<html>
<head>
<base href="https://bugs.freedesktop.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - [radeon][TTM] Contention when evicting large buffers between VRAM and GTT"
href="https://bugs.freedesktop.org/show_bug.cgi?id=92775#c2">Comment # 2</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - [radeon][TTM] Contention when evicting large buffers between VRAM and GTT"
href="https://bugs.freedesktop.org/show_bug.cgi?id=92775">bug 92775</a>
from <span class="vcard"><a class="email" href="mailto:david1.zhou@amd.com" title="david1.zhou@amd.com <david1.zhou@amd.com>"> <span class="fn">david1.zhou@amd.com</span></a>
</span></b>
<pre>(In reply to Michel Dänzer from <a href="show_bug.cgi?id=92775#c1">comment #1</a>)
<span class="quote">> (In reply to Shawn Starr from <a href="show_bug.cgi?id=92775#c0">comment #0</a>)
> > One solution discussed is to split up the transfer into smaller chunks in
> > radeon_ttm.
>
> Specifically, here's how I think a fallback could be implemented in the
> kernel driver which can never fail because of fragmentation or resource
> starvation:
>
> During initialization, reserve some pinned GTT memory for bounce buffers.
> When a BO can't be bound to GTT for eviction as in the case reported here,
> instead do the eviction directly from VRAM to CPU domain in one or several
> passes of:
> 1. Copy part of the BO from VRAM to one of the reserved bounce buffers in
> GTT using the GPU.
> 2. Copy that part of the BO from the bounce buffer to the BO's system RAM
> pages using the CPU.</span >
if we can split a large BO to two parts, one is in VRAM, one is in GTT, seems
also to be helpful for this case.</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>