unable to handle kernel paging request at ffffb1f144f52090

Hello,

I’m confused by these errors:

[ 1719.653554] BUG: unable to handle kernel paging request at ffffb24b8479d000
[ 1719.653569] IP: [<ffffffffc0263ee8>] os_unlock_user_pages+0x28/0x60 [nvidia]
[ 1719.654834] PGD 17fd77067 PUD 17fd78067 PMD 504563067 PTE 0
[ 1719.654839] Oops: 0000 [#1] SMP
[ 1719.654858] Modules linked in: xt_comment xt_mark nvidia_uvm(OE) veth nfsv3 rpcsec_gss_krb5 nfsv4 dns_resolver nfs fscache fuse ipt_MASQUERADE nf_nat_masquerade_ipv4 nf_conntrack_netlink nfnetlink iptable_nat xt_addrtype iptable_filter xt_conntrack br_netfilter bridge stp llc overlay(T) vport_vxlan vxlan ip6_udp_tunnel udp_tunnel openvswitch nf_conntrack_ipv6 nf_nat_ipv6 nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_defrag_ipv6 nf_nat nf_conntrack ppdev mlx4_ib iosf_mbi crc32_pclmul snd_hda_codec_generic ghash_clmulni_intel ib_core mlx4_en ptp pps_core aesni_intel lrw gf128mul glue_helper ablk_helper cryptd snd_hda_intel joydev snd_hda_codec parport_pc parport snd_hda_core snd_hwdep snd_seq snd_seq_device pcspkr mlx4_core sg virtio_balloon snd_pcm qxl ttm snd_timer snd devlink soundcore i2c_piix4
[ 1719.654907] nfsd auth_rpcgss nfs_acl lockd grace sunrpc ip_tables xfs libcrc32c nvidia_drm(POE) nvidia_modeset(POE) nvidia(POE) sr_mod cdrom virtio_blk ata_generic pata_acpi virtio_console drm_kms_helper crct10dif_pclmul crct10dif_common crc32c_intel syscopyarea sysfillrect sysimgblt fb_sys_fops drm floppy ata_piix libata serio_raw drm_panel_orientation_quirks ipmi_devintf virtio_pci ipmi_msghandler virtio_ring virtio dm_mirror dm_region_hash dm_log dm_mod
[ 1719.654935] CPU: 5 PID: 10931 Comm: python Kdump: loaded Tainted: P OE ------------ T 3.10.0-957.el7.x86_64 #1
[ 1719.654937] Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011
[ 1719.654939] task: ffff9cf83a27c100 ti: ffff9cf82b37c000 task.ti: ffff9cf82b37c000
[ 1719.654941] RIP: 0010:[<ffffffffc0263ee8>] [<ffffffffc0263ee8>] os_unlock_user_pages+0x28/0x60 [nvidia]
[ 1719.655012] RSP: 0018:ffff9cf82b37fcb0 EFLAGS: 00010216
[ 1719.655014] RAX: 00000000000001ff RBX: 0000000000000200 RCX: 0000000000000034
[ 1719.655016] RDX: 0000000000000000 RSI: fffffe7a8a3b3d40 RDI: fffffe7a8a3b3d40
[ 1719.655017] RBP: ffff9cf82b37fcd0 R08: ea00000000000000 R09: f9ea28ecf5000000
[ 1719.655019] R10: 0615d58d953b3d40 R11: 000000000000ffff R12: ffffb24b8479d000
[ 1719.655021] R13: ffffb24b8479c008 R14: 0000000000040000 R15: 0000000000000027
[ 1719.655024] FS: 00007f6ed7fff700(0000) GS:ffff9cf921d40000(0000) knlGS:0000000000000000
[ 1719.655026] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1719.655027] CR2: ffffb24b8479d000 CR3: 00000004f33f0000 CR4: 00000000003606e0
[ 1719.655037] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 1719.655039] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 1719.655041] Call Trace:
[ 1719.655255] [<ffffffffc098d4a9>] _nv000870rm+0xa9/0xe0 [nvidia]
[ 1719.655380] [<ffffffffc098dd62>] ? _nv000940rm+0x882/0x900 [nvidia]
[ 1719.655523] [<ffffffffc0998254>] ? rm_ioctl+0x54/0xb0 [nvidia]
[ 1719.655569] [<ffffffffa8bf09ff>] ? vma_set_page_prot+0x2f/0x50
[ 1719.655585] [<ffffffffa8c3e001>] ? __check_object_size+0x191/0x250
[ 1719.655649] [<ffffffffc0258859>] ? nvidia_ioctl+0x609/0x7d0 [nvidia]
[ 1719.655710] [<ffffffffc0253083>] ? nvidia_frontend_unlocked_ioctl+0x43/0x50 [nvidia]
[ 1719.655735] [<ffffffffa8c56210>] ? do_vfs_ioctl+0x3a0/0x5a0
[ 1719.655738] [<ffffffffa8c564b1>] ? SyS_ioctl+0xa1/0xc0
[ 1719.655769] [<ffffffffa9174ddb>] ? system_call_fastpath+0x22/0x27
[ 1719.655770] Code: 00 00 00 0f 1f 44 00 00 55 31 c0 48 89 e5 41 56 49 89 fe 41 55 49 89 f5 41 54 53 31 db 48 85 ff 74 22 90 4d 8d 64 c5 00 83 c3 01 <49> 8b 3c 24 e8 1f da 95 e8 49 8b 3c 24 e8 56 06 96 e8 89 d8 4c
[ 1719.655806] RIP [<ffffffffc0263ee8>] os_unlock_user_pages+0x28/0x60 [nvidia]
[ 1719.655870] RSP <ffff9cf82b37fcb0>
[ 1719.655871] CR2: ffffb24b8479d000

Host : CentOS 7.6
VM : CentOS 7.6
GPU : Tesla P40
vGPU : P40-8C

There is no support for CentOS. So what do you expect here? You should open a support ticket with the stack trace but this won’t be possible with CentOS. Try to reproduce with Redhat.