<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p>OK, according to
<a class="moz-txt-link-freetext" href="https://www.kernel.org/doc/Documentation/memory-barriers.txt">https://www.kernel.org/doc/Documentation/memory-barriers.txt</a>,
SLEEP AND WAKE-UP FUNCTIONS sections implies that.</p>
<p>Will respin the patch.<br>
</p>
<p>Andrey<br>
</p>
<br>
<div class="moz-cite-prefix">On 06/20/2018 11:02 AM, Andrey
Grodzovsky wrote:<br>
</div>
<blockquote type="cite"
cite="mid:6f9d3c60-2050-f13c-df72-bff240d669b9@amd.com">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<p>Are you referring to insert_work->smp_mb(); ?</p>
<p>Andrey<br>
</p>
<br>
<div class="moz-cite-prefix">On 06/20/2018 10:50 AM, Christian
König wrote:<br>
</div>
<blockquote type="cite"
cite="mid:d96fc78e-ef3d-fb3a-8548-7a6a7cc3d52f@gmail.com">
<blockquote type="cite" style="color: #000000;">+ * Try to
submit work so SQ_EDC_INFO can be accessed from <br>
+ * BH. If previous work submission hasn't finished yet <br>
+ * just print whatever info is possible directly from the
ISR. <br>
+ */ <br>
+ if (work_pending(&adev->gfx.sq_work.work)) { <br>
+ gfx_v8_0_parse_sq_irq(adev, ih_data); <br>
+ } else { <br>
+ adev->gfx.sq_work.ih_data = ih_data; <br>
+ /* Verify the new value visible in BH handler */ <br>
+ smp_wmb(); <br>
</blockquote>
<br>
You can drop the barrier here and in
gfx_v8_0_sq_irq_work_func(), the schedule_work() function is a
barrier itself for both reads and writes anyway. <br>
<br>
Apart from that the both patches are Acked-by: Christian König <a
class="moz-txt-link-rfc2396E"
href="mailto:christian.koenig@amd.com" moz-do-not-send="true"><christian.koenig@amd.com></a>.
<br>
<br>
Regards, <br>
Christian. </blockquote>
<br>
</blockquote>
<br>
</body>
</html>