summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2024-10-30kasan: remove vmalloc_percpu testAndrey Konovalov
Commit 1a2473f0cbc0 ("kasan: improve vmalloc tests") added the vmalloc_percpu KASAN test with the assumption that __alloc_percpu always uses vmalloc internally, which is tagged by KASAN. However, __alloc_percpu might allocate memory from the first per-CPU chunk, which is not allocated via vmalloc(). As a result, the test might fail. Remove the test until proper KASAN annotation for the per-CPU allocated are added; tracked in https://bugzilla.kernel.org/show_bug.cgi?id=215019. Link: https://lkml.kernel.org/r/20241022160706.38943-1-andrey.konovalov@linux.dev Fixes: 1a2473f0cbc0 ("kasan: improve vmalloc tests") Signed-off-by: Andrey Konovalov <andreyknvl@gmail.com> Reported-by: Samuel Holland <samuel.holland@sifive.com> Link: https://lore.kernel.org/all/4a245fff-cc46-44d1-a5f9-fd2f1c3764ae@sifive.com/ Reported-by: Sabyrzhan Tasbolatov <snovitoll@gmail.com> Link: https://lore.kernel.org/all/CACzwLxiWzNqPBp4C1VkaXZ2wDwvY3yZeetCi1TLGFipKW77drA@mail.gmail.com/ Cc: Alexander Potapenko <glider@google.com> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com> Cc: Dmitry Vyukov <dvyukov@google.com> Cc: Marco Elver <elver@google.com> Cc: Sabyrzhan Tasbolatov <snovitoll@gmail.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-10-30tools/mm: -Werror fixes in page-types/slabinfoWladislav Wiebe
Commit e6d2c436ff693 ("tools/mm: allow users to provide additional cflags/ldflags") passes now CFLAGS to Makefile. With this, build systems with default -Werror enabled found: slabinfo.c:1300:25: error: ignoring return value of 'chdir' declared with attribute 'warn_unused_result' [-Werror=unused-result]                          chdir("..");                          ^~~~~~~~~~~ page-types.c:397:35: error: format '%lu' expects argument of type 'long unsigned int', but argument 2 has type 'uint64_t' {aka 'long long unsigned int'} [-Werror=format=]                          printf("%lu\t", mapcnt0);                                  ~~^     ~~~~~~~ .. Fix page-types by using PRIu64 for uint64_t prints and check in slabinfo for return code on chdir(".."). Link: https://lkml.kernel.org/r/c1ceb507-94bc-461c-934d-c19b77edd825@gmail.com Fixes: e6d2c436ff69 ("tools/mm: allow users to provide additional cflags/ldflags") Signed-off-by: Wladislav Wiebe <wladislav.kw@gmail.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Herton R. Krzesinski <herton@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-10-30mm, swap: avoid over reclaim of full clustersKairui Song
When running low on usable slots, cluster allocator will try to reclaim the full clusters aggressively to reclaim HAS_CACHE slots. This guarantees that as long as there are any usable slots, HAS_CACHE or not, the swap device will be usable and workload won't go OOM early. Before the cluster allocator, swap allocator fails easily if device is filled up with reclaimable HAS_CACHE slots. Which can be easily reproduced with following simple program: #include <stdio.h> #include <string.h> #include <linux/mman.h> #include <sys/mman.h> #define SIZE 8192UL * 1024UL * 1024UL int main(int argc, char **argv) { long tmp; char *p = mmap(NULL, SIZE, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); memset(p, 0, SIZE); madvise(p, SIZE, MADV_PAGEOUT); for (unsigned long i = 0; i < SIZE; ++i) tmp += p[i]; getchar(); /* Pause */ return 0; } Setup an 8G non ramdisk swap, the first run of the program will swapout 8G ram successfully. But run same program again after the first run paused, the second run can't swapout all 8G memory as now half of the swap device is pinned by HAS_CACHE. There was a random scan in the old allocator that may reclaim part of the HAS_CACHE by luck, but it's unreliable. The new allocator's added reclaim of full clusters when device is low on usable slots. But when multiple CPUs are seeing the device is low on usable slots at the same time, they ran into a thundering herd problem. This is an observable problem on large machine with mass parallel workload, as full cluster reclaim is slower on large swap device and higher number of CPUs will also make things worse. Testing using a 128G ZRAM on a 48c96t system. When the swap device is very close to full (eg. 124G / 128G), running build linux kernel with make -j96 in a 1G memory cgroup will hung (not a softlockup though) spinning in full cluster reclaim for about ~5min before go OOM. To solve this, split the full reclaim into two parts: - Instead of do a synchronous aggressively reclaim when device is low, do only one aggressively reclaim when device is strictly full with a kworker. This still ensures in worst case the device won't be unusable because of HAS_CACHE slots. - To avoid allocation (especially higher order) suffer from HAS_CACHE filling up clusters and kworker not responsive enough, do one synchronous scan every time the free list is drained, and only scan one cluster. This is kind of similar to the random reclaim before, keeps the full clusters rotated and has a minimal latency. This should provide a fair reclaim strategy suitable for most workloads. Link: https://lkml.kernel.org/r/20241022175512.10398-1-ryncsn@gmail.com Fixes: 2cacbdfdee65 ("mm: swap: add a adaptive full cluster cache reclaim") Signed-off-by: Kairui Song <kasong@tencent.com> Cc: Barry Song <v-songbaohua@oppo.com> Cc: Chris Li <chrisl@kernel.org> Cc: "Huang, Ying" <ying.huang@intel.com> Cc: Hugh Dickins <hughd@google.com> Cc: Kalesh Singh <kaleshsingh@google.com> Cc: Ryan Roberts <ryan.roberts@arm.com> Cc: Yosry Ahmed <yosryahmed@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-10-30mm: fix PSWPIN counter for large folios swap-inBarry Song
Similar to PSWPOUT, we should count the number of base pages instead of large folios. Link: https://lkml.kernel.org/r/20241023210201.2798-1-21cnbao@gmail.com Fixes: 242d12c98174 ("mm: support large folios swap-in for sync io devices") Signed-off-by: Barry Song <v-songbaohua@oppo.com> Acked-by: David Hildenbrand <david@redhat.com> Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com> Cc: Chris Li <chrisl@kernel.org> Cc: Yosry Ahmed <yosryahmed@google.com> Cc: "Huang, Ying" <ying.huang@intel.com> Cc: Kairui Song <kasong@tencent.com> Cc: Ryan Roberts <ryan.roberts@arm.com> Cc: Kanchana P Sridhar <kanchana.p.sridhar@intel.com> Cc: Usama Arif <usamaarif642@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-10-30mm: avoid VM_BUG_ON when try to map an anon large folio to zero page.Zi Yan
An anonymous large folio can be split into non order-0 folios, try_to_map_unused_to_zeropage() should not VM_BUG_ON compound pages but just return false. This fixes the crash when splitting anonymous large folios to non order-0 folios. Link: https://lkml.kernel.org/r/20241023171236.1122535-1-ziy@nvidia.com Fixes: b1f202060afe ("mm: remap unused subpages to shared zeropage when splitting isolated thp") Signed-off-by: Zi Yan <ziy@nvidia.com> Acked-by: David Hildenbrand <david@redhat.com> Acked-by: Usama Arif <usamaarif642@gmail.com> Cc: Barry Song <baohua@kernel.org> Cc: Domenico Cerasuolo <cerasuolodomenico@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Matthew Wilcox <willy@infradead.org> Cc: Mike Rapoport (Microsoft) <rppt@kernel.org> Cc: Nico Pache <npache@redhat.com> Cc: Rik van Riel <riel@surriel.com> Cc: Roman Gushchin <roman.gushchin@linux.dev> Cc: Ryan Roberts <ryan.roberts@arm.com> Cc: Shakeel Butt <shakeel.butt@linux.dev> Cc: Yu Zhao <yuzhao@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-10-30mm/codetag: fix null pointer check logic for ref and tagHao Ge
When we compile and load lib/slub_kunit.c,it will cause a panic. The root cause is that __kmalloc_cache_noprof was directly called instead of kmem_cache_alloc,which resulted in no alloc_tag being allocated.This caused current->alloc_tag to be null,leading to a null pointer dereference in alloc_tag_ref_set. Despite the fact that my colleague Pei Xiao will later fix the code in slub_kunit.c,we still need fix null pointer check logic for ref and tag to avoid panic caused by a null pointer dereference. Here is the log for the panic: [ 74.779373][ T2158] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 [ 74.780130][ T2158] Mem abort info: [ 74.780406][ T2158] ESR = 0x0000000096000004 [ 74.780756][ T2158] EC = 0x25: DABT (current EL), IL = 32 bits [ 74.781225][ T2158] SET = 0, FnV = 0 [ 74.781529][ T2158] EA = 0, S1PTW = 0 [ 74.781836][ T2158] FSC = 0x04: level 0 translation fault [ 74.782288][ T2158] Data abort info: [ 74.782577][ T2158] ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000 [ 74.783068][ T2158] CM = 0, WnR = 0, TnD = 0, TagAccess = 0 [ 74.783533][ T2158] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0 [ 74.784010][ T2158] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000105f34000 [ 74.784586][ T2158] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000 [ 74.785293][ T2158] Internal error: Oops: 0000000096000004 [#1] SMP [ 74.785805][ T2158] Modules linked in: slub_kunit kunit ip6t_rpfilter ip6t_REJECT nf_reject_ipv6 ipt_REJECT nf_reject_ipv4 xt_conntrack ebtable_nat ebtable_broute ip6table_nat ip6table_mangle 4 [ 74.790661][ T2158] CPU: 0 UID: 0 PID: 2158 Comm: kunit_try_catch Kdump: loaded Tainted: G W N 6.12.0-rc3+ #2 [ 74.791535][ T2158] Tainted: [W]=WARN, [N]=TEST [ 74.791889][ T2158] Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015 [ 74.792479][ T2158] pstate: 40400005 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 74.793101][ T2158] pc : alloc_tagging_slab_alloc_hook+0x120/0x270 [ 74.793607][ T2158] lr : alloc_tagging_slab_alloc_hook+0x120/0x270 [ 74.794095][ T2158] sp : ffff800084d33cd0 [ 74.794418][ T2158] x29: ffff800084d33cd0 x28: 0000000000000000 x27: 0000000000000000 [ 74.795095][ T2158] x26: 0000000000000000 x25: 0000000000000012 x24: ffff80007b30e314 [ 74.795822][ T2158] x23: ffff000390ff6f10 x22: 0000000000000000 x21: 0000000000000088 [ 74.796555][ T2158] x20: ffff000390285840 x19: fffffd7fc3ef7830 x18: ffffffffffffffff [ 74.797283][ T2158] x17: ffff8000800e63b4 x16: ffff80007b33afc4 x15: ffff800081654c00 [ 74.798011][ T2158] x14: 0000000000000000 x13: 205d383531325420 x12: 5b5d383734363537 [ 74.798744][ T2158] x11: ffff800084d337e0 x10: 000000000000005d x9 : 00000000ffffffd0 [ 74.799476][ T2158] x8 : 7f7f7f7f7f7f7f7f x7 : ffff80008219d188 x6 : c0000000ffff7fff [ 74.800206][ T2158] x5 : ffff0003fdbc9208 x4 : ffff800081edd188 x3 : 0000000000000001 [ 74.800932][ T2158] x2 : 0beaa6dee1ac5a00 x1 : 0beaa6dee1ac5a00 x0 : ffff80037c2cb000 [ 74.801656][ T2158] Call trace: [ 74.801954][ T2158] alloc_tagging_slab_alloc_hook+0x120/0x270 [ 74.802494][ T2158] __kmalloc_cache_noprof+0x148/0x33c [ 74.802976][ T2158] test_kmalloc_redzone_access+0x4c/0x104 [slub_kunit] [ 74.803607][ T2158] kunit_try_run_case+0x70/0x17c [kunit] [ 74.804124][ T2158] kunit_generic_run_threadfn_adapter+0x2c/0x4c [kunit] [ 74.804768][ T2158] kthread+0x10c/0x118 [ 74.805141][ T2158] ret_from_fork+0x10/0x20 [ 74.805540][ T2158] Code: b9400a80 11000400 b9000a80 97ffd858 (f94012d3) [ 74.806176][ T2158] SMP: stopping secondary CPUs [ 74.808130][ T2158] Starting crashdump kernel... Link: https://lkml.kernel.org/r/20241020070819.307944-1-hao.ge@linux.dev Fixes: e0a955bf7f61 ("mm/codetag: add pgalloc_tag_copy()") Signed-off-by: Hao Ge <gehao@kylinos.cn> Acked-by: Suren Baghdasaryan <surenb@google.com> Suggested-by: Suren Baghdasaryan <surenb@google.com> Acked-by: Yu Zhao <yuzhao@google.com> Cc: Kent Overstreet <kent.overstreet@linux.dev> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-10-30mm/gup: stop leaking pinned pages in low memory conditionsJohn Hubbard
If a driver tries to call any of the pin_user_pages*(FOLL_LONGTERM) family of functions, and requests "too many" pages, then the call will erroneously leave pages pinned. This is visible in user space as an actual memory leak. Repro is trivial: just make enough pin_user_pages(FOLL_LONGTERM) calls to exhaust memory. The root cause of the problem is this sequence, within __gup_longterm_locked(): __get_user_pages_locked() rc = check_and_migrate_movable_pages() ...which gets retried in a loop. The loop error handling is incomplete, clearly due to a somewhat unusual and complicated tri-state error API. But anyway, if -ENOMEM, or in fact, any unexpected error is returned from check_and_migrate_movable_pages(), then __gup_longterm_locked() happily returns the error, while leaving the pages pinned. In the failed case, which is an app that requests (via a device driver) 30720000000 bytes to be pinned, and then exits, I see this: $ grep foll /proc/vmstat nr_foll_pin_acquired 7502048 nr_foll_pin_released 2048 And after applying this patch, it returns to balanced pins: $ grep foll /proc/vmstat nr_foll_pin_acquired 7502048 nr_foll_pin_released 7502048 Note that the child routine, check_and_migrate_movable_folios(), avoids this problem, by unpinning any folios in the **folios argument, before returning an error. Fix this by making check_and_migrate_movable_pages() behave in exactly the same way as check_and_migrate_movable_folios(): unpin all pages in **pages, before returning an error. Also, documentation was an aggravating factor, so: 1) Consolidate the documentation for these two routines, now that they have identical external behavior. 2) Rewrite the consolidated documentation: a) Clearly list the three return code cases, and what happens in each case. b) Mention that one of the cases unpins the pages or folios, before returning an error code. Link: https://lkml.kernel.org/r/20241018223411.310331-1-jhubbard@nvidia.com Fixes: 24a95998e9ba ("mm/gup.c: simplify and fix check_and_migrate_movable_pages() return codes") Signed-off-by: John Hubbard <jhubbard@nvidia.com> Reviewed-by: Alistair Popple <apopple@nvidia.com> Suggested-by: David Hildenbrand <david@redhat.com> Cc: Shigeru Yoshida <syoshida@redhat.com> Cc: Jason Gunthorpe <jgg@nvidia.com> Cc: Minchan Kim <minchan@kernel.org> Cc: Pasha Tatashin <pasha.tatashin@soleen.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2024-10-30Merge branch 'add-ethernet-dts-schema-for-qcs615-qcs8300'Jakub Kicinski
Yijie Yang says: ==================== Add ethernet dts schema for qcs615/qcs8300 Document the ethernet and SerDes compatible for qcs8300. This platform shares the same EMAC and SerDes as sa8775p, so the compatible fallback to it. Document the ethernet compatible for qcs615. This platform shares the same EMAC as sm8150, so the compatible fallback to it. Document the compatible for revision 2 of the qcs8300-ride board. v2: https://lore.kernel.org/20241017-schema-v2-0-2320f68dc126@quicinc.com v1: https://lore.kernel.org/132a8e29-3be7-422a-bc83-d6be00fac3e8@kernel.org ==================== Link: https://patch.msgid.link/20241029-schema-v3-0-fbde519eaf00@quicinc.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30dt-bindings: net: qcom,ethqos: add description for qcs8300Yijie Yang
Add compatible for the MAC controller on qcs8300 platforms. Since qcs8300 shares the same EMAC as sa8775p, so it fallback to the compatible. Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by: Yijie Yang <quic_yijiyang@quicinc.com> Link: https://patch.msgid.link/20241029-schema-v3-2-fbde519eaf00@quicinc.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30dt-bindings: net: qcom,ethqos: add description for qcs615Yijie Yang
Add compatible for the MAC controller on qcs615 platform. Since qcs615 shares the same EMAC as sm8150, so it fallback to that compatible. Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by: Yijie Yang <quic_yijiyang@quicinc.com> Link: https://patch.msgid.link/20241029-schema-v3-1-fbde519eaf00@quicinc.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30Merge branch 'bna-remove-error-checking-for-debugfs-create-apis'Jakub Kicinski
Zhen Lei says: ==================== bna: Remove error checking for debugfs create APIs 1. Fix the incorrect return value check for debugfs_create_dir() and debugfs_create_file(), which returns ERR_PTR(-ERROR) instead of NULL when it fails. 2. Remove field bnad_dentry_files[] in struct bnad. When a directory is deleted, files in the directory are automatically deleted. Therefore, there is need to record these files. ==================== Link: https://patch.msgid.link/20241028020943.507-1-thunder.leizhen@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30bna: Remove field bnad_dentry_files[] in struct bnadZhen Lei
Function debugfs_remove() recursively removes a directory, include all files created by debugfs_create_file(). Therefore, there is no need to explicitly record each file with member ->bnad_dentry_files[] and explicitly delete them at the end. Remove field bnad_dentry_files[] and its related processing codes for simplification. Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20241028020943.507-3-thunder.leizhen@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30bna: Remove error checking for debugfs create APIsZhen Lei
Driver bna can work fine even if any previous call to debugfs create APIs failed. All return value checks of them should be dropped, as debugfs APIs say. Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20241028020943.507-2-thunder.leizhen@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30rtnetlink: Fix an error handling path in rtnl_newlink()Christophe JAILLET
When some code has been moved in the commit in Fixes, some "return err;" have correctly been changed in goto <some_where_in_the_error_handling_path> but this one was missed. Should "ops->maxtype > RTNL_MAX_TYPE" happen, then some resources would leak. Go through the error handling path to fix these leaks. Fixes: 0d3008d1a9ae ("rtnetlink: Move ops->validate to rtnl_newlink().") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://patch.msgid.link/eca90eeb4d9e9a0545772b68aeaab883d9fe2279.1729952228.git.christophe.jaillet@wanadoo.fr Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: ethernet: mtk_wed: fix path of MT7988 WO firmwareDaniel Golle
linux-firmware commit 808cba84 ("mtk_wed: add firmware for mt7988 Wireless Ethernet Dispatcher") added mt7988_wo_{0,1}.bin in the 'mediatek/mt7988' directory while driver current expects the files in the 'mediatek' directory. Change path in the driver header now that the firmware has been added. Fixes: e2f64db13aa1 ("net: ethernet: mtk_wed: introduce WED support for MT7988") Signed-off-by: Daniel Golle <daniel@makrotopia.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://patch.msgid.link/Zxz0GWTR5X5LdWPe@pidgin.makrotopia.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30Merge branch 'mlxsw-fixes'Jakub Kicinski
Petr Machata says: ==================== mlxsw: Fixes In this patchset: - Tx header should be pushed for each packet which is transmitted via Spectrum ASICs. Patch #1 adds a missing call to skb_cow_head() to make sure that there is both enough room to push the Tx header and that the SKB header is not cloned and can be modified. - Commit b5b60bb491b2 ("mlxsw: pci: Use page pool for Rx buffers allocation") converted mlxsw to use page pool for Rx buffers allocation. Sync for CPU and for device should be done for Rx pages. In patches #2 and #3, add the missing calls to sync pages for, respectively, CPU and the device. - Patch #4 then fixes a bug to IPv6 GRE forwarding offload. Patch #5 adds a generic forwarding test that fails with mlxsw ports prior to the fix. ==================== Link: https://patch.msgid.link/cover.1729866134.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30selftests: forwarding: Add IPv6 GRE remote change testsIdo Schimmel
Test that after changing the remote address of an ip6gre net device traffic is forwarded as expected. Test with both flat and hierarchical topologies and with and without an input / output keys. Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Petr Machata <petrm@nvidia.com> Signed-off-by: Petr Machata <petrm@nvidia.com> Link: https://patch.msgid.link/02b05246d2cdada0cf2fccffc0faa8a424d0f51b.1729866134.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30mlxsw: spectrum_ipip: Fix memory leak when changing remote IPv6 addressIdo Schimmel
The device stores IPv6 addresses that are used for encapsulation in linear memory that is managed by the driver. Changing the remote address of an ip6gre net device never worked properly, but since cited commit the following reproducer [1] would result in a warning [2] and a memory leak [3]. The problem is that the new remote address is never added by the driver to its hash table (and therefore the device) and the old address is never removed from it. Fix by programming the new address when the configuration of the ip6gre net device changes and removing the old one. If the address did not change, then the above would result in increasing the reference count of the address and then decreasing it. [1] # ip link add name bla up type ip6gre local 2001:db8:1::1 remote 2001:db8:2::1 tos inherit ttl inherit # ip link set dev bla type ip6gre remote 2001:db8:3::1 # ip link del dev bla # devlink dev reload pci/0000:01:00.0 [2] WARNING: CPU: 0 PID: 1682 at drivers/net/ethernet/mellanox/mlxsw/spectrum.c:3002 mlxsw_sp_ipv6_addr_put+0x140/0x1d0 Modules linked in: CPU: 0 UID: 0 PID: 1682 Comm: ip Not tainted 6.12.0-rc3-custom-g86b5b55bc835 #151 Hardware name: Nvidia SN5600/VMOD0013, BIOS 5.13 05/31/2023 RIP: 0010:mlxsw_sp_ipv6_addr_put+0x140/0x1d0 [...] Call Trace: <TASK> mlxsw_sp_router_netdevice_event+0x55f/0x1240 notifier_call_chain+0x5a/0xd0 call_netdevice_notifiers_info+0x39/0x90 unregister_netdevice_many_notify+0x63e/0x9d0 rtnl_dellink+0x16b/0x3a0 rtnetlink_rcv_msg+0x142/0x3f0 netlink_rcv_skb+0x50/0x100 netlink_unicast+0x242/0x390 netlink_sendmsg+0x1de/0x420 ____sys_sendmsg+0x2bd/0x320 ___sys_sendmsg+0x9a/0xe0 __sys_sendmsg+0x7a/0xd0 do_syscall_64+0x9e/0x1a0 entry_SYSCALL_64_after_hwframe+0x77/0x7f [3] unreferenced object 0xffff898081f597a0 (size 32): comm "ip", pid 1626, jiffies 4294719324 hex dump (first 32 bytes): 20 01 0d b8 00 02 00 00 00 00 00 00 00 00 00 01 ............... 21 49 61 83 80 89 ff ff 00 00 00 00 01 00 00 00 !Ia............. backtrace (crc fd9be911): [<00000000df89c55d>] __kmalloc_cache_noprof+0x1da/0x260 [<00000000ff2a1ddb>] mlxsw_sp_ipv6_addr_kvdl_index_get+0x281/0x340 [<000000009ddd445d>] mlxsw_sp_router_netdevice_event+0x47b/0x1240 [<00000000743e7757>] notifier_call_chain+0x5a/0xd0 [<000000007c7b9e13>] call_netdevice_notifiers_info+0x39/0x90 [<000000002509645d>] register_netdevice+0x5f7/0x7a0 [<00000000c2e7d2a9>] ip6gre_newlink_common.isra.0+0x65/0x130 [<0000000087cd6d8d>] ip6gre_newlink+0x72/0x120 [<000000004df7c7cc>] rtnl_newlink+0x471/0xa20 [<0000000057ed632a>] rtnetlink_rcv_msg+0x142/0x3f0 [<0000000032e0d5b5>] netlink_rcv_skb+0x50/0x100 [<00000000908bca63>] netlink_unicast+0x242/0x390 [<00000000cdbe1c87>] netlink_sendmsg+0x1de/0x420 [<0000000011db153e>] ____sys_sendmsg+0x2bd/0x320 [<000000003b6d53eb>] ___sys_sendmsg+0x9a/0xe0 [<00000000cae27c62>] __sys_sendmsg+0x7a/0xd0 Fixes: cf42911523e0 ("mlxsw: spectrum_ipip: Use common hash table for IPv6 address mapping") Reported-by: Maksym Yaremchuk <maksymy@nvidia.com> Signed-off-by: Ido Schimmel <idosch@nvidia.com> Reviewed-by: Petr Machata <petrm@nvidia.com> Signed-off-by: Petr Machata <petrm@nvidia.com> Link: https://patch.msgid.link/e91012edc5a6cb9df37b78fd377f669381facfcb.1729866134.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30mlxsw: pci: Sync Rx buffers for deviceAmit Cohen
Non-coherent architectures, like ARM, may require invalidating caches before the device can use the DMA mapped memory, which means that before posting pages to device, drivers should sync the memory for device. Sync for device can be configured as page pool responsibility. Set the relevant flag and define max_len for sync. Cc: Jiri Pirko <jiri@resnulli.us> Fixes: b5b60bb491b2 ("mlxsw: pci: Use page pool for Rx buffers allocation") Signed-off-by: Amit Cohen <amcohen@nvidia.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Signed-off-by: Petr Machata <petrm@nvidia.com> Link: https://patch.msgid.link/92e01f05c4f506a4f0a9b39c10175dcc01994910.1729866134.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30mlxsw: pci: Sync Rx buffers for CPUAmit Cohen
When Rx packet is received, drivers should sync the pages for CPU, to ensure the CPU reads the data written by the device and not stale data from its cache. Add the missing sync call in Rx path, sync the actual length of data for each fragment. Cc: Jiri Pirko <jiri@resnulli.us> Fixes: b5b60bb491b2 ("mlxsw: pci: Use page pool for Rx buffers allocation") Signed-off-by: Amit Cohen <amcohen@nvidia.com> Reviewed-by: Ido Schimmel <idosch@nvidia.com> Signed-off-by: Petr Machata <petrm@nvidia.com> Link: https://patch.msgid.link/461486fac91755ca4e04c2068c102250026dcd0b.1729866134.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30mlxsw: spectrum_ptp: Add missing verification before pushing Tx headerAmit Cohen
Tx header should be pushed for each packet which is transmitted via Spectrum ASICs. The cited commit moved the call to skb_cow_head() from mlxsw_sp_port_xmit() to functions which handle Tx header. In case that mlxsw_sp->ptp_ops->txhdr_construct() is used to handle Tx header, and txhdr_construct() is mlxsw_sp_ptp_txhdr_construct(), there is no call for skb_cow_head() before pushing Tx header size to SKB. This flow is relevant for Spectrum-1 and Spectrum-4, for PTP packets. Add the missing call to skb_cow_head() to make sure that there is both enough room to push the Tx header and that the SKB header is not cloned and can be modified. An additional set will be sent to net-next to centralize the handling of the Tx header by pushing it to every packet just before transmission. Cc: Richard Cochran <richardcochran@gmail.com> Fixes: 24157bc69f45 ("mlxsw: Send PTP packets as data packets to overcome a limitation") Signed-off-by: Amit Cohen <amcohen@nvidia.com> Signed-off-by: Petr Machata <petrm@nvidia.com> Link: https://patch.msgid.link/5145780b07ebbb5d3b3570f311254a3a2d554a44.1729866134.git.petrm@nvidia.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30dt-bindings: net: renesas,ether: Add iommus propertyGeert Uytterhoeven
make dtbs_check: arch/arm64/boot/dts/renesas/r8a77980-condor.dtb: ethernet@e7400000: 'iommus' does not match any of the regexes: '@[0-9a-f]$', 'pinctrl-[0-9]+' from schema $id: http://devicetree.org/schemas/net/renesas,ether.yaml# Ethernet Controllers on R-Car Gen2/Gen3 SoCs can make use of the IOMMU, so add the missing iommus property. Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be> Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Reviewed-by: Sergei Shtylyov <sergei.shtylyov@gmail.com> Acked-by: Rob Herring (Arm) <robh@kernel.org> Link: https://patch.msgid.link/2ca890323477a21c22e13f6a1328288f4ee816f9.1729868894.git.geert+renesas@glider.be Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: skip offload for NETIF_F_IPV6_CSUM if ipv6 header contains extensionBenoît Monin
As documented in skbuff.h, devices with NETIF_F_IPV6_CSUM capability can only checksum TCP and UDP over IPv6 if the IP header does not contains extension. This is enforced for UDP packets emitted from user-space to an IPv6 address as they go through ip6_make_skb(), which calls __ip6_append_data() where a check is done on the header size before setting CHECKSUM_PARTIAL. But the introduction of UDP encapsulation with fou6 added a code-path where it is possible to get an skb with a partial UDP checksum and an IPv6 header with extension: * fou6 adds a UDP header with a partial checksum if the inner packet does not contains a valid checksum. * ip6_tunnel adds an IPv6 header with a destination option extension header if encap_limit is non-zero (the default value is 4). The thread linked below describes in more details how to reproduce the problem with GRE-in-UDP tunnel. Add a check on the network header size in skb_csum_hwoffload_help() to make sure no IPv6 packet with extension header is handed to a network device with NETIF_F_IPV6_CSUM capability. Link: https://lore.kernel.org/netdev/26548921.1r3eYUQgxm@benoit.monin/T/#u Fixes: aa3463d65e7b ("fou: Add encap ops for IPv6 tunnels") Signed-off-by: Benoît Monin <benoit.monin@gmx.fr> Reviewed-by: Willem de Bruijn <willemb@google.com> Link: https://patch.msgid.link/5fbeecfc311ea182aa1d1c771725ab8b4cac515e.1729778144.git.benoit.monin@gmx.fr Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30Merge branch 'net-sparx5-add-support-for-lan969x-switch-device'Jakub Kicinski
Daniel Machon says: ==================== net: sparx5: add support for lan969x switch device == Description: This series is the second of a multi-part series, that prepares and adds support for the new lan969x switch driver. The upstreaming efforts is split into multiple series (might change a bit as we go along): 1) Prepare the Sparx5 driver for lan969x (merged) --> 2) add support lan969x (same basic features as Sparx5 provides excl. FDMA and VCAP). 3) Add support for lan969x VCAP, FDMA and RGMII == Lan969x in short: The lan969x Ethernet switch family [1] provides a rich set of switching features and port configurations (up to 30 ports) from 10Mbps to 10Gbps, with support for RGMII, SGMII, QSGMII, USGMII, and USXGMII, ideal for industrial & process automation infrastructure applications, transport, grid automation, power substation automation, and ring & intra-ring topologies. The LAN969x family is hardware and software compatible and scalable supporting 46Gbps to 102Gbps switch bandwidths. == Preparing Sparx5 for lan969x: The main preparation work for lan969x has already been merged [1]. After this series is applied, lan969x will have the same functionality as Sparx5, except for VCAP and FDMA support. QoS features that requires the VCAP (e.g. PSFP, port mirroring) will obviously not work until VCAP support is added later. == Patch breakdown: Patch #1-#4 do some preparation work for lan969x Patch #5 adds new registers required by lan969x Patch #6 adds initial match data for all lan969x targets Patch #7 defines the lan969x register differences Patch #8 adds lan969x constants to match data Patch #9 adds some lan969x ops in bulk Patch #10 adds PTP function to ops Patch #11 adds lan969x_calendar.c for calculating the calendar Patch #12 makes additional use of the is_sparx5() macro to branch out in certain places. Patch #13 documents lan969x in the dt-bindings Patch #14 adds lan969x compatible string to sparx5 driver Patch #15 introduces new concept of per-target features [1] https://lore.kernel.org/netdev/20241004-b4-sparx5-lan969x-switch-driver-v2-0-d3290f581663@microchip.com/ v1: https://lore.kernel.org/20241021-sparx5-lan969x-switch-driver-2-v1-0-c8c49ef21e0f@microchip.com ==================== Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-0-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: sparx5: add feature supportDaniel Machon
Lan969x supports a number of different features, depending on the target. Add new field sparx5->features and initialize the features based on the target. Also add the function sparx5_has_feature() and use it throughout. For now, we only need to handle features: PSFP and PTP - more will come in the future. [1] https://www.microchip.com/en-us/product/lan9698 Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-15-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: sparx5: add compatible string for lan969xDaniel Machon
Add lan9691-switch compatible string to mchp_sparx5_match. Guard it with IS_ENABLED(CONFIG_LAN969X_SWITCH) to make sure Sparx5 can be compiled on its own. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-14-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30dt-bindings: net: add compatible strings for lan969x targetsDaniel Machon
Add compatible strings for the twelve different lan969x targets that we support. Either a sparx5-switch or lan9691-switch compatible string provided on their own, or any lan969x-switch compatible string with a fallback to lan9691-switch. Also, add myself as a maintainer. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-13-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: sparx5: use is_sparx5() macro throughoutDaniel Machon
Use the is_sparx5() macro (introduced in earlier series [1]), in places where we need to handle things a bit differently on lan969x. These places are: - in sparx5_dsm_calendar_update() we need to switch the calendar from a to b on lan969x. - in sparx5_start() we need to make sure the HSCH_SYS_CLK_PER register is only touched on Sparx5. - in sparx5_start() we need to disable VCAP and FDMA for lan969x (will come in later series). - in sparx5_mirror_port_get() we must make sure the ANA_AC_PROBE_PORT_CFG1 register is only read on Sparx5. - sparx5_netdev.c and sparx5_packet.c we need to use different IFH (Internal Frame Header) offsets for lan969x. - in sparx5_port_fifo_sz() we must bail out on lan969x. - in sparx5_port_config_low_set() we must configure the phase detection registers. - in sparx5_port_config() and sparx5_port_init() we must do some additional configuration of the port devices. - in sparx5_dwrr_conf_set() we must derive the scheduling layer [1] https://lore.kernel.org/netdev/20241004-b4-sparx5-lan969x-switch-driver-v2-8-d3290f581663@microchip.com/ Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-12-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: lan969x: add function for calculating the DSM calendarDaniel Machon
Lan969x has support for RedBox / HSR / PRP (not implemented yet). In order to accommodate for this in the future, we need to give lan969x it's own function for calculating the DSM calendar. The function calculates the calendar for each taxi bus. The calendar is used for bandwidth allocation towards the ports attached to the taxi bus. A calendar configuration consists of up-to 64 slots, which may be allocated to ports or left unused. Each slot accounts for 1 clock cycle. Also expose sparx5_cal_speed_to_value(), sparx5_get_port_cal_speed, sparx5_cal_bw and SPX5_DSM_CAL_EMPTY for use by lan969x. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-11-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: lan969x: add PTP handler functionDaniel Machon
Add PTP IRQ handler for lan969x. This is required, as the PTP registers are placed in two different targets on Sparx5 and lan969x. The implementation is otherwise the same as on Sparx5. Also, expose sparx5_get_hwtimestamp() for use by lan969x. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-10-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: lan969x: add lan969x ops to match dataDaniel Machon
Add a bunch of small lan969x ops in bulk. These ops are explained in detail in a previous series [1]. [1] https://lore.kernel.org/netdev/20241004-b4-sparx5-lan969x-switch-driver-v2-8-d3290f581663@microchip.com/ Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-9-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: lan969x: add constants to match dataDaniel Machon
Add the lan969x constants to match data. These are already used throughout the Sparx5 code (introduced in earlier series [1]), so no need to update any code use. [1] https://lore.kernel.org/netdev/20241004-b4-sparx5-lan969x-switch-driver-v2-0-d3290f581663@microchip.com/ Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-8-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: lan969x: add register diffs to match dataDaniel Machon
Add new file lan969x_regs.c that defines all the register differences for lan969x, and add it to the lan969x match data. GW_DEV2G5_PHASE_DETECTOR_CTRL, FP_DEV2G5_PHAD_CTRL_PHAD_ENA and FP_DEV2G5_PHAD_CTRL_PHAD_FAILED are required by the new register macros which was introduced earlier. Add these for Sparx5 also. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-7-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: lan969x: add match data for lan969xDaniel Machon
Add match data for lan969x, with initial fields for iomap, iomap_size and ioranges. Add new Kconfig symbol CONFIG_LAN969X_CONFIG for compiling the lan969x driver. It has been decided to give lan969x its own Kconfig symbol, as a considerable amount of code is needed, beside the Sparx5 code, to add full chip support (and more will be added in future series). Also this makes it possible to compile Sparx5 without lan969x. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-6-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: sparx5: add registers required by lan969xDaniel Machon
Lan969x will require a few additional registers for certain operations. Some are shared, some are not. Add these. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-5-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: sparx5: add sparx5 context pointer to a few functionsDaniel Machon
In preparation for lan969x, add the sparx5 context pointer to certain IFH (Internal Frame Header) functions. This is required, as the is_sparx5() function will be used here in a subsequent patch. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-4-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: sparx5: change frequency calculation for SDLB'sDaniel Machon
In preparation for lan969x, rework the function that calculates the SDLB (Service Dual Leacky Bucket) clock. This is required, as the HSCH_SYS_CLK_PER register is Sparx5-exclusive. Instead derive the clock from the core clock, using the sparx5_clk_period() function. The clock stays the same before and after this patch, only now, sparx5_sdlb_clk_hz_get() can be used for lan969x too. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-3-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: sparx5: change spx5_wr to spx5_rmw in cal update()Daniel Machon
In preparation for lan969x, use spx5_rmw() for enabling the update of the calendar. This is required to not overwrite the DSM_TAXI_CAL_CFG register, as an additional write will be added before this one, in a subsequent patch. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-2-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: sparx5: add support for lan969x targets and core clockDaniel Machon
In preparation for lan969x, add lan969x targets to sparx5_target_chiptype and set the core clock frequency for these throughout. Lan969x only supports a core clock frequency of 328MHz. Also, set the policer update internal (pol_upd_int) matching the 328 MHz frequency of the lan969x targets. Reviewed-by: Steen Hegelund <Steen.Hegelund@microchip.com> Signed-off-by: Daniel Machon <daniel.machon@microchip.com> Link: https://patch.msgid.link/20241024-sparx5-lan969x-switch-driver-2-v2-1-a0b5fae88a0f@microchip.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30gve: change to use page_pool_put_full_page when recycling pagesHarshitha Ramamurthy
The driver currently uses page_pool_put_page() to recycle page pool pages. Since gve uses split pages, if the fragment being recycled is not the last fragment in the page, there is no dma sync operation. When the last fragment is recycled, dma sync is performed by page pool infra according to the value passed as dma_sync_size which right now is set to the size of fragment. But the correct thing to do is to dma sync the entire page when the last fragment is recycled. Hence change to using page_pool_put_full_page(). Link: https://lore.kernel.org/netdev/89d7ce83-cc1d-4791-87b5-6f7af29a031d@huawei.com/ Suggested-by: Yunsheng Lin <linyunsheng@huawei.com> Reviewed-by: Praveen Kaligineedi <pkaligineedi@google.com> Reviewed-by: Willem de Bruijn <willemb@google.com> Signed-off-by: Harshitha Ramamurthy <hramamurthy@google.com> Reviewed-by: Yunsheng Lin <linyunsheng@huawei.com> Fixes: ebdfae0d377b ("gve: adopt page pool for DQ RDA mode") Link: https://patch.msgid.link/20241023221141.3008011-1-pkaligineedi@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30Merge branch 'refactoring-rvu-nic-driver'Jakub Kicinski
Geetha sowjanya says: ==================== Refactoring RVU NIC driver This is a preparation pathset for follow-up "Introducing RVU representors driver" patches. The RVU representor driver creates representor netdev of each rvu device when switch dev mode is enabled. RVU representor and NIC have a similar set of HW resources(NIX_LF,RQ/SQ/CQ) and implements a subset of NIC functionality. This patch set groups hw resources and queue configuration code into single API and export the existing functions so, that code can be shared between NIC and representor drivers. These patches are part of "Introduce RVU representors" patchset. https://lore.kernel.org/all/ZsdJ-w00yCI4NQ8T@nanopsycho.orion/T/ As suggested by "Jiri Pirko", submitting as separate patchset. ==================== Link: https://patch.msgid.link/20241023161843.15543-1-gakula@marvell.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30octeontx2-pf: Move shared APIs to header fileGeetha sowjanya
Move mbox, hw resources and interrupt configuration functions to common header file. So, that they can be used later by the RVU representor driver. Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20241023161843.15543-5-gakula@marvell.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30octeontx2-pf: Reuse PF max mtu valueGeetha sowjanya
Reuse the maximum support HW MTU value that is fetch during probe. Instead of fetching through mbox each time mtu is changed as the value is fixed for interface. Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20241023161843.15543-4-gakula@marvell.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30octeontx2-pf: Add new APIs for queue memory alloc/free.Geetha sowjanya
Group the queue(RX/TX/CQ) memory allocation and free code to single APIs. Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20241023161843.15543-3-gakula@marvell.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30octeontx2-pf: Define common API for HW resources configurationGeetha sowjanya
Define new API "otx2_init_rsrc" and move the HW blocks NIX/NPA resources configuration code under this API. So, that it can be used by the RVU representor driver that has similar resources of RVU NIC. Signed-off-by: Geetha sowjanya <gakula@marvell.com> Reviewed-by: Jiri Pirko <jiri@resnulli.us> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://patch.msgid.link/20241023161843.15543-2-gakula@marvell.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30Merge branch 'mirroring-to-dsa-cpu-port'Jakub Kicinski
Vladimir Oltean says: ==================== Mirroring to DSA CPU port Users of the NXP LS1028A SoC (drivers/net/dsa/ocelot L2 switch inside) have requested to mirror packets from the ingress of a switch port to software. Both port-based and flow-based mirroring is required. The simplest way I could come up with was to set up tc mirred actions towards a dummy net_device, and make the offloading of that be accepted by the driver. Currently, the pattern in drivers is to reject mirred towards ports they don't know about, but I'm now permitting that, precisely by mirroring "to the CPU". For testers, this series depends on commit 34d35b4edbbe ("net/sched: act_api: deny mismatched skip_sw/skip_hw flags for actions created by classifiers") from net/main, which is absent from net-next as of the day of posting (Oct 23). Without the bug fix it is possible to create invalid configurations which are not rejected by the kernel. v2: https://lore.kernel.org/20241017165215.3709000-1-vladimir.oltean@nxp.com RFC: https://lore.kernel.org/20240913152915.2981126-1-vladimir.oltean@nxp.com For historical purposes, link to a much older (and much different) attempt: https://lore.kernel.org/20191002233750.13566-1-olteanv@gmail.com ==================== Link: https://patch.msgid.link/20241023135251.1752488-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: mscc: ocelot: allow tc-flower mirred action towards foreign interfacesVladimir Oltean
Debugging certain flows in the offloaded switch data path can be done by installing two tc-mirred filters for mirroring: one in the hardware data path, which copies the frames to the CPU, and one which takes the frame from there and mirrors it to a virtual interface like a dummy device, where it can be seen with tcpdump. The effect of having 2 filters run on the same packet can be obtained by default using tc, by not specifying either the 'skip_sw' or 'skip_hw' keywords. Instead of refusing to offload mirroring/redirecting packets towards interfaces that aren't switch ports, just treat every other destination for what it is: something that is handled in software, behind the CPU port. Usage: $ ip link add dummy0 type dummy; ip link set dummy0 up $ tc qdisc add dev swp0 clsact $ tc filter add dev swp0 ingress protocol ip flower action mirred ingress mirror dev dummy0 Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20241023135251.1752488-7-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: dsa: allow matchall mirroring rules towards the CPUVladimir Oltean
If the CPU bandwidth capacity permits, it may be useful to mirror the entire ingress of a user port to software. This is in fact possible to express even if there is no net_device representation for the CPU port. In fact, that approach was already exhausted and that representation wouldn't have even helped [1]. The idea behind implementing this is that currently, we refuse to offload any mirroring towards a non-DSA target net_device. But if we acknowledge the fact that to reach any foreign net_device, the switch must send the packet to the CPU anyway, then we can simply offload just that part, and let the software do the rest. There is only one condition we need to uphold: the filter needs to be present in the software data path as well (no skip_sw). There are 2 actions to consider: FLOW_ACTION_MIRRED (redirect to egress of target interface) and FLOW_ACTION_MIRRED_INGRESS (redirect to ingress of target interface). We don't have the ability/API to offload FLOW_ACTION_MIRRED_INGRESS when the target port is also a DSA user port, but we could also permit that through mirred to the CPU + software. Example: $ ip link add dummy0 type dummy; ip link set dummy0 up $ tc qdisc add dev swp0 clsact $ tc filter add dev swp0 ingress matchall action mirred ingress mirror dev dummy0 Any DSA driver with a ds->ops->port_mirror_add() implementation can now make use of this with no additional change. [1] https://lore.kernel.org/netdev/20191002233750.13566-1-olteanv@gmail.com/ Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20241023135251.1752488-6-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: dsa: add more extack messages in dsa_user_add_cls_matchall_mirred()Vladimir Oltean
Do not leave -EOPNOTSUPP errors without an explanation. It is confusing for the user to figure out what is wrong otherwise. Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20241023135251.1752488-5-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-10-30net: dsa: use "extack" as argument to flow_action_basic_hw_stats_check()Vladimir Oltean
We already have an "extack" stack variable in dsa_user_add_cls_matchall_police() and dsa_user_add_cls_matchall_mirred(), there is no need to retrieve it again from cls->common.extack. Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Link: https://patch.msgid.link/20241023135251.1752488-4-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
generated by cgit (git 2.34.1) at 2025-09-16 01:37:21 +0000