Re: [Intel-gfx] [PATCH] drm/i915: Flush ggtt writes through the old fenced vma before changing fences

[Lahtinen, Joonas]

On Tue, 2018-01-30 at 16:44 +, Chris Wilson wrote:
> This is a precautionary measure as I have no evidence to suggest we've
> hit a bug here (I was hoping this might explain gdg's odd behaviour, but
> alas), but given that we have a function to flush the ggtt writes it
> seems prudent to use it prior to changing the fence register. Due to the
> intrinsic nature of the GTT often operating as an independent mmio path,
> we should not just rely on the write to the fence acting as a full flush
> for GTT writes.
> 
> Signed-off-by: Chris Wilson 
> Cc: Joonas Lahtinen 

Reviewed-by: Joonas Lahtinen 

Regards, Joonas
-
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki 
Business Identity Code: 0357606 - 4 
Domiciled in Helsinki 

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH v11 5/6] drm/i915: add query uAPI

[Lahtinen, Joonas]

On Mon, 2018-01-29 at 16:24 +, Lionel Landwerlin wrote:
>
> Thanks all,
> 
> I don't really read much opposition to the current patch series. If 
> anything we could actually want to do more it seems.
> It would be good to have the green light and land that.
> I've played quickly with a Chris' idea and will add a couple of RFC 
> patches for discussion.

Where is the latest iteration of the less generic IOCTL?

I'm still not sold on this, the resulting uAPI is quite horrendous with
the macros to interpret the results.

Regards, Joonas
-
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki 
Business Identity Code: 0357606 - 4 
Domiciled in Helsinki 

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [PATCH v3] drm/i915: Do not use ggtt_view with (aliasing) PPGTT

[Lahtinen, Joonas]

On ti, 2015-03-17 at 07:31 +0200, Joonas Lahtinen wrote:
 On ma, 2015-03-16 at 18:50 +0100, Daniel Vetter wrote:
  On Mon, Mar 16, 2015 at 02:11:13PM +0200, Joonas Lahtinen wrote:
  [snip]
int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level 
   cache_level,
   u32 flags);
   @@ -2800,60 +2796,48 @@ struct dma_buf *i915_gem_prime_export(struct 
   drm_device *dev,

void i915_gem_restore_fences(struct drm_device *dev);

   -unsigned long i915_gem_obj_offset_view(struct drm_i915_gem_object *o,
   -struct i915_address_space *vm,
   -enum i915_ggtt_view_type view);
   -static inline
   -unsigned long i915_gem_obj_offset(struct drm_i915_gem_object *o,
   -   struct i915_address_space *vm)
   +static inline bool i915_is_ggtt(struct i915_address_space *vm);

Indeed it is. Thanks!

Regards, Joonas

PS. Sorry for spam, the enter key is so close to the space bar.

  
  This forward decl seems unneeded leftover from earlier patch iterations.
  I've removed it and merged the patch.
  
  Thanks, Daniel
  
  [snip]

-
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki 
Business Identity Code: 0357606 - 4 
Domiciled in Helsinki 

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

[Intel-gfx] lib: Fix out of tree build of version.h

[Lahtinen, Joonas]

Currently out of tree build fails because the version.h.tmp is generated
into the source directory instead of the build directory where it is
later looked for. This commit fixes it.

From 2c0617e21101d69e7219c6660936c0015f93f8ee Mon Sep 17 00:00:00 2001
From: Joonas Lahtinen joonas.lahti...@linux.intel.com
Date: Thu, 11 Dec 2014 15:05:11 +0200
Subject: [PATCH] lib: Fix out of tree build of version.h

Write the version.h.tmp file into the build directory instead of source
directory. This allows out of tree building when those two are not the
same.

Signed-off-by: Joonas Lahtinen joonas.lahti...@linux.intel.com
---
 lib/Makefile.sources |   12 ++--
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/lib/Makefile.sources b/lib/Makefile.sources
index 819b21a..34a3d31 100644
--- a/lib/Makefile.sources
+++ b/lib/Makefile.sources
@@ -55,16 +55,16 @@ libintel_tools_la_SOURCES = \
 $(IGT_LIB_PATH)/version.h.tmp:
@touch $@
@if test -d $(GPU_TOOLS_PATH)/.git; then \
-   if which git  /dev/null 21; then cd $(@D); \
+   if which git  /dev/null 21; then \
+   cd $(GPU_TOOLS_PATH); \
git log -n 1 --oneline | \
-   sed 's/^\([^ ]*\) .*/#define IGT_GIT_SHA1 g\1/' \
-$(@F) ; \
+   sed 's/^\([^ ]*\) .*/#define IGT_GIT_SHA1 g\1/' ; \
else \
-   echo '#define IGT_GIT_SHA1 NO-GIT'  $@ ; \
+   echo '#define IGT_GIT_SHA1 NO-GIT' ; \
fi \
else \
-   echo '#define IGT_GIT_SHA1 NOT-GIT'  $@ ; \
-   fi
+   echo '#define IGT_GIT_SHA1 NOT-GIT' ; \
+   fi  $@
 
 
 $(IGT_LIB_PATH)/version.h: $(IGT_LIB_PATH)/version.h.tmp
-- 
1.7.9.5
-
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki 
Business Identity Code: 0357606 - 4 
Domiciled in Helsinki 

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx