<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
On 23.02.2012 18:00, Jerome Glisse wrote:
<blockquote
cite="mid:CAH3drwYGubUtLhFGsq7W6+E06Av2WpVyWZsYzJbsrU_4RqVLrw@mail.gmail.com"
type="cite"><br>
<div class="gmail_quote">2012/2/23 Christian König <span
dir="ltr"><<a moz-do-not-send="true"
href="mailto:deathsimple@vodafone.de">deathsimple@vodafone.de</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
So don't confuse devs by doing so.<br>
<br>
Signed-off-by: Christian König <<a moz-do-not-send="true"
href="mailto:deathsimple@vodafone..de">deathsimple@vodafone.de</a>><br>
---<br>
drivers/gpu/drm/radeon/r600.c | 15 +--------------<br>
1 files changed, 1 insertions(+), 14 deletions(-)<br>
<br>
diff --git a/drivers/gpu/drm/radeon/r600.c
b/drivers/gpu/drm/radeon/r600.c<br>
index 4f08e5e..4a4ac8f 100644<br>
--- a/drivers/gpu/drm/radeon/r600.c<br>
+++ b/drivers/gpu/drm/radeon/r600.c<br>
@@ -2719,20 +2719,7 @@ int r600_ib_test(struct radeon_device
*rdev, int ring)<br>
ib->ptr[0] = PACKET3(PACKET3_SET_CONFIG_REG, 1);<br>
ib->ptr[1] = ((scratch -
PACKET3_SET_CONFIG_REG_OFFSET) >> 2);<br>
ib->ptr[2] = 0xDEADBEEF;<br>
- ib->ptr[3] = PACKET2(0);<br>
- ib->ptr[4] = PACKET2(0);<br>
- ib->ptr[5] = PACKET2(0);<br>
- ib->ptr[6] = PACKET2(0);<br>
- ib->ptr[7] = PACKET2(0);<br>
- ib->ptr[8] = PACKET2(0);<br>
- ib->ptr[9] = PACKET2(0);<br>
- ib->ptr[10] = PACKET2(0);<br>
- ib->ptr[11] = PACKET2(0);<br>
- ib->ptr[12] = PACKET2(0);<br>
- ib->ptr[13] = PACKET2(0);<br>
- ib->ptr[14] = PACKET2(0);<br>
- ib->ptr[15] = PACKET2(0);<br>
- ib->length_dw = 16;<br>
+ ib->length_dw = 3;<br>
r = radeon_ib_schedule(rdev, ib);<br>
if (r) {<br>
radeon_scratch_free(rdev, scratch);<br>
<span class="HOEnZb"><font color="#888888">--<br>
1.7.5.4</font></span><br>
</blockquote>
</div>
<br>
You sure about that ? I remember this helped with GPU lockup and i
also seen fglrx aligning IB.<br>
</blockquote>
Yeah, pretty much. Well I searched for halve an hour for the
corresponding IB alignment in mesa/the CS ioctl until I finally
figured out that there isn't any. <br>
<br>
So IBs submitted by usermode aren't aligned in any way.... So it
really seems to work fine and I couldn't find any reason why we
should align an IB for the GFX ring in our docs also.<br>
<br>
Christian.<br>
</body>
</html>