Re: [Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2015-04-07 Thread Daniel Vetter
On Fri, Apr 03, 2015 at 10:43:29AM +0100, Chris Wilson wrote:
 On Thu, Nov 20, 2014 at 09:26:30AM +0100, Daniel Vetter wrote:
  diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c 
  b/drivers/gpu/drm/i915/i915_gem_tiling.c
  index 749ab485569e..03c675a4476e 100644
  --- a/drivers/gpu/drm/i915/i915_gem_tiling.c
  +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
  @@ -178,6 +178,15 @@ i915_gem_detect_bit_6_swizzle(struct drm_device *dev)
  }
  break;
  }
  +
  +   /* check for L-shaped memory aka modified enhanced addressing */
  +   if (IS_GEN4(dev)) {
  +   uint32_t ddc2 = I915_READ(DCC2);
  +
  +   if (!(ddc2  DCC2_MODIFIED_ENHANCED_DISABLE))
  +   dev_priv-quirks |= QUIRK_PIN_SWIZZLED_PAGES;
  +   }
  +
  if (dcc == 0x) {
  DRM_ERROR(Couldn't read from MCHBAR.  
Disabling tiling.\n);
 
 I resurrected my dual-channel evenly loaded gm45 and to my surprise
 found it reporting a L-shaped memory layout.

Well my gm45 is also dual-channel and evenly loaded and still mangles
tiled buffers when going through swap. Is gem_tiled_swapping happy on your
box?

 Perhaps the if ((ddc2  DISABLE) == 0) is inverted or is not the truly
 magic bit?

It was the best one I could find - given how old this all is now I think
some false positives are acceptable in the test. I did try a few other
things which did all fail.
-Daniel
 
 ickle@x200s $ sudo cat /sys/kernel/debug/dri/0/i915_swizzle_info
 bit6 swizzle for X-tiling = bit9/bit10/bit11
 bit6 swizzle for Y-tiling = bit9/bit11
 DDC = 0x000f0002
 DDC2 = 0x0010
 C0DRB3 = 0x
 C1DRB3 = 0x
 L-shaped memory detected
 -Chris
 
 -- 
 Chris Wilson, Intel Open Source Technology Centre

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


Re: [Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2015-04-07 Thread Chris Wilson
On Tue, Apr 07, 2015 at 10:05:39AM +0200, Daniel Vetter wrote:
 On Fri, Apr 03, 2015 at 10:43:29AM +0100, Chris Wilson wrote:
  I resurrected my dual-channel evenly loaded gm45 and to my surprise
  found it reporting a L-shaped memory layout.
 
 Well my gm45 is also dual-channel and evenly loaded and still mangles
 tiled buffers when going through swap. Is gem_tiled_swapping happy on your
 box?

ickle@x200s ~/intel-gpu-tools/tests $ sudo ./gem_tiled_swapping 
IGT-Version: 1.10-ge809895 (x86_64) (Linux: 4.0.0-rc6+ x86_64)
Using 640 1MiB objects (available RAM: 343/3847, swap: 3989)
Subtest non-threaded: SUCCESS (23.344s)
Subtest threaded: SUCCESS (110.801s)

  Perhaps the if ((ddc2  DISABLE) == 0) is inverted or is not the truly
  magic bit?
 
 It was the best one I could find - given how old this all is now I think
 some false positives are acceptable in the test. I did try a few other
 things which did all fail.

It's the false-negative I worry about. I am just worried this is not the
bit we are looking for. Time will tell if we get a new bug report with
broken swizzling after swapping.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


Re: [Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2015-04-03 Thread Chris Wilson
On Thu, Nov 20, 2014 at 09:26:30AM +0100, Daniel Vetter wrote:
 diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c 
 b/drivers/gpu/drm/i915/i915_gem_tiling.c
 index 749ab485569e..03c675a4476e 100644
 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c
 +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
 @@ -178,6 +178,15 @@ i915_gem_detect_bit_6_swizzle(struct drm_device *dev)
   }
   break;
   }
 +
 + /* check for L-shaped memory aka modified enhanced addressing */
 + if (IS_GEN4(dev)) {
 + uint32_t ddc2 = I915_READ(DCC2);
 +
 + if (!(ddc2  DCC2_MODIFIED_ENHANCED_DISABLE))
 + dev_priv-quirks |= QUIRK_PIN_SWIZZLED_PAGES;
 + }
 +
   if (dcc == 0x) {
   DRM_ERROR(Couldn't read from MCHBAR.  
 Disabling tiling.\n);

I resurrected my dual-channel evenly loaded gm45 and to my surprise
found it reporting a L-shaped memory layout.

Perhaps the if ((ddc2  DISABLE) == 0) is inverted or is not the truly
magic bit?

ickle@x200s $ sudo cat /sys/kernel/debug/dri/0/i915_swizzle_info
bit6 swizzle for X-tiling = bit9/bit10/bit11
bit6 swizzle for Y-tiling = bit9/bit11
DDC = 0x000f0002
DDC2 = 0x0010
C0DRB3 = 0x
C1DRB3 = 0x
L-shaped memory detected
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


[Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2014-11-20 Thread Daniel Vetter
Let's just throw in the towel on this one and take the cheap way out.

Based on a patch from Chris Wilson, but checking for a different bit.
Chris' patch checked for even bank layout, this one here for a magic
bit. Given the evidence we've gathered (not much) both work I think,
but checking for the magic bit might be more accurate.

Anyway, works on my gm45 here.

For paranoi restrict to gen4 (and mobile), since we've only ever seen
this on gm45 and i965gm.

Also add some debugfs output so that we can skip the tiled swapping
tests properly in these cases.

v2: Clean up the quirk'ed pin count in free_object to avoid upsetting
the WARN_ON. Spotted by Chris.

Cc: Chris Wilson ch...@chris-wilson.co.uk
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=28813
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=45092
Signed-off-by: Daniel Vetter daniel.vet...@intel.com
---
 drivers/gpu/drm/i915/i915_debugfs.c|  6 ++
 drivers/gpu/drm/i915/i915_drv.h|  1 +
 drivers/gpu/drm/i915/i915_gem.c| 19 +++
 drivers/gpu/drm/i915/i915_gem_tiling.c | 18 ++
 drivers/gpu/drm/i915/i915_reg.h|  2 ++
 5 files changed, 46 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c 
b/drivers/gpu/drm/i915/i915_debugfs.c
index 0a6981399642..a43124e4a77b 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1975,6 +1975,8 @@ static int i915_swizzle_info(struct seq_file *m, void 
*data)
if (IS_GEN3(dev) || IS_GEN4(dev)) {
seq_printf(m, DDC = 0x%08x\n,
   I915_READ(DCC));
+   seq_printf(m, DDC2 = 0x%08x\n,
+  I915_READ(DCC2));
seq_printf(m, C0DRB3 = 0x%04x\n,
   I915_READ16(C0DRB3));
seq_printf(m, C1DRB3 = 0x%04x\n,
@@ -1997,6 +1999,10 @@ static int i915_swizzle_info(struct seq_file *m, void 
*data)
seq_printf(m, DISP_ARB_CTL = 0x%08x\n,
   I915_READ(DISP_ARB_CTL));
}
+
+   if (dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES)
+   seq_puts(m, L-shaped memory detected\n);
+
intel_runtime_pm_put(dev_priv);
mutex_unlock(dev-struct_mutex);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 54691bcf1822..d5accc481082 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -747,6 +747,7 @@ enum intel_sbi_destination {
 #define QUIRK_INVERT_BRIGHTNESS (12)
 #define QUIRK_BACKLIGHT_PRESENT (13)
 #define QUIRK_PIPEB_FORCE (14)
+#define QUIRK_PIN_SWIZZLED_PAGES (15)
 
 struct intel_fbdev;
 struct intel_fbc_work;
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 1de94cc63517..8d04e0979d0a 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -2120,6 +2120,10 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object 
*obj)
if (i915_gem_object_needs_bit17_swizzle(obj))
i915_gem_object_do_bit_17_swizzle(obj);
 
+   if (obj-tiling_mode != I915_TILING_NONE 
+   dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES)
+   i915_gem_object_pin_pages(obj);
+
return 0;
 
 err_pages:
@@ -4300,6 +4304,7 @@ int
 i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
   struct drm_file *file_priv)
 {
+   struct drm_i915_private *dev_priv = dev-dev_private;
struct drm_i915_gem_madvise *args = data;
struct drm_i915_gem_object *obj;
int ret;
@@ -4327,6 +4332,15 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void 
*data,
goto out;
}
 
+   if (obj-pages 
+   obj-tiling_mode != I915_TILING_NONE 
+   dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES) {
+   if (obj-madv == I915_MADV_WILLNEED)
+   i915_gem_object_unpin_pages(obj);
+   if (args-madv == I915_MADV_WILLNEED)
+   i915_gem_object_pin_pages(obj);
+   }
+
if (obj-madv != __I915_MADV_PURGED)
obj-madv = args-madv;
 
@@ -4478,6 +4492,11 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 
WARN_ON(obj-frontbuffer_bits);
 
+   if (obj-pages  obj-madv == I915_MADV_WILLNEED 
+   dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES 
+   obj-tiling_mode != I915_TILING_NONE)
+   i915_gem_object_unpin_pages(obj);
+
if (WARN_ON(obj-pages_pin_count))
obj-pages_pin_count = 0;
if (discard_backing_storage(obj))
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c 
b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 749ab485569e..03c675a4476e 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -178,6 +178,15 @@ i915_gem_detect_bit_6_swizzle(struct drm_device *dev)
}
break;
}
+
+  

Re: [Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2014-11-20 Thread Chris Wilson
On Thu, Nov 20, 2014 at 09:26:30AM +0100, Daniel Vetter wrote:
 Let's just throw in the towel on this one and take the cheap way out.
 
 Based on a patch from Chris Wilson, but checking for a different bit.
 Chris' patch checked for even bank layout, this one here for a magic
 bit. Given the evidence we've gathered (not much) both work I think,
 but checking for the magic bit might be more accurate.
 
 Anyway, works on my gm45 here.
 
 For paranoi restrict to gen4 (and mobile), since we've only ever seen
 this on gm45 and i965gm.
 
 Also add some debugfs output so that we can skip the tiled swapping
 tests properly in these cases.
 
 v2: Clean up the quirk'ed pin count in free_object to avoid upsetting
 the WARN_ON. Spotted by Chris.
 
 Cc: Chris Wilson ch...@chris-wilson.co.uk
 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=28813
 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=45092
 Signed-off-by: Daniel Vetter daniel.vet...@intel.com

I want to complain about the nearly duplicated repeated checks, but I
failed to rewrite them compactly. :|

Reviewed-by: Chris Wilson ch...@chris-wilson.co.uk

Oh, one more request: can I haz decoded quirks in the error state?
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


Re: [Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2014-11-20 Thread Daniel Vetter
On Thu, Nov 20, 2014 at 08:37:36AM +, Chris Wilson wrote:
 On Thu, Nov 20, 2014 at 09:26:30AM +0100, Daniel Vetter wrote:
  Let's just throw in the towel on this one and take the cheap way out.
  
  Based on a patch from Chris Wilson, but checking for a different bit.
  Chris' patch checked for even bank layout, this one here for a magic
  bit. Given the evidence we've gathered (not much) both work I think,
  but checking for the magic bit might be more accurate.
  
  Anyway, works on my gm45 here.
  
  For paranoi restrict to gen4 (and mobile), since we've only ever seen
  this on gm45 and i965gm.
  
  Also add some debugfs output so that we can skip the tiled swapping
  tests properly in these cases.
  
  v2: Clean up the quirk'ed pin count in free_object to avoid upsetting
  the WARN_ON. Spotted by Chris.
  
  Cc: Chris Wilson ch...@chris-wilson.co.uk
  Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=28813
  Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=45092
  Signed-off-by: Daniel Vetter daniel.vet...@intel.com
 
 I want to complain about the nearly duplicated repeated checks, but I
 failed to rewrite them compactly. :|
 
 Reviewed-by: Chris Wilson ch...@chris-wilson.co.uk

Queued for -next, thanks for the review.

 Oh, one more request: can I haz decoded quirks in the error state?

Hm, I haven't ever yet wanted to see them. Not even sure whether this
might be useful here really ...
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


Re: [Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2014-11-20 Thread Chris Wilson
On Thu, Nov 20, 2014 at 10:22:42AM +0100, Daniel Vetter wrote:
 On Thu, Nov 20, 2014 at 08:37:36AM +, Chris Wilson wrote:
  Oh, one more request: can I haz decoded quirks in the error state?
 
 Hm, I haven't ever yet wanted to see them. Not even sure whether this
 might be useful here really ...

Here, I think just logging obj-pages_pin_count would answer some
worries. At the moment our quirks concern interaction with the display
hardware (backlight, PIPEA, ssc), if we start adding more that are
focused on memory management or GPU interaction, then we will went to
see the quirk list.

Thinking of which, there are quite a few cache/execution mode registers
we should be dumping upon error.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


[Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2014-11-18 Thread Daniel Vetter
Let's just throw in the towel on this one and take the cheap way out.

Based on a patch from Chris Wilson, but checking for a different bit.
Chris' patch checked for even bank layout, this one here for a magic
bit. Given the evidence we've gathered (not much) both work I think,
but checking for the magic bit might be more accurate.

Anyway, works on my gm45 here.

For paranoi restrict to gen4 (and mobile), since we've only ever seen
this on gm45 and i965gm.

Also add some debugfs output so that we can skip the tiled swapping
tests properly in these cases.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=28813
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=45092
Signed-off-by: Daniel Vetter daniel.vet...@intel.com
---
 drivers/gpu/drm/i915/i915_debugfs.c|  6 ++
 drivers/gpu/drm/i915/i915_drv.h|  1 +
 drivers/gpu/drm/i915/i915_gem.c| 14 ++
 drivers/gpu/drm/i915/i915_gem_tiling.c | 18 ++
 drivers/gpu/drm/i915/i915_reg.h|  2 ++
 5 files changed, 41 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c 
b/drivers/gpu/drm/i915/i915_debugfs.c
index 0a6981399642..a43124e4a77b 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1975,6 +1975,8 @@ static int i915_swizzle_info(struct seq_file *m, void 
*data)
if (IS_GEN3(dev) || IS_GEN4(dev)) {
seq_printf(m, DDC = 0x%08x\n,
   I915_READ(DCC));
+   seq_printf(m, DDC2 = 0x%08x\n,
+  I915_READ(DCC2));
seq_printf(m, C0DRB3 = 0x%04x\n,
   I915_READ16(C0DRB3));
seq_printf(m, C1DRB3 = 0x%04x\n,
@@ -1997,6 +1999,10 @@ static int i915_swizzle_info(struct seq_file *m, void 
*data)
seq_printf(m, DISP_ARB_CTL = 0x%08x\n,
   I915_READ(DISP_ARB_CTL));
}
+
+   if (dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES)
+   seq_puts(m, L-shaped memory detected\n);
+
intel_runtime_pm_put(dev_priv);
mutex_unlock(dev-struct_mutex);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 54691bcf1822..d5accc481082 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -747,6 +747,7 @@ enum intel_sbi_destination {
 #define QUIRK_INVERT_BRIGHTNESS (12)
 #define QUIRK_BACKLIGHT_PRESENT (13)
 #define QUIRK_PIPEB_FORCE (14)
+#define QUIRK_PIN_SWIZZLED_PAGES (15)
 
 struct intel_fbdev;
 struct intel_fbc_work;
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 1de94cc63517..196ebd07d118 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -2120,6 +2120,10 @@ i915_gem_object_get_pages_gtt(struct drm_i915_gem_object 
*obj)
if (i915_gem_object_needs_bit17_swizzle(obj))
i915_gem_object_do_bit_17_swizzle(obj);
 
+   if (obj-tiling_mode != I915_TILING_NONE 
+   dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES)
+   i915_gem_object_pin_pages(obj);
+
return 0;
 
 err_pages:
@@ -4300,6 +4304,7 @@ int
 i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
   struct drm_file *file_priv)
 {
+   struct drm_i915_private *dev_priv = dev-dev_private;
struct drm_i915_gem_madvise *args = data;
struct drm_i915_gem_object *obj;
int ret;
@@ -4327,6 +4332,15 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void 
*data,
goto out;
}
 
+   if (obj-pages 
+   obj-tiling_mode != I915_TILING_NONE 
+   dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES) {
+   if (obj-madv == I915_MADV_WILLNEED)
+   i915_gem_object_unpin_pages(obj);
+   if (args-madv == I915_MADV_WILLNEED)
+   i915_gem_object_pin_pages(obj);
+   }
+
if (obj-madv != __I915_MADV_PURGED)
obj-madv = args-madv;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c 
b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 749ab485569e..03c675a4476e 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -178,6 +178,15 @@ i915_gem_detect_bit_6_swizzle(struct drm_device *dev)
}
break;
}
+
+   /* check for L-shaped memory aka modified enhanced addressing */
+   if (IS_GEN4(dev)) {
+   uint32_t ddc2 = I915_READ(DCC2);
+
+   if (!(ddc2  DCC2_MODIFIED_ENHANCED_DISABLE))
+   dev_priv-quirks |= QUIRK_PIN_SWIZZLED_PAGES;
+   }
+
if (dcc == 0x) {
DRM_ERROR(Couldn't read from MCHBAR.  
  Disabling tiling.\n);
@@ -393,6 +402,15 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,
}
 
 

Re: [Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2014-11-18 Thread Chris Wilson
On Tue, Nov 18, 2014 at 02:41:22PM +0100, Daniel Vetter wrote:
  struct intel_fbdev;
  struct intel_fbc_work;
 diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
 index 1de94cc63517..196ebd07d118 100644
 --- a/drivers/gpu/drm/i915/i915_gem.c
 +++ b/drivers/gpu/drm/i915/i915_gem.c
 @@ -2120,6 +2120,10 @@ i915_gem_object_get_pages_gtt(struct 
 drm_i915_gem_object *obj)
   if (i915_gem_object_needs_bit17_swizzle(obj))
   i915_gem_object_do_bit_17_swizzle(obj);
  
 + if (obj-tiling_mode != I915_TILING_NONE 
 + dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES)
 + i915_gem_object_pin_pages(obj);

Note that we have a WARN_ON(obj-pages_pin_count) in
i915_gem_free_object() now.

if (dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES)
  obj-pages_pin_count = 0;

or we could just throw away the WARN_ON().
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


Re: [Intel-gfx] [PATCH] drm/i915: Pin tiled objects for L-shaped configs

2014-11-18 Thread Daniel Vetter
On Tue, Nov 18, 2014 at 02:01:37PM +, Chris Wilson wrote:
 On Tue, Nov 18, 2014 at 02:41:22PM +0100, Daniel Vetter wrote:
   struct intel_fbdev;
   struct intel_fbc_work;
  diff --git a/drivers/gpu/drm/i915/i915_gem.c 
  b/drivers/gpu/drm/i915/i915_gem.c
  index 1de94cc63517..196ebd07d118 100644
  --- a/drivers/gpu/drm/i915/i915_gem.c
  +++ b/drivers/gpu/drm/i915/i915_gem.c
  @@ -2120,6 +2120,10 @@ i915_gem_object_get_pages_gtt(struct 
  drm_i915_gem_object *obj)
  if (i915_gem_object_needs_bit17_swizzle(obj))
  i915_gem_object_do_bit_17_swizzle(obj);
   
  +   if (obj-tiling_mode != I915_TILING_NONE 
  +   dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES)
  +   i915_gem_object_pin_pages(obj);
 
 Note that we have a WARN_ON(obj-pages_pin_count) in
 i915_gem_free_object() now.
 
 if (dev_priv-quirks  QUIRK_PIN_SWIZZLED_PAGES)
   obj-pages_pin_count = 0;
 
 or we could just throw away the WARN_ON().

Shame on me for not noticing that. I think I'll just drop the pinning for
tiled objects manually, that way we can keep the WARN_ON. Given all the
tricks being pulled with pin refcounts I think that check is useful.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx