From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 107898] "kfd: Failed to resume IOMMU for device 1002:15dd" on Raven Ridge Date: Thu, 20 Sep 2018 09:00:17 +0000 Message-ID: References: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============1375401090==" Return-path: Received: from culpepper.freedesktop.org (culpepper.freedesktop.org [131.252.210.165]) by gabe.freedesktop.org (Postfix) with ESMTP id 7A29F6E56A for ; Thu, 20 Sep 2018 09:00:17 +0000 (UTC) In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org --===============1375401090== Content-Type: multipart/alternative; boundary="15374340171.9d9b9.15100" Content-Transfer-Encoding: 7bit --15374340171.9d9b9.15100 Date: Thu, 20 Sep 2018 09:00:17 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated https://bugs.freedesktop.org/show_bug.cgi?id=3D107898 --- Comment #9 from Marvin Damschen --- ROCm 1.9 runs OpenCL on GPU on top of mainline kfd and seems stable. Howeve= r: - CPU is not detected as a compute device (rocminfo attached) - Performance, at least in darktable, is quite low (the "bench.SRW" benchma= rk in OpenCL on GPU takes more than 3 times longer than on CPU without OpenCL). The problem could be that memory buffers are too small, clinfo reports: "Max memory allocation 268435456 (256MiB)" which seems quite small to me (?). Are these problems a result of incorrect information in CRAT? Best regards Marvin --=20 You are receiving this mail because: You are the assignee for the bug.= --15374340171.9d9b9.15100 Date: Thu, 20 Sep 2018 09:00:17 +0000 MIME-Version: 1.0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated

Commen= t # 9 on bug 10789= 8 from Marvin Damschen 
ROCm 1.9 runs OpenCL on GPU on top of mainline kfd and seems s=
table. However:
- CPU is not detected as a compute device (rocminfo attached)
- Performance, at least in darktable, is quite low (the "bench.SRW&quo=
t; benchmark
in OpenCL on GPU takes more than 3 times longer than on CPU without OpenCL).
The problem could be that memory buffers are too small, clinfo reports:
"Max memory allocation                           268435456 (256MiB)&qu=
ot;
which seems quite small to me (?).

Are these problems a result of incorrect information in CRAT?

Best regards
Marvin

You are receiving this mail because:
  • You are the assignee for the bug.
= --15374340171.9d9b9.15100-- --===============1375401090== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZHJpLWRldmVs IG1haWxpbmcgbGlzdApkcmktZGV2ZWxAbGlzdHMuZnJlZWRlc2t0b3Aub3JnCmh0dHBzOi8vbGlz dHMuZnJlZWRlc2t0b3Aub3JnL21haWxtYW4vbGlzdGluZm8vZHJpLWRldmVsCg== --===============1375401090==--