Backport: TCP Devmem #1
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
page_pool_alloc_netmem (without an s) was the mirror of page_pool_alloc_pages (with an s), which was confusing. Rename to page_pool_alloc_netmems so it's the mirror of page_pool_alloc_pages. Signed-off-by: Mina Almasry <[email protected]> Acked-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Create page_pool_alloc_netmem to be the mirror of page_pool_alloc. This enables drivers that want currently use page_pool_alloc to transition to netmem by converting the call sites to page_pool_alloc_netmem. Signed-off-by: Mina Almasry <[email protected]> Acked-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Move the `dma_map` and `dma_sync` checks to `page_pool_init` to make them generic. Set dma_sync to false for devmem memory provider because the dma_sync APIs should not be used for dma_buf backed devmem memory provider. Cc: Jason Gunthorpe <[email protected]> Signed-off-by: Samiullah Khawaja <[email protected]> Signed-off-by: Mina Almasry <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
dmabuf dma-addresses should not be dma_sync'd for CPU/device. Typically its the driver responsibility to dma_sync for CPU, but the driver should not dma_sync for CPU if the netmem is actually coming from a dmabuf memory provider. The page_pool already exposes a helper for dma_sync_for_cpu: page_pool_dma_sync_for_cpu. Upgrade this existing helper to handle netmem, and have it skip dma_sync if the memory is from a dmabuf memory provider. Drivers should migrate to using this helper when adding support for netmem. Also minimize the impact on the dma syncing performance for pages. Special case the dma-sync path for pages to not go through the overhead checks for dma-syncing and conversion to netmem. Cc: Alexander Lobakin <[email protected]> Cc: Jason Gunthorpe <[email protected]> Signed-off-by: Mina Almasry <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
…t/tnguy/net-queue Tony Nguyen says: ==================== ice: fix Rx data path for heavy 9k MTU traffic Maciej Fijalkowski says: This patchset fixes a pretty nasty issue that was reported by RedHat folks which occurred after ~30 minutes (this value varied, just trying here to state that it was not observed immediately but rather after a considerable longer amount of time) when ice driver was tortured with jumbo frames via mix of iperf traffic executed simultaneously with wrk/nginx on client/server sides (HTTP and TCP workloads basically). The reported splats were spanning across all the bad things that can happen to the state of page - refcount underflow, use-after-free, etc. One of these looked as follows: [ 2084.019891] BUG: Bad page state in process swapper/34 pfn:97fcd0 [ 2084.025990] page:00000000a60ee772 refcount:-1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x97fcd0 [ 2084.035462] flags: 0x17ffffc0000000(node=0|zone=2|lastcpupid=0x1fffff) [ 2084.041990] raw: 0017ffffc0000000 dead000000000100 dead000000000122 0000000000000000 [ 2084.049730] raw: 0000000000000000 0000000000000000 ffffffffffffffff 0000000000000000 [ 2084.057468] page dumped because: nonzero _refcount [ 2084.062260] Modules linked in: bonding tls sunrpc intel_rapl_msr intel_rapl_common intel_uncore_frequency intel_uncore_frequency_common i10nm_edac nfit libnvdimm x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm mgag200 irqd [ 2084.137829] CPU: 34 PID: 0 Comm: swapper/34 Kdump: loaded Not tainted 5.14.0-427.37.1.el9_4.x86_64 #1 [ 2084.147039] Hardware name: Dell Inc. PowerEdge R750/0216NK, BIOS 1.13.2 12/19/2023 [ 2084.154604] Call Trace: [ 2084.157058] <IRQ> [ 2084.159080] dump_stack_lvl+0x34/0x48 [ 2084.162752] bad_page.cold+0x63/0x94 [ 2084.166333] check_new_pages+0xb3/0xe0 [ 2084.170083] rmqueue_bulk+0x2d2/0x9e0 [ 2084.173749] ? ktime_get+0x35/0xa0 [ 2084.177159] rmqueue_pcplist+0x13b/0x210 [ 2084.181081] rmqueue+0x7d3/0xd40 [ 2084.184316] ? xas_load+0x9/0xa0 [ 2084.187547] ? xas_find+0x183/0x1d0 [ 2084.191041] ? xa_find_after+0xd0/0x130 [ 2084.194879] ? intel_iommu_iotlb_sync_map+0x89/0xe0 [ 2084.199759] get_page_from_freelist+0x11f/0x530 [ 2084.204291] __alloc_pages+0xf2/0x250 [ 2084.207958] ice_alloc_rx_bufs+0xcc/0x1c0 [ice] [ 2084.212543] ice_clean_rx_irq+0x631/0xa20 [ice] [ 2084.217111] ice_napi_poll+0xdf/0x2a0 [ice] [ 2084.221330] __napi_poll+0x27/0x170 [ 2084.224824] net_rx_action+0x233/0x2f0 [ 2084.228575] __do_softirq+0xc7/0x2ac [ 2084.232155] __irq_exit_rcu+0xa1/0xc0 [ 2084.235821] common_interrupt+0x80/0xa0 [ 2084.239662] </IRQ> [ 2084.241768] <TASK> The fix is mostly about reverting what was done in commit 1dc1a7e ("ice: Centrallize Rx buffer recycling") followed by proper timing on page_count() storage and then removing the ice_rx_buf::act related logic (which was mostly introduced for purposes from cited commit). Special thanks to Xu Du for providing reproducer and Jacob Keller for initial extensive analysis. * '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue: ice: stop storing XDP verdict within ice_rx_buf ice: gather page_count()'s of each frag right before XDP prog call ice: put Rx buffers after being done with current frame ==================== Link: https://patch.msgid.link/[email protected] Signed-off-by: Jakub Kicinski <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
Filesystems can write to disk from page reclaim with __GFP_FS set. Marc found a case where scsi_realloc_sdev_budget_map() ends up in page reclaim with GFP_KERNEL, where it could try to take filesystem locks again, leading to a deadlock. WARNING: possible circular locking dependency detected 6.13.0 #1 Not tainted ------------------------------------------------------ kswapd0/70 is trying to acquire lock: ffff8881025d5d78 (&q->q_usage_counter(io)){++++}-{0:0}, at: blk_mq_submit_bio+0x461/0x6e0 but task is already holding lock: ffffffff81ef5f40 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x9f/0x760 The full lockdep splat can be found in Marc's report: https://lkml.org/lkml/2025/1/24/1101 Avoid the potential deadlock by doing the allocation with GFP_NOIO, which prevents both filesystem and block layer recursion. Reported-by: Marc Aurèle La France <[email protected]> Signed-off-by: Rik van Riel <[email protected]> Link: https://lore.kernel.org/r/20250129104525.0ae8421e@fangorn Reviewed-by: Christoph Hellwig <[email protected]> Signed-off-by: Martin K. Petersen <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
Protected mode assumes that at minimum vgic-v3 is present, however KVM fails to actually enforce this at the time of initialization. As such, when running protected mode in a half-baked state on GICv2 hardware we see the hyp go belly up at vcpu_load() when it tries to restore the vgic-v3 cpuif: $ ./arch_timer_edge_cases [ 130.599140] kvm [4518]: nVHE hyp panic at: [<ffff800081102b58>] __kvm_nvhe___vgic_v3_restore_vmcr_aprs+0x8/0x84! [ 130.603685] kvm [4518]: Cannot dump pKVM nVHE stacktrace: !CONFIG_PROTECTED_NVHE_STACKTRACE [ 130.611962] kvm [4518]: Hyp Offset: 0xfffeca95ed000000 [ 130.617053] Kernel panic - not syncing: HYP panic: [ 130.617053] PS:800003c9 PC:0000b56a94102b58 ESR:0000000002000000 [ 130.617053] FAR:ffff00007b98d4d0 HPFAR:00000000007b98d0 PAR:0000000000000000 [ 130.617053] VCPU:0000000000000000 [ 130.638013] CPU: 0 UID: 0 PID: 4518 Comm: arch_timer_edge Tainted: G C 6.13.0-rc3-00009-gf7d03fcbf1f4 #1 [ 130.648790] Tainted: [C]=CRAP [ 130.651721] Hardware name: Libre Computer AML-S905X-CC (DT) [ 130.657242] Call trace: [ 130.659656] show_stack+0x18/0x24 (C) [ 130.663279] dump_stack_lvl+0x38/0x90 [ 130.666900] dump_stack+0x18/0x24 [ 130.670178] panic+0x388/0x3e8 [ 130.673196] nvhe_hyp_panic_handler+0x104/0x208 [ 130.677681] kvm_arch_vcpu_load+0x290/0x548 [ 130.681821] vcpu_load+0x50/0x80 [ 130.685013] kvm_arch_vcpu_ioctl_run+0x30/0x868 [ 130.689498] kvm_vcpu_ioctl+0x2e0/0x974 [ 130.693293] __arm64_sys_ioctl+0xb4/0xec [ 130.697174] invoke_syscall+0x48/0x110 [ 130.700883] el0_svc_common.constprop.0+0x40/0xe0 [ 130.705540] do_el0_svc+0x1c/0x28 [ 130.708818] el0_svc+0x30/0xd0 [ 130.711837] el0t_64_sync_handler+0x10c/0x138 [ 130.716149] el0t_64_sync+0x198/0x19c [ 130.719774] SMP: stopping secondary CPUs [ 130.723660] Kernel Offset: disabled [ 130.727103] CPU features: 0x000,00000800,02800000,0200421b [ 130.732537] Memory Limit: none [ 130.735561] ---[ end Kernel panic - not syncing: HYP panic: [ 130.735561] PS:800003c9 PC:0000b56a94102b58 ESR:0000000002000000 [ 130.735561] FAR:ffff00007b98d4d0 HPFAR:00000000007b98d0 PAR:0000000000000000 [ 130.735561] VCPU:0000000000000000 ]--- Fix it by failing KVM initialization if the system doesn't implement vgic-v3, as protected mode will never do anything useful on such hardware. Reported-by: Mark Brown <[email protected]> Closes: https://lore.kernel.org/kvmarm/[email protected]/ Signed-off-by: Oliver Upton <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Marc Zyngier <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
…/kernel/git/kvmarm/kvmarm into HEAD KVM/arm64 fixes for 6.14, take #1 - Correctly clean the BSS to the PoC before allowing EL2 to access it on nVHE/hVHE/protected configurations - Propagate ownership of debug registers in protected mode after the rework that landed in 6.14-rc1 - Stop pretending that we can run the protected mode without a GICv3 being present on the host - Fix a use-after-free situation that can occur if a vcpu fails to initialise the NV shadow S2 MMU contexts - Always evaluate the need to arm a background timer for fully emulated guest timers - Fix the emulation of EL1 timers in the absence of FEAT_ECV - Correctly handle the EL2 virtual timer, specially when HCR_EL2.E2H==0
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
commit c8347f9 ("gpu: host1x: Fix boot regression for Tegra") caused a use of uninitialized mutex leading to below warning when CONFIG_DEBUG_MUTEXES and CONFIG_DEBUG_LOCK_ALLOC are enabled. [ 41.662843] ------------[ cut here ]------------ [ 41.663012] DEBUG_LOCKS_WARN_ON(lock->magic != lock) [ 41.663035] WARNING: CPU: 4 PID: 794 at kernel/locking/mutex.c:587 __mutex_lock+0x670/0x878 [ 41.663458] Modules linked in: rtw88_8822c(+) bluetooth(+) rtw88_pci rtw88_core mac80211 aquantia libarc4 crc_itu_t cfg80211 tegra194_cpufreq dwmac_tegra(+) arm_dsu_pmu stmmac_platform stmmac pcs_xpcs rfkill at24 host1x(+) tegra_bpmp_thermal ramoops reed_solomon fuse loop nfnetlink xfs mmc_block rpmb_core ucsi_ccg ina3221 crct10dif_ce xhci_tegra ghash_ce lm90 sha2_ce sha256_arm64 sha1_ce sdhci_tegra pwm_fan sdhci_pltfm sdhci gpio_keys rtc_tegra cqhci mmc_core phy_tegra_xusb i2c_tegra tegra186_gpc_dma i2c_tegra_bpmp spi_tegra114 dm_mirror dm_region_hash dm_log dm_mod [ 41.665078] CPU: 4 UID: 0 PID: 794 Comm: (udev-worker) Not tainted 6.11.0-29.31_1538613708.el10.aarch64+debug #1 [ 41.665838] Hardware name: NVIDIA NVIDIA Jetson AGX Orin Developer Kit/Jetson, BIOS 36.3.0-gcid-35594366 02/26/2024 [ 41.672555] pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 41.679636] pc : __mutex_lock+0x670/0x878 [ 41.683834] lr : __mutex_lock+0x670/0x878 [ 41.688035] sp : ffff800084b77090 [ 41.691446] x29: ffff800084b77160 x28: ffffdd4bebf7b000 x27: ffffdd4be96b1000 [ 41.698799] x26: 1fffe0002308361c x25: 1ffff0001096ee18 x24: 0000000000000000 [ 41.706149] x23: 0000000000000000 x22: 0000000000000002 x21: ffffdd4be6e3c7a0 [ 41.713500] x20: ffff800084b770f0 x19: ffff00011841b1e8 x18: 0000000000000000 [ 41.720675] x17: 0000000000000000 x16: 0000000000000000 x15: 0720072007200720 [ 41.728023] x14: 0000000000000000 x13: 0000000000000001 x12: ffff6001a96eaab3 [ 41.735375] x11: 1fffe001a96eaab2 x10: ffff6001a96eaab2 x9 : ffffdd4be4838bbc [ 41.742723] x8 : 00009ffe5691554e x7 : ffff000d4b755593 x6 : 0000000000000001 [ 41.749985] x5 : ffff000d4b755590 x4 : 1fffe0001d88f001 x3 : dfff800000000000 [ 41.756988] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff0000ec478000 [ 41.764251] Call trace: [ 41.766695] __mutex_lock+0x670/0x878 [ 41.770373] mutex_lock_nested+0x2c/0x40 [ 41.774134] host1x_intr_start+0x54/0xf8 [host1x] [ 41.778863] host1x_runtime_resume+0x150/0x228 [host1x] [ 41.783935] pm_generic_runtime_resume+0x84/0xc8 [ 41.788485] __rpm_callback+0xa0/0x478 [ 41.792422] rpm_callback+0x15c/0x1a8 [ 41.795922] rpm_resume+0x698/0xc08 [ 41.799597] __pm_runtime_resume+0xa8/0x140 [ 41.803621] host1x_probe+0x810/0xbc0 [host1x] [ 41.807909] platform_probe+0xcc/0x1a8 [ 41.811845] really_probe+0x188/0x800 [ 41.815347] __driver_probe_device+0x164/0x360 [ 41.819810] driver_probe_device+0x64/0x1a8 [ 41.823834] __driver_attach+0x180/0x490 [ 41.827773] bus_for_each_dev+0x104/0x1a0 [ 41.831797] driver_attach+0x44/0x68 [ 41.835296] bus_add_driver+0x23c/0x4e8 [ 41.839235] driver_register+0x15c/0x3a8 [ 41.843170] __platform_register_drivers+0xa4/0x208 [ 41.848159] tegra_host1x_init+0x4c/0xff8 [host1x] [ 41.853147] do_one_initcall+0xd4/0x380 [ 41.856997] do_init_module+0x1dc/0x698 [ 41.860758] load_module+0xc70/0x1300 [ 41.864435] __do_sys_init_module+0x1a8/0x1d0 [ 41.868721] __arm64_sys_init_module+0x74/0xb0 [ 41.873183] invoke_syscall.constprop.0+0xdc/0x1e8 [ 41.877997] do_el0_svc+0x154/0x1d0 [ 41.881671] el0_svc+0x54/0x140 [ 41.884820] el0t_64_sync_handler+0x120/0x130 [ 41.889285] el0t_64_sync+0x1a4/0x1a8 [ 41.892960] irq event stamp: 69737 [ 41.896370] hardirqs last enabled at (69737): [<ffffdd4be6d7768c>] _raw_spin_unlock_irqrestore+0x44/0xe8 [ 41.905739] hardirqs last disabled at (69736): [<ffffdd4be59dcd40>] clk_enable_lock+0x98/0x198 [ 41.914314] softirqs last enabled at (68082): [<ffffdd4be466b1d0>] handle_softirqs+0x4c8/0x890 [ 41.922977] softirqs last disabled at (67945): [<ffffdd4be44f02a4>] __do_softirq+0x1c/0x28 [ 41.931289] ---[ end trace 0000000000000000 ]--- Inside the probe function when pm_runtime_enable() is called, the PM core invokes a resume callback if the device Host1x is in a suspended state. As it can be seen in the logs above, this leads to host1x_intr_start() function call which is trying to acquire a mutex lock. But, the function host_intr_init() only gets called after the pm_runtime_enable() where mutex is initialised leading to the use of mutex prior to its initialisation. Fix this by moving the mutex initialisation prior to the runtime PM enablement function pm_runtime_enable() in probe. Fixes: c8347f9 ("gpu: host1x: Fix boot regression for Tegra") Signed-off-by: Rupinderjit Singh <[email protected]> Reviewed-by: Jon Hunter <[email protected]> Tested-by: Jon Hunter <[email protected]> Signed-off-by: Thierry Reding <[email protected]> Link: https://patchwork.ozlabs.org/project/linux-tegra/patch/[email protected]/
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
vxlan_init() must check vxlan_vnigroup_init() success otherwise a crash happens later, spotted by syzbot. Oops: general protection fault, probably for non-canonical address 0xdffffc000000002c: 0000 [#1] PREEMPT SMP KASAN NOPTI KASAN: null-ptr-deref in range [0x0000000000000160-0x0000000000000167] CPU: 0 UID: 0 PID: 7313 Comm: syz-executor147 Not tainted 6.14.0-rc1-syzkaller-00276-g69b54314c975 #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 RIP: 0010:vxlan_vnigroup_uninit+0x89/0x500 drivers/net/vxlan/vxlan_vnifilter.c:912 Code: 00 48 8b 44 24 08 4c 8b b0 98 41 00 00 49 8d 86 60 01 00 00 48 89 c2 48 89 44 24 10 48 b8 00 00 00 00 00 fc ff df 48 c1 ea 03 <80> 3c 02 00 0f 85 4d 04 00 00 49 8b 86 60 01 00 00 48 ba 00 00 00 RSP: 0018:ffffc9000cc1eea8 EFLAGS: 00010202 RAX: dffffc0000000000 RBX: 0000000000000001 RCX: ffffffff8672effb RDX: 000000000000002c RSI: ffffffff8672ecb9 RDI: ffff8880461b4f18 RBP: ffff8880461b4ef4 R08: 0000000000000001 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000020000 R13: ffff8880461b0d80 R14: 0000000000000000 R15: dffffc0000000000 FS: 00007fecfa95d6c0(0000) GS:ffff88806a600000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fecfa95cfb8 CR3: 000000004472c000 CR4: 0000000000352ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> vxlan_uninit+0x1ab/0x200 drivers/net/vxlan/vxlan_core.c:2942 unregister_netdevice_many_notify+0x12d6/0x1f30 net/core/dev.c:11824 unregister_netdevice_many net/core/dev.c:11866 [inline] unregister_netdevice_queue+0x307/0x3f0 net/core/dev.c:11736 register_netdevice+0x1829/0x1eb0 net/core/dev.c:10901 __vxlan_dev_create+0x7c6/0xa30 drivers/net/vxlan/vxlan_core.c:3981 vxlan_newlink+0xd1/0x130 drivers/net/vxlan/vxlan_core.c:4407 rtnl_newlink_create net/core/rtnetlink.c:3795 [inline] __rtnl_newlink net/core/rtnetlink.c:3906 [inline] Fixes: f9c4bb0 ("vxlan: vni filtering support on collect metadata device") Reported-by: [email protected] Closes: https://lore.kernel.org/netdev/[email protected]/T/#u Signed-off-by: Eric Dumazet <[email protected]> Cc: Roopa Prabhu <[email protected]> Reviewed-by: Ido Schimmel <[email protected]> Link: https://patch.msgid.link/[email protected] Signed-off-by: Jakub Kicinski <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
bch2_nocow_write_convert_unwritten is already in transaction context: 00191 ========= TEST generic/648 00242 kernel BUG at fs/bcachefs/btree_iter.c:3332! 00242 Internal error: Oops - BUG: 00000000f2000800 [#1] PREEMPT SMP 00242 Modules linked in: 00242 CPU: 4 UID: 0 PID: 2593 Comm: fsstress Not tainted 6.13.0-rc3-ktest-g345af8f855b7 #14403 00242 Hardware name: linux,dummy-virt (DT) 00242 pstate: 60001005 (nZCv daif -PAN -UAO -TCO -DIT +SSBS BTYPE=--) 00242 pc : __bch2_trans_get+0x120/0x410 00242 lr : __bch2_trans_get+0xcc/0x410 00242 sp : ffffff80d89af600 00242 x29: ffffff80d89af600 x28: ffffff80ddb23000 x27: 00000000fffff705 00242 x26: ffffff80ddb23028 x25: ffffff80d8903fe0 x24: ffffff80ebb30168 00242 x23: ffffff80c8aeb500 x22: 000000000000005d x21: ffffff80d8904078 00242 x20: ffffff80d8900000 x19: ffffff80da9e8000 x18: 0000000000000000 00242 x17: 64747568735f6c61 x16: 6e72756f6a20726f x15: 0000000000000028 00242 x14: 0000000000000004 x13: 000000000000f787 x12: ffffffc081bbcdc8 00242 x11: 0000000000000000 x10: 0000000000000003 x9 : ffffffc08094efbc 00242 x8 : 000000001092c111 x7 : 000000000000000c x6 : ffffffc083c31fc4 00242 x5 : ffffffc083c31f28 x4 : ffffff80c8aeb500 x3 : ffffff80ebb30000 00242 x2 : 0000000000000001 x1 : 0000000000000a21 x0 : 000000000000028e 00242 Call trace: 00242 __bch2_trans_get+0x120/0x410 (P) 00242 bch2_inum_offset_err_msg+0x48/0xb0 00242 bch2_nocow_write_convert_unwritten+0x3d0/0x530 00242 bch2_nocow_write+0xeb0/0x1000 00242 __bch2_write+0x330/0x4e8 00242 bch2_write+0x1f0/0x530 00242 bch2_direct_write+0x530/0xc00 00242 bch2_write_iter+0x160/0xbe0 00242 vfs_write+0x1cc/0x360 00242 ksys_write+0x5c/0xf0 00242 __arm64_sys_write+0x20/0x30 00242 invoke_syscall.constprop.0+0x54/0xe8 00242 do_el0_svc+0x44/0xc0 00242 el0_svc+0x34/0xa0 00242 el0t_64_sync_handler+0x104/0x130 00242 el0t_64_sync+0x154/0x158 00242 Code: 6b01001f 54ffff01 79408460 3617fec0 (d4210000) 00242 ---[ end trace 0000000000000000 ]--- 00242 Kernel panic - not syncing: Oops - BUG: Fatal exception Signed-off-by: Alan Huang <[email protected]> Signed-off-by: Kent Overstreet <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
We have several places across the kernel where we want to access another
task's syscall arguments, such as ptrace(2), seccomp(2), etc., by making
a call to syscall_get_arguments().
This works for register arguments right away by accessing the task's
`regs' member of `struct pt_regs', however for stack arguments seen with
32-bit/o32 kernels things are more complicated. Technically they ought
to be obtained from the user stack with calls to an access_remote_vm(),
but we have an easier way available already.
So as to be able to access syscall stack arguments as regular function
arguments following the MIPS calling convention we copy them over from
the user stack to the kernel stack in arch/mips/kernel/scall32-o32.S, in
handle_sys(), to the current stack frame's outgoing argument space at
the top of the stack, which is where the handler called expects to see
its incoming arguments. This area is also pointed at by the `pt_regs'
pointer obtained by task_pt_regs().
Make the o32 stack argument space a proper member of `struct pt_regs'
then, by renaming the existing member from `pad0' to `args' and using
generated offsets to access the space. No functional change though.
With the change in place the o32 kernel stack frame layout at the entry
to a syscall handler invoked by handle_sys() is therefore as follows:
$sp + 68 -> | ... | <- pt_regs.regs[9]
+---------------------+
$sp + 64 -> | $t0 | <- pt_regs.regs[8]
+---------------------+
$sp + 60 -> | $a3/argument #4 | <- pt_regs.regs[7]
+---------------------+
$sp + 56 -> | $a2/argument #3 | <- pt_regs.regs[6]
+---------------------+
$sp + 52 -> | $a1/argument #2 | <- pt_regs.regs[5]
+---------------------+
$sp + 48 -> | $a0/argument #1 | <- pt_regs.regs[4]
+---------------------+
$sp + 44 -> | $v1 | <- pt_regs.regs[3]
+---------------------+
$sp + 40 -> | $v0 | <- pt_regs.regs[2]
+---------------------+
$sp + 36 -> | $at | <- pt_regs.regs[1]
+---------------------+
$sp + 32 -> | $zero | <- pt_regs.regs[0]
+---------------------+
$sp + 28 -> | stack argument torvalds#8 | <- pt_regs.args[7]
+---------------------+
$sp + 24 -> | stack argument torvalds#7 | <- pt_regs.args[6]
+---------------------+
$sp + 20 -> | stack argument torvalds#6 | <- pt_regs.args[5]
+---------------------+
$sp + 16 -> | stack argument #5 | <- pt_regs.args[4]
+---------------------+
$sp + 12 -> | psABI space for $a3 | <- pt_regs.args[3]
+---------------------+
$sp + 8 -> | psABI space for $a2 | <- pt_regs.args[2]
+---------------------+
$sp + 4 -> | psABI space for $a1 | <- pt_regs.args[1]
+---------------------+
$sp + 0 -> | psABI space for $a0 | <- pt_regs.args[0]
+---------------------+
holding user data received and with the first 4 frame slots reserved by
the psABI for the compiler to spill the incoming arguments from $a0-$a3
registers (which it sometimes does according to its needs) and the next
4 frame slots designated by the psABI for any stack function arguments
that follow. This data is also available for other tasks to peek/poke
at as reqired and where permitted.
Signed-off-by: Maciej W. Rozycki <[email protected]>
Signed-off-by: Thomas Bogendoerfer <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
This makes ptrace/get_syscall_info selftest pass on mips o32 and mips64 o32 by fixing the following two test assertions: 1. get_syscall_info test assertion on mips o32: # get_syscall_info.c:218:get_syscall_info:Expected exp_args[5] (3134521044) == info.entry.args[4] (4911432) # get_syscall_info.c:219:get_syscall_info:wait #1: entry stop mismatch 2. get_syscall_info test assertion on mips64 o32: # get_syscall_info.c:209:get_syscall_info:Expected exp_args[2] (3134324433) == info.entry.args[1] (18446744072548908753) # get_syscall_info.c:210:get_syscall_info:wait #1: entry stop mismatch The first assertion happens due to mips_get_syscall_arg() trying to access another task's context but failing to do it properly because get_user() it calls just peeks at the current task's context. It usually does not crash because the default user stack always gets assigned the same VMA, but it is pure luck which mips_get_syscall_arg() wouldn't have if e.g. the stack was switched (via setcontext(3) or however) or a non-default process's thread peeked at, and in any case irrelevant data is obtained just as observed with the test case. mips_get_syscall_arg() ought to be using access_remote_vm() instead to retrieve the other task's stack contents, but given that the data has been already obtained and saved in `struct pt_regs' it would be an overkill. The first assertion is fixed for mips o32 by using struct pt_regs.args instead of get_user() to obtain syscall arguments. This approach works due to this piece in arch/mips/kernel/scall32-o32.S: /* * Ok, copy the args from the luser stack to the kernel stack. */ .set push .set noreorder .set nomacro load_a4: user_lw(t5, 16(t0)) # argument #5 from usp load_a5: user_lw(t6, 20(t0)) # argument torvalds#6 from usp load_a6: user_lw(t7, 24(t0)) # argument torvalds#7 from usp load_a7: user_lw(t8, 28(t0)) # argument torvalds#8 from usp loads_done: sw t5, PT_ARG4(sp) # argument #5 to ksp sw t6, PT_ARG5(sp) # argument torvalds#6 to ksp sw t7, PT_ARG6(sp) # argument torvalds#7 to ksp sw t8, PT_ARG7(sp) # argument torvalds#8 to ksp .set pop .section __ex_table,"a" PTR_WD load_a4, bad_stack_a4 PTR_WD load_a5, bad_stack_a5 PTR_WD load_a6, bad_stack_a6 PTR_WD load_a7, bad_stack_a7 .previous arch/mips/kernel/scall64-o32.S has analogous code for mips64 o32 that allows fixing the issue by obtaining syscall arguments from struct pt_regs.regs[4..11] instead of the erroneous use of get_user(). The second assertion is fixed by truncating 64-bit values to 32-bit syscall arguments. Fixes: c0ff3c5 ("MIPS: Enable HAVE_ARCH_TRACEHOOK.") Signed-off-by: Dmitry V. Levin <[email protected]> Signed-off-by: Thomas Bogendoerfer <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
This fixes the following trace by reworking the locking of l2cap_conn so instead of only locking when changing the chan_l list this promotes chan_lock to a general lock of l2cap_conn so whenever it is being held it would prevents the likes of l2cap_conn_del to run: list_del corruption, ffff888021297e00->prev is LIST_POISON2 (dead000000000122) ------------[ cut here ]------------ kernel BUG at lib/list_debug.c:61! Oops: invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI CPU: 1 UID: 0 PID: 5896 Comm: syz-executor213 Not tainted 6.14.0-rc1-next-20250204-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 12/27/2024 RIP: 0010:__list_del_entry_valid_or_report+0x12c/0x190 lib/list_debug.c:59 Code: 8c 4c 89 fe 48 89 da e8 32 8c 37 fc 90 0f 0b 48 89 df e8 27 9f 14 fd 48 c7 c7 a0 c0 60 8c 4c 89 fe 48 89 da e8 15 8c 37 fc 90 <0f> 0b 4c 89 e7 e8 0a 9f 14 fd 42 80 3c 2b 00 74 08 4c 89 e7 e8 cb RSP: 0018:ffffc90003f6f998 EFLAGS: 00010246 RAX: 000000000000004e RBX: dead000000000122 RCX: 01454d423f7fbf00 RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000 RBP: dffffc0000000000 R08: ffffffff819f077c R09: 1ffff920007eded0 R10: dffffc0000000000 R11: fffff520007eded1 R12: dead000000000122 R13: dffffc0000000000 R14: ffff8880352248d8 R15: ffff888021297e00 FS: 00007f7ace6686c0(0000) GS:ffff8880b8700000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f7aceeeb1d0 CR3: 000000003527c000 CR4: 00000000003526f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> __list_del_entry_valid include/linux/list.h:124 [inline] __list_del_entry include/linux/list.h:215 [inline] list_del_rcu include/linux/rculist.h:168 [inline] hci_chan_del+0x70/0x1b0 net/bluetooth/hci_conn.c:2858 l2cap_conn_free net/bluetooth/l2cap_core.c:1816 [inline] kref_put include/linux/kref.h:65 [inline] l2cap_conn_put+0x70/0xe0 net/bluetooth/l2cap_core.c:1830 l2cap_sock_shutdown+0xa8a/0x1020 net/bluetooth/l2cap_sock.c:1377 l2cap_sock_release+0x79/0x1d0 net/bluetooth/l2cap_sock.c:1416 __sock_release net/socket.c:642 [inline] sock_close+0xbc/0x240 net/socket.c:1393 __fput+0x3e9/0x9f0 fs/file_table.c:448 task_work_run+0x24f/0x310 kernel/task_work.c:227 ptrace_notify+0x2d2/0x380 kernel/signal.c:2522 ptrace_report_syscall include/linux/ptrace.h:415 [inline] ptrace_report_syscall_exit include/linux/ptrace.h:477 [inline] syscall_exit_work+0xc7/0x1d0 kernel/entry/common.c:173 syscall_exit_to_user_mode_prepare kernel/entry/common.c:200 [inline] __syscall_exit_to_user_mode_work kernel/entry/common.c:205 [inline] syscall_exit_to_user_mode+0x24a/0x340 kernel/entry/common.c:218 do_syscall_64+0x100/0x230 arch/x86/entry/common.c:89 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f7aceeaf449 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 41 19 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b0 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f7ace668218 EFLAGS: 00000246 ORIG_RAX: 000000000000002a RAX: fffffffffffffffc RBX: 00007f7acef39328 RCX: 00007f7aceeaf449 RDX: 000000000000000e RSI: 0000000020000100 RDI: 0000000000000004 RBP: 00007f7acef39320 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000003 R13: 0000000000000004 R14: 00007f7ace668670 R15: 000000000000000b </TASK> Modules linked in: ---[ end trace 0000000000000000 ]--- RIP: 0010:__list_del_entry_valid_or_report+0x12c/0x190 lib/list_debug.c:59 Code: 8c 4c 89 fe 48 89 da e8 32 8c 37 fc 90 0f 0b 48 89 df e8 27 9f 14 fd 48 c7 c7 a0 c0 60 8c 4c 89 fe 48 89 da e8 15 8c 37 fc 90 <0f> 0b 4c 89 e7 e8 0a 9f 14 fd 42 80 3c 2b 00 74 08 4c 89 e7 e8 cb RSP: 0018:ffffc90003f6f998 EFLAGS: 00010246 RAX: 000000000000004e RBX: dead000000000122 RCX: 01454d423f7fbf00 RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000 RBP: dffffc0000000000 R08: ffffffff819f077c R09: 1ffff920007eded0 R10: dffffc0000000000 R11: fffff520007eded1 R12: dead000000000122 R13: dffffc0000000000 R14: ffff8880352248d8 R15: ffff888021297e00 FS: 00007f7ace6686c0(0000) GS:ffff8880b8600000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f7acef05b08 CR3: 000000003527c000 CR4: 00000000003526f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Reported-by: [email protected] Tested-by: [email protected] Fixes: b4f82f9 ("Bluetooth: L2CAP: Fix slab-use-after-free Read in l2cap_send_cmd") Signed-off-by: Luiz Augusto von Dentz <[email protected]> Signed-off-by: Dan Carpenter <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
The behaviour of kthread_create_worker() was recently changed to align with the one of kthread_create(). The kthread worker is created but not awaken by default. This is to allow the use of kthread_affine_preferred() and kthread_bind[_mask]() with kthread workers. In order to keep the old behaviour and wake the kthread up, kthread_run_worker() must be used. All the pre-existing users have been converted, except for UVC that was introduced in the same merge window as the API change. This results in hangs: INFO: task UVCG:82 blocked for more than 491 seconds. Tainted: G T 6.13.0-rc2-00014-gb04e317b5226 #1 task:UVCG state:D stack:0 pid:82 Call Trace: __schedule schedule schedule_preempt_disabled kthread ? kthread_flush_work ret_from_fork ret_from_fork_asm entry_INT80_32 Fix this with converting UVCG kworker to the new API. Reported-by: kernel test robot <[email protected]> Closes: https://lore.kernel.org/oe-lkp/[email protected] Fixes: f0bbfbd ("usb: gadget: uvc: rework to enqueue in pump worker from encoded queue") Cc: stable <[email protected]> Cc: Michael Grzeschik <[email protected]> Signed-off-by: Frederic Weisbecker <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Greg Kroah-Hartman <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
tcf_exts_miss_cookie_base_alloc() calls xa_alloc_cyclic() which can return 1 if the allocation succeeded after wrapping. This was treated as an error, with value 1 returned to caller tcf_exts_init_ex() which sets exts->actions to NULL and returns 1 to caller fl_change(). fl_change() treats err == 1 as success, calling tcf_exts_validate_ex() which calls tcf_action_init() with exts->actions as argument, where it is dereferenced. Example trace: BUG: kernel NULL pointer dereference, address: 0000000000000000 CPU: 114 PID: 16151 Comm: handler114 Kdump: loaded Not tainted 5.14.0-503.16.1.el9_5.x86_64 #1 RIP: 0010:tcf_action_init+0x1f8/0x2c0 Call Trace: tcf_action_init+0x1f8/0x2c0 tcf_exts_validate_ex+0x175/0x190 fl_change+0x537/0x1120 [cls_flower] Fixes: 80cd22c ("net/sched: cls_api: Support hardware miss to tc action") Signed-off-by: Pierre Riteau <[email protected]> Reviewed-by: Michal Swiatkowski <[email protected]> Link: https://patch.msgid.link/[email protected] Signed-off-by: Jakub Kicinski <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
When trying to mmap a trace instance buffer that is attached to reserve_mem, it would crash: BUG: unable to handle page fault for address: ffffe97bd00025c8 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 2862f3067 P4D 2862f3067 PUD 0 Oops: Oops: 0000 [#1] PREEMPT_RT SMP PTI CPU: 4 UID: 0 PID: 981 Comm: mmap-rb Not tainted 6.14.0-rc2-test-00003-g7f1a5e3fbf9e-dirty torvalds#233 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 RIP: 0010:validate_page_before_insert+0x5/0xb0 Code: e2 01 89 d0 c3 cc cc cc cc 66 66 2e 0f 1f 84 00 00 00 00 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 44 00 00 <48> 8b 46 08 a8 01 75 67 66 90 48 89 f0 8b 50 34 85 d2 74 76 48 89 RSP: 0018:ffffb148c2f3f968 EFLAGS: 00010246 RAX: ffff9fa5d3322000 RBX: ffff9fa5ccff9c08 RCX: 00000000b879ed29 RDX: ffffe97bd00025c0 RSI: ffffe97bd00025c0 RDI: ffff9fa5ccff9c08 RBP: ffffb148c2f3f9f0 R08: 0000000000000004 R09: 0000000000000004 R10: 0000000000000000 R11: 0000000000000200 R12: 0000000000000000 R13: 00007f16a18d5000 R14: ffff9fa5c48db6a8 R15: 0000000000000000 FS: 00007f16a1b54740(0000) GS:ffff9fa73df00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffe97bd00025c8 CR3: 00000001048c6006 CR4: 0000000000172ef0 Call Trace: <TASK> ? __die_body.cold+0x19/0x1f ? __die+0x2e/0x40 ? page_fault_oops+0x157/0x2b0 ? search_module_extables+0x53/0x80 ? validate_page_before_insert+0x5/0xb0 ? kernelmode_fixup_or_oops.isra.0+0x5f/0x70 ? __bad_area_nosemaphore+0x16e/0x1b0 ? bad_area_nosemaphore+0x16/0x20 ? do_kern_addr_fault+0x77/0x90 ? exc_page_fault+0x22b/0x230 ? asm_exc_page_fault+0x2b/0x30 ? validate_page_before_insert+0x5/0xb0 ? vm_insert_pages+0x151/0x400 __rb_map_vma+0x21f/0x3f0 ring_buffer_map+0x21b/0x2f0 tracing_buffers_mmap+0x70/0xd0 __mmap_region+0x6f0/0xbd0 mmap_region+0x7f/0x130 do_mmap+0x475/0x610 vm_mmap_pgoff+0xf2/0x1d0 ksys_mmap_pgoff+0x166/0x200 __x64_sys_mmap+0x37/0x50 x64_sys_call+0x1670/0x1d70 do_syscall_64+0xbb/0x1d0 entry_SYSCALL_64_after_hwframe+0x77/0x7f The reason was that the code that maps the ring buffer pages to user space has: page = virt_to_page((void *)cpu_buffer->subbuf_ids[s]); And uses that in: vm_insert_pages(vma, vma->vm_start, pages, &nr_pages); But virt_to_page() does not work with vmap()'d memory which is what the persistent ring buffer has. It is rather trivial to allow this, but for now just disable mmap() of instances that have their ring buffer from the reserve_mem option. If an mmap() is performed on a persistent buffer it will return -ENODEV just like it would if the .mmap field wasn't defined in the file_operations structure. Cc: [email protected] Cc: Masami Hiramatsu <[email protected]> Cc: Mathieu Desnoyers <[email protected]> Cc: Vincent Donnefort <[email protected]> Link: https://lore.kernel.org/[email protected] Fixes: 9b7bdf6 ("tracing: Have trace_printk not use binary prints if boot buffer") Signed-off-by: Steven Rostedt (Google) <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
In Inline mode, the journal is unused, and journal_sectors is zero. Calculating the journal watermark requires dividing by journal_sectors, which should be done only if the journal is configured. Otherwise, a simple table query (dmsetup table) can cause OOPS. This bug did not show on some systems, perhaps only due to compiler optimization. On my 32-bit testing machine, this reliably crashes with the following: : Oops: divide error: 0000 [#1] PREEMPT SMP : CPU: 0 UID: 0 PID: 2450 Comm: dmsetup Not tainted 6.14.0-rc2+ torvalds#959 : EIP: dm_integrity_status+0x2f8/0xab0 [dm_integrity] ... Signed-off-by: Milan Broz <[email protected]> Signed-off-by: Mikulas Patocka <[email protected]> Fixes: fb09876 ("dm-integrity: introduce the Inline mode") Cc: [email protected] # 6.11+
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
Syzkaller reports the following bug: BUG: spinlock bad magic on CPU#1, syz-executor.0/7995 lock: 0xffff88805303f3e0, .magic: 00000000, .owner: <none>/-1, .owner_cpu: 0 CPU: 1 PID: 7995 Comm: syz-executor.0 Tainted: G E 5.10.209+ #1 Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 11/12/2020 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x119/0x179 lib/dump_stack.c:118 debug_spin_lock_before kernel/locking/spinlock_debug.c:83 [inline] do_raw_spin_lock+0x1f6/0x270 kernel/locking/spinlock_debug.c:112 __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:117 [inline] _raw_spin_lock_irqsave+0x50/0x70 kernel/locking/spinlock.c:159 reset_per_cpu_data+0xe6/0x240 [drop_monitor] net_dm_cmd_trace+0x43d/0x17a0 [drop_monitor] genl_family_rcv_msg_doit+0x22f/0x330 net/netlink/genetlink.c:739 genl_family_rcv_msg net/netlink/genetlink.c:783 [inline] genl_rcv_msg+0x341/0x5a0 net/netlink/genetlink.c:800 netlink_rcv_skb+0x14d/0x440 net/netlink/af_netlink.c:2497 genl_rcv+0x29/0x40 net/netlink/genetlink.c:811 netlink_unicast_kernel net/netlink/af_netlink.c:1322 [inline] netlink_unicast+0x54b/0x800 net/netlink/af_netlink.c:1348 netlink_sendmsg+0x914/0xe00 net/netlink/af_netlink.c:1916 sock_sendmsg_nosec net/socket.c:651 [inline] __sock_sendmsg+0x157/0x190 net/socket.c:663 ____sys_sendmsg+0x712/0x870 net/socket.c:2378 ___sys_sendmsg+0xf8/0x170 net/socket.c:2432 __sys_sendmsg+0xea/0x1b0 net/socket.c:2461 do_syscall_64+0x30/0x40 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x62/0xc7 RIP: 0033:0x7f3f9815aee9 Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b0 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f3f972bf0c8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 00007f3f9826d050 RCX: 00007f3f9815aee9 RDX: 0000000020000000 RSI: 0000000020001300 RDI: 0000000000000007 RBP: 00007f3f981b63bd R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 000000000000006e R14: 00007f3f9826d050 R15: 00007ffe01ee6768 If drop_monitor is built as a kernel module, syzkaller may have time to send a netlink NET_DM_CMD_START message during the module loading. This will call the net_dm_monitor_start() function that uses a spinlock that has not yet been initialized. To fix this, let's place resource initialization above the registration of a generic netlink family. Found by InfoTeCS on behalf of Linux Verification Center (linuxtesting.org) with Syzkaller. Fixes: 9a8afc8 ("Network Drop Monitor: Adding drop monitor implementation & Netlink protocol") Cc: [email protected] Signed-off-by: Ilia Gavrilov <[email protected]> Reviewed-by: Ido Schimmel <[email protected]> Link: https://patch.msgid.link/[email protected] Signed-off-by: Jakub Kicinski <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
A softlockup issue was found with stress test: watchdog: BUG: soft lockup - CPU#27 stuck for 26s! [migration/27:181] CPU: 27 UID: 0 PID: 181 Comm: migration/27 6.14.0-rc2-next-20250210 #1 Stopper: multi_cpu_stop <- stop_machine_from_inactive_cpu RIP: 0010:stop_machine_yield+0x2/0x10 RSP: 0000:ff4a0dcecd19be48 EFLAGS: 00000246 RAX: ffffffff89c0108f RBX: ff4a0dcec03afe44 RCX: 0000000000000000 RDX: ff1cdaaf6eba5808 RSI: 0000000000000282 RDI: ff1cda80c1775a40 RBP: 0000000000000001 R08: 00000011620096c6 R09: 7fffffffffffffff R10: 0000000000000001 R11: 0000000000000100 R12: ff1cda80c1775a40 R13: 0000000000000000 R14: 0000000000000001 R15: ff4a0dcec03afe20 FS: 0000000000000000(0000) GS:ff1cdaaf6eb80000(0000) CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 00000025e2c2a001 CR4: 0000000000773ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: multi_cpu_stop+0x8f/0x100 cpu_stopper_thread+0x90/0x140 smpboot_thread_fn+0xad/0x150 kthread+0xc2/0x100 ret_from_fork+0x2d/0x50 The stress test involves CPU hotplug operations and memory control group (memcg) operations. The scenario can be described as follows: echo xx > memory.max cache_ap_online oom_reaper (CPU23) (CPU50) xx < usage stop_machine_from_inactive_cpu for(;;) // all active cpus trigger OOM queue_stop_cpus_work // waiting oom_reaper multi_cpu_stop(migration/xx) // sync all active cpus ack // waiting cpu23 ack // CPU50 loops in multi_cpu_stop waiting cpu50 Detailed explanation: 1. When the usage is larger than xx, an OOM may be triggered. If the process does not handle with ths kill signal immediately, it will loop in the memory_max_write. 2. When cache_ap_online is triggered, the multi_cpu_stop is queued to the active cpus. Within the multi_cpu_stop function, it attempts to synchronize the CPU states. However, the CPU23 didn't acknowledge because it is stuck in a loop within the for(;;). 3. The oom_reaper process is blocked because CPU50 is in a loop, waiting for CPU23 to acknowledge the synchronization request. 4. Finally, it formed cyclic dependency and lead to softlockup and dead loop. To fix this issue, add cond_resched() in the memory_max_write, so that it will not block migration task. Link: https://lkml.kernel.org/r/[email protected] Fixes: b6e6edc ("mm: memcontrol: reclaim and OOM kill when shrinking memory.max below usage") Signed-off-by: Chen Ridong <[email protected]> Acked-by: Michal Hocko <[email protected]> Cc: Roman Gushchin <[email protected]> Cc: Johannes Weiner <[email protected]> Cc: Shakeel Butt <[email protected]> Cc: Muchun Song <[email protected]> Cc: Wang Weiyang <[email protected]> Signed-off-by: Andrew Morton <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
The namespace percpu counter protects pending I/O, and we can only safely diable the namespace once the counter drop to zero. Otherwise we end up with a crash when running blktests/nvme/058 (eg for loop transport): [ 2352.930426] [ T53909] Oops: general protection fault, probably for non-canonical address 0xdffffc0000000005: 0000 [#1] PREEMPT SMP KASAN PTI [ 2352.930431] [ T53909] KASAN: null-ptr-deref in range [0x0000000000000028-0x000000000000002f] [ 2352.930434] [ T53909] CPU: 3 UID: 0 PID: 53909 Comm: kworker/u16:5 Tainted: G W 6.13.0-rc6 torvalds#232 [ 2352.930438] [ T53909] Tainted: [W]=WARN [ 2352.930440] [ T53909] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-3.fc41 04/01/2014 [ 2352.930443] [ T53909] Workqueue: nvmet-wq nvme_loop_execute_work [nvme_loop] [ 2352.930449] [ T53909] RIP: 0010:blkcg_set_ioprio+0x44/0x180 as the queue is already torn down when calling submit_bio(); So we need to init the percpu counter in nvmet_ns_enable(), and wait for it to drop to zero in nvmet_ns_disable() to avoid having I/O pending after the namespace has been disabled. Fixes: 74d1696 ("nvmet-loop: avoid using mutex in IO hotpath") Signed-off-by: Hannes Reinecke <[email protected]> Reviewed-by: Nilay Shroff <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Chaitanya Kulkarni <[email protected]> Tested-by: Shin'ichiro Kawasaki <[email protected]> Signed-off-by: Keith Busch <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
The delayed work item function nvmet_pci_epf_poll_sqs_work() polls all submission queues and keeps running in a loop as long as commands are being submitted by the host. Depending on the preemption configuration of the kernel, under heavy command workload, this function can thus run for more than RCU_CPU_STALL_TIMEOUT seconds, leading to a RCU stall: rcu: INFO: rcu_sched self-detected stall on CPU rcu: 5-....: (20998 ticks this GP) idle=4244/1/0x4000000000000000 softirq=301/301 fqs=5132 rcu: (t=21000 jiffies g=-443 q=12 ncpus=8) CPU: 5 UID: 0 PID: 82 Comm: kworker/5:1 Not tainted 6.14.0-rc2 #1 Hardware name: Radxa ROCK 5B (DT) Workqueue: events nvmet_pci_epf_poll_sqs_work [nvmet_pci_epf] pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : dw_edma_device_tx_status+0xb8/0x130 lr : dw_edma_device_tx_status+0x9c/0x130 sp : ffff800080b5bbb0 x29: ffff800080b5bbb0 x28: ffff0331c5c78400 x27: ffff0331c1cd1960 x26: ffff0331c0e39010 x25: ffff0331c20e4000 x24: ffff0331c20e4a90 x23: 0000000000000000 x22: 0000000000000001 x21: 00000000005aca33 x20: ffff800080b5bc30 x19: ffff0331c123e370 x18: 000000000ab29e62 x17: ffffb2a878c9c118 x16: ffff0335bde82040 x15: 0000000000000000 x14: 000000000000017b x13: 00000000ee601780 x12: 0000000000000018 x11: 0000000000000000 x10: 0000000000000001 x9 : 0000000000000040 x8 : 00000000ee601780 x7 : 0000000105c785c0 x6 : ffff0331c1027d80 x5 : 0000000001ee7ad6 x4 : ffff0335bdea16c0 x3 : ffff0331c123e438 x2 : 00000000005aca33 x1 : 0000000000000000 x0 : ffff0331c123e410 Call trace: dw_edma_device_tx_status+0xb8/0x130 (P) dma_sync_wait+0x60/0xbc nvmet_pci_epf_dma_transfer+0x128/0x264 [nvmet_pci_epf] nvmet_pci_epf_poll_sqs_work+0x2a0/0x2e0 [nvmet_pci_epf] process_one_work+0x144/0x390 worker_thread+0x27c/0x458 kthread+0xe8/0x19c ret_from_fork+0x10/0x20 The solution for this is simply to explicitly allow rescheduling using cond_resched(). However, since doing so for every loop of nvmet_pci_epf_poll_sqs_work() significantly degrades performance (for 4K random reads using 4 I/O queues, the maximum IOPS goes down from 137 KIOPS to 110 KIOPS), call cond_resched() every second to avoid the RCU stalls. Fixes: 0faa0fe ("nvmet: New NVMe PCI endpoint function target driver") Signed-off-by: Damien Le Moal <[email protected]> Reviewed-by: Christoph Hellwig <[email protected]> Reviewed-by: Sagi Grimberg <[email protected]> Signed-off-by: Keith Busch <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
Brad Spengler reported the list_del() corruption splat in gtp_net_exit_batch_rtnl(). [0] Commit eb28fd7 ("gtp: Destroy device along with udp socket's netns dismantle.") added the for_each_netdev() loop in gtp_net_exit_batch_rtnl() to destroy devices in each netns as done in geneve and ip tunnels. However, this could trigger ->dellink() twice for the same device during ->exit_batch_rtnl(). Say we have two netns A & B and gtp device B that resides in netns B but whose UDP socket is in netns A. 1. cleanup_net() processes netns A and then B. 2. gtp_net_exit_batch_rtnl() finds the device B while iterating netns A's gn->gtp_dev_list and calls ->dellink(). [ device B is not yet unlinked from netns B as unregister_netdevice_many() has not been called. ] 3. gtp_net_exit_batch_rtnl() finds the device B while iterating netns B's for_each_netdev() and calls ->dellink(). gtp_dellink() cleans up the device's hash table, unlinks the dev from gn->gtp_dev_list, and calls unregister_netdevice_queue(). Basically, calling gtp_dellink() multiple times is fine unless CONFIG_DEBUG_LIST is enabled. Let's remove for_each_netdev() in gtp_net_exit_batch_rtnl() and delegate the destruction to default_device_exit_batch() as done in bareudp. [0]: list_del corruption, ffff8880aaa62c00->next (autoslab_size_M_dev_P_net_core_dev_11127_8_1328_8_S_4096_A_64_n_139+0xc00/0x1000 [slab object]) is LIST_POISON1 (ffffffffffffff02) (prev is 0xffffffffffffff04) kernel BUG at lib/list_debug.c:58! Oops: invalid opcode: 0000 [#1] PREEMPT SMP KASAN CPU: 1 UID: 0 PID: 1804 Comm: kworker/u8:7 Tainted: G T 6.12.13-grsec-full-20250211091339 #1 Tainted: [T]=RANDSTRUCT Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 Workqueue: netns cleanup_net RIP: 0010:[<ffffffff84947381>] __list_del_entry_valid_or_report+0x141/0x200 lib/list_debug.c:58 Code: c2 76 91 31 c0 e8 9f b1 f7 fc 0f 0b 4d 89 f0 48 c7 c1 02 ff ff ff 48 89 ea 48 89 ee 48 c7 c7 e0 c2 76 91 31 c0 e8 7f b1 f7 fc <0f> 0b 4d 89 e8 48 c7 c1 04 ff ff ff 48 89 ea 48 89 ee 48 c7 c7 60 RSP: 0018:fffffe8040b4fbd0 EFLAGS: 00010283 RAX: 00000000000000cc RBX: dffffc0000000000 RCX: ffffffff818c4054 RDX: ffffffff84947381 RSI: ffffffff818d1512 RDI: 0000000000000000 RBP: ffff8880aaa62c00 R08: 0000000000000001 R09: fffffbd008169f32 R10: fffffe8040b4f997 R11: 0000000000000001 R12: a1988d84f24943e4 R13: ffffffffffffff02 R14: ffffffffffffff04 R15: ffff8880aaa62c08 RBX: kasan shadow of 0x0 RCX: __wake_up_klogd.part.0+0x74/0xe0 kernel/printk/printk.c:4554 RDX: __list_del_entry_valid_or_report+0x141/0x200 lib/list_debug.c:58 RSI: vprintk+0x72/0x100 kernel/printk/printk_safe.c:71 RBP: autoslab_size_M_dev_P_net_core_dev_11127_8_1328_8_S_4096_A_64_n_139+0xc00/0x1000 [slab object] RSP: process kstack fffffe8040b4fbd0+0x7bd0/0x8000 [kworker/u8:7+netns 1804 ] R09: kasan shadow of process kstack fffffe8040b4f990+0x7990/0x8000 [kworker/u8:7+netns 1804 ] R10: process kstack fffffe8040b4f997+0x7997/0x8000 [kworker/u8:7+netns 1804 ] R15: autoslab_size_M_dev_P_net_core_dev_11127_8_1328_8_S_4096_A_64_n_139+0xc08/0x1000 [slab object] FS: 0000000000000000(0000) GS:ffff888116000000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000748f5372c000 CR3: 0000000015408000 CR4: 00000000003406f0 shadow CR4: 00000000003406f0 Stack: 0000000000000000 ffffffff8a0c35e7 ffffffff8a0c3603 ffff8880aaa62c00 ffff8880aaa62c00 0000000000000004 ffff88811145311c 0000000000000005 0000000000000001 ffff8880aaa62000 fffffe8040b4fd40 ffffffff8a0c360d Call Trace: <TASK> [<ffffffff8a0c360d>] __list_del_entry_valid include/linux/list.h:131 [inline] fffffe8040b4fc28 [<ffffffff8a0c360d>] __list_del_entry include/linux/list.h:248 [inline] fffffe8040b4fc28 [<ffffffff8a0c360d>] list_del include/linux/list.h:262 [inline] fffffe8040b4fc28 [<ffffffff8a0c360d>] gtp_dellink+0x16d/0x360 drivers/net/gtp.c:1557 fffffe8040b4fc28 [<ffffffff8a0d0404>] gtp_net_exit_batch_rtnl+0x124/0x2c0 drivers/net/gtp.c:2495 fffffe8040b4fc88 [<ffffffff8e705b24>] cleanup_net+0x5a4/0xbe0 net/core/net_namespace.c:635 fffffe8040b4fcd0 [<ffffffff81754c97>] process_one_work+0xbd7/0x2160 kernel/workqueue.c:3326 fffffe8040b4fd88 [<ffffffff81757195>] process_scheduled_works kernel/workqueue.c:3407 [inline] fffffe8040b4fec0 [<ffffffff81757195>] worker_thread+0x6b5/0xfa0 kernel/workqueue.c:3488 fffffe8040b4fec0 [<ffffffff817782a0>] kthread+0x360/0x4c0 kernel/kthread.c:397 fffffe8040b4ff78 [<ffffffff814d8594>] ret_from_fork+0x74/0xe0 arch/x86/kernel/process.c:172 fffffe8040b4ffb8 [<ffffffff8110f509>] ret_from_fork_asm+0x29/0xc0 arch/x86/entry/entry_64.S:399 fffffe8040b4ffe8 </TASK> Modules linked in: Fixes: eb28fd7 ("gtp: Destroy device along with udp socket's netns dismantle.") Reported-by: Brad Spengler <[email protected]> Signed-off-by: Kuniyuki Iwashima <[email protected]> Link: https://patch.msgid.link/[email protected] Signed-off-by: Jakub Kicinski <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
…umers While using nvme target with use_srq on, below kernel panic is noticed. [ 549.698111] bnxt_en 0000:41:00.0 enp65s0np0: FEC autoneg off encoding: Clause 91 RS(544,514) [ 566.393619] Oops: divide error: 0000 [#1] PREEMPT SMP NOPTI .. [ 566.393799] <TASK> [ 566.393807] ? __die_body+0x1a/0x60 [ 566.393823] ? die+0x38/0x60 [ 566.393835] ? do_trap+0xe4/0x110 [ 566.393847] ? bnxt_qplib_alloc_init_hwq+0x1d4/0x580 [bnxt_re] [ 566.393867] ? bnxt_qplib_alloc_init_hwq+0x1d4/0x580 [bnxt_re] [ 566.393881] ? do_error_trap+0x7c/0x120 [ 566.393890] ? bnxt_qplib_alloc_init_hwq+0x1d4/0x580 [bnxt_re] [ 566.393911] ? exc_divide_error+0x34/0x50 [ 566.393923] ? bnxt_qplib_alloc_init_hwq+0x1d4/0x580 [bnxt_re] [ 566.393939] ? asm_exc_divide_error+0x16/0x20 [ 566.393966] ? bnxt_qplib_alloc_init_hwq+0x1d4/0x580 [bnxt_re] [ 566.393997] bnxt_qplib_create_srq+0xc9/0x340 [bnxt_re] [ 566.394040] bnxt_re_create_srq+0x335/0x3b0 [bnxt_re] [ 566.394057] ? srso_return_thunk+0x5/0x5f [ 566.394068] ? __init_swait_queue_head+0x4a/0x60 [ 566.394090] ib_create_srq_user+0xa7/0x150 [ib_core] [ 566.394147] nvmet_rdma_queue_connect+0x7d0/0xbe0 [nvmet_rdma] [ 566.394174] ? lock_release+0x22c/0x3f0 [ 566.394187] ? srso_return_thunk+0x5/0x5f Page size and shift info is set only for the user space SRQs. Set page size and page shift for kernel space SRQs also. Fixes: 0c4dcd6 ("RDMA/bnxt_re: Refactor hardware queue memory allocation") Signed-off-by: Kashyap Desai <[email protected]> Signed-off-by: Selvin Xavier <[email protected]> Link: https://patch.msgid.link/[email protected] Signed-off-by: Leon Romanovsky <[email protected]>
mina
pushed a commit
that referenced
this pull request
Feb 28, 2025
We have recently seen reports of lockdep circular lock dependency warnings
when loading the iAVF driver:
[ 1504.790308] ======================================================
[ 1504.790309] WARNING: possible circular locking dependency detected
[ 1504.790310] 6.13.0 #net_next_rt.c2933b2befe2.el9 Not tainted
[ 1504.790311] ------------------------------------------------------
[ 1504.790312] kworker/u128:0/13566 is trying to acquire lock:
[ 1504.790313] ffff97d0e4738f18 (&dev->lock){+.+.}-{4:4}, at: register_netdevice+0x52c/0x710
[ 1504.790320]
[ 1504.790320] but task is already holding lock:
[ 1504.790321] ffff97d0e47392e8 (&adapter->crit_lock){+.+.}-{4:4}, at: iavf_finish_config+0x37/0x240 [iavf]
[ 1504.790330]
[ 1504.790330] which lock already depends on the new lock.
[ 1504.790330]
[ 1504.790330]
[ 1504.790330] the existing dependency chain (in reverse order) is:
[ 1504.790331]
[ 1504.790331] -> #1 (&adapter->crit_lock){+.+.}-{4:4}:
[ 1504.790333] __lock_acquire+0x52d/0xbb0
[ 1504.790337] lock_acquire+0xd9/0x330
[ 1504.790338] mutex_lock_nested+0x4b/0xb0
[ 1504.790341] iavf_finish_config+0x37/0x240 [iavf]
[ 1504.790347] process_one_work+0x248/0x6d0
[ 1504.790350] worker_thread+0x18d/0x330
[ 1504.790352] kthread+0x10e/0x250
[ 1504.790354] ret_from_fork+0x30/0x50
[ 1504.790357] ret_from_fork_asm+0x1a/0x30
[ 1504.790361]
[ 1504.790361] -> #0 (&dev->lock){+.+.}-{4:4}:
[ 1504.790364] check_prev_add+0xf1/0xce0
[ 1504.790366] validate_chain+0x46a/0x570
[ 1504.790368] __lock_acquire+0x52d/0xbb0
[ 1504.790370] lock_acquire+0xd9/0x330
[ 1504.790371] mutex_lock_nested+0x4b/0xb0
[ 1504.790372] register_netdevice+0x52c/0x710
[ 1504.790374] iavf_finish_config+0xfa/0x240 [iavf]
[ 1504.790379] process_one_work+0x248/0x6d0
[ 1504.790381] worker_thread+0x18d/0x330
[ 1504.790383] kthread+0x10e/0x250
[ 1504.790385] ret_from_fork+0x30/0x50
[ 1504.790387] ret_from_fork_asm+0x1a/0x30
[ 1504.790389]
[ 1504.790389] other info that might help us debug this:
[ 1504.790389]
[ 1504.790389] Possible unsafe locking scenario:
[ 1504.790389]
[ 1504.790390] CPU0 CPU1
[ 1504.790391] ---- ----
[ 1504.790391] lock(&adapter->crit_lock);
[ 1504.790393] lock(&dev->lock);
[ 1504.790394] lock(&adapter->crit_lock);
[ 1504.790395] lock(&dev->lock);
[ 1504.790397]
[ 1504.790397] *** DEADLOCK ***
This appears to be caused by the change in commit 5fda3f3 ("net: make
netdev_lock() protect netdev->reg_state"), which added a netdev_lock() in
register_netdevice.
The iAVF driver calls register_netdevice() from iavf_finish_config(), as a
final stage of its state machine post-probe. It currently takes the RTNL
lock, then the netdev lock, and then the device critical lock. This pattern
is used throughout the driver. Thus there is a strong dependency that the
crit_lock should not be acquired before the net device lock. The change to
register_netdevice creates an ABBA lock order violation because the iAVF
driver is holding the crit_lock while calling register_netdevice, which
then takes the netdev_lock.
It seems likely that future refactors could result in netdev APIs which
hold the netdev_lock while calling into the driver. This means that we
should not re-order the locks so that netdev_lock is acquired after the
device private crit_lock.
Instead, notice that we already release the netdev_lock prior to calling
the register_netdevice. This flow only happens during the early driver
initialization as we transition through the __IAVF_STARTUP,
__IAVF_INIT_VERSION_CHECK, __IAVF_INIT_GET_RESOURCES, etc.
Analyzing the places where we take crit_lock in the driver there are two
sources:
a) several of the work queue tasks including adminq_task, watchdog_task,
reset_task, and the finish_config task.
b) various callbacks which ultimately stem back to .ndo operations or
ethtool operations.
The latter cannot be triggered until after the netdevice registration is
completed successfully.
The iAVF driver uses alloc_ordered_workqueue, which is an unbound workqueue
that has a max limit of 1, and thus guarantees that only a single work item
on the queue is executing at any given time, so none of the other work
threads could be executing due to the ordered workqueue guarantees.
The iavf_finish_config() function also does not do anything else after
register_netdevice, unless it fails. It seems unlikely that the driver
private crit_lock is protecting anything that register_netdevice() itself
touches.
Thus, to fix this ABBA lock violation, lets simply release the
adapter->crit_lock as well as netdev_lock prior to calling
register_netdevice(). We do still keep holding the RTNL lock as required by
the function. If we do fail to register the netdevice, then we re-acquire
the adapter critical lock to finish the transition back to
__IAVF_INIT_CONFIG_ADAPTER.
This ensures every call where both netdev_lock and the adapter->crit_lock
are acquired under the same ordering.
Fixes: afc6649 ("eth: iavf: extend the netdev_lock usage")
Signed-off-by: Jacob Keller <[email protected]>
Tested-by: Przemek Kitszel <[email protected]>
Reviewed-by: Przemek Kitszel <[email protected]>
Reviewed-by: Jakub Kicinski <[email protected]>
Tested-by: Rafal Romanowski <[email protected]>
Signed-off-by: Tony Nguyen <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>
Document expectations from drivers looking to add support for device memory tcp or other netmem based features. Signed-off-by: Mina Almasry <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
b79225d to
1710b85
Compare
a6928a8 to
16f7903
Compare
Add documentation outlining the usage and details of the devmem TCP TX API. Signed-off-by: Mina Almasry <[email protected]> Acked-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Drivers need to make sure not to pass netmem dma-addrs to the dma-mapping API in order to support netmem TX. Add helpers and netmem_dma_*() helpers that enables special handling of netmem dma-addrs that drivers can use. Document in netmem.rst what drivers need to do to support netmem TX. Signed-off-by: Mina Almasry <[email protected]> Acked-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Use netmem_dma_*() helpers in gve_tx_dqo.c DQO-RDA paths to enable netmem TX support in that mode. Declare support for netmem TX in GVE DQO-RDA mode. Signed-off-by: Mina Almasry <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
We should not enable netmem TX for drivers that don't declare support. Check for driver netmem TX support during devmem TX binding and fail if the driver does not have the functionality. Check for driver support in validate_xmit_skb as well. Signed-off-by: Mina Almasry <[email protected]> Acked-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
That should make it possible to do expected payload validation on the caller side. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
So we can plug the other ones in the future if needed. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
There is a bunch of places where error() calls look out of place. Use the same error(1, errno, ...) pattern everywhere. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Support 3-tuple filtering by making client_ip optional. When -c is not passed, don't specify src-ip/src-port in the filter. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
To make it clear what's required and what's not. Also, some of the values don't seem like a good defaults; for example eth1. Move the invocation comment to the top, add missing -s to the client and cleanup the client invocation a bit to make more readable. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Use dualstack socket to support both v4 and v6. v4-mapped-v6 address can be used to do v4. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
ntuple off/on might be not enough to do it on all NICs. Add a bunch of shell crap to explicitly remove the rules. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
In the next patch the hard-coded queue numbers are gonna be removed. So introduce some initial support for ethtool YNL and use it to enable header split. Also, tcp-data-split requires latest ethtool which is unlikely to be present in the distros right now. (ideally, we should not shell out to ethtool at all). Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Use single last queue of the device and probe it dynamically. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
… has been provided This will be used as a 'probe' mode in the selftest to check whether the device supports the devmem or not. Use hard-coded queue layout (two last queues) and prevent user from passing custom -q and/or -t. Reviewed-by: Mina Almasry <[email protected]> Reviewed-by: Joe Damato <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
This is where all the tests that depend on the HW functionality live in and this is where the automated test is gonna be added in the next patch. Reviewed-by: Mina Almasry <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Only RX side for now and small message to test the setup.
In the future, we can extend it to TX side and to testing
both sides with a couple of megs of data.
make \
-C tools/testing/selftests \
TARGETS="drivers/hw/net" \
install INSTALL_PATH=~/tmp/ksft
scp ~/tmp/ksft ${HOST}:
scp ~/tmp/ksft ${PEER}:
cfg+="NETIF=${DEV}\n"
cfg+="LOCAL_V6=${HOST_IP}\n"
cfg+="REMOTE_V6=${PEER_IP}\n"
cfg+="REMOTE_TYPE=ssh\n"
cfg+="REMOTE_ARGS=root@${PEER}\n"
echo -e "$cfg" | ssh root@${HOST} "cat > ksft/drivers/net/net.config"
ssh root@${HOST} "cd ksft && ./run_kselftest.sh -t
drivers/net:devmem.py"
Reviewed-by: Mina Almasry <[email protected]>
Signed-off-by: Stanislav Fomichev <[email protected]>
Signed-off-by: Pranjal Shrivastava <[email protected]>
Since YNL doesn't depend on libmnl either, any more, it's actually possible to build selftests without having libmnl installed. Note: This is required to let the ncdevmem test to build for 6.12 Signed-off-by: Jakub Kicinski <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Add support for devmem TX in ncdevmem. This is a combination of the ncdevmem from the devmem TCP series RFCv1 which included the TX path, and work by Stan to include the netlink API and refactored on top of his generic memory_provider support. Signed-off-by: Mina Almasry <[email protected]> Signed-off-by: Stanislav Fomichev <[email protected]> Acked-by: Stanislav Fomichev <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Not all the devices have the capability for the driver to query for the registered RSS configuration. The driver can discover this by checking the relevant device option during setup. If it cannot, the driver needs to store the RSS config cache and directly return such cache when queried by the ethtool. RSS config is inited when driver probes. Also the default RSS config will be adjusted when there is RX queue count change. At this point, only keys of GVE_RSS_KEY_SIZE and indirection tables of GVE_RSS_INDIR_SIZE are supported. Signed-off-by: Ziwei Xiao <[email protected]> Reviewed-by: Harshitha Ramamurthy <[email protected]> Reviewed-by: Willem de Bruijn <[email protected]> Signed-off-by: Praveen Kaligineedi <[email protected]> Signed-off-by: Jeroen de Borst <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
In preparation for the upcoming page pool adoption for DQO raw addressing mode, move RX buffer management code to a new file. In the follow on patches, page pool code will be added to this file. No functional change, just movement of code. Reviewed-by: Praveen Kaligineedi <[email protected]> Reviewed-by: Shailend Chand <[email protected]> Reviewed-by: Willem de Bruijn <[email protected]> Signed-off-by: Harshitha Ramamurthy <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
For DQ queue format in raw DMA addressing(RDA) mode, implement page pool recycling of buffers by leveraging a few helper functions. DQ QPL mode will continue to use the exisiting recycling logic. This is because in QPL mode, the pages come from a constant set of pages that the driver pre-allocates and registers with the device. Reviewed-by: Praveen Kaligineedi <[email protected]> Reviewed-by: Shailend Chand <[email protected]> Reviewed-by: Willem de Bruijn <[email protected]> Signed-off-by: Harshitha Ramamurthy <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Implement netdev_stats_ops to export basic per-queue stats. With page pool support for DQO added in the previous patches, rx-alloc-fail captures failures in page pool allocations as well since the rx_buf_alloc_fail stat tracked in the driver is incremented when gve_alloc_buffer returns error. Reviewed-by: Praveen Kaligineedi <[email protected]> Reviewed-by: Willem de Bruijn <[email protected]> Signed-off-by: Harshitha Ramamurthy <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
…ages 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(). Signed-off-by: Pranjal Shrivastava <[email protected]>
16f7903 to
b54d4a3
Compare
When a queue is stopped using the ndo queue API, before destroying its page pool, the associated NAPI instance needs to be unlinked to avoid warnings. Handle this by calling page_pool_disable_direct_recycling() when stopping a queue. Fixes: ebdfae0 ("gve: adopt page pool for DQ RDA mode") Reviewed-by: Praveen Kaligineedi <[email protected]> Signed-off-by: Harshitha Ramamurthy <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
Use netmem_ref instead of struct page to track pages for DQO RDA mode which will support devmem TCP. To achieve this, track netmem_ref in struct gve_rx_slot_page_info for DQO raw addressing mode while page is used for QPL mode. Change-Id: I40da83100b765e4af60fe23d33e92963794e93fe Signed-off-by: Harshitha Ramamurthy <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
GVE driver changes for TCP devmem Signed-off-by: Mina Almasry <[email protected]> Signed-off-by: Pranjal Shrivastava <[email protected]>
pran005
pushed a commit
to pran005/linux
that referenced
this pull request
Mar 6, 2025
into HEAD KVM/riscv fixes for 6.14, take mina#1 - Fix hart status check in SBI HSM extension - Fix hart suspend_type usage in SBI HSM extension - Fix error returned by SBI IPI and TIME extensions for unsupported function IDs - Fix suspend_type usage in SBI SUSP extension - Remove unnecessary vcpu kick after injecting interrupt via IMSIC guest file
pran005
pushed a commit
to pran005/linux
that referenced
this pull request
Mar 6, 2025
The customer reports that there is a soft lockup issue related to the i2c driver. After checking, the i2c module was doing a tx transfer and the bmc machine reboots in the middle of the i2c transaction, the i2c module keeps the status without being reset. Due to such an i2c module status, the i2c irq handler keeps getting triggered since the i2c irq handler is registered in the kernel booting process after the bmc machine is doing a warm rebooting. The continuous triggering is stopped by the soft lockup watchdog timer. Disable the interrupt enable bit in the i2c module before calling devm_request_irq to fix this issue since the i2c relative status bit is read-only. Here is the soft lockup log. [ 28.176395] watchdog: BUG: soft lockup - CPU#0 stuck for 26s! [swapper/0:1] [ 28.183351] Modules linked in: [ 28.186407] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.15.120-yocto-s-dirty-bbebc78 mina#1 [ 28.201174] pstate: 40000005 (nZcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 28.208128] pc : __do_softirq+0xb0/0x368 [ 28.212055] lr : __do_softirq+0x70/0x368 [ 28.215972] sp : ffffff8035ebca00 [ 28.219278] x29: ffffff8035ebca00 x28: 0000000000000002 x27: ffffff80071a3780 [ 28.226412] x26: ffffffc008bdc000 x25: ffffffc008bcc640 x24: ffffffc008be50c0 [ 28.233546] x23: ffffffc00800200c x22: 0000000000000000 x21: 000000000000001b [ 28.240679] x20: 0000000000000000 x19: ffffff80001c3200 x18: ffffffffffffffff [ 28.247812] x17: ffffffc02d2e0000 x16: ffffff8035eb8b40 x15: 00001e8480000000 [ 28.254945] x14: 02c3647e37dbfcb6 x13: 02c364f2ab14200c x12: 0000000002c364f2 [ 28.262078] x11: 00000000fa83b2da x10: 000000000000b67e x9 : ffffffc008010250 [ 28.269211] x8 : 000000009d983d00 x7 : 7fffffffffffffff x6 : 0000036d74732434 [ 28.276344] x5 : 00ffffffffffffff x4 : 0000000000000015 x3 : 0000000000000198 [ 28.283476] x2 : ffffffc02d2e0000 x1 : 00000000000000e0 x0 : ffffffc008bdcb40 [ 28.290611] Call trace: [ 28.293052] __do_softirq+0xb0/0x368 [ 28.296625] __irq_exit_rcu+0xe0/0x100 [ 28.300374] irq_exit+0x14/0x20 [ 28.303513] handle_domain_irq+0x68/0x90 [ 28.307440] gic_handle_irq+0x78/0xb0 [ 28.311098] call_on_irq_stack+0x20/0x38 [ 28.315019] do_interrupt_handler+0x54/0x5c [ 28.319199] el1_interrupt+0x2c/0x4c [ 28.322777] el1h_64_irq_handler+0x14/0x20 [ 28.326872] el1h_64_irq+0x74/0x78 [ 28.330269] __setup_irq+0x454/0x780 [ 28.333841] request_threaded_irq+0xd0/0x1b4 [ 28.338107] devm_request_threaded_irq+0x84/0x100 [ 28.342809] npcm_i2c_probe_bus+0x188/0x3d0 [ 28.346990] platform_probe+0x6c/0xc4 [ 28.350653] really_probe+0xcc/0x45c [ 28.354227] __driver_probe_device+0x8c/0x160 [ 28.358578] driver_probe_device+0x44/0xe0 [ 28.362670] __driver_attach+0x124/0x1d0 [ 28.366589] bus_for_each_dev+0x7c/0xe0 [ 28.370426] driver_attach+0x28/0x30 [ 28.373997] bus_add_driver+0x124/0x240 [ 28.377830] driver_register+0x7c/0x124 [ 28.381662] __platform_driver_register+0x2c/0x34 [ 28.386362] npcm_i2c_init+0x3c/0x5c [ 28.389937] do_one_initcall+0x74/0x230 [ 28.393768] kernel_init_freeable+0x24c/0x2b4 [ 28.398126] kernel_init+0x28/0x130 [ 28.401614] ret_from_fork+0x10/0x20 [ 28.405189] Kernel panic - not syncing: softlockup: hung tasks [ 28.411011] SMP: stopping secondary CPUs [ 28.414933] Kernel Offset: disabled [ 28.418412] CPU features: 0x00000000,00000802 [ 28.427644] Rebooting in 20 seconds.. Fixes: 56a1485 ("i2c: npcm7xx: Add Nuvoton NPCM I2C controller driver") Signed-off-by: Tyrone Ting <[email protected]> Cc: <[email protected]> # v5.8+ Reviewed-by: Tali Perry <[email protected]> Signed-off-by: Andi Shyti <[email protected]> Link: https://lore.kernel.org/r/[email protected]
pran005
pushed a commit
to pran005/linux
that referenced
this pull request
Mar 6, 2025
Commit <d74169ceb0d2> ("iommu/vt-d: Allocate DMAR fault interrupts
locally") moved the call to enable_drhd_fault_handling() to a code
path that does not hold any lock while traversing the drhd list. Fix
it by ensuring the dmar_global_lock lock is held when traversing the
drhd list.
Without this fix, the following warning is triggered:
=============================
WARNING: suspicious RCU usage
6.14.0-rc3 torvalds#55 Not tainted
-----------------------------
drivers/iommu/intel/dmar.c:2046 RCU-list traversed in non-reader section!!
other info that might help us debug this:
rcu_scheduler_active = 1, debug_locks = 1
2 locks held by cpuhp/1/23:
#0: ffffffff84a67c50 (cpu_hotplug_lock){++++}-{0:0}, at: cpuhp_thread_fun+0x87/0x2c0
mina#1: ffffffff84a6a380 (cpuhp_state-up){+.+.}-{0:0}, at: cpuhp_thread_fun+0x87/0x2c0
stack backtrace:
CPU: 1 UID: 0 PID: 23 Comm: cpuhp/1 Not tainted 6.14.0-rc3 torvalds#55
Call Trace:
<TASK>
dump_stack_lvl+0xb7/0xd0
lockdep_rcu_suspicious+0x159/0x1f0
? __pfx_enable_drhd_fault_handling+0x10/0x10
enable_drhd_fault_handling+0x151/0x180
cpuhp_invoke_callback+0x1df/0x990
cpuhp_thread_fun+0x1ea/0x2c0
smpboot_thread_fn+0x1f5/0x2e0
? __pfx_smpboot_thread_fn+0x10/0x10
kthread+0x12a/0x2d0
? __pfx_kthread+0x10/0x10
ret_from_fork+0x4a/0x60
? __pfx_kthread+0x10/0x10
ret_from_fork_asm+0x1a/0x30
</TASK>
Holding the lock in enable_drhd_fault_handling() triggers a lockdep splat
about a possible deadlock between dmar_global_lock and cpu_hotplug_lock.
This is avoided by not holding dmar_global_lock when calling
iommu_device_register(), which initiates the device probe process.
Fixes: d74169c ("iommu/vt-d: Allocate DMAR fault interrupts locally")
Reported-and-tested-by: Ido Schimmel <[email protected]>
Closes: https://lore.kernel.org/linux-iommu/[email protected]/
Tested-by: Breno Leitao <[email protected]>
Cc: [email protected]
Signed-off-by: Lu Baolu <[email protected]>
Reviewed-by: Kevin Tian <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Tested-by: Ido Schimmel <[email protected]>
Signed-off-by: Joerg Roedel <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Backport TCP Devmem fixes on 6.12 and Tx path along with the required kselftest dependencies for cos.
Note: The current version includes driver changes for Tx path but may be not have the relevant driver changes for the Rx path yet.
Test: Builds with x86_64 & arm64 defconfigs.
Test: Builds with lakitu_defconfig
Test: Kselftests build:
make -C tools/testing/selftests/ TARGETS="drivers/net/hw" install INSTALL_PATH=~/tmp/ksft