On 15 June 2011 12:43, Ben Widawsky wrote:
> On Wed, Jun 15, 2011 at 12:51:47AM +0800, Daniel J Blueman wrote:
>> On 14 June 2011 13:23, Eric Anholt wrote:
>> > On Tue, 14 Jun 2011 12:18:36 +0800, Daniel J Blueman
>> > wrote:
>> >> Hi Eric,
>> >>
>> >> The frequent ~1.5s pauses I hit with SNB h
On Wed, Jun 15, 2011 at 12:51:47AM +0800, Daniel J Blueman wrote:
> On 14 June 2011 13:23, Eric Anholt wrote:
> > On Tue, 14 Jun 2011 12:18:36 +0800, Daniel J Blueman
> > wrote:
> >> Hi Eric,
> >>
> >> The frequent ~1.5s pauses I hit with SNB hardware in the gnome3 UI (eg
> >> whenever you hit t
On 15 June 2011 10:06, Eric Anholt wrote:
> On Wed, 15 Jun 2011 00:51:47 +0800, Daniel J Blueman
> wrote:
>> On 14 June 2011 13:23, Eric Anholt wrote:
>> > On Tue, 14 Jun 2011 12:18:36 +0800, Daniel J Blueman
>> > wrote:
>> >> Hi Eric,
>> >>
>> >> The frequent ~1.5s pauses I hit with SNB hard
On Wed, 15 Jun 2011 00:51:47 +0800, Daniel J Blueman
wrote:
> On 14 June 2011 13:23, Eric Anholt wrote:
> > On Tue, 14 Jun 2011 12:18:36 +0800, Daniel J Blueman
> > wrote:
> >> Hi Eric,
> >>
> >> The frequent ~1.5s pauses I hit with SNB hardware in the gnome3 UI (eg
> >> whenever you hit the t
This reverts commit 4a684a4117abd756291969336af454e8a958802f.
Userland has always been required to set the object's domain to GTT
before using it through a GTT mapping, it's not something that the
kernel is supposed to enforce. (The pagefault support is so that we
can handle multiple mappings with
Preliminary code for shader debugging. This provides a trace like
implementation which can get EU architectural state at each instruction.
The code currently only supports the most basic set of state and
features.
Signed-off-by: Ben Widawsky
---
configure.ac |1 +
Signed-off-by: Ben Widawsky
---
lib/intel_reg.h |5 +
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/lib/intel_reg.h b/lib/intel_reg.h
index 48d8f66..ef9cb63 100644
--- a/lib/intel_reg.h
+++ b/lib/intel_reg.h
@@ -3488,5 +3488,10 @@ typedef enum {
#define TRANS_DP_HSYNC_
This app is required for debug features which seem to (undocumented)
reset themselves if/when the GT goes to sleep.
Signed-off-by: Ben Widawsky
---
tools/Makefile.am |1 +
tools/forcewaked.c | 45 +
2 files changed, 46 insertions(+), 0 deletions
We can deprecate the old code by using the non-safe flag in the new API.
The safe flag should allow the previous behavior to continue.
The code also adds some range checking on register access. This code is
gives hooks to prevent tools from doing bad things.
Signed-off-by: Ben Widawsky
---
lib/
The following patches enable GEN hardware shader debugging features. This is
for >= GEN6 only.
The patches are not usable without patches to the DDX as well as Mesa.
Furthermore currently a kernel ioctl is required (more on that later). As
these are all in tools, I figure I would post these now w
On Tue, 14 Jun 2011 19:22:33 +0100
Chris Wilson wrote:
> On Tue, 14 Jun 2011 11:13:08 -0700, Jesse Barnes
> wrote:
> > Use the blit ring for submitting flips since the render ring doesn't
> > generate flip complete interrupts.
> >
> > Signed-off-by: Jesse Barnes
> > ---
> > drivers/gpu/drm/i
Use the blit ring for submitting flips since the render ring doesn't
generate flip complete interrupts.
Signed-off-by: Jesse Barnes
---
drivers/gpu/drm/i915/intel_display.c | 26 ++
1 files changed, 26 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/i915/int
This makes things a little clearer and prevents us from running old code
on a new chipset that may not be supported.
Signed-off-by: Jesse Barnes
---
drivers/gpu/drm/i915/i915_drv.h |3 +
drivers/gpu/drm/i915/intel_display.c | 202 +-
2 files changed, 151
On Tue, 14 Jun 2011 11:13:08 -0700, Jesse Barnes
wrote:
> Use the blit ring for submitting flips since the render ring doesn't
> generate flip complete interrupts.
>
> Signed-off-by: Jesse Barnes
> ---
> drivers/gpu/drm/i915/intel_display.c | 25 +
> 1 files changed,
On Tue, 14 Jun 2011 11:13:07 -0700, Jesse Barnes
wrote:
> @@ -6343,7 +6458,12 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
> /* Offset into the new buffer for cases of shared fbs between CRTCs */
> offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8;
>
> -
This makes things a little clearer and prevents us from running old code
on a new chipset that may not be supported.
Signed-off-by: Jesse Barnes
---
drivers/gpu/drm/i915/i915_drv.h |3 +
drivers/gpu/drm/i915/intel_display.c | 196 -
2 files changed, 145
Use the blit ring for submitting flips since the render ring doesn't
generate flip complete interrupts.
Signed-off-by: Jesse Barnes
---
drivers/gpu/drm/i915/intel_display.c | 25 +
1 files changed, 25 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/i915/inte
On Wed, 15 Jun 2011 00:51:47 +0800, Daniel J Blueman
wrote:
> True; that is quite heavy handed delay looping.
>
> It's a pity the usual Intel font didn't make it to the programmer's
> reference manuals. Anyway, unmasking the blitter user interrupt in the
> hardware
> status mask register addres
On 14 June 2011 13:23, Eric Anholt wrote:
> On Tue, 14 Jun 2011 12:18:36 +0800, Daniel J Blueman
> wrote:
>> Hi Eric,
>>
>> The frequent ~1.5s pauses I hit with SNB hardware in the gnome3 UI (eg
>> whenever you hit the top-left of the screen to show all windows) are
>> nicely addressed by your r
On Mon, 13 Jun 2011 16:00:53 -0700, Jesse Barnes
wrote:
> Yeah, it's just an ugly hack.
Let's also put a default: return -ENODEV in there for future sanity.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
___
Intel-gfx mailing list
Intel-
20 matches
Mail list logo
