On Tue, Nov 30, 2021 at 11:19 AM Ignat Korchagin <ignat@xxxxxxxxxxxxxx> wrote: > > On Tue, Nov 30, 2021 at 11:11 AM Paolo Bonzini <pbonzini@xxxxxxxxxx> wrote: > > > > On 11/30/21 11:58, Ignat Korchagin wrote: > > > I have managed to reliably reproduce the issue on a QEMU VM (on a host > > > with nested virtualisation enabled). Here are the steps: > > > > > > 1. Install gvisor as per > > > https://gvisor.dev/docs/user_guide/install/#install-latest > > > 2. Run > > > $ for i in $(seq 1 100); do sudo runsc --platform=kvm --network=none > > > do echo ok; done > > > > > > I've tried to recompile the kernel with the above patch, but > > > unfortunately it does fix the issue. I'm happy to try other > > > patches/fixes queued for 5.16-rc4 > > > > You can find them already in the "for-linus" tag of kvm.git as well as > > in the master branch, but there isn't much else. > > > > Paolo > > Thanks. I've tried to compile the kernel from kvm.git "for-linus" tag, > but the issue is still there, so probably no commits address the > problem. > Will keep digging. > > Ignat I have also noticed another new warning, when running this on the kernel from kvm.git branch: [ 70.284354][ T2928] WARNING: CPU: 4 PID: 2928 at arch/x86/kvm/x86.c:9886 kvm_arch_vcpu_ioctl_run+0x126c/0x17d0 [ 70.284354][ T2928] Modules linked in: [ 70.284354][ T2928] CPU: 4 PID: 2928 Comm: exe Not tainted 5.16.0-rc2 #2 [ 70.284354][ T2928] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015 [ 70.284354][ T2928] RIP: 0010:kvm_arch_vcpu_ioctl_run+0x126c/0x17d0 [ 70.284354][ T2928] Code: 49 89 b7 f8 01 00 00 e9 8e ee ff ff 49 8b 87 80 00 00 00 45 31 e4 c7 40 08 07 00 00 00 49 83 87 b8 20 00 00 01 e9 35 f2 ff ff <0f> 0b 4c 89 ff e8 ea 72 03 00 83 f8 01 41 89 c4 0f 85 47 f9 ff ff [ 70.284354][ T2928] RSP: 0018:ffffb09fc0653d60 EFLAGS: 00010002 [ 70.284354][ T2928] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff9d9083929cc0 [ 70.284354][ T2928] RDX: ffff9d9083929c01 RSI: ffffffff92f2e509 RDI: ffffffff92e8010e [ 70.284354][ T2928] RBP: ffffb09fc0653df0 R08: 0000000000000000 R09: ffffb09fc052c340 [ 70.284354][ T2928] R10: ffff9d91fffde000 R11: 0000000000034800 R12: 0000000000000000 [ 70.284354][ T2928] R13: ffffb09fc052c440 R14: ffff9d90839fc038 R15: ffff9d90839fc000 [ 70.284354][ T2928] FS: 0000000001cc6c30(0000) GS:ffff9d91f7d00000(0000) knlGS:0000000000000000 [ 70.284354][ T2928] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 70.284354][ T2928] CR2: 000000c000316000 CR3: 0000000102b4c006 CR4: 0000000000172ee0 [ 70.284354][ T2928] Call Trace: [ 70.284354][ T2928] <TASK> [ 70.284354][ T2928] ? memcg_slab_free_hook+0xcc/0x190 [ 70.284354][ T2928] ? kmem_cache_free+0x264/0x2b0 [ 70.284354][ T2928] kvm_vcpu_ioctl+0x274/0x680 [ 70.284354][ T2928] ? _raw_spin_lock_irq+0x14/0x2f [ 70.284354][ T2928] ? _raw_spin_unlock_irq+0x13/0x30 [ 70.284354][ T2928] ? signal_setup_done+0xe9/0x160 [ 70.284354][ T2928] ? fpregs_mark_activate+0x32/0x90 [ 70.284354][ T2928] ? arch_do_signal_or_restart+0x525/0x6b0 [ 70.284354][ T2928] __x64_sys_ioctl+0x40a/0x950 [ 70.284354][ T2928] do_syscall_64+0x3b/0x90 [ 70.284354][ T2928] entry_SYSCALL_64_after_hwframe+0x44/0xae [ 70.284354][ T2928] RIP: 0033:0x489516 [ 70.284354][ T2928] Code: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc 48 8b 7c 24 10 48 8b 74 24 18 48 8b 54 24 20 48 8b 44 24 08 0f 05 <48> 3d 01 f0 ff ff 76 1b 48 c7 44 24 28 ff ff ff ff 48 c7 44 24 30 [ 70.284354][ T2928] RSP: 002b:000000c000009a10 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [ 70.284354][ T2928] RAX: ffffffffffffffda RBX: 000000c0002fa480 RCX: 0000000000489516 [ 70.284354][ T2928] RDX: 0000000000000000 RSI: 000000000000ae80 RDI: 0000000000000008 [ 70.284354][ T2928] RBP: 000000c000009aa0 R08: 0000000000000001 R09: 0000000000000000 [ 70.284354][ T2928] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 [ 70.639977][ T2928] R13: 0000000000000000 R14: 000000000142fb48 R15: 0000000000000000 [ 70.639977][ T2928] </TASK> [ 70.639977][ T2928] ---[ end trace a3a88c91ba4a4df8 ]--- Ignat
