linux-kernel - [PATCH] [138/139] Revert drm/radeon/kms: properly compute group

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue,  1 Feb 2011 16:45:38 -0800 (PST)
From:	Andi Kleen <andi@...stfloor.org>
To:	sconklin@...onical.com, linux-kernel@...r.kernel.org,
	stable@...nel.org
Subject: [PATCH] [138/139] Revert drm/radeon/kms: properly compute group_size on 6xx/7xx

2.6.35-longterm review patch.  If anyone has any objections, please let me know.

------------------

From: Steve Conklin <sconklin@...onical.com>

We discovered a regression for Radeon users in our latest proposed
kernel for 2.6.35 (Maverick), and have isolated it to this patch:

http://git.kernel.org/?p=linux/kernel/git/longterm/linux-2.6.35.y.git;a=commit;h=b8e9a4a45f8427837f4dba89
+bda4d4e3f3a5c726

We took that patch as part of 2.6.35.10, and one of our testers has
reported that our build of that kernel also exhibits the problem.

These are mainline kernels built with the Ubuntu configs.
http://kernel.ubuntu.com/~kernel-ppa/mainline/v2.6.35.10-maverick/

Our bug report is here:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/703553

Upstream bug report:

https://bugzilla.kernel.org/show_bug.cgi?id=24802

Index: linux-2.6.35.y/drivers/gpu/drm/radeon/r600.c
===================================================================
--- linux-2.6.35.y.orig/drivers/gpu/drm/radeon/r600.c
+++ linux-2.6.35.y/drivers/gpu/drm/radeon/r600.c
@@ -1597,11 +1597,8 @@ void r600_gpu_init(struct radeon_device 
 	rdev->config.r600.tiling_npipes = rdev->config.r600.max_tile_pipes;
 	rdev->config.r600.tiling_nbanks = 4 << ((ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT);
 	tiling_config |= BANK_TILING((ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT);
-	tiling_config |= GROUP_SIZE((ramcfg & BURSTLENGTH_MASK) >> BURSTLENGTH_SHIFT);
-	if ((ramcfg & BURSTLENGTH_MASK) >> BURSTLENGTH_SHIFT)
-		rdev->config.r600.tiling_group_size = 512;
-	else
-		rdev->config.r600.tiling_group_size = 256;
+	tiling_config |= GROUP_SIZE(0);
+	rdev->config.r600.tiling_group_size = 256;
 	tmp = (ramcfg & NOOFROWS_MASK) >> NOOFROWS_SHIFT;
 	if (tmp > 3) {
 		tiling_config |= ROW_TILING(3);
Index: linux-2.6.35.y/drivers/gpu/drm/radeon/rv770.c
===================================================================
--- linux-2.6.35.y.orig/drivers/gpu/drm/radeon/rv770.c
+++ linux-2.6.35.y/drivers/gpu/drm/radeon/rv770.c
@@ -627,11 +627,10 @@ static void rv770_gpu_init(struct radeon
 	else
 		gb_tiling_config |= BANK_TILING((mc_arb_ramcfg & NOOFBANK_MASK) >> NOOFBANK_SHIFT);
 	rdev->config.rv770.tiling_nbanks = 4 << ((gb_tiling_config >> 4) & 0x3);
-	gb_tiling_config |= GROUP_SIZE((mc_arb_ramcfg & BURSTLENGTH_MASK) >> BURSTLENGTH_SHIFT);
-	if ((mc_arb_ramcfg & BURSTLENGTH_MASK) >> BURSTLENGTH_SHIFT)
-		rdev->config.rv770.tiling_group_size = 512;
-	else
-		rdev->config.rv770.tiling_group_size = 256;
+
+	gb_tiling_config |= GROUP_SIZE(0);
+	rdev->config.rv770.tiling_group_size = 256;
+
 	if (((mc_arb_ramcfg & NOOFROWS_MASK) >> NOOFROWS_SHIFT) > 3) {
 		gb_tiling_config |= ROW_TILING(3);
 		gb_tiling_config |= SAMPLE_SPLIT(3);
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/