<html>
    <head>
      <base href="https://bugs.freedesktop.org/">
    </head>
    <body>
      <p>
        <div>
            <b><a class="bz_bug_link 
          bz_status_NEEDINFO "
   title="NEEDINFO - AMD Xorg start failes with non-4K page sizes"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=106258#c21">Comment # 21</a>
              on <a class="bz_bug_link 
          bz_status_NEEDINFO "
   title="NEEDINFO - AMD Xorg start failes with non-4K page sizes"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=106258">bug 106258</a>
              from <span class="vcard"><a class="email" href="mailto:foxbat@ruin.net" title="foxbat@ruin.net">foxbat@ruin.net</a>
</span></b>
        <pre>I am still getting amdgpu crashes on POWER9 (ppc64le) using the latest AMD
firmware and on kernel 4.17 (release). My card is a WX5100 - Polaris 10. I can
reliably reproduce this by having spectacle attempt to capture an area of the
screen. However I also experience crashes randomly while moving a window.
Here is the dmesg output:

[   75.701274] CPU: 37 PID: 3391 Comm: spectacle Not tainted 4.17.0-foxbat-64k
#1
[   75.701275] NIP:  c00800001748cd08 LR: c008000017495944 CTR:
0000000000000000
[   75.701278] REGS: c000001f8ea53230 TRAP: 0700   Not tainted 
(4.17.0-foxbat-64k)
[   75.701278] MSR:  9000000000029033 <SF,HV,EE,ME,IR,DR,RI,LE>  CR: 24828848 
XER: 20040000
[   75.701284] CFAR: c00800001748c724 SOFTE: 0
               GPR00: c008000017495944 c000001f8ea534b0 c0080000176e6400
c000201cb2303418
               GPR04: c000201ca9cfce60 0000000009fff138 0000000000000100
00000000000005a3
               GPR08: 3817b594461fff5f 0000000000100000 0000000000000000
c008000017649dd0
               GPR12: 0000000000008000 c000001ffffd5600 c000201cb2300000
c000001fe89d9d20
               GPR16: 00000000027ffc4e c000201ca9cfce60 00000000fffe2000
ffffffffffffffef
               GPR20: 0000000009fff138 0000000000000100 c000201cb23066a8
0000000000000000
               GPR24: 00000000027ffc4e 00000000000e5fff c000201cb2303418
00000001fe000000
               GPR28: c000201cb2300000 0000000000000000 c000201ca9cfce60
0000000009fff138
[   75.701323] NIP [c00800001748cd08] amdgpu_sa_bo_new+0x640/0x6c0 [amdgpu]
[   75.701347] LR [c008000017495944] amdgpu_ib_get+0x8c/0x120 [amdgpu]
[   75.701348] Call Trace:
[   75.701370] [c000001f8ea534b0] [c00800001748c7f8]
amdgpu_sa_bo_new+0x130/0x6c0 [amdgpu] (unreliable)
[   75.701395] [c000001f8ea53710] [c008000017495944] amdgpu_ib_get+0x8c/0x120
[amdgpu]
[   75.701421] [c000001f8ea53790] [c008000017572558]
amdgpu_job_alloc_with_ib+0x90/0x110 [amdgpu]
[   75.701442] [c000001f8ea537d0] [c008000017492124]
amdgpu_vm_bo_update_mapping+0x35c/0x480 [amdgpu]
[   75.701469] [c000001f8ea538c0] [c0080000174925c8]
amdgpu_vm_bo_update+0x380/0x740 [amdgpu]
[   75.701489] [c000001f8ea539d0] [c008000017476d10]
amdgpu_gem_va_ioctl+0x5f8/0x620 [amdgpu]
[   75.701496] [c000001f8ea53b20] [c008000015348ce8]
drm_ioctl_kernel+0xa0/0x140 [drm]
[   75.701502] [c000001f8ea53b70] [c0080000153491b4] drm_ioctl+0x1ac/0x4d0
[drm]
[   75.701518] [c000001f8ea53cb0] [c008000017450078] amdgpu_drm_ioctl+0x70/0xd0
[amdgpu]
[   75.701521] [c000001f8ea53d00] [c0000000003e216c] do_vfs_ioctl+0xdc/0x8a0
[   75.701523] [c000001f8ea53da0] [c0000000003e2a34] ksys_ioctl+0x104/0x120
[   75.701525] [c000001f8ea53df0] [c0000000003e2a90] sys_ioctl+0x40/0xa0
[   75.701528] [c000001f8ea53e30] [c00000000000b9e0] system_call+0x58/0x6c
[   75.701529] Instruction dump:
[   75.701531] 7fc3f378 38210260 e8010010 81810008 ea21ff88 ea81ffa0 eaa1ffa8
eb41ffd0
[   75.701535] ebc1fff0 7c0803a6 7d908120 4e800020 <0fe00000> 3bc0ffea 7fc3f378
38210260
[   75.701540] ---[ end trace 838930e3a806a76d ]---
[   75.701557] amdgpu 0000:01:00.0: failed to get a new IB (-22)
[   75.701637] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA
(-22)
[   75.702847] amdgpu 0000:01:00.0: failed to get a new IB (-22)
[   75.736208] amdgpu 0000:01:00.0: failed to get a new IB (-22)
[   75.741527] amdgpu 0000:01:00.0: failed to get a new IB (-22)
[   75.741602] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA
(-22)
[   75.743120] amdgpu 0000:01:00.0: failed to get a new IB (-22)
[   75.743182] [drm:amdgpu_gem_va_ioctl [amdgpu]] *ERROR* Couldn't update BO_VA
(-22)</pre>
        </div>
      </p>


      <hr>
      <span>You are receiving this mail because:</span>

      <ul>
          <li>You are the assignee for the bug.</li>
      </ul>
    </body>
</html>