July 2024 - Linux-kselftest-mirror

[PATCH v5 00/28] NT synchronization primitive driver

by Elizabeth Figura

This patch series implements a new char misc driver, /dev/ntsync, which is used to implement Windows NT synchronization primitives. NT synchronization primitives are unique in that the wait functions both are vectored, operate on multiple types of object with different behaviour (mutex, semaphore, event), and affect the state of the objects they wait on. This model is not compatible with existing kernel synchronization objects or interfaces, and therefore the ntsync driver implements its own wait queues and locking. This patch series is rebased against the "char-misc-next" branch of gregkh/char-misc.git. == Background == The Wine project emulates the Windows API in user space. One particular part of that API, namely the NT synchronization primitives, have historically been implemented via RPC to a dedicated "kernel" process. However, more recent applications use these APIs more strenuously, and the overhead of RPC has become a bottleneck. The NT synchronization APIs are too complex to implement on top of existing primitives without sacrificing correctness. Certain operations, such as NtPulseEvent() or the "wait-for-all" mode of NtWaitForMultipleObjects(), require direct control over the underlying wait queue, and implementing a wait queue sufficiently robust for Wine in user space is not possible. This proposed driver, therefore, implements the problematic interfaces directly in the Linux kernel. This driver was presented at Linux Plumbers Conference 2023. For those further interested in the history of synchronization in Wine and past attempts to solve this problem in user space, a recording of the presentation can be viewed here: https://www.youtube.com/watch?v=NjU4nyWyhU8 == Performance == The performance measurements described below are copied from earlier versions of the patch set. While some of the code has changed, I do not currently anticipate that it has changed drastically enough to affect those measurements. The gain in performance varies wildly depending on the application in question and the user's hardware. For some games NT synchronization is not a bottleneck and no change can be observed, but for others frame rate improvements of 50 to 150 percent are not atypical. The following table lists frame rate measurements from a variety of games on a variety of hardware, taken by users Dmitry Skvortsov, FuzzyQuils, OnMars, and myself: Game Upstream ntsync improvement =========================================================================== Anger Foot 69 99 43% Call of Juarez 99.8 224.1 125% Dirt 3 110.6 860.7 678% Forza Horizon 5 108 160 48% Lara Croft: Temple of Osiris 141 326 131% Metro 2033 164.4 199.2 21% Resident Evil 2 26 77 196% The Crew 26 51 96% Tiny Tina's Wonderlands 130 360 177% Total War Saga: Troy 109 146 34% =========================================================================== == Patches == The intended semantics of the patches are broadly intended to match those of the corresponding Windows functions. For those not already familiar with the Windows functions (or their undocumented behaviour), patch 27/28 provides a detailed specification, and individual patches also include a brief description of the API they are implementing. The patches making use of this driver in Wine can be retrieved or browsed here: https://repo.or.cz/wine/zf.git/shortlog/refs/heads/ntsync5 == Previous versions == Changes from v4: * Rework wait-all locking code to avoid taking more than one spinlock at a time, and also to fix a race where the wait-all lock would not be not correctly taken. The new locking mechanism involves taking a simple spinlock for normal "any" waits, and taking a device-wide mutex for "all" waits or when locking any object that is involved in an "all" wait. The mechanism was written by Peter Zijlstra. * Try to reword or clarify various parts of the documentation (patch 27), per Peter Zijlstra. * I did not rename NTSYNC_IOC_SEM_POST to RELEASE (like NT) although this was suggested by Peter Zijlstra, mostly because it's not clear to me that renaming an already committed ioctl would be fine. The API committed isn't actually usable yet, though, so if altering it would be fine on those grounds, I can revise this series to rename the function accordingly. * Similarly, I did not change the create ioctls to return the fd directly, although this was suggested and would be a bit simpler and cleaner, because NTSYNC_IOC_CREATE_SEM already exists upstream and returns the fd through a struct. I can make this change in the next revision if that'd be preferable. I also still would appreciate a clarification on the advice in [1]. [1] https://docs.kernel.org/driver-api/ioctl.html#return-code * Link to v4: https://lore.kernel.org/lkml/20240416010837.333694-1-zfigura@codeweavers.co… * Link to v3: https://lore.kernel.org/lkml/20240329000621.148791-1-zfigura@codeweavers.co… * Link to v2: https://lore.kernel.org/lkml/20240219223833.95710-1-zfigura@codeweavers.com/ * Link to v1: https://lore.kernel.org/lkml/20240214233645.9273-1-zfigura@codeweavers.com/ * Link to RFC v2: https://lore.kernel.org/lkml/20240131021356.10322-1-zfigura@codeweavers.com/ * Link to RFC v1: https://lore.kernel.org/lkml/20240124004028.16826-1-zfigura@codeweavers.com/ Elizabeth Figura (28): ntsync: Introduce NTSYNC_IOC_WAIT_ANY. ntsync: Introduce NTSYNC_IOC_WAIT_ALL. ntsync: Introduce NTSYNC_IOC_CREATE_MUTEX. ntsync: Introduce NTSYNC_IOC_MUTEX_UNLOCK. ntsync: Introduce NTSYNC_IOC_MUTEX_KILL. ntsync: Introduce NTSYNC_IOC_CREATE_EVENT. ntsync: Introduce NTSYNC_IOC_EVENT_SET. ntsync: Introduce NTSYNC_IOC_EVENT_RESET. ntsync: Introduce NTSYNC_IOC_EVENT_PULSE. ntsync: Introduce NTSYNC_IOC_SEM_READ. ntsync: Introduce NTSYNC_IOC_MUTEX_READ. ntsync: Introduce NTSYNC_IOC_EVENT_READ. ntsync: Introduce alertable waits. selftests: ntsync: Add some tests for semaphore state. selftests: ntsync: Add some tests for mutex state. selftests: ntsync: Add some tests for NTSYNC_IOC_WAIT_ANY. selftests: ntsync: Add some tests for NTSYNC_IOC_WAIT_ALL. selftests: ntsync: Add some tests for wakeup signaling with WINESYNC_IOC_WAIT_ANY. selftests: ntsync: Add some tests for wakeup signaling with WINESYNC_IOC_WAIT_ALL. selftests: ntsync: Add some tests for manual-reset event state. selftests: ntsync: Add some tests for auto-reset event state. selftests: ntsync: Add some tests for wakeup signaling with events. selftests: ntsync: Add tests for alertable waits. selftests: ntsync: Add some tests for wakeup signaling via alerts. selftests: ntsync: Add a stress test for contended waits. maintainers: Add an entry for ntsync. docs: ntsync: Add documentation for the ntsync uAPI. ntsync: No longer depend on BROKEN. Documentation/userspace-api/index.rst | 1 + Documentation/userspace-api/ntsync.rst | 398 +++++ MAINTAINERS | 9 + drivers/misc/Kconfig | 1 - drivers/misc/ntsync.c | 989 +++++++++++- include/uapi/linux/ntsync.h | 39 + tools/testing/selftests/Makefile | 1 + .../selftests/drivers/ntsync/.gitignore | 1 + .../testing/selftests/drivers/ntsync/Makefile | 7 + tools/testing/selftests/drivers/ntsync/config | 1 + .../testing/selftests/drivers/ntsync/ntsync.c | 1407 +++++++++++++++++ 11 files changed, 2850 insertions(+), 4 deletions(-) create mode 100644 Documentation/userspace-api/ntsync.rst create mode 100644 tools/testing/selftests/drivers/ntsync/.gitignore create mode 100644 tools/testing/selftests/drivers/ntsync/Makefile create mode 100644 tools/testing/selftests/drivers/ntsync/config create mode 100644 tools/testing/selftests/drivers/ntsync/ntsync.c base-commit: f5b335dc025cfee90957efa90dc72fada0d5abb4 -- 2.43.0

10 months

1
30
0 0

[PATCH net-next v17 00/14] Device Memory TCP

by Mina Almasry

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. 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> v10: https://patchwork.kernel.org/project/netdevbpf/list/?series=852422&state=* ==== Major Changes: -------------- v9 was sent right before the merge window closed (sorry!). v10 is almost a re-send of the series now that the merge window re-opened. Only rebased to latest net-next and addressed some minor iterative comments received on v9. As usual, the full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v10/ Detailed changelog: ------------------- - Fixed tokens leaking in DONTNEED setsockopt (Nikolay). - Moved net_iov_dma_addr() to devmem.c and made it a devmem specific helpers (David). - Rename hook alloc_pages to alloc_netmems as alloc_pages is now preprocessor macro defined and causes a build error. v9: === Major Changes: -------------- GVE queue API has been merged. Submitting this version as non-RFC after rebasing on top of the merged API, and dropped the out of tree queue API I was carrying on github. Addressed the little feedback v8 has received. Detailed changelog: ------------------ - Added new patch from David Wei to this series for netdev_rx_queue_restart() - Fixed sparse error. - Removed CONFIG_ checks in netmem_is_net_iov() - Flipped skb->readable to skb->unreadable - Minor fixes to selftests & docs. RFC v8: ======= Major Changes: -------------- - Fixed build error generated by patch-by-patch build. - Applied docs suggestions from Randy. RFC v7: ======= Major Changes: -------------- This revision largely rebases on top of net-next and addresses the feedback RFCv6 received from folks, namely Jakub, Yunsheng, Arnd, David, & Pavel. The series remains in RFC because the queue-API ndos defined in this series are not yet implemented. I have a GVE implementation I carry out of tree for my testing. A upstreamable GVE implementation is in the works. Aside from that, in my estimation all the patches are ready for review/merge. Please do take a look. As usual the full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v7/ Detailed changelog: - Use admin-perm in netlink API. - Addressed feedback from Jakub with regards to netlink API implementation. - Renamed devmem.c functions to something more appropriate for that file. - Improve the performance seen through the page_pool benchmark. - Fix the value definition of all the SO_DEVMEM_* uapi. - Various fixes to documentation. Perf - page-pool benchmark: --------------------------- Improved performance of bench_page_pool_simple.ko tests compared to v6: https://pastebin.com/raw/v5dYRg8L net-next base: 8 cycle fast path. RFC v6: 10 cycle fast path. RFC v7: 9 cycle fast path. RFC v7 with CONFIG_DMA_SHARED_BUFFER disabled: 8 cycle fast path, same as baseline. Perf - Devmem TCP benchmark: --------------------- Perf is about the same regardless of the changes in v7, namely the removal of the static_branch_unlikely to improve the page_pool benchmark performance: 189/200gbps bi-directional throughput with RX devmem TCP and regular TCP TX i.e. ~95% line rate. RFC v6: ======= Major Changes: -------------- This revision largely rebases on top of net-next and addresses the little feedback RFCv5 received. The series remains in RFC because the queue-API ndos defined in this series are not yet implemented. I have a GVE implementation I carry out of tree for my testing. A upstreamable GVE implementation is in the works. Aside from that, in my estimation all the patches are ready for review/merge. Please do take a look. As usual the full devmem TCP changes including the full GVE driver implementation is here: https://github.com/mina/linux/commits/tcpdevmem-v6/ This version also comes with some performance data recorded in the cover letter (see below changelog). Detailed changelog: - Rebased on top of the merged netmem_ref changes. - Converted skb->dmabuf to skb->readable (Pavel). Pavel's original suggestion was to remove the skb->dmabuf flag entirely, but when I looked into it closely, I found the issue that if we remove the flag we have to dereference the shinfo(skb) pointer to obtain the first frag to tell whether an skb is readable or not. This can cause a performance regression if it dirties the cache line when the shinfo(skb) was not really needed. Instead, I converted the skb->dmabuf flag into a generic skb->readable flag which can be re-used by io_uring 0-copy RX. - Squashed a few locking optimizations from Eric Dumazet in the RX path and the DEVMEM_DONTNEED setsockopt. - Expanded the tests a bit. Added validation for invalid scenarios and added some more coverage. Perf - page-pool benchmark: --------------------------- bench_page_pool_simple.ko tests with and without these changes: https://pastebin.com/raw/ncHDwAbn AFAIK the number that really matters in the perf tests is the 'tasklet_page_pool01_fast_path Per elem'. This one measures at about 8 cycles without the changes but there is some 1 cycle noise in some results. With the patches this regresses to 9 cycles with the changes but there is 1 cycle noise occasionally running this test repeatedly. Lastly I tried disable the static_branch_unlikely() in netmem_is_net_iov() check. To my surprise disabling the static_branch_unlikely() check reduces the fast path back to 8 cycles, but the 1 cycle noise remains. Perf - Devmem TCP benchmark: --------------------- 189/200gbps bi-directional throughput with RX devmem TCP and regular TCP TX i.e. ~95% line rate. Major changes in RFC v5: ======================== 1. Rebased on top of 'Abstract page from net stack' series and used the new netmem type to refer to LSB set pointers instead of re-using struct page. 2. Downgraded this series back to RFC and called it RFC v5. This is because this series is now dependent on 'Abstract page from net stack'[1] and the queue API. Both are removed from the series to reduce the patch # and those bits are fairly independent or pre-requisite work. 3. Reworked the page_pool devmem support to use netmem and for some more unified handling. 4. Reworked the reference counting of net_iov (renamed from page_pool_iov) to use pp_ref_count for refcounting. The full changes including the dependent series and GVE page pool support is here: https://github.com/mina/linux/commits/tcpdevmem-rfcv5/ [1] https://patchwork.kernel.org/project/netdevbpf/list/?series=810774 Major changes in v1: ==================== 1. Implemented MVP queue API ndos to remove the userspace-visible driver reset. 2. Fixed issues in the napi_pp_put_page() devmem frag unref path. 3. Removed RFC tag. Many smaller addressed comments across all the patches (patches have individual change log). Full tree including the rest of the GVE driver changes: https://github.com/mina/linux/commits/tcpdevmem-v1 Changes in RFC v3: ================== 1. Pulled in the memory-provider dependency from Jakub's RFC[1] to make the series reviewable and mergeable. 2. Implemented multi-rx-queue binding which was a todo in v2. 3. Fix to cmsg handling. The sticking point in RFC v2[2] was the device reset required to refill the device rx-queues after the dmabuf bind/unbind. The solution suggested as I understand is a subset of the per-queue management ops Jakub suggested or similar: https://lore.kernel.org/netdev/20230815171638.4c057dcd@kernel.org/ This is not addressed in this revision, because: 1. This point was discussed at netconf & netdev and there is openness to using the current approach of requiring a device reset. 2. Implementing individual queue resetting seems to be difficult for my test bed with GVE. My prototype to test this ran into issues with the rx-queues not coming back up properly if reset individually. At the moment I'm unsure if it's a mistake in the POC or a genuine issue in the virtualization stack behind GVE, which currently doesn't test individual rx-queue restart. 3. Our usecases are not bothered by requiring a device reset to refill the buffer queues, and we'd like to support NICs that run into this limitation with resetting individual queues. My thought is that drivers that have trouble with per-queue configs can use the support in this series, while drivers that support new netdev ops to reset individual queues can automatically reset the queue as part of the dma-buf bind/unbind. The same approach with device resets is presented again for consideration with other sticking points addressed. This proposal includes the rx devmem path only proposed for merge. For a snapshot of my entire tree which includes the GVE POC page pool support & device memory support: https://github.com/torvalds/linux/compare/master...mina:linux:tcpdevmem-v3 [1] https://lore.kernel.org/netdev/f8270765-a27b-6ccf-33ea-cda097168d79@redhat.… [2] https://lore.kernel.org/netdev/CAHS8izOVJGJH5WF68OsRWFKJid1_huzzUK+hpKbLcL4… Changes in RFC v2: ================== The sticking point in RFC v1[1] was the dma-buf pages approach we used to deliver the device memory to the TCP stack. RFC v2 is a proof-of-concept that attempts to resolve this by implementing scatterlist support in the networking stack, such that we can import the dma-buf scatterlist directly. This is the approach proposed at a high level here[2]. Detailed changes: 1. Replaced dma-buf pages approach with importing scatterlist into the page pool. 2. Replace the dma-buf pages centric API with a netlink API. 3. Removed the TX path implementation - there is no issue with implementing the TX path with scatterlist approach, but leaving out the TX path makes it easier to review. 4. Functionality is tested with this proposal, but I have not conducted perf testing yet. I'm not sure there are regressions, but I removed perf claims from the cover letter until they can be re-confirmed. 5. Added Signed-off-by: contributors to the implementation. 6. Fixed some bugs with the RX path since RFC v1. Any feedback welcome, but specifically the biggest pending questions needing feedback IMO are: 1. Feedback on the scatterlist-based approach in general. 2. Netlink API (Patch 1 & 2). 3. Approach to handle all the drivers that expect to receive pages from the page pool (Patch 6). [1] https://lore.kernel.org/netdev/dfe4bae7-13a0-3c5d-d671-f61b375cb0b4@gmail.c… [2] https://lore.kernel.org/netdev/CAHS8izPm6XRS54LdCDZVd0C75tA1zHSu6jLVO8nzTLX… ================== * TL;DR: Device memory TCP (devmem TCP) is a proposal for transferring data to and/or from device memory efficiently, without bouncing the data to a host memory buffer. * Problem: A large amount of data transfers have device memory as the source and/or destination. Accelerators drastically increased the volume of such transfers. Some examples include: - ML accelerators transferring large amounts of training data from storage into GPU/TPU memory. In some cases ML training setup time can be as long as 50% of TPU compute time, improving data transfer throughput & efficiency can help improving GPU/TPU utilization. - Distributed training, where ML accelerators, such as GPUs on different hosts, exchange data among them. - Distributed raw block storage applications transfer large amounts of data with remote SSDs, much of this data does not require host processing. Today, the majority of the Device-to-Device data transfers the network are implemented as the following low level operations: Device-to-Host copy, Host-to-Host network transfer, and Host-to-Device copy. The implementation is suboptimal, especially for bulk data transfers, and can put significant strains on system resources, such as host memory bandwidth, PCIe bandwidth, etc. One important reason behind the current state is the kernel’s lack of semantics to express device to network transfers. * Proposal: In this patch series we attempt to optimize this use case by implementing socket APIs that enable the user to: 1. send device memory across the network directly, and 2. receive incoming network packets directly into device memory. Packet _payloads_ go directly from the NIC to device memory for receive and from device memory to NIC for transmit. Packet _headers_ go to/from host memory and are processed by the TCP/IP stack normally. The NIC _must_ support header split to achieve this. Advantages: - Alleviate host memory bandwidth pressure, compared to existing network-transfer + device-copy semantics. - Alleviate PCIe BW pressure, by limiting data transfer to the lowest level of the PCIe tree, compared to traditional path which sends data through the root complex. * Patch overview: ** Part 1: netlink API Gives user ability to bind dma-buf to an RX queue. ** Part 2: scatterlist support Currently the standard for device memory sharing is DMABUF, which doesn't generate struct pages. On the other hand, networking stack (skbs, drivers, and page pool) operate on pages. We have 2 options: 1. Generate struct pages for dmabuf device memory, or, 2. Modify the networking stack to process scatterlist. Approach #1 was attempted in RFC v1. RFC v2 implements approach #2. ** part 3: page pool support We piggy back on page pool memory providers proposal: https://github.com/kuba-moo/linux/tree/pp-providers It allows the page pool to define a memory provider that provides the page allocation and freeing. It helps abstract most of the device memory TCP changes from the driver. ** part 4: support for unreadable skb frags Page pool iovs are not accessible by the host; we implement changes throughput the networking stack to correctly handle skbs with unreadable frags. ** Part 5: recvmsg() APIs We define user APIs for the user to send and receive device memory. Not included with this series is the GVE devmem TCP support, just to simplify the review. Code available here if desired: https://github.com/mina/linux/tree/tcpdevmem This series is built on top of net-next with Jakub's pp-providers changes cherry-picked. * NIC dependencies: 1. (strict) Devmem TCP require the NIC to support header split, i.e. the capability to split incoming packets into a header + payload and to put each into a separate buffer. Devmem TCP works by using device memory for the packet payload, and host memory for the packet headers. 2. (optional) Devmem TCP works better with flow steering support & RSS support, i.e. the NIC's ability to steer flows into certain rx queues. This allows the sysadmin to enable devmem TCP on a subset of the rx queues, and steer devmem TCP traffic onto these queues and non devmem TCP elsewhere. The NIC I have access to with these properties is the GVE with DQO support running in Google Cloud, but any NIC that supports these features would suffice. I may be able to help reviewers bring up devmem TCP on their NICs. * Testing: The series includes a udmabuf kselftest that show a simple use case of devmem TCP and validates the entire data path end to end without a dependency on a specific dmabuf provider. ** Test Setup Kernel: net-next with this series and memory provider API cherry-picked locally. Hardware: Google Cloud A3 VMs. NIC: GVE with header split & RSS & flow steering support. Cc: Pavel Begunkov <asml.silence(a)gmail.com> Cc: David Wei <dw(a)davidwei.uk> Cc: Jason Gunthorpe <jgg(a)ziepe.ca> Cc: Yunsheng Lin <linyunsheng(a)huawei.com> Cc: Shailend Chand <shailend(a)google.com> Cc: Harshitha Ramamurthy <hramamurthy(a)google.com> Cc: Shakeel Butt <shakeel.butt(a)linux.dev> Cc: Jeroen de Borst <jeroendb(a)google.com> Cc: Praveen Kaligineedi <pkaligineedi(a)google.com> Mina Almasry (14): 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: move dmaddr helpers to .c file 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/skbuff.h | 61 ++- include/linux/skbuff_ref.h | 9 +- include/linux/socket.h | 1 + include/net/devmem.h | 128 ++++++ include/net/mp_dmabuf_devmem.h | 44 ++ include/net/netdev_rx_queue.h | 5 + include/net/netmem.h | 164 +++++++- include/net/page_pool/helpers.h | 42 +- include/net/page_pool/types.h | 8 + include/net/sock.h | 2 + include/net/tcp.h | 5 +- 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/core/Makefile | 3 +- net/core/datagram.c | 6 + net/core/dev.c | 7 +- net/core/devmem.c | 378 +++++++++++++++++ net/core/gro.c | 3 +- net/core/netdev-genl-gen.c | 23 + net/core/netdev-genl-gen.h | 6 + net/core/netdev-genl.c | 111 +++++ net/core/netdev_rx_queue.c | 74 ++++ net/core/netmem_priv.h | 36 ++ net/core/page_pool.c | 147 +++++-- net/core/page_pool_priv.h | 3 + net/core/page_pool_user.c | 4 + net/core/skbuff.c | 77 +++- net/core/sock.c | 68 +++ net/ipv4/esp4.c | 3 +- net/ipv4/tcp.c | 261 +++++++++++- net/ipv4/tcp_input.c | 13 +- net/ipv4/tcp_ipv4.c | 16 + net/ipv4/tcp_minisocks.c | 2 + net/ipv4/tcp_output.c | 5 +- net/ipv6/esp6.c | 3 +- net/packet/af_packet.c | 4 +- tools/include/uapi/linux/netdev.h | 13 + tools/testing/selftests/net/.gitignore | 1 + tools/testing/selftests/net/Makefile | 9 + tools/testing/selftests/net/ncdevmem.c | 536 ++++++++++++++++++++++++ 49 files changed, 2563 insertions(+), 121 deletions(-) create mode 100644 Documentation/networking/devmem.rst create mode 100644 include/net/devmem.h create mode 100644 include/net/mp_dmabuf_devmem.h create mode 100644 net/core/devmem.c create mode 100644 net/core/netdev_rx_queue.c create mode 100644 net/core/netmem_priv.h create mode 100644 tools/testing/selftests/net/ncdevmem.c -- 2.46.0.rc1.232.g9752f9e123-goog

10 months, 1 week

5
21
0 0

[PATCH] KVM: selftests: arm64: Correct feature test for S1PIE in get-reg-list

by Mark Brown

The ID register for S1PIE is ID_AA64MMFR3_EL1.S1PIE which is bits 11:8 but get-reg-list uses a shift of 4, checking SCTLRX instead. Use a shift of 8 instead. Fixes: 5f0419a0083b ("KVM: selftests: get-reg-list: add Permission Indirection registers") Signed-off-by: Mark Brown <broonie(a)kernel.org> --- tools/testing/selftests/kvm/aarch64/get-reg-list.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/kvm/aarch64/get-reg-list.c b/tools/testing/selftests/kvm/aarch64/get-reg-list.c index 709d7d721760..4abebde78187 100644 --- a/tools/testing/selftests/kvm/aarch64/get-reg-list.c +++ b/tools/testing/selftests/kvm/aarch64/get-reg-list.c @@ -32,13 +32,13 @@ static struct feature_id_reg feat_id_regs[] = { { ARM64_SYS_REG(3, 0, 10, 2, 2), /* PIRE0_EL1 */ ARM64_SYS_REG(3, 0, 0, 7, 3), /* ID_AA64MMFR3_EL1 */ - 4, + 8, 1 }, { ARM64_SYS_REG(3, 0, 10, 2, 3), /* PIR_EL1 */ ARM64_SYS_REG(3, 0, 0, 7, 3), /* ID_AA64MMFR3_EL1 */ - 4, + 8, 1 } }; --- base-commit: 8400291e289ee6b2bf9779ff1c83a291501f017b change-id: 20240731-kvm-arm64-fix-s1pie-test-a40b6be58d7b Best regards, -- Mark Brown <broonie(a)kernel.org>

10 months, 1 week

4
8
0 0

[RFC PATCH 0/1] Add kselftest to detect boot event slowdowns

by Laura Nao

Add a new kselftest to detect and report slowdowns in key boot events. The test uses ftrace to track timings for specific boot events and compares these timestamps against reference values provided in YAML format. The test includes the following files: - `bootconfig` file: configures ftrace and lists reference key boot events. - `config` fragment: enables boot time tracing and attaches the bootconfig file to the kernel image. - `kprobe_timestamps_to_yaml.py` script: parses the current trace file to extract event names and timestamps and writes them to a YAML file. The script is intended to be run once to generate initial reference values; the generated file is not meant to be stored in the kernel sources but should be provided as input to the test itself. YAML format was chosen to allow easy integration with per-platform data used in other tests, such as the discoverable devices probe test in tools/testing/selftests/devices. Another option is to use JSON, as the file is not intended for manual editing and JSON is already supported by the Python standard library. - `test_boot_time.py` script: parses the current trace file and compares timestamps against the values in the YAML file provided as input. Reports a failure if any timestamp differs from the reference value by more than the specified delta. - `trace_utils.py` file: utility functions to mount debugfs and parse the trace file to extract relevant information. The bootconfig file provided is an initial draft with some reference kprobe events to showcase how the test works. I would appreciate feedback from those interested in running this test on which boot events should be added. Different key events might be relevant depending on the platform and its boot time requirements. This file should serve as a common ground and be populated with critical events and functions common to different platforms. Feedback on the overall approach of this test and suggestions for additional boot events to trace would be greatly appreciated. Example output with a deliberately small delta of 0.01 to demonstrate failures: TAP version 13 1..4 ok 1 populate_rootfs_begin # 'run_init_process_begin' differs by 0.033990 seconds. not ok 2 run_init_process_begin # 'run_init_process_end' differs by 0.033796 seconds. not ok 3 run_init_process_end ok 4 unpack_to_rootfs_begin # Totals: pass:2 fail:2 xfail:0 xpass:0 skip:0 error:0 This patch depends on "kselftest: Move ksft helper module to common directory": https://lore.kernel.org/all/20240705-dev-err-log-selftest-v2-2-163b9cd7b3c1… which was picked through the usb tree and is queued for 6.11-rc1. Best, Laura Laura Nao (1): kselftests: Add test to detect boot event slowdowns tools/testing/selftests/Makefile | 1 + tools/testing/selftests/boot-time/Makefile | 17 ++++ tools/testing/selftests/boot-time/bootconfig | 8 ++ tools/testing/selftests/boot-time/config | 4 + .../boot-time/kprobe_timestamps_to_yaml.py | 55 +++++++++++ .../selftests/boot-time/test_boot_time.py | 94 +++++++++++++++++++ .../selftests/boot-time/trace_utils.py | 63 +++++++++++++ 7 files changed, 242 insertions(+) create mode 100644 tools/testing/selftests/boot-time/Makefile create mode 100644 tools/testing/selftests/boot-time/bootconfig create mode 100644 tools/testing/selftests/boot-time/config create mode 100755 tools/testing/selftests/boot-time/kprobe_timestamps_to_yaml.py create mode 100755 tools/testing/selftests/boot-time/test_boot_time.py create mode 100644 tools/testing/selftests/boot-time/trace_utils.py -- 2.30.2

10 months, 1 week

2
5
0 0

Converting kselftest test modules to kunit

by Muhammad Usama Anjum

Hi Kees and All, There are several tests in kselftest subsystem which load modules to tests the internals of the kernel. Most of these test modules are just loaded by the kselftest, their status isn't read and reported to the user logs. Hence they don't provide benefit of executing those tests. I've found patches from Kees where he has been converting such kselftests to kunit tests [1]. The probable motivation is to move tests output of kselftest subsystem which only triggers tests without correctly reporting the results. On the other hand, kunit is there to test the kernel's internal functions which can't be done by userspace. Kselftest: Test user facing APIs from userspace Kunit: Test kernel's internal functions from kernelspace This brings me to conclusion that kselftest which are loading modules to test kernelspace should be converted to kunit tests. I've noted several such kselftests. This is just my understanding. Please mention if I'm correct above or more reasons to support kselftest test modules transformation into kunit test. [1] https://lore.kernel.org/all/20221018082824.never.845-kees@kernel.org/ -- BR, Muhammad Usama Anjum

10 months, 1 week

6
17
2 0

[PATCH RFT v7 0/9] fork: Support shadow stacks in clone3()

by Mark Brown

The kernel has recently added support for shadow stacks, currently x86 only using their CET feature but both arm64 and RISC-V have equivalent features (GCS and Zicfiss respectively), I am actively working on GCS[1]. With shadow stacks the hardware maintains an additional stack containing only the return addresses for branch instructions which is not generally writeable by userspace and ensures that any returns are to the recorded addresses. This provides some protection against ROP attacks and making it easier to collect call stacks. These shadow stacks are allocated in the address space of the userspace process. Our API for shadow stacks does not currently offer userspace any flexiblity for managing the allocation of shadow stacks for newly created threads, instead the kernel allocates a new shadow stack with the same size as the normal stack whenever a thread is created with the feature enabled. The stacks allocated in this way are freed by the kernel when the thread exits or shadow stacks are disabled for the thread. This lack of flexibility and control isn't ideal, in the vast majority of cases the shadow stack will be over allocated and the implicit allocation and deallocation is not consistent with other interfaces. As far as I can tell the interface is done in this manner mainly because the shadow stack patches were in development since before clone3() was implemented. Since clone3() is readily extensible let's add support for specifying a shadow stack when creating a new thread or process in a similar manner to how the normal stack is specified, keeping the current implicit allocation behaviour if one is not specified either with clone3() or through the use of clone(). The user must provide a shadow stack address and size, this must point to memory mapped for use as a shadow stackby map_shadow_stack() with a shadow stack token at the top of the stack. Please note that the x86 portions of this code are build tested only, I don't appear to have a system that can run CET avaible to me, I have done testing with an integration into my pending work for GCS. There is some possibility that the arm64 implementation may require the use of clone3() and explicit userspace allocation of shadow stacks, this is still under discussion. Please further note that the token consumption done by clone3() is not currently implemented in an atomic fashion, Rick indicated that he would look into fixing this if people are OK with the implementation. A new architecture feature Kconfig option for shadow stacks is added as here, this was suggested as part of the review comments for the arm64 GCS series and since we need to detect if shadow stacks are supported it seemed sensible to roll it in here. [1] https://lore.kernel.org/r/20231009-arm64-gcs-v6-0-78e55deaa4dd@kernel.org/ Signed-off-by: Mark Brown <broonie(a)kernel.org> --- Changes in v7: - Rebase onto v6.11-rc1. - Typo fixes. - Link to v6: https://lore.kernel.org/r/20240623-clone3-shadow-stack-v6-0-9ee7783b1fb9@ke… Changes in v6: - Rebase onto v6.10-rc3. - Ensure we don't try to free the parent shadow stack in error paths of x86 arch code. - Spelling fixes in userspace API document. - Additional cleanups and improvements to the clone3() tests to support the shadow stack tests. - Link to v5: https://lore.kernel.org/r/20240203-clone3-shadow-stack-v5-0-322c69598e4b@ke… Changes in v5: - Rebase onto v6.8-rc2. - Rework ABI to have the user allocate the shadow stack memory with map_shadow_stack() and a token. - Force inlining of the x86 shadow stack enablement. - Move shadow stack enablement out into a shared header for reuse by other tests. - Link to v4: https://lore.kernel.org/r/20231128-clone3-shadow-stack-v4-0-8b28ffe4f676@ke… Changes in v4: - Formatting changes. - Use a define for minimum shadow stack size and move some basic validation to fork.c. - Link to v3: https://lore.kernel.org/r/20231120-clone3-shadow-stack-v3-0-a7b8ed3e2acc@ke… Changes in v3: - Rebase onto v6.7-rc2. - Remove stale shadow_stack in internal kargs. - If a shadow stack is specified unconditionally use it regardless of CLONE_ parameters. - Force enable shadow stacks in the selftest. - Update changelogs for RISC-V feature rename. - Link to v2: https://lore.kernel.org/r/20231114-clone3-shadow-stack-v2-0-b613f8681155@ke… Changes in v2: - Rebase onto v6.7-rc1. - Remove ability to provide preallocated shadow stack, just specify the desired size. - Link to v1: https://lore.kernel.org/r/20231023-clone3-shadow-stack-v1-0-d867d0b5d4d0@ke… --- Mark Brown (9): Documentation: userspace-api: Add shadow stack API documentation selftests: Provide helper header for shadow stack testing mm: Introduce ARCH_HAS_USER_SHADOW_STACK fork: Add shadow stack support to clone3() selftests/clone3: Remove redundant flushes of output streams selftests/clone3: Factor more of main loop into test_clone3() selftests/clone3: Explicitly handle child exits due to signals selftests/clone3: Allow tests to flag if -E2BIG is a valid error code selftests/clone3: Test shadow stack support Documentation/userspace-api/index.rst | 1 + Documentation/userspace-api/shadow_stack.rst | 41 ++++ arch/x86/Kconfig | 1 + arch/x86/include/asm/shstk.h | 11 +- arch/x86/kernel/process.c | 2 +- arch/x86/kernel/shstk.c | 104 +++++++--- fs/proc/task_mmu.c | 2 +- include/linux/mm.h | 2 +- include/linux/sched/task.h | 13 ++ include/uapi/linux/sched.h | 13 +- kernel/fork.c | 76 ++++++-- mm/Kconfig | 6 + tools/testing/selftests/clone3/clone3.c | 224 ++++++++++++++++++---- tools/testing/selftests/clone3/clone3_selftests.h | 40 +++- tools/testing/selftests/ksft_shstk.h | 63 ++++++ 15 files changed, 511 insertions(+), 88 deletions(-) --- base-commit: 8400291e289ee6b2bf9779ff1c83a291501f017b change-id: 20231019-clone3-shadow-stack-15d40d2bf536 Best regards, -- Mark Brown <broonie(a)kernel.org>

10 months, 1 week

3
22
0 0

[PATCH] tracing/selftests: Run the ownership test twice

by Steven Rostedt

From: "Steven Rostedt (Google)" <rostedt(a)goodmis.org> A regression happened where running the ownership test passes on the first iteration but fails running it a second time. This was caught and fixed, but a later change brought it back. The regression was missed because the automated tests only run the tests once per boot. Change the ownership test to iterate through the tests twice, as this will catch the regression with a single run. Signed-off-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> --- .../ftrace/test.d/00basic/test_ownership.tc | 34 +++++++++++-------- 1 file changed, 20 insertions(+), 14 deletions(-) diff --git a/tools/testing/selftests/ftrace/test.d/00basic/test_ownership.tc b/tools/testing/selftests/ftrace/test.d/00basic/test_ownership.tc index c45094d1e1d2..71e43a92352a 100644 --- a/tools/testing/selftests/ftrace/test.d/00basic/test_ownership.tc +++ b/tools/testing/selftests/ftrace/test.d/00basic/test_ownership.tc @@ -83,32 +83,38 @@ run_tests() { done } -mount -o remount,"$new_options" . +# Run the tests twice as leftovers can cause issues +for loop in 1 2 ; do -run_tests + echo "Running iteration $loop" -mount -o remount,"$mount_options" . + mount -o remount,"$new_options" . -for d in "." "events" "events/sched" "events/sched/sched_switch" "events/sched/sched_switch/enable" $canary; do - test "$d" $original_group -done + run_tests + + mount -o remount,"$mount_options" . + + for d in "." "events" "events/sched" "events/sched/sched_switch" "events/sched/sched_switch/enable" $canary; do + test "$d" $original_group + done # check instances as well -chgrp $other_group instances + chgrp $other_group instances -instance="$(mktemp -u test-XXXXXX)" + instance="$(mktemp -u test-XXXXXX)" -mkdir instances/$instance + mkdir instances/$instance -cd instances/$instance + cd instances/$instance -run_tests + run_tests -cd ../.. + cd ../.. -rmdir instances/$instance + rmdir instances/$instance -chgrp $original_group instances + chgrp $original_group instances +done exit 0 -- 2.43.0

10 months, 1 week

3
4
0 0

[PATCH] selftests/uprobes: Add a basic uprobe testcase

by Masami Hiramatsu (Google)

From: Masami Hiramatsu (Google) <mhiramat(a)kernel.org> Add a basic uprobe testcase which checks whether add/remove/trace operations works on /bin/sh. Signed-off-by: Masami Hiramatsu (Google) <mhiramat(a)kernel.org> --- .../ftrace/test.d/dynevent/add_remove_uprobe.tc | 26 ++++++++++++++++++++ 1 file changed, 26 insertions(+) create mode 100644 tools/testing/selftests/ftrace/test.d/dynevent/add_remove_uprobe.tc diff --git a/tools/testing/selftests/ftrace/test.d/dynevent/add_remove_uprobe.tc b/tools/testing/selftests/ftrace/test.d/dynevent/add_remove_uprobe.tc new file mode 100644 index 000000000000..a275decdc880 --- /dev/null +++ b/tools/testing/selftests/ftrace/test.d/dynevent/add_remove_uprobe.tc @@ -0,0 +1,26 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0 +# description: Generic dynamic event - add/remove/test uprobe events +# requires: uprobe_events + +echo 0 > events/enable +echo > dynamic_events + +echo 'cat /proc/$$/maps' | /bin/sh | \ + grep "r-xp .*/bin/.*sh$" | \ + awk '{printf "p:myevent %s:0x%s\n", $6,$3 }' >> uprobe_events + +grep -q myevent uprobe_events +test -d events/uprobes/myevent + +echo 1 > events/uprobes/myevent/enable +echo 'ls' | /bin/sh > /dev/null +echo 0 > events/uprobes/myevent/enable +grep -q myevent trace + +echo "-:myevent" >> uprobe_events +! grep -q myevent uprobe_events + +echo > uprobe_events + +clear_trace

10 months, 1 week

3
2
0 0

[PATCH v7] mm, memcg: cg2 memory{.swap,}.peak write handlers

by David Finkel

This revision only updates the tests from the previous revision[1], and integrates an Acked-by[2] and a Reviewed-By[3] into the first commit message. Documentation/admin-guide/cgroup-v2.rst | 22 ++- include/linux/cgroup-defs.h | 5 + include/linux/cgroup.h | 3 + include/linux/memcontrol.h | 5 + include/linux/page_counter.h | 11 +- kernel/cgroup/cgroup-internal.h | 2 + kernel/cgroup/cgroup.c | 7 + mm/memcontrol.c | 116 +++++++++++++-- mm/page_counter.c | 30 +++- tools/testing/selftests/cgroup/cgroup_util.c | 22 +++ tools/testing/selftests/cgroup/cgroup_util.h | 2 + tools/testing/selftests/cgroup/test_memcontrol.c | 264 ++++++++++++++++++++++++++++++++- 12 files changed, 454 insertions(+), 35 deletions(-) [1]: https://lore.kernel.org/cgroups/20240729143743.34236-1-davidf@vimeo.com/T/ [2]: https://lore.kernel.org/cgroups/20240729143743.34236-1-davidf@vimeo.com/T/#… [3]: https://lore.kernel.org/cgroups/20240729143743.34236-1-davidf@vimeo.com/T/#… Thank you all for the support and reviews so far! David Finkel Senior Principal Software Engineer Vimeo Inc.

10 months, 1 week

4
9
0 0

[PATCH v2 0/2] selftests: harness: refactor __constructor_order

by Masahiro Yamada

This series refactors __constructor_order because __constructor_order_last() is unneeded. No code change since v1. I reworded "reverse-order" to "backward-order" in commit description. Masahiro Yamada (2): selftests: harness: remove unneeded __constructor_order_last() selftests: harness: rename __constructor_order for clarification .../drivers/s390x/uvdevice/test_uvdevice.c | 6 ------ tools/testing/selftests/hid/hid_bpf.c | 6 ------ tools/testing/selftests/kselftest_harness.h | 18 ++++-------------- tools/testing/selftests/rtc/rtctest.c | 7 ------- 4 files changed, 4 insertions(+), 33 deletions(-) -- 2.43.0

10 months, 1 week

3
4
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror July 2024