[PATCH] agp/uninorth: Fix lockups with radeon KMS and >1x.

Subsystems: agpgart driver, the rest

STALE5573d

2 messages, 2 authors, 2011-05-20 · open the first message on its own page

[PATCH] agp/uninorth: Fix lockups with radeon KMS and >1x.

From: Michel Dänzer <hidden>
Date: 2011-05-19 14:08:54

From: Michel Dänzer <redacted>

This was based on a description by Ben Herrenschmidt:
I've removed that SBA reset from the normal TLB invalidation path and
left it only once after turning AGP on.
About six months ago, he said:
I did it a bit differently, but yeah, you get the idea. I'm doing a
patch series so don't bother pushing things too hard yet.
But I haven't seen anything from him about this since then, and people are
regularly hitting these lockups, so here we are...

Signed-off-by: Michel Dänzer <redacted>
---
 drivers/char/agp/uninorth-agp.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/char/agp/uninorth-agp.c b/drivers/char/agp/uninorth-agp.c
index 47c2218..55af723 100644
--- a/drivers/char/agp/uninorth-agp.c
+++ b/drivers/char/agp/uninorth-agp.c
@@ -80,7 +80,7 @@ static void uninorth_tlbflush(struct agp_memory *mem)
 			       ctrl | UNI_N_CFG_GART_INVAL);
 	pci_write_config_dword(agp_bridge->dev, UNI_N_CFG_GART_CTRL, ctrl);
 
-	if (uninorth_rev <= 0x30) {
+	if (!mem && uninorth_rev <= 0x30) {
 		pci_write_config_dword(agp_bridge->dev, UNI_N_CFG_GART_CTRL,
 				       ctrl | UNI_N_CFG_GART_2xRESET);
 		pci_write_config_dword(agp_bridge->dev, UNI_N_CFG_GART_CTRL,
-- 
1.7.5.1

Re: [PATCH] agp/uninorth: Fix lockups with radeon KMS and >1x.

From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: 2011-05-20 01:10:33

On Thu, 2011-05-19 at 16:08 +0200, Michel Dänzer wrote:
From: Michel Dänzer <redacted>

This was based on a description by Ben Herrenschmidt:
quoted
I've removed that SBA reset from the normal TLB invalidation path and
left it only once after turning AGP on.
About six months ago, he said:
quoted
I did it a bit differently, but yeah, you get the idea. I'm doing a
patch series so don't bother pushing things too hard yet.
But I haven't seen anything from him about this since then, and people are
regularly hitting these lockups, so here we are...

Signed-off-by: Michel Dänzer <redacted>
Oops. I do have a pile of patches, but I never got something "stable"
enough and got distracted by more important stuff. Dave, please merge
this for now.

Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

Thanks !

Cheers,
Ben.
quoted hunk
---
 drivers/char/agp/uninorth-agp.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/char/agp/uninorth-agp.c b/drivers/char/agp/uninorth-agp.c
index 47c2218..55af723 100644
--- a/drivers/char/agp/uninorth-agp.c
+++ b/drivers/char/agp/uninorth-agp.c
@@ -80,7 +80,7 @@ static void uninorth_tlbflush(struct agp_memory *mem)
 			       ctrl | UNI_N_CFG_GART_INVAL);
 	pci_write_config_dword(agp_bridge->dev, UNI_N_CFG_GART_CTRL, ctrl);
 
-	if (uninorth_rev <= 0x30) {
+	if (!mem && uninorth_rev <= 0x30) {
 		pci_write_config_dword(agp_bridge->dev, UNI_N_CFG_GART_CTRL,
 				       ctrl | UNI_N_CFG_GART_2xRESET);
 		pci_write_config_dword(agp_bridge->dev, UNI_N_CFG_GART_CTRL,
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help