non-zero after the + * cred_guard_mutex was acquired. + */ + struct mutex cred_guard_mutex; /* guard against foreign influences on * credential calculations * (notably. ptrace) + * Held while execve runs, except when + * a sibling thread is being traced. * Deprecated do not use in new code. * Use exec_update_lock instead. */ diff --git a/kernel/cred.c b/kernel/cred.c index 98cb4eca23fb..586cb6c7cf6b 100644 --- a/kernel/cred.c +++ b/kernel/cred.c @@ -433,6 +433,28 @@ static bool cred_cap_issubset(const struct cred *set, const struct cred *subset) return false; } +/** + * is_dumpability_changed - Will changing creds from old to new + * affect the dumpability in commit_creds? + * + * Return: false - dumpability will not be changed in commit_creds. + * Return: true - dumpability will be changed to non-dumpable. + * + * @old: The old credentials + * @new: The new credentials + */ +bool is_dumpability_changed(const struct cred *old, const struct cred *new) +{ + if (!uid_eq(old->euid, new->euid) || + !gid_eq(old->egid, new->egid) || + !uid_eq(old->fsuid, new->fsuid) || + !gid_eq(old->fsgid, new->fsgid) || + !cred_cap_issubset(old, new)) + return true; + + return false; +} + /** * commit_creds - Install new credentials upon the current task * @new: The credentials to be assigned @@ -467,11 +489,7 @@ int commit_creds(struct cred *new) get_cred(new); /* we will require a ref for the subj creds too */ /* dumpability changes */ - if (!uid_eq(old->euid, new->euid) || - !gid_eq(old->egid, new->egid) || - !uid_eq(old->fsuid, new->fsuid) || - !gid_eq(old->fsgid, new->fsgid) || - !cred_cap_issubset(old, new)) { + if (is_dumpability_changed(old, new)) { if (task->mm) set_dumpable(task->mm, suid_dumpable); task->pdeath_signal = 0; diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 443057bee87c..eb1c450bb7d7 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -20,6 +20,7 @@ #include <linux/pagemap.h> #include <linux/ptrace.h> #include <linux/security.h> +#include <linux/binfmts.h> #include <linux/signal.h> #include <linux/uio.h> #include <linux/audit.h> @@ -435,6 +436,28 @@ static int ptrace_attach(struct task_struct *task, long request, if (retval) goto unlock_creds; + if (unlikely(task->in_execve)) { + struct linux_binprm *bprm = task->signal->exec_bprm; + const struct cred __rcu *old_cred; + struct mm_struct *old_mm; + + retval = down_write_killable(&task->signal->exec_update_lock); + if (retval) + goto unlock_creds; + task_lock(task); + old_cred = task->real_cred; + old_mm = task->mm; + rcu_assign_pointer(task->real_cred, bprm->cred); + task->mm = bprm->mm; + retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH_REALCREDS); + rcu_assign_pointer(task->real_cred, old_cred); + task->mm = old_mm; + task_unlock(task); + up_write(&task->signal->exec_update_lock); + if (retval) + goto unlock_creds; + } + write_lock_irq(&tasklist_lock); retval = -EPERM; if (unlikely(task->exit_state)) @@ -508,6 +531,14 @@ static int ptrace_traceme(void) { int ret = -EPERM; + if (mutex_lock_interruptible(&current->signal->cred_guard_mutex)) + return -ERESTARTNOINTR; + + if (unlikely(current->signal->exec_bprm)) { + mutex_unlock(&current->signal->cred_guard_mutex); + return -ERESTARTNOINTR; + } + write_lock_irq(&tasklist_lock); /* Are we already being traced? */ if (!current->ptrace) { @@ -523,6 +554,7 @@ static int ptrace_traceme(void) } } write_unlock_irq(&tasklist_lock); + mutex_unlock(&current->signal->cred_guard_mutex); return ret; } diff --git a/kernel/seccomp.c b/kernel/seccomp.c index 255999ba9190..b29bbfa0b044 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -1955,9 +1955,15 @@ static long seccomp_set_mode_filter(unsigned int flags, * Make sure we cannot change seccomp or nnp state via TSYNC * while another thread is in the middle of calling exec. */ - if (flags & SECCOMP_FILTER_FLAG_TSYNC && - mutex_lock_killable(&current->signal->cred_guard_mutex)) - goto out_put_fd; + if (flags & SECCOMP_FILTER_FLAG_TSYNC) { + if (mutex_lock_killable(&current->signal->cred_guard_mutex)) + goto out_put_fd; + + if (unlikely(current->signal->exec_bprm)) { + mutex_unlock(&current->signal->cred_guard_mutex); + goto out_put_fd; + } + } spin_lock_irq(&current->sighand->siglock); diff --git a/tools/testing/selftests/ptrace/vmaccess.c b/tools/testing/selftests/ptrace/vmaccess.c index 4db327b44586..3b7d81fb99bb 100644 --- a/tools/testing/selftests/ptrace/vmaccess.c +++ b/tools/testing/selftests/ptrace/vmaccess.c @@ -39,8 +39,15 @@ TEST(vmaccess) f = open(mm, O_RDONLY); ASSERT_GE(f, 0); close(f); - f = kill(pid, SIGCONT); - ASSERT_EQ(f, 0); + f = waitpid(-1, NULL, 0); + ASSERT_NE(f, -1); + ASSERT_NE(f, 0); + ASSERT_NE(f, pid); + f = waitpid(-1, NULL, 0); + ASSERT_EQ(f, pid); + f = waitpid(-1, NULL, 0); + ASSERT_EQ(f, -1); + ASSERT_EQ(errno, ECHILD); } TEST(attach) @@ -57,22 +64,24 @@ TEST(attach) sleep(1); k = ptrace(PTRACE_ATTACH, pid, 0L, 0L); - ASSERT_EQ(errno, EAGAIN); - ASSERT_EQ(k, -1); + ASSERT_EQ(k, 0); k = waitpid(-1, &s, WNOHANG); ASSERT_NE(k, -1); ASSERT_NE(k, 0); ASSERT_NE(k, pid); ASSERT_EQ(WIFEXITED(s), 1); ASSERT_EQ(WEXITSTATUS(s), 0); - sleep(1); - k = ptrace(PTRACE_ATTACH, pid, 0L, 0L); + k = waitpid(-1, &s, 0); + ASSERT_EQ(k, pid); + ASSERT_EQ(WIFSTOPPED(s), 1); + ASSERT_EQ(WSTOPSIG(s), SIGTRAP); + k = ptrace(PTRACE_CONT, pid, 0L, 0L); ASSERT_EQ(k, 0); k = waitpid(-1, &s, 0); ASSERT_EQ(k, pid); ASSERT_EQ(WIFSTOPPED(s), 1); ASSERT_EQ(WSTOPSIG(s), SIGSTOP); - k = ptrace(PTRACE_DETACH, pid, 0L, 0L); + k = ptrace(PTRACE_CONT, pid, 0L, 0L); ASSERT_EQ(k, 0); k = waitpid(-1, &s, 0); ASSERT_EQ(k, pid); -- 2.39.2

2 days

17
71
0 0

[PATCH] selftests/ftrace: Add test dependency

by Thibault Ferrante

test_duplicates miss a running dependency and leads to test failures on kernel with specific configuration. Signed-off-by: Thibault Ferrante <thibault.ferrante(a)canonical.com> --- .../testing/selftests/ftrace/test.d/dynevent/test_duplicates.tc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/ftrace/test.d/dynevent/test_duplicates.tc b/tools/testing/selftests/ftrace/test.d/dynevent/test_duplicates.tc index d3a79da215c8..0b5e4543e70b 100644 --- a/tools/testing/selftests/ftrace/test.d/dynevent/test_duplicates.tc +++ b/tools/testing/selftests/ftrace/test.d/dynevent/test_duplicates.tc @@ -1,7 +1,7 @@ #!/bin/sh # SPDX-License-Identifier: GPL-2.0 # description: Generic dynamic event - check if duplicate events are caught -# requires: dynamic_events "e[:[<group>/][<event>]] <attached-group>.<attached-event> [<args>]":README +# requires: dynamic_events events/syscalls/sys_enter_openat "e[:[<group>/][<event>]] <attached-group>.<attached-event> [<args>]":README echo 0 > events/enable -- 2.39.2

2 days, 8 hours

2
1
0 0

[PATCH] selftests/ftrace: Test toplevel-enable for instance

by Zheng Yejian

'available_events' is actually not required by 'test.d/event/toplevel-enable.tc' and its Existence has been tested in 'test.d/00basic/basic4.tc'. So the require of 'available_events' can be dropped and then we can add 'instance' flag to test 'test.d/event/toplevel-enable.tc' for instance. Test result show as below: # ./ftracetest test.d/event/toplevel-enable.tc === Ftrace unit tests === [1] event tracing - enable/disable with top level files [PASS] [2] (instance) event tracing - enable/disable with top level files [PASS] # of passed: 2 # of failed: 0 # of unresolved: 0 # of untested: 0 # of unsupported: 0 # of xfailed: 0 # of undefined(test bug): 0 Signed-off-by: Zheng Yejian <zhengyejian1(a)huawei.com> --- tools/testing/selftests/ftrace/test.d/event/toplevel-enable.tc | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/ftrace/test.d/event/toplevel-enable.tc b/tools/testing/selftests/ftrace/test.d/event/toplevel-enable.tc index 93c10ea42a68..8b8e1aea985b 100644 --- a/tools/testing/selftests/ftrace/test.d/event/toplevel-enable.tc +++ b/tools/testing/selftests/ftrace/test.d/event/toplevel-enable.tc @@ -1,7 +1,8 @@ #!/bin/sh # SPDX-License-Identifier: GPL-2.0 # description: event tracing - enable/disable with top level files -# requires: available_events set_event events/enable +# requires: set_event events/enable +# flags: instance do_reset() { echo > set_event -- 2.25.1

2 days, 8 hours

2
4
0 0

[PATCH v3] selftests/ftrace: traceonoff_triggers: strip off names

by Yipeng Zou

The func_traceonoff_triggers.tc sometimes goes to fail on my board, Kunpeng-920. [root@localhost]# ./ftracetest ./test.d/ftrace/func_traceonoff_triggers.tc -l fail.log === Ftrace unit tests === [1] ftrace - test for function traceon/off triggers [FAIL] [2] (instance) ftrace - test for function traceon/off triggers [UNSUPPORTED] I look up the log, and it shows that the md5sum is different between csum1 and csum2. ++ cnt=611 ++ sleep .1 +++ cnt_trace +++ grep -v '^#' trace +++ wc -l ++ cnt2=611 ++ '[' 611 -ne 611 ']' +++ cat tracing_on ++ on=0 ++ '[' 0 '!=' 0 ']' +++ md5sum trace ++ csum1='76896aa74362fff66a6a5f3cf8a8a500 trace' ++ sleep .1 +++ md5sum trace ++ csum2='ee8625a21c058818fc26e45c1ed3f6de trace' ++ '[' '76896aa74362fff66a6a5f3cf8a8a500 trace' '!=' 'ee8625a21c058818fc26e45c1ed3f6de trace' ']' ++ fail 'Tracing file is still changing' ++ echo Tracing file is still changing Tracing file is still changing ++ exit_fail ++ exit 1 So I directly dump the trace file before md5sum, the diff shows that: [root@localhost]# diff trace_1.log trace_2.log -y --suppress-common-lines dockerd-12285 [036] d.... 18385.510290: sched_stat | <...>-12285 [036] d.... 18385.510290: sched_stat dockerd-12285 [036] d.... 18385.510291: sched_swit | <...>-12285 [036] d.... 18385.510291: sched_swit <...>-740 [044] d.... 18385.602859: sched_stat | kworker/44:1-740 [044] d.... 18385.602859: sched_stat <...>-740 [044] d.... 18385.602860: sched_swit | kworker/44:1-740 [044] d.... 18385.602860: sched_swit And we can see that <...> filed be filled with names. We can strip off the names there to fix that. After strip off the names: kworker/u257:0-12 [019] d..2. 2528.758910: sched_stat | -12 [019] d..2. 2528.758910: sched_stat_runtime: comm=k kworker/u257:0-12 [019] d..2. 2528.758912: sched_swit | -12 [019] d..2. 2528.758912: sched_switch: prev_comm=kw <idle>-0 [000] d.s5. 2528.762318: sched_waki | -0 [000] d.s5. 2528.762318: sched_waking: comm=sshd pi <idle>-0 [037] dNh2. 2528.762326: sched_wake | -0 [037] dNh2. 2528.762326: sched_wakeup: comm=sshd pi <idle>-0 [037] d..2. 2528.762334: sched_swit | -0 [037] d..2. 2528.762334: sched_switch: prev_comm=sw Fixes: d87b29179aa0 ("selftests: ftrace: Use md5sum to take less time of checking logs") Suggested-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> Signed-off-by: Yipeng Zou <zouyipeng(a)huawei.com> Acked-by: Masami Hiramatsu (Google) <mhiramat(a)kernel.org> Reviewed-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> --- .../ftrace/test.d/ftrace/func_traceonoff_triggers.tc | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/ftrace/test.d/ftrace/func_traceonoff_triggers.tc b/tools/testing/selftests/ftrace/test.d/ftrace/func_traceonoff_triggers.tc index aee22289536b..1b57771dbfdf 100644 --- a/tools/testing/selftests/ftrace/test.d/ftrace/func_traceonoff_triggers.tc +++ b/tools/testing/selftests/ftrace/test.d/ftrace/func_traceonoff_triggers.tc @@ -90,9 +90,10 @@ if [ $on != "0" ]; then fail "Tracing is not off" fi -csum1=`md5sum trace` +# Cannot rely on names being around as they are only cached, strip them +csum1=`cat trace | sed -e 's/^ *[^ ]*$-[0-9][0-9]*$/\1/' | md5sum` sleep $SLEEP_TIME -csum2=`md5sum trace` +csum2=`cat trace | sed -e 's/^ *[^ ]*$-[0-9][0-9]*$/\1/' | md5sum` if [ "$csum1" != "$csum2" ]; then fail "Tracing file is still changing" -- 2.34.1

2 days, 8 hours

2
3
0 0

[PATCH v3 0/4] selftests/resctrl: Enable MBM and MBA tests on AMD

by Babu Moger

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

2 months, 1 week

5
15
0 0

[PATCH v7 0/6] mm/memfd: introduce MFD_NOEXEC_SEAL and MFD_EXEC

by jeffxu＠chromium.org

From: Jeff Xu <jeffxu(a)google.com> Since Linux introduced the memfd feature, memfd have always had their execute bit set, and the memfd_create() syscall doesn't allow setting it differently. However, in a secure by default system, such as ChromeOS, (where all executables should come from the rootfs, which is protected by Verified boot), this executable nature of memfd opens a door for NoExec bypass and enables “confused deputy attack”. E.g, in VRP bug [1]: cros_vm process created a memfd to share the content with an external process, however the memfd is overwritten and used for executing arbitrary code and root escalation. [2] lists more VRP in this kind. On the other hand, executable memfd has its legit use, runc uses memfd’s seal and executable feature to copy the contents of the binary then execute them, for such system, we need a solution to differentiate runc's use of executable memfds and an attacker's [3]. To address those above, this set of patches add following: 1> Let memfd_create() set X bit at creation time. 2> Let memfd to be sealed for modifying X bit. 3> A new pid namespace sysctl: vm.memfd_noexec to control the behavior of X bit.For example, if a container has vm.memfd_noexec=2, then memfd_create() without MFD_NOEXEC_SEAL will be rejected. 4> A new security hook in memfd_create(). This make it possible to a new LSM, which rejects or allows executable memfd based on its security policy. Change history: v7: - patch 2/6: remove #ifdef and MAX_PATH (memfd_test.c). - patch 3/6: check capability (CAP_SYS_ADMIN) from userns instead of global ns (pid_sysctl.h). Add a tab (pid_namespace.h). - patch 5/6: remove #ifdef (memfd_test.c) - patch 6/6: remove unneeded security_move_mount(security.c). v6:https://lore.kernel.org/lkml/20221206150233.1963717-1-jeffxu@google.com/ - Address comment and move "#ifdef CONFIG_" from .c file to pid_sysctl.h v5:https://lore.kernel.org/lkml/20221206152358.1966099-1-jeffxu@google.com/ - Pass vm.memfd_noexec from current ns to child ns. - Fix build issue detected by kernel test robot. - Add missing security.c v3:https://lore.kernel.org/lkml/20221202013404.163143-1-jeffxu@google.com/ - Address API design comments in v2. - Let memfd_create() to set X bit at creation time. - A new pid namespace sysctl: vm.memfd_noexec to control behavior of X bit. - A new security hook in memfd_create(). v2:https://lore.kernel.org/lkml/20220805222126.142525-1-jeffxu@google.com/ - address comments in V1. - add sysctl (vm.mfd_noexec) to set the default file permissions of memfd_create to be non-executable. v1:https://lwn.net/Articles/890096/ [1] https://crbug.com/1305411 [2] https://bugs.chromium.org/p/chromium/issues/list?q=type%3Dbug-security%20me… [3] https://lwn.net/Articles/781013/ Daniel Verkamp (2): mm/memfd: add F_SEAL_EXEC selftests/memfd: add tests for F_SEAL_EXEC Jeff Xu (4): mm/memfd: add MFD_NOEXEC_SEAL and MFD_EXEC mm/memfd: Add write seals when apply SEAL_EXEC to executable memfd selftests/memfd: add tests for MFD_NOEXEC_SEAL MFD_EXEC mm/memfd: security hook for memfd_create include/linux/lsm_hook_defs.h | 1 + include/linux/lsm_hooks.h | 4 + include/linux/pid_namespace.h | 19 ++ include/linux/security.h | 6 + include/uapi/linux/fcntl.h | 1 + include/uapi/linux/memfd.h | 4 + kernel/pid_namespace.c | 5 + kernel/pid_sysctl.h | 59 ++++ mm/memfd.c | 61 +++- mm/shmem.c | 6 + security/security.c | 5 + tools/testing/selftests/memfd/fuse_test.c | 1 + tools/testing/selftests/memfd/memfd_test.c | 341 ++++++++++++++++++++- 13 files changed, 510 insertions(+), 3 deletions(-) create mode 100644 kernel/pid_sysctl.h base-commit: eb7081409f94a9a8608593d0fb63a1aa3d6f95d8 -- 2.39.0.rc1.256.g54fd8350bd-goog

3 months

9
25
0 0

[PATCH v2] selftests/ptrace/get_syscall_info: fix for MIPS n32

by Dmitry V. Levin

MIPS n32 is one of two ILP32 architectures supported by the kernel that have 64-bit syscall arguments (another one is x32). When this test passed 32-bit arguments to syscall(), they were sign-extended in libc, PTRACE_GET_SYSCALL_INFO reported these sign-extended 64-bit values, and the test complained about the mismatch. Fix this by passing arguments of the appropriate type to syscall(), which is "unsigned long long" on MIPS n32, and __kernel_ulong_t on other architectures. As a side effect, this also extends the test on all 64-bit architectures by choosing constants that don't fit into 32-bit integers. Signed-off-by: Dmitry V. Levin <ldv(a)strace.io> --- v2: Fixed MIPS #ifdef. .../selftests/ptrace/get_syscall_info.c | 53 +++++++++++-------- 1 file changed, 32 insertions(+), 21 deletions(-) diff --git a/tools/testing/selftests/ptrace/get_syscall_info.c b/tools/testing/selftests/ptrace/get_syscall_info.c index 5bcd1c7b5be6..2970f72d66d3 100644 --- a/tools/testing/selftests/ptrace/get_syscall_info.c +++ b/tools/testing/selftests/ptrace/get_syscall_info.c @@ -11,8 +11,19 @@ #include <err.h> #include <signal.h> #include <asm/unistd.h> +#include <linux/types.h> #include "linux/ptrace.h" +#if defined(_MIPS_SIM) && _MIPS_SIM == _MIPS_SIM_NABI32 +/* + * MIPS N32 is the only architecture where __kernel_ulong_t + * does not match the bitness of syscall arguments. + */ +typedef unsigned long long kernel_ulong_t; +#else +typedef __kernel_ulong_t kernel_ulong_t; +#endif + static int kill_tracee(pid_t pid) { @@ -42,37 +53,37 @@ sys_ptrace(int request, pid_t pid, unsigned long addr, unsigned long data) TEST(get_syscall_info) { - static const unsigned long args[][7] = { + const kernel_ulong_t args[][7] = { /* a sequence of architecture-agnostic syscalls */ { __NR_chdir, - (unsigned long) "", - 0xbad1fed1, - 0xbad2fed2, - 0xbad3fed3, - 0xbad4fed4, - 0xbad5fed5 + (uintptr_t) "", + (kernel_ulong_t) 0xdad1bef1bad1fed1ULL, + (kernel_ulong_t) 0xdad2bef2bad2fed2ULL, + (kernel_ulong_t) 0xdad3bef3bad3fed3ULL, + (kernel_ulong_t) 0xdad4bef4bad4fed4ULL, + (kernel_ulong_t) 0xdad5bef5bad5fed5ULL }, { __NR_gettid, - 0xcaf0bea0, - 0xcaf1bea1, - 0xcaf2bea2, - 0xcaf3bea3, - 0xcaf4bea4, - 0xcaf5bea5 + (kernel_ulong_t) 0xdad0bef0caf0bea0ULL, + (kernel_ulong_t) 0xdad1bef1caf1bea1ULL, + (kernel_ulong_t) 0xdad2bef2caf2bea2ULL, + (kernel_ulong_t) 0xdad3bef3caf3bea3ULL, + (kernel_ulong_t) 0xdad4bef4caf4bea4ULL, + (kernel_ulong_t) 0xdad5bef5caf5bea5ULL }, { __NR_exit_group, 0, - 0xfac1c0d1, - 0xfac2c0d2, - 0xfac3c0d3, - 0xfac4c0d4, - 0xfac5c0d5 + (kernel_ulong_t) 0xdad1bef1fac1c0d1ULL, + (kernel_ulong_t) 0xdad2bef2fac2c0d2ULL, + (kernel_ulong_t) 0xdad3bef3fac3c0d3ULL, + (kernel_ulong_t) 0xdad4bef4fac4c0d4ULL, + (kernel_ulong_t) 0xdad5bef5fac5c0d5ULL } }; - const unsigned long *exp_args; + const kernel_ulong_t *exp_args; pid_t pid = fork(); @@ -154,7 +165,7 @@ TEST(get_syscall_info) } ASSERT_LT(0, (rc = sys_ptrace(PTRACE_GET_SYSCALL_INFO, pid, size, - (unsigned long) &info))) { + (uintptr_t) &info))) { LOG_KILL_TRACEE("PTRACE_GET_SYSCALL_INFO: %m"); } ASSERT_EQ(expected_none_size, rc) { @@ -177,7 +188,7 @@ TEST(get_syscall_info) case SIGTRAP | 0x80: ASSERT_LT(0, (rc = sys_ptrace(PTRACE_GET_SYSCALL_INFO, pid, size, - (unsigned long) &info))) { + (uintptr_t) &info))) { LOG_KILL_TRACEE("PTRACE_GET_SYSCALL_INFO: %m"); } switch (ptrace_stop) { -- ldv

6 months, 3 weeks

3
7
0 0

[RFC PATCH 00/39] 1G page support for guest_memfd

by Ackerley Tng

Hello, This patchset is our exploration of how to support 1G pages in guest_memfd, and how the pages will be used in Confidential VMs. The patchset covers: + How to get 1G pages + Allowing mmap() of guest_memfd to userspace so that both private and shared memory can use the same physical pages + Splitting and reconstructing pages to support conversions and mmap() + How the VM, userspace and guest_memfd interact to support conversions + Selftests to test all the above + Selftests also demonstrate the conversion flow between VM, userspace and guest_memfd. Why 1G pages in guest memfd? Bring guest_memfd to performance and memory savings parity with VMs that are backed by HugeTLBfs. + Performance is improved with 1G pages by more TLB hits and faster page walks on TLB misses. + Memory savings from 1G pages comes from HugeTLB Vmemmap Optimization (HVO). Options for 1G page support: 1. HugeTLB 2. Contiguous Memory Allocator (CMA) 3. Other suggestions are welcome! Comparison between options: 1. HugeTLB + Refactor HugeTLB to separate allocator from the rest of HugeTLB + Pro: Graceful transition for VMs backed with HugeTLB to guest_memfd + Near term: Allows co-tenancy of HugeTLB and guest_memfd backed VMs + Pro: Can provide iterative steps toward new future allocator + Unexplored: Managing userspace-visible changes + e.g. HugeTLB's free_hugepages will decrease if HugeTLB is used, but not when future allocator is used 2. CMA + Port some HugeTLB features to be applied on CMA + Pro: Clean slate What would refactoring HugeTLB involve? (Some refactoring was done in this RFC, more can be done.) 1. Broadly involves separating the HugeTLB allocator from the rest of HugeTLB + Brings more modularity to HugeTLB + No functionality change intended + Likely step towards HugeTLB's integration into core-mm 2. guest_memfd will use just the allocator component of HugeTLB, not including the complex parts of HugeTLB like + Userspace reservations (resv_map) + Shared PMD mappings + Special page walkers What features would need to be ported to CMA? + Improved allocation guarantees + Per NUMA node pool of huge pages + Subpools per guest_memfd + Memory savings + Something like HugeTLB Vmemmap Optimization + Configuration/reporting features + Configuration of number of pages available (and per NUMA node) at and after host boot + Reporting of memory usage/availability statistics at runtime HugeTLB was picked as the source of 1G pages for this RFC because it allows a graceful transition, and retains memory savings from HVO. To illustrate this, if a host machine uses HugeTLBfs to back VMs, and a confidential VM were to be scheduled on that host, some HugeTLBfs pages would have to be given up and returned to CMA for guest_memfd pages to be rebuilt from that memory. This requires memory to be reserved for HVO to be removed and reapplied on the new guest_memfd memory. This not only slows down memory allocation but also trims the benefits of HVO. Memory would have to be reserved on the host to facilitate these transitions. Improving how guest_memfd uses the allocator in a future revision of this RFC: To provide an easier transition away from HugeTLB, guest_memfd's use of HugeTLB should be limited to these allocator functions: + reserve(node, page_size, num_pages) => opaque handle + Used when a guest_memfd inode is created to reserve memory from backend allocator + allocate(handle, mempolicy, page_size) => folio + To allocate a folio from guest_memfd's reservation + split(handle, folio, target_page_size) => void + To take a huge folio, and split it to smaller folios, restore to filemap + reconstruct(handle, first_folio, nr_pages) => void + To take a folio, and reconstruct a huge folio out of nr_pages from the first_folio + free(handle, folio) => void + To return folio to guest_memfd's reservation + error(handle, folio) => void + To handle memory errors + unreserve(handle) => void + To return guest_memfd's reservation to allocator backend Userspace should only provide a page size when creating a guest_memfd and should not have to specify HugeTLB. Overview of patches: + Patches 01-12 + Many small changes to HugeTLB, mostly to separate HugeTLBfs concepts from HugeTLB, and to expose HugeTLB functions. + Patches 13-16 + Letting guest_memfd use HugeTLB + Creation of each guest_memfd reserves pages from HugeTLB's global hstate and puts it into the guest_memfd inode's subpool + Each folio allocation takes a page from the guest_memfd inode's subpool + Patches 17-21 + Selftests for new HugeTLB features in guest_memfd + Patches 22-24 + More small changes on the HugeTLB side to expose functions needed by guest_memfd + Patch 25: + Uses the newly available functions from patches 22-24 to split HugeTLB pages. In this patch, HugeTLB folios are always split to 4K before any usage, private or shared. + Patches 26-28 + Allow mmap() in guest_memfd and faulting in shared pages + Patch 29 + Enables conversion between private/shared pages + Patch 30 + Required to zero folios after conversions to avoid leaking initialized kernel memory + Patch 31-38 + Add selftests to test mapping pages to userspace, guest/host memory sharing and update conversions tests + Patch 33 illustrates the conversion flow between VM/userspace/guest_memfd + Patch 39 + Dynamically split and reconstruct HugeTLB pages instead of always splitting before use. All earlier selftests are expected to still pass. TODOs: + Add logic to wait for safe_refcount [1] + Look into lazy splitting/reconstruction of pages + Currently, when the KVM_SET_MEMORY_ATTRIBUTES is invoked, not only is the mem_attr_array and faultability updated, the pages in the requested range are also split/reconstructed as necessary. We want to look into delaying splitting/reconstruction to fault time. + Solve race between folios being faulted in and being truncated + When running private_mem_conversions_test with more than 1 vCPU, a folio getting truncated may get faulted in by another process, causing elevated mapcounts when the folio is freed (VM_BUG_ON_FOLIO). + Add intermediate splits (1G should first split to 2M and not split directly to 4K) + Use guest's lock instead of hugetlb_lock + Use multi-index xarray/replace xarray with some other data struct for faultability flag + Refactor HugeTLB better, present generic allocator interface Please let us know your thoughts on: + HugeTLB as the choice of transitional allocator backend + Refactoring HugeTLB to provide generic allocator interface + Shared/private conversion flow + Requiring user to request kernel to unmap pages from userspace using madvise(MADV_DONTNEED) + Failing conversion on elevated mapcounts/pincounts/refcounts + Process of splitting/reconstructing page + Anything else! [1] https://lore.kernel.org/all/20240829-guest-memfd-lib-v2-0-b9afc1ff3656@quic… Ackerley Tng (37): mm: hugetlb: Simplify logic in dequeue_hugetlb_folio_vma() mm: hugetlb: Refactor vma_has_reserves() to should_use_hstate_resv() mm: hugetlb: Remove unnecessary check for avoid_reserve mm: mempolicy: Refactor out policy_node_nodemask() mm: hugetlb: Refactor alloc_buddy_hugetlb_folio_with_mpol() to interpret mempolicy instead of vma mm: hugetlb: Refactor dequeue_hugetlb_folio_vma() to use mpol mm: hugetlb: Refactor out hugetlb_alloc_folio mm: truncate: Expose preparation steps for truncate_inode_pages_final mm: hugetlb: Expose hugetlb_subpool_{get,put}_pages() mm: hugetlb: Add option to create new subpool without using surplus mm: hugetlb: Expose hugetlb_acct_memory() mm: hugetlb: Move and expose hugetlb_zero_partial_page() KVM: guest_memfd: Make guest mem use guest mem inodes instead of anonymous inodes KVM: guest_memfd: hugetlb: initialization and cleanup KVM: guest_memfd: hugetlb: allocate and truncate from hugetlb KVM: guest_memfd: Add page alignment check for hugetlb guest_memfd KVM: selftests: Add basic selftests for hugetlb-backed guest_memfd KVM: selftests: Support various types of backing sources for private memory KVM: selftests: Update test for various private memory backing source types KVM: selftests: Add private_mem_conversions_test.sh KVM: selftests: Test that guest_memfd usage is reported via hugetlb mm: hugetlb: Expose vmemmap optimization functions mm: hugetlb: Expose HugeTLB functions for promoting/demoting pages mm: hugetlb: Add functions to add/move/remove from hugetlb lists KVM: guest_memfd: Track faultability within a struct kvm_gmem_private KVM: guest_memfd: Allow mmapping guest_memfd files KVM: guest_memfd: Use vm_type to determine default faultability KVM: Handle conversions in the SET_MEMORY_ATTRIBUTES ioctl KVM: guest_memfd: Handle folio preparation for guest_memfd mmap KVM: selftests: Allow vm_set_memory_attributes to be used without asserting return value of 0 KVM: selftests: Test using guest_memfd memory from userspace KVM: selftests: Test guest_memfd memory sharing between guest and host KVM: selftests: Add notes in private_mem_kvm_exits_test for mmap-able guest_memfd KVM: selftests: Test that pinned pages block KVM from setting memory attributes to PRIVATE KVM: selftests: Refactor vm_mem_add to be more flexible KVM: selftests: Add helper to perform madvise by memslots KVM: selftests: Update private_mem_conversions_test for mmap()able guest_memfd Vishal Annapurve (2): KVM: guest_memfd: Split HugeTLB pages for guest_memfd use KVM: guest_memfd: Dynamically split/reconstruct HugeTLB page fs/hugetlbfs/inode.c | 35 +- include/linux/hugetlb.h | 54 +- include/linux/kvm_host.h | 1 + include/linux/mempolicy.h | 2 + include/linux/mm.h | 1 + include/uapi/linux/kvm.h | 26 + include/uapi/linux/magic.h | 1 + mm/hugetlb.c | 346 ++-- mm/hugetlb_vmemmap.h | 11 - mm/mempolicy.c | 36 +- mm/truncate.c | 26 +- tools/include/linux/kernel.h | 4 +- tools/testing/selftests/kvm/Makefile | 3 + .../kvm/guest_memfd_hugetlb_reporting_test.c | 222 +++ .../selftests/kvm/guest_memfd_pin_test.c | 104 ++ .../selftests/kvm/guest_memfd_sharing_test.c | 160 ++ .../testing/selftests/kvm/guest_memfd_test.c | 238 ++- .../testing/selftests/kvm/include/kvm_util.h | 45 +- .../testing/selftests/kvm/include/test_util.h | 18 + tools/testing/selftests/kvm/lib/kvm_util.c | 443 +++-- tools/testing/selftests/kvm/lib/test_util.c | 99 ++ .../kvm/x86_64/private_mem_conversions_test.c | 158 +- .../x86_64/private_mem_conversions_test.sh | 91 + .../kvm/x86_64/private_mem_kvm_exits_test.c | 11 +- virt/kvm/guest_memfd.c | 1563 ++++++++++++++++- virt/kvm/kvm_main.c | 17 + virt/kvm/kvm_mm.h | 16 + 27 files changed, 3288 insertions(+), 443 deletions(-) create mode 100644 tools/testing/selftests/kvm/guest_memfd_hugetlb_reporting_test.c create mode 100644 tools/testing/selftests/kvm/guest_memfd_pin_test.c create mode 100644 tools/testing/selftests/kvm/guest_memfd_sharing_test.c create mode 100755 tools/testing/selftests/kvm/x86_64/private_mem_conversions_test.sh -- 2.46.0.598.g6f2099f65c-goog

7 months, 1 week

17
129
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror January 2025