linux-kernel - [119/129] drm/radeon/kms/evergreen: fix gpu hangs in userspace accel code

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: Windows password security audit tool. GUI, reports in PDF.

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

Date:	Sat, 18 Sep 2010 12:13:37 -0700
From:	Greg KH <gregkh@...e.de>
To:	linux-kernel@...r.kernel.org, stable@...nel.org
Cc:	stable-review@...nel.org, torvalds@...ux-foundation.org,
	akpm@...ux-foundation.org, alan@...rguk.ukuu.org.uk,
	Alex Deucher <alexdeucher@...il.com>,
	Dave Airlie <airlied@...hat.com>
Subject: [119/129] drm/radeon/kms/evergreen: fix gpu hangs in userspace accel code

2.6.35-stable review patch.  If anyone has any objections, please let us know.

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

From: Alex Deucher <alexdeucher@...il.com>

commit 7e7b41d2ff30ed7ad4bf401d18566e6f38e42e4f upstream.

These VGT regs need to be programmed via the ring rather than
MMIO as on previous asics (r6xx/r7xx).

Signed-off-by: Alex Deucher <alexdeucher@...il.com>
Signed-off-by: Dave Airlie <airlied@...hat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@...e.de>

---
 drivers/gpu/drm/radeon/evergreen.c |   39 ++++++++++++++++++++++++++++++++++++-
 drivers/gpu/drm/radeon/r600.c      |    5 ----
 2 files changed, 39 insertions(+), 5 deletions(-)

--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -658,6 +658,43 @@ static int evergreen_cp_load_microcode(s
 	return 0;
 }
 
+static int evergreen_cp_start(struct radeon_device *rdev)
+{
+	int r;
+	uint32_t cp_me;
+
+	r = radeon_ring_lock(rdev, 7);
+	if (r) {
+		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
+		return r;
+	}
+	radeon_ring_write(rdev, PACKET3(PACKET3_ME_INITIALIZE, 5));
+	radeon_ring_write(rdev, 0x1);
+	radeon_ring_write(rdev, 0x0);
+	radeon_ring_write(rdev, rdev->config.evergreen.max_hw_contexts - 1);
+	radeon_ring_write(rdev, PACKET3_ME_INITIALIZE_DEVICE_ID(1));
+	radeon_ring_write(rdev, 0);
+	radeon_ring_write(rdev, 0);
+	radeon_ring_unlock_commit(rdev);
+
+	cp_me = 0xff;
+	WREG32(CP_ME_CNTL, cp_me);
+
+	r = radeon_ring_lock(rdev, 4);
+	if (r) {
+		DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r);
+		return r;
+	}
+	/* init some VGT regs */
+	radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2));
+	radeon_ring_write(rdev, (VGT_VERTEX_REUSE_BLOCK_CNTL - PACKET3_SET_CONTEXT_REG_START) >> 2);
+	radeon_ring_write(rdev, 0xe);
+	radeon_ring_write(rdev, 0x10);
+	radeon_ring_unlock_commit(rdev);
+
+	return 0;
+}
+
 int evergreen_cp_resume(struct radeon_device *rdev)
 {
 	u32 tmp;
@@ -702,7 +739,7 @@ int evergreen_cp_resume(struct radeon_de
 	rdev->cp.rptr = RREG32(CP_RB_RPTR);
 	rdev->cp.wptr = RREG32(CP_RB_WPTR);
 
-	r600_cp_start(rdev);
+	evergreen_cp_start(rdev);
 	rdev->cp.ready = true;
 	r = radeon_ring_test(rdev);
 	if (r) {
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -2104,10 +2104,7 @@ int r600_cp_start(struct radeon_device *
 	}
 	radeon_ring_write(rdev, PACKET3(PACKET3_ME_INITIALIZE, 5));
 	radeon_ring_write(rdev, 0x1);
-	if (rdev->family >= CHIP_CEDAR) {
-		radeon_ring_write(rdev, 0x0);
-		radeon_ring_write(rdev, rdev->config.evergreen.max_hw_contexts - 1);
-	} else if (rdev->family >= CHIP_RV770) {
+	if (rdev->family >= CHIP_RV770) {
 		radeon_ring_write(rdev, 0x0);
 		radeon_ring_write(rdev, rdev->config.rv770.max_hw_contexts - 1);
 	} else {


--
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/