du81692468
发件人： du81692468
发送时间： 2012-08-22 14:41
收件人： xorg
主题： help
Hello,
System Environment: fedora 14
intel driver version: xf86-video-intel-2.17.0
Graphics: onboard intel graphics + NVIDIA
Screen: 3
Unable to set a resolution of 800 × 600, start startx when there
On Wed, 22 Aug 2012 16:40:57 -0700, Tejun Heo wrote:
> This is an equivalent conversion and will ease scheduled removal of
> WQ_NON_REENTRANT.
>
> Signed-off-by: Tejun Heo
Reviewed-by: Chris Wilson
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
___
On Thu, Aug 23, 2012 at 08:56:37AM +0100, Chris Wilson wrote:
> On Wed, 22 Aug 2012 16:40:57 -0700, Tejun Heo wrote:
> > This is an equivalent conversion and will ease scheduled removal of
> > WQ_NON_REENTRANT.
> >
> > Signed-off-by: Tejun Heo
> Reviewed-by: Chris Wilson
Acked-by: Daniel Vette
On Thu, Aug 23, 2012 at 12:08:57PM +0530, Vijay Purushothaman wrote:
> This is already fixed for ILK and SNB in the below commit but somehow
> IVB is missed.
>
> commit ab2f9df10dd955f1fc0a8650e377588c98f1c029
> Author: Jesse Barnes
> Date: Mon Feb 27 12:40:10 2012 -0800
>
>
On Wed, 22 Aug 2012, Ben Widawsky wrote:
> INSTDONE is used in many places, and it varies from generation to
> generation. This provides a good reason for us to extract the logic to
> read the relevant information.
>
> The patch has no functional change. It's prep for some new stuff.
>
> v2: move
The principal use for set-to-domain is for userspace to serialise
operations with a particular buffer, for example to maintain coherency
with a CPU map or to ratelimit its rendering by waiting on all previous
operations before continuing. As such we tend to hold the struct_mutex
for long periods du
This prevents the case of unbinding the object in order to process the
relocations through the GTT and then rebinding it only to then proceed
to use cpu relocations as the object is now in the CPU write domain. By
choosing to use cpu relocations up front, we can therefore avoid the
rebind penalty.
If we need to stall in order to complete the pin_and_fence operation
during execbuffer reservation, there is a high likelihood that the
operation will be interrupted by a signal (thanks X!). In order to
simplify the cleanup along that error path, the object was
unconditionally unbound and the error
On Wed, Aug 22, 2012 at 10:21 PM, Daniel Vetter wrote:
> Please test the for-damien branch in my personal fdo git repo. I'm rather
> positive that the patch there should fix this (but in the least it should
> unearth the real culprit).
The branch works, but so does HEAD^ on that branch.
After ch
From: Damien Lespiau
What seems to be a copy/paste typo.
Signed-off-by: Damien Lespiau
---
tools/intel_reg_dumper.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/intel_reg_dumper.c b/tools/intel_reg_dumper.c
index a994f00..f6392e2 100644
--- a/tools/intel_reg_dumper
From: Damien Lespiau
This reports which lanes are locked.
Signed-off-by: Damien Lespiau
---
lib/intel_reg.h | 8
tools/intel_reg_dumper.c | 21 +
2 files changed, 29 insertions(+)
diff --git a/lib/intel_reg.h b/lib/intel_reg.h
index ffded64..0796cb5 1006
From: Damien Lespiau
This register gives information about which lanes are symbol locked
along with a bunch of possible training failures.
Signed-off-by: Damien Lespiau
---
lib/intel_reg.h | 36 +
tools/intel_reg_dumper.c | 84 ++
From: Damien Lespiau
Signed-off-by: Damien Lespiau
---
lib/intel_reg.h | 2 ++
tools/intel_reg_dumper.c | 6 +-
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/lib/intel_reg.h b/lib/intel_reg.h
index fa9e866..42b684d 100644
--- a/lib/intel_reg.h
+++ b/lib/intel_reg.h
Am Tuesday 21 August 2012 03:26:44 schrieb du81692468:
> Hello, I have a question:
> System Environment: fedora 14
> Graphics card: intel integrated graphics + NVIDIA
> When I installed the xf86-video-intel-2.17.0 driver, I can not set the
> screen resolution of 800 * 600, when I downgraded to
> xf
INSTDONE is used in many places, and it varies from generation to
generation. This provides a good reason for us to extract the logic to
read the relevant information.
The patch has no functional change. It's prep for some new stuff.
v2: move the memset inside of i915_get_extra_instdone (Jani)
v3
On Thu, Aug 23, 2012 at 01:26:39PM +0100, Lespiau, Damien wrote:
> On Wed, Aug 22, 2012 at 10:21 PM, Daniel Vetter wrote:
> > Please test the for-damien branch in my personal fdo git repo. I'm rather
> > positive that the patch there should fix this (but in the least it should
> > unearth the real
On Thu, Aug 23, 2012 at 12:22:27PM -0700, Tejun Heo wrote:
> Hello,
>
> On Thu, Aug 23, 2012 at 10:43:25AM +0200, Daniel Vetter wrote:
> > On Thu, Aug 23, 2012 at 08:56:37AM +0100, Chris Wilson wrote:
> > > On Wed, 22 Aug 2012 16:40:57 -0700, Tejun Heo wrote:
> > > > This is an equivalent convers
On 2012-08-23 06:04, Damien Lespiau wrote:
From: Damien Lespiau
This reports which lanes are locked.
Signed-off-by: Damien Lespiau
---
lib/intel_reg.h | 8
tools/intel_reg_dumper.c | 21 +
2 files changed, 29 insertions(+)
diff --git a/lib/intel_reg.h
On 2012-08-23 06:04, Damien Lespiau wrote:
From: Damien Lespiau
This register gives information about which lanes are symbol locked
along with a bunch of possible training failures.
Signed-off-by: Damien Lespiau
I don't know where a bunch of these bit definitions came from, I cannot
find the
On Sat, Aug 11, 2012 at 03:41:06PM +0100, Chris Wilson wrote:
> As we wish to create specialised object constructions in the near
> future that share the same basic GEM object struct, export the default
> initializer.
>
> Signed-off-by: Chris Wilson
I've slurped in the 3 patches up to here to -ne
On Thu, Aug 23, 2012 at 01:12:51PM +0100, Chris Wilson wrote:
> The principal use for set-to-domain is for userspace to serialise
> operations with a particular buffer, for example to maintain coherency
> with a CPU map or to ratelimit its rendering by waiting on all previous
> operations before co
On Thu, Aug 23, 2012 at 01:12:52PM +0100, Chris Wilson wrote:
> This prevents the case of unbinding the object in order to process the
> relocations through the GTT and then rebinding it only to then proceed
> to use cpu relocations as the object is now in the CPU write domain. By
> choosing to use
22 matches
Mail list logo
