Bug 565074

Summary:	sys-kernel/gentoo-sources: radeon gpu hardlocks.
Product:	Gentoo Linux	Reporter:	DaggyStyle <daggs>
Component:	[OLD] Core system	Assignee:	Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel>
Status:	RESOLVED NEEDINFO
Severity:	major	CC:	ansla80
Priority:	Normal
Version:	unspecified
Hardware:	All
OS:	Linux
Whiteboard:
Package list:		Runtime testing required:	---
Attachments:	kernel config emerge --info

Description DaggyStyle 2015-11-07 11:30:49 UTC

I have two radeon gpus on my system, see:
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Park [Mobility Radeon HD 5430]
06:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] RV635 [Radeon HD 3650/3750/4570/4580]

the latter is not used currently.
since upgrading to sys-kernel/gentoo-sources-4.2.0 my main gpu hard locks almost every boot. the only way to fix it is to reboot the system.

see dmesg of the lock below:
Nov  6 19:59:29 NCC-5001-D kernel: radeon 0000:01:00.0: ring 0 stalled for more than 10006msec
Nov  6 19:59:29 NCC-5001-D kernel: radeon 0000:01:00.0: GPU lockup (current fence id 0x0000000000001ae7 last fence id 0x0000000000001ae8 on ring 0)
Nov  6 19:59:29 NCC-5001-D kernel: [drm:rv770_stop_dpm [radeon]] *ERROR* Could not force DPM to low.
Nov  6 19:59:29 NCC-5001-D kernel: BUG: unable to handle kernel paging request at ffffc9040141fffc
Nov  6 19:59:29 NCC-5001-D kernel: IP: [<ffffffffa0538d48>] radeon_ring_backup+0xc8/0x170 [radeon]
Nov  6 19:59:29 NCC-5001-D kernel: PGD 41d00c067 PUD 0
Nov  6 19:59:29 NCC-5001-D kernel: Oops: 0000 [#1] PREEMPT SMP
Nov  6 19:59:29 NCC-5001-D kernel: Modules linked in: nfsd auth_rpcgss oid_registry nfsv3 nfs lockd grace sunrpc it87 hwmon_vid cfg80211 xfs libcrc32c br_netfilter bridge stp ipv6 llc snd_hda_codec_realtek snd_hda_codec_generic radeon iTCO_wdt drm_kms_helper ttm drm snd_hda_codec_hdmi snd_hda_intel snd_hda_codec snd_hwdep snd_hda_core i2c_algo_bit snd_pcm snd_timer agpgart r8169 lpc_ich snd i2c_i801 mei_me mii mei mfd_core soundcore x86_pkg_temp_thermal coretemp
Nov  6 19:59:29 NCC-5001-D kernel: CPU: 1 PID: 3504 Comm: X Not tainted 4.3.0-gentoo #1
Nov  6 19:59:29 NCC-5001-D kernel: Hardware name: Gigabyte Technology Co., Ltd. Z68P-DS3/Z68P-DS3, BIOS F9 03/21/2012
Nov  6 19:59:29 NCC-5001-D kernel: task: ffff880419ce6780 ti: ffff8804115d4000 task.ti: ffff8804115d4000
Nov  6 19:59:29 NCC-5001-D kernel: RIP: 0010:[<ffffffffa0538d48>]  [<ffffffffa0538d48>] radeon_ring_backup+0xc8/0x170 [radeon]
Nov  6 19:59:29 NCC-5001-D kernel: RSP: 0018:ffff8804115d7cc8  EFLAGS: 00010202
Nov  6 19:59:29 NCC-5001-D kernel: RAX: ffffc900005dc000 RBX: 00000000ffffffff RCX: 0000000000000000
Nov  6 19:59:29 NCC-5001-D kernel: RDX: 0000000000000000 RSI: ffffc9040141fffc RDI: 00000000000d7880
Nov  6 19:59:29 NCC-5001-D kernel: RBP: ffff8804150653b8 R08: 0000000000000001 R09: ffffc900006b4000
Nov  6 19:59:29 NCC-5001-D kernel: R10: 00003ffffffff000 R11: ffffc900006b3fff R12: ffff880415065390
Nov  6 19:59:29 NCC-5001-D kernel: R13: 0000000000035e21 R14: ffff8804115d7d30 R15: ffff880415064018
Nov  6 19:59:29 NCC-5001-D kernel: FS:  00007fe8f89778c0(0000) GS:ffff88042f440000(0000) knlGS:0000000000000000
Nov  6 19:59:29 NCC-5001-D kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Nov  6 19:59:29 NCC-5001-D kernel: CR2: ffffc9040141fffc CR3: 0000000411658000 CR4: 00000000000406e0
Nov  6 19:59:29 NCC-5001-D kernel: Stack:
Nov  6 19:59:29 NCC-5001-D kernel:  ffff880415064018 ffff880415064000 ffff8804150653b8 ffff8804115d7d30
Nov  6 19:59:29 NCC-5001-D kernel:  ffff8804150653b8 0000000000000000 ffffffffa050a7d9 ffff880415064738
Nov  6 19:59:29 NCC-5001-D kernel:  0000000100089540 0000000000000000 ffff8800b7f451f0 0000000000000001
Nov  6 19:59:29 NCC-5001-D kernel: Call Trace:
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffffa050a7d9>] ? radeon_gpu_reset+0xa9/0x2e0 [radeon]
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffff813b6cfa>] ? reservation_object_wait_timeout_rcu+0x1da/0x290
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffffa0536e65>] ? radeon_gem_handle_lockup.part.3+0x5/0x10 [radeon]
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffffa02bbfe8>] ? drm_ioctl+0xf8/0x430 [drm]
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffffa0537ad0>] ? radeon_gem_busy_ioctl+0xc0/0xc0 [radeon]
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffffa0508041>] ? radeon_drm_ioctl+0x41/0x70 [radeon]
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffff811549ae>] ? do_vfs_ioctl+0x2be/0x490
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffff8115d84b>] ? __fget+0x6b/0xb0
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffff81154bb6>] ? SyS_ioctl+0x36/0x70
Nov  6 19:59:29 NCC-5001-D kernel:  [<ffffffff8153b5db>] ? entry_SYSCALL_64_fastpath+0x16/0x6e
Nov  6 19:59:29 NCC-5001-D kernel: Code: 85 c0 49 89 06 0f 84 84 00 00 00 41 8d 7d ff 31 c9 48 c1 e7 02 eb 07 49 8b 06 48 83 c1 04 48 8b 75 08 8d 53 01 89 db 48 8d 34 9e <8b> 36 89 34 08 23 55 54 48 39 f9 89 d3 75 dd 4c 89 e7 e8 31 0b
Nov  6 19:59:29 NCC-5001-D kernel: RIP  [<ffffffffa0538d48>] radeon_ring_backup+0xc8/0x170 [radeon]
Nov  6 19:59:29 NCC-5001-D kernel:  RSP <ffff8804115d7cc8>
Nov  6 19:59:29 NCC-5001-D kernel: CR2: ffffc9040141fffc
Nov  6 19:59:29 NCC-5001-D kernel: ---[ end trace 7bf1d8e3d87f944b ]---

emerge --info and kernel config is attached.
this renders my system rather useless so I hope this will get fixed fast.

thanks.

Reproducible: Always

Comment 1 DaggyStyle 2015-11-07 11:33:20 UTC

Created attachment 416236 [details]
kernel config

Comment 2 DaggyStyle 2015-11-07 11:33:44 UTC

Created attachment 416238 [details]
emerge --info

Comment 3 Mike Pagano gentoo-dev

2017-03-02 00:34:53 UTC

Is this still an issue with later kernels?