<html>
<head>
<base href="https://bugs.freedesktop.org/">
</head>
<body><span class="vcard"><a class="email" href="mailto:tele42k3@hotmail.com" title="tele42k3@hotmail.com">tele42k3@hotmail.com</a>
</span> changed
<a class="bz_bug_link
bz_status_NEW "
title="NEW - obs-studio + latest mesa on amdgpu/vega64 leaks kernel memory rapidly"
href="https://bugs.freedesktop.org/show_bug.cgi?id=111527">bug 111527</a>
<br>
<table border="1" cellspacing="0" cellpadding="8">
<tr>
<th>What</th>
<th>Removed</th>
<th>Added</th>
</tr>
<tr>
<td style="text-align:right;">CC</td>
<td>
</td>
<td>tele42k3@hotmail.com
</td>
</tr></table>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - obs-studio + latest mesa on amdgpu/vega64 leaks kernel memory rapidly"
href="https://bugs.freedesktop.org/show_bug.cgi?id=111527#c2">Comment # 2</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - obs-studio + latest mesa on amdgpu/vega64 leaks kernel memory rapidly"
href="https://bugs.freedesktop.org/show_bug.cgi?id=111527">bug 111527</a>
from <span class="vcard"><a class="email" href="mailto:tele42k3@hotmail.com" title="tele42k3@hotmail.com">tele42k3@hotmail.com</a>
</span></b>
<pre>Thanks for the clear steps to reproduce this issue. I managed to reproduce this
on my RX 480 and it bisected to:
commit 11a3679e3aba3524cf987f1f808d92c25f16e080
Author: Michel Dänzer <<a href="mailto:michel.daenzer@amd.com">michel.daenzer@amd.com</a>>
Date: Fri Jun 28 18:35:56 2019 +0200
winsys/amdgpu: Make KMS handles valid for original DRM file descriptor
Getting a DMA-buf fd and converting that to a handle using our duplicate
of that file descriptor (getting at which requires passing a
radeon_winsys pointer to the buffer_get_handle hook) makes sure of this,
since duplicated file descriptors reference the same file description
and therefore the same GEM handle namespace.
This is necessary because libdrm_amdgpu may use a different DRM file
descriptor with a separate handle namespace internally, e.g. because it
always reuses any existing amdgpu_device_handle for the same device.
amdgpu_bo_export returns a handle which is valid for that internal
file descriptor.
Bugzilla: <a href="https://bugs.freedesktop.org/110903">https://bugs.freedesktop.org/110903</a>
Reviewed-by: Marek Olšák <<a href="mailto:marek.olsak@amd.com">marek.olsak@amd.com</a>>
Tested-by: Pierre-Eric Pelloux-Prayer <<a href="mailto:pierre-eric.pelloux-prayer@amd.com">pierre-eric.pelloux-prayer@amd.com</a>>
While testing I saw a .8 to 1 MB/s slow leak which appeared immediately on
opening OBS with the test scene. It felt like it consistently hit some obscured
value like 64MB before the major memory leak started, which helped bisect the
issue.
I reverted the commit on top of f8887909c6683986990474b61afd6d4335a69e41 with
good results.</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>