[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#775264: mesa: Please apply patch from freedesktop bug #85647



Source: mesa
Version: 10.3.2-1
Severity: important
Tags: patch

I have had problems with my TAHITI radeon videocard for the last year or so,
where it would lock up randomly between 10 minutes to a few hours from booting.
So much so that I have needed to switch to fglrx to get my system reliable.

This problem apparently only manifests on certain chipsets, like the TAHITI
variant I have and not all radeonsi cards out there.

I have compiled the current mesa 10.3.2-1 which is the latest available in
jessie and sid, with a patch posted in freedesktop bug #85647, and my system is
stable in the 3 weeks I'm running this custom version.

This patch is already included in mesa 10.3.4, but since a new release has not
been forthcoming in the last weeks and the current version must hurt others out
there. I'm filing this bug report to nudge the maintainer to either include
this patch in 10.3.2, or upgrade to 10.3.4.

bug: https://bugs.freedesktop.org/show_bug.cgi?id=85647
patch: https://bugs.freedesktop.org/attachment.cgi?id=109261



-- System Information:
Debian Release: 8.0
  APT prefers testing
  APT policy: (750, 'testing'), (700, 'stable'), (600, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.16.0-4-amd64 (SMP w/8 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
diff --git a/src/gallium/drivers/radeonsi/si_dma.c b/src/gallium/drivers/radeonsi/si_dma.c
index b1bd5e7..3da2d26 100644
--- a/src/gallium/drivers/radeonsi/si_dma.c
+++ b/src/gallium/drivers/radeonsi/si_dma.c
@@ -250,6 +250,9 @@ void si_dma_copy(struct pipe_context *ctx,
 		return;
 	}
 
+	/* XXX: Asynchronous DMA causes lockups for some */
+	goto fallback;
+
 	if (src->format != dst->format || src_box->depth > 1 ||
 	    rdst->dirty_level_mask != 0 ||
 	    rdst->cmask.size || rdst->fmask.size ||

Reply to: