search for: trace_hardirqs_off

On Tue, Aug 11, 2020 at 11:20:54AM +0200, peterz at infradead.org wrote: > On Tue, Aug 11, 2020 at 10:38:50AM +0200, J?rgen Gro? wrote: > > In case you don't want to do it I can send the patch for the Xen > > variants. > > I might've opened a whole new can of worms here. I'm not sure we > can/want to fix the entire fallout this release :/ > > Let me

On Tue, Aug 11, 2020 at 11:20:54AM +0200, peterz at infradead.org wrote: > On Tue, Aug 11, 2020 at 10:38:50AM +0200, J?rgen Gro? wrote: > > In case you don't want to do it I can send the patch for the Xen > > variants. > > I might've opened a whole new can of worms here. I'm not sure we > can/want to fix the entire fallout this release :/ > > Let me

...514005] <IRQ> [<ffffffff811259a2>] lockdep_rcu_suspicious+0xe2/0x130 [ 2.514107] [<ffffffff810ea10c>] __atomic_notifier_call_chain+0x12c/0x140 [ 2.514169] [<ffffffff810e9fe0>] ? atomic_notifier_chain_unregister+0x90/0x90 [ 2.514258] [<ffffffff811216cd>] ? trace_hardirqs_off+0xd/0x10 [ 2.514318] [<ffffffff810ea136>] atomic_notifier_call_chain+0x16/0x20 [ 2.514381] [<ffffffff810777c3>] exit_idle+0x43/0x50 [ 2.514441] [<ffffffff81568865>] xen_evtchn_do_upcall+0x25/0x50 [ 2.514503] [<ffffffff81aa690e>] xen_do_hypervisor_callback+0x1...

...514005] <IRQ> [<ffffffff811259a2>] lockdep_rcu_suspicious+0xe2/0x130 [ 2.514107] [<ffffffff810ea10c>] __atomic_notifier_call_chain+0x12c/0x140 [ 2.514169] [<ffffffff810e9fe0>] ? atomic_notifier_chain_unregister+0x90/0x90 [ 2.514258] [<ffffffff811216cd>] ? trace_hardirqs_off+0xd/0x10 [ 2.514318] [<ffffffff810ea136>] atomic_notifier_call_chain+0x16/0x20 [ 2.514381] [<ffffffff810777c3>] exit_idle+0x43/0x50 [ 2.514441] [<ffffffff81568865>] xen_evtchn_do_upcall+0x25/0x50 [ 2.514503] [<ffffffff81aa690e>] xen_do_hypervisor_callback+0x1...

...e0 ? drm_calc_vbltimestamp_from_scanoutpos+0x625/0x740 (drivers/gpu/drm/drm_irq.c:743) __asan_report_load4_noabort+0x2e/0x30 (mm/kasan/report.c:331) drm_calc_vbltimestamp_from_scanoutpos+0x625/0x740 (drivers/gpu/drm/drm_irq.c:743) ? drm_irq_install+0x570/0x570 (drivers/gpu/drm/drm_irq.c:459) ? trace_hardirqs_off+0xd/0x10 (kernel/locking/lockdep.c:2780) ? _raw_spin_unlock_irqrestore+0x4b/0x50 (kernel/locking/spinlock.c:190) ? try_to_wake_up+0xc6/0xd00 (kernel/sched/core.c:2010) ? debug_check_no_locks_freed+0x280/0x280 (kernel/locking/lockdep.c:4270) ? debug_check_no_locks_freed+0x280/0x280 (kernel/locki...

...xattr+0x77/0x82 [btrfs] [ 1998.478275] [<ffffffff811f3044>] cap_inode_need_killpriv+0x2d/0x37 [ 1998.478275] [<ffffffff810d8309>] file_remove_suid+0x27/0x64 [ 1998.478275] [<ffffffffa0240e14>] btrfs_file_aio_write+0x159/0x49a [btrfs] [ 1998.478275] [<ffffffff8107bfd2>] ? trace_hardirqs_off+0xd/0xf [ 1998.478275] [<ffffffff81071f42>] ? local_clock+0x36/0x4d [ 1998.478275] [<ffffffff810800ab>] ? lock_release_non_nested+0xdb/0x263 [ 1998.478275] [<ffffffff81121996>] do_sync_write+0xcb/0x108 [ 1998.478275] [<ffffffff810f33a4>] ? might_fault+0x5c/0xac [ 1998.47...

...+0x3b/0x90 [nouveau] [ 37.438649] [<ffffffffa02707b2>] nouveau_channel_idle+0x42/0xb0 [nouveau] [ 37.438673] [<ffffffffa026bb3f>] nouveau_fbcon_sync+0x7f/0xb0 [nouveau] [ 37.438677] [<ffffffff8148998a>] cfb_imageblit+0x9a/0x4d0 [ 37.438681] [<ffffffff810d22ef>] ? trace_hardirqs_off_caller+0x1f/0xc0 [ 37.438693] [<ffffffffa013d30e>] drm_fb_helper_cfb_imageblit+0xe/0x10 [drm_kms_helper] [ 37.438717] [<ffffffffa026bbc1>] nouveau_fbcon_imageblit+0x51/0xd0 [nouveau] [ 37.438719] [<ffffffff8147c02c>] bit_putcs+0x2dc/0x530 [ 37.438721] [<ffffffff810d...

...02b330a7f8 ffff8802b330a820 ffff8802b330a7f8 0000000000000000 0000000000000001 Call Trace: <IRQ> [<ffffffff81755f39>] dump_stack+0x19/0x1b [<ffffffff8174f526>] print_circular_bug+0x1fb/0x20c [<ffffffff810b65c3>] __lock_acquire+0x1c43/0x1d30 [<ffffffff810b1f8d>] ? trace_hardirqs_off+0xd/0x10 [<ffffffff8102b1a8>] ? flat_send_IPI_mask+0x88/0xa0 [<ffffffff810b6d62>] lock_acquire+0x92/0x1f0 [<ffffffffa0086269>] ? drm_handle_vblank+0x69/0x400 [drm] [<ffffffff8175e926>] _raw_spin_lock_irqsave+0x46/0x60 [<ffffffffa0086269>] ? drm_handle_vblank+0x69/...

...79/0x480 [ 36.076895] vhost_vq_init_access+0xca/0x540 [ 36.081276] vhost_net_ioctl+0xee0/0x1920 [ 36.085397] ? vhost_net_stop_vq+0xf0/0xf0 [ 36.089603] ? avc_ss_reset+0x110/0x110 [ 36.093547] ? __handle_mm_fault+0x5ba/0x38c0 [ 36.098012] ? __pmd_alloc+0x4e0/0x4e0 [ 36.101869] ? trace_hardirqs_off+0x10/0x10 [ 36.106161] ? __fd_install+0x25f/0x740 [ 36.110106] ? find_held_lock+0x35/0x1d0 [ 36.114145] ? check_same_owner+0x320/0x320 [ 36.118438] ? rcu_note_context_switch+0x710/0x710 [ 36.123334] ? __do_page_fault+0x5f7/0xc90 [ 36.127540] ? vhost_net_stop_vq+0xf0/0xf0 [ 36.13...

...ister_driver+0x2a/0x90 [ 468.542476] [<ffffffffa004bd87>] virtio_pci_driver_exit+0x10/0x289 [virtio_pci] [ 468.545189] [<ffffffff811b5230>] SyS_delete_module+0x260/0x2b0 [ 468.547626] [<ffffffff811b4fd0>] ? free_module+0x450/0x450 [ 468.550061] [<ffffffff8115314d>] ? trace_hardirqs_off+0xd/0x10 [ 468.552521] [<ffffffff81157866>] ? trace_hardirqs_on_caller+0x16/0x290 [ 468.555142] [<ffffffff81c4f972>] entry_SYSCALL_64_fastpath+0x12/0x76 [ 468.557138] Memory state around the buggy address: [ 468.558965] ffff880035364400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb...

...ister_driver+0x2a/0x90 [ 468.542476] [<ffffffffa004bd87>] virtio_pci_driver_exit+0x10/0x289 [virtio_pci] [ 468.545189] [<ffffffff811b5230>] SyS_delete_module+0x260/0x2b0 [ 468.547626] [<ffffffff811b4fd0>] ? free_module+0x450/0x450 [ 468.550061] [<ffffffff8115314d>] ? trace_hardirqs_off+0xd/0x10 [ 468.552521] [<ffffffff81157866>] ? trace_hardirqs_on_caller+0x16/0x290 [ 468.555142] [<ffffffff81c4f972>] entry_SYSCALL_64_fastpath+0x12/0x76 [ 468.557138] Memory state around the buggy address: [ 468.558965] ffff880035364400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb...

...90 > [ 468.542476] [<ffffffffa004bd87>] virtio_pci_driver_exit+0x10/0x289 [virtio_pci] > [ 468.545189] [<ffffffff811b5230>] SyS_delete_module+0x260/0x2b0 > [ 468.547626] [<ffffffff811b4fd0>] ? free_module+0x450/0x450 > [ 468.550061] [<ffffffff8115314d>] ? trace_hardirqs_off+0xd/0x10 > [ 468.552521] [<ffffffff81157866>] ? trace_hardirqs_on_caller+0x16/0x290 > [ 468.555142] [<ffffffff81c4f972>] entry_SYSCALL_64_fastpath+0x12/0x76 > [ 468.557138] Memory state around the buggy address: > [ 468.558965] ffff880035364400: fb fb fb fb fb fb fb f...

This series was originally motivated by a deadlock, introduced in commit 1d7c71a3e2f77336df536855b0efd2dc5bdeb41b 'drm/nouveau/disp: port vblank handling to event interface', due to inverted lock order between nouveau_drm_vblank_enable() and nouveau_drm_vblank_handler() (the complete lockdep report is included in the patch 4/5 changelog). Because this series fixes the vblank event

Setting the IF flag can cause an VM exit. So we should avoid touching the IF flag until absolutely necessary. This patch change the way the paravirt arch_local_irq_restore() works by checking the previous flags value and call arch_local_irq_enable() only if the IF flag was set previously. On a 32 vCPUs KVM guest running the AIM7 five-sec workload, the performance increased slightly from 302136.32

Setting the IF flag can cause an VM exit. So we should avoid touching the IF flag until absolutely necessary. This patch change the way the paravirt arch_local_irq_restore() works by checking the previous flags value and call arch_local_irq_enable() only if the IF flag was set previously. On a 32 vCPUs KVM guest running the AIM7 five-sec workload, the performance increased slightly from 302136.32