On Thu, Aug 07, 2014 at 12:14:24PM +0200, Bruno Jiménez wrote:
This patch adds a new struct: r600_transfer_global. It will
act as a wrapper around an r600_resource_global and an r600_transfer.
It will be used for calling r600_compute_global_transfer_unmap when
transfer_unmap is called. And
On Tue, Aug 12, 2014 at 11:14:06AM -0700, Jason Ekstrand wrote:
The link error occurs because the static libraries are linked in the wrong
order. This fixes it.
Reviewed-by: Tom Stellard thomas.stell...@amd.com
Signed-off-by: Jason Ekstrand jason.ekstr...@intel.com
Bugzilla: https
This will decrement the reference count for buffers referenced in the
command stream will prevent us from leaking them.
CC: 10.2 mesa-sta...@lists.freedesktop.org
---
src/gallium/drivers/radeonsi/si_compute.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
This is required by the spec.
CC: 10.2 mesa-sta...@lists.freedesktop.org
---
src/gallium/state_trackers/clover/api/queue.cpp | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/clover/api/queue.cpp
b/src/gallium/state_trackers/clover/api/queue.cpp
We were leaking the input buffer used for kernel arguments and since
we were allocating it using si_upload_const_buffer() we were leaking
1 MB per kernel invocation.
CC: 10.2 mesa-sta...@lists.freedesktop.org
---
src/gallium/drivers/radeonsi/si_compute.c | 22 ++
1 file
There is a hard limit in older kernels of 256 MB for buffer allocations,
so report this value as MAX_MEM_ALLOC_SIZE and adjust MAX_GLOBAL_SIZE
to statisfy requirements of OpenCL.
CC: 10.2 mesa-sta...@lists.freedesktop.org
---
src/gallium/drivers/radeon/r600_pipe_common.c | 32
CC: 10.2 mesa-sta...@lists.freedesktop.org
---
src/gallium/drivers/radeonsi/si_compute.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_compute.c
b/src/gallium/drivers/radeonsi/si_compute.c
index e8fc8eb..dff5ddd 100644
---
CC: 10.2 mesa-sta...@lists.freedesktop.org
---
src/gallium/drivers/r600/r600_pipe.c | 11 ++-
src/gallium/drivers/radeonsi/si_pipe.c | 7 +++
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git a/src/gallium/drivers/r600/r600_pipe.c
b/src/gallium/drivers/r600/r600_pipe.c
CC: 10.2 mesa-sta...@lists.freedesktop.org
---
src/gallium/drivers/radeonsi/si_compute.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_compute.c
b/src/gallium/drivers/radeonsi/si_compute.c
index 42e4fec..482d475 100644
---
Hi,
This series contains fixes for applications which allocate large amounts
of memory.
The first two patches fix the values reported for
PIPE_COMPUTE_CAP_MAX_GLOBAL_SIZE, PIPE_COMPUTE_CAP_MAX_MEM_ALLOC_SIZE,
and PIPE_SHADER_CAP_MAX_CONST_BUFFER_SIZE so that applications don't
allocate more
On Thu, Aug 07, 2014 at 04:02:40PM +0300, Francisco Jerez wrote:
Jan Vesely jan.ves...@rutgers.edu writes:
This respin includes Francisco's approach of providing implicit
in the arg vector passed from clover, and Tom's idea of appending
implicit args after the kernel args.
Hmmm...
On Fri, Aug 08, 2014 at 04:53:46AM +0300, Kertesz Laszlo wrote:
Is the r600 llvm backend supposed to behave exactly the same as the
radeonsi llvm?
Support for r600 in LLVM is experimental, but for radeonsi it should
work well.
-Tom
--
O zi buna,
Kertesz Laszlo
On Fri, Aug 01, 2014 at 03:55:56PM +0200, Marek Olšák wrote:
From: David Heidelberger david.heidelber...@ixit.cz
Get rid of error Failed to build loop info by fixing failure in cases
like
4: SGE temp[2].x, temp[0]., const[0].;
5: CMP temp[1].x, -temp[2]., const[0].,
On Wed, Jul 30, 2014 at 07:11:35PM -0400, Jan Vesely wrote:
Make the function static.
No need to cc llvm-commits on these mesa patches. Reviewers follow both
lists.
This needs corresponding change in LLVM otherwise it breaks parameter passing
CC: Tom Stellard t...@stellard.net
CC: Matt
On Thu, Jul 31, 2014 at 01:28:45PM -0400, Jan Vesely wrote:
On Thu, 2014-07-31 at 11:02 -0400, Tom Stellard wrote:
On Wed, Jul 30, 2014 at 07:11:35PM -0400, Jan Vesely wrote:
Make the function static.
No need to cc llvm-commits on these mesa patches. Reviewers follow both
lists
On Tue, Jul 29, 2014 at 09:50:23AM -0500, Aaron Watry wrote:
Hi list,
I was starting to look into getting cppamp-driver-ng working with
mesa/clover, and I quickly ran into a question...
cppamp-driver-ng explicitly lists which platforms are supported in its
OpenCL back-end's source code. I
On Sun, Jul 27, 2014 at 01:56:15PM +0200, Bruno Jiménez wrote:
v2: Rebased on top of master
I've pushed both of these patches, thanks!
-Tom
---
src/gallium/drivers/r600/compute_memory_pool.c | 59
+-
src/gallium/drivers/r600/compute_memory_pool.h | 58
On Sat, Jul 19, 2014 at 07:35:51PM +0200, Bruno Jiménez wrote:
This allows us two things: we now need less item copies when we have
to defrag+grow the pool (to just one copy per item) and, even in the
case where we don't need to defrag the pool, we reduce the data copied
to just the useful
On Fri, Jul 25, 2014 at 10:33:42AM -0400, Jan Vesely wrote:
The iteration variables go from 0 anyway.
Signed-off-by: Jan Vesely jan.ves...@rutgers.edu
Thanks, I pushed patch #1.
-Tom
---
A collection of fixes for gcc warnings I ran across.
src/gallium/drivers/r600/evergreen_compute.c
On Fri, Jul 25, 2014 at 11:28:19PM +0200, Bruno Jimenez wrote:
On Fri, 2014-07-25 at 12:46 -0400, Tom Stellard wrote:
On Sat, Jul 19, 2014 at 07:35:51PM +0200, Bruno Jiménez wrote:
This allows us two things: we now need less item copies when we have
to defrag+grow the pool (to just one
Most image functions are required to return a CL_INVALID_OPERATION
error when used on devices without image support.
v2:
- Simplified the code
---
src/gallium/state_trackers/clover/api/memory.cpp | 6 ++
src/gallium/state_trackers/clover/api/sampler.cpp | 3 +++
of 4 x 32-bit
constants). I can use this instead.
-Tom
Marek
On Thu, Jul 24, 2014 at 3:05 AM, Tom Stellard thomas.stell...@amd.com wrote:
---
src/gallium/docs/source/screen.rst | 2 ++
src/gallium/include/pipe/p_defines.h | 3 ++-
2 files changed, 4 insertions(+), 1 deletion
On Fri, Jul 18, 2014 at 12:35:24PM -0400, Alex Deucher wrote:
On Fri, Jul 18, 2014 at 3:54 AM, Glenn Kennard glenn.kenn...@gmail.com
wrote:
Piglit quick tests including sqrt pass, no other regressions,
tested on radeon 6670.
---
Should be slightly more precise than the invsqrt/recip/mul
On Fri, Jul 18, 2014 at 01:09:03PM +0200, Bruno Jimenez wrote:
On Thu, 2014-07-17 at 22:56 -0400, Tom Stellard wrote:
On Wed, Jul 16, 2014 at 11:12:42PM +0200, Bruno Jiménez wrote:
Hi,
This series finally adds support for defragmenting the pool for
OpenCL buffers in the r600g
On Tue, Jul 22, 2014 at 12:36:33AM +0200, Grigori Goronzy wrote:
On 17.07.2014 21:24, Tom Stellard wrote:
On Thu, Jul 17, 2014 at 06:44:25PM +0200, Grigori Goronzy wrote:
Accuracy of some operations was recently improved in the R600 backend,
at the cost of slower code. This is required
On Sat, Jul 19, 2014 at 07:35:49PM +0200, Bruno Jiménez wrote:
---
src/gallium/drivers/r600/compute_memory_pool.c | 43
++
src/gallium/drivers/r600/compute_memory_pool.h | 1 +
2 files changed, 25 insertions(+), 19 deletions(-)
diff --git
On Sat, Jul 19, 2014 at 07:35:50PM +0200, Bruno Jiménez wrote:
This will be used in the following patch to avoid duplicated code
---
Reviewed-by: Tom Stellard thomas.stell...@amd.com
src/gallium/drivers/r600/compute_memory_pool.c | 11 ++-
src/gallium/drivers/r600
data that the items use.
Note: The fallback path is a bit ugly now, but hopefully we won't need
it much.
Reviewed-by: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/r600/compute_memory_pool.c | 40
--
src/gallium/drivers/r600/compute_memory_pool.h | 2
Most image functions are required to return a CL_INVALID_OPERATION
error when used on devices without image support.
---
src/gallium/state_trackers/clover/api/memory.cpp | 6 ++
src/gallium/state_trackers/clover/api/sampler.cpp | 3 +++
src/gallium/state_trackers/clover/api/transfer.cpp
---
src/gallium/docs/source/screen.rst| 2 ++
src/gallium/drivers/radeon/r600_pipe_common.c | 7 +++
src/gallium/include/pipe/p_defines.h | 3 ++-
3 files changed, 11 insertions(+), 1 deletion(-)
diff --git a/src/gallium/docs/source/screen.rst
---
src/gallium/state_trackers/clover/api/device.cpp | 2 +-
src/gallium/state_trackers/clover/core/device.cpp | 6 ++
src/gallium/state_trackers/clover/core/device.hpp | 1 +
3 files changed, 8 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/clover/api/device.cpp
---
src/gallium/docs/source/screen.rst | 2 ++
src/gallium/include/pipe/p_defines.h | 3 ++-
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/gallium/docs/source/screen.rst
b/src/gallium/docs/source/screen.rst
index 830a1a5..219c9f9 100644
---
---
src/gallium/drivers/radeon/r600_pipe_common.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/gallium/drivers/radeon/r600_pipe_common.c
b/src/gallium/drivers/radeon/r600_pipe_common.c
index bf0585d..2ea8f3d 100644
--- a/src/gallium/drivers/radeon/r600_pipe_common.c
+++
---
src/gallium/state_trackers/clover/core/device.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/gallium/state_trackers/clover/core/device.cpp
b/src/gallium/state_trackers/clover/core/device.cpp
index 63aa193..ada5267 100644
---
---
src/gallium/drivers/radeonsi/si_compute.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/src/gallium/drivers/radeonsi/si_compute.c
b/src/gallium/drivers/radeonsi/si_compute.c
index 3a9f00f..a7d61e7 100644
--- a/src/gallium/drivers/radeonsi/si_compute.c
+++
---
src/gallium/winsys/radeon/drm/radeon_drm_winsys.c | 6 ++
src/gallium/winsys/radeon/drm/radeon_winsys.h | 2 ++
2 files changed, 8 insertions(+)
diff --git a/src/gallium/winsys/radeon/drm/radeon_drm_winsys.c
b/src/gallium/winsys/radeon/drm/radeon_drm_winsys.c
index 576fea5..7cda70a
The scratch buffer will be used for private memory and also register
spilling.
---
src/gallium/drivers/radeonsi/si_compute.c | 85 ++-
src/gallium/drivers/radeonsi/si_shader.c | 5 ++
src/gallium/drivers/radeonsi/si_shader.h | 2 +
3 files changed, 90
The is used for programs that have arrays of constants that
are accessed using dynamic indices. The shader will compute
the base address of the constants and then access them using
SMRD instructions.
---
src/gallium/drivers/radeon/r600_pipe_common.h | 5 +
v2:
- Preserve word boundaries.
---
src/gallium/auxiliary/util/u_math.h | 17 +
1 file changed, 17 insertions(+)
diff --git a/src/gallium/auxiliary/util/u_math.h
b/src/gallium/auxiliary/util/u_math.h
index b9ed197..5de181a 100644
--- a/src/gallium/auxiliary/util/u_math.h
+++
---
src/gallium/drivers/radeonsi/si_descriptors.c | 4 +---
src/gallium/drivers/radeonsi/si_shader.c | 8 +---
2 files changed, 2 insertions(+), 10 deletions(-)
diff --git a/src/gallium/drivers/radeonsi/si_descriptors.c
b/src/gallium/drivers/radeonsi/si_descriptors.c
index
---
src/gallium/drivers/r600/evergreen_compute.h | 13 -
src/gallium/drivers/radeon/r600_pipe_common.h | 5 +
src/gallium/drivers/radeonsi/si_compute.c | 5 +
3 files changed, 10 insertions(+), 13 deletions(-)
diff --git a/src/gallium/drivers/r600/evergreen_compute.h
v2:
- Preserve word boundaries.
v3:
- Use const and restrict.
- Fix indentation.
---
src/gallium/auxiliary/util/u_math.h | 17 +
1 file changed, 17 insertions(+)
diff --git a/src/gallium/auxiliary/util/u_math.h
b/src/gallium/auxiliary/util/u_math.h
index b9ed197..f6dcb22
The scratch buffer will be used for private memory and also register
spilling.
v2:
- Code cleanups
---
I had some uncommitted changes left in my tree when I generated v1 of this
patch.
src/gallium/drivers/radeonsi/si_compute.c | 80 ++-
but possibly less accurate code.
Piglit didn't indicate any regressions.
Both patches are:
Reviewed-by: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/radeon/radeon_llvm_emit.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/src/gallium/drivers/radeon/radeon_llvm_emit.c
b
On Wed, Jul 16, 2014 at 11:12:42PM +0200, Bruno Jiménez wrote:
Hi,
This series finally adds support for defragmenting the pool for
OpenCL buffers in the r600g driver. It is mostly a rewritten of
the series that I wrote some months ago.
For defragmenting the pool I have thought of two
Also change the wait parameter from false to true.
---
I'm really not sure what is correct here, but this patch fixes event profiling
on SI.
src/gallium/state_trackers/clover/core/timestamp.cpp | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git
We use PIPE_QUERY_TIMESTAMP for profiling events, and gallium specifies
that the timestamp be in nanoseconds.
---
src/gallium/state_trackers/clover/api/device.cpp | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/clover/api/device.cpp
v2:
- Move the end_query() call into the timestamp constructor.
- Still pass false as the wait parameter to get_query_result().
---
src/gallium/state_trackers/clover/core/timestamp.cpp | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/gallium/state_trackers/clover/core/timestamp.cpp
---
src/gallium/auxiliary/util/u_math.h | 22 ++
src/gallium/drivers/radeonsi/si_shader.c | 8 +---
2 files changed, 23 insertions(+), 7 deletions(-)
diff --git a/src/gallium/auxiliary/util/u_math.h
b/src/gallium/auxiliary/util/u_math.h
index b9ed197..cd3cf04
The is used for programs that have arrays of constants that
are accessed using dynamic indices. The shader will compute
the base address of the constants and then access them using
SMRD instructions.
---
src/gallium/drivers/radeon/r600_pipe_common.h | 5 +
We were trying to print the error message after disposing the
message object.
---
src/gallium/drivers/radeon/radeon_llvm_emit.c | 11 ---
1 file changed, 4 insertions(+), 7 deletions(-)
diff --git a/src/gallium/drivers/radeon/radeon_llvm_emit.c
On Fri, Jul 11, 2014 at 10:20:54AM +0200, Bruno Jiménez wrote:
v2: Add information about the item's starting point and size
Reviewed-by: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/r600/compute_memory_pool.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/src
On Fri, Jul 11, 2014 at 01:00:34AM +0200, Marek Olšák wrote:
I have just tested it and it works with LLVM 3.4.2.
Ok, thanks.
Both patches are:
Reviewed-by: Tom Stellard thomas.stell...@amd.com
Marek
On Thu, Jul 10, 2014 at 5:11 PM, Tom Stellard t...@stellard.net wrote:
On Tue, Jul 08
On Tue, Jul 08, 2014 at 03:37:02AM +0200, Marek Olšák wrote:
From: Marek Olšák marek.ol...@amd.com
Needed by ARB_draw_indirect.
I think we should come up with a rule for how long we should support
older versions of LLVM. Do you have any thoughts about this? I was
thinking we could have each
On Tue, Jul 08, 2014 at 01:37:02AM +0200, Marek Olšák wrote:
From: Marek Olšák marek.ol...@amd.com
This is a follow-up to the commit which adds texture fetches with offsets.
Will this still work with LLVM 3.4.2 ?
-Tom
---
src/gallium/drivers/radeonsi/si_shader.c | 29
On Tue, Jul 08, 2014 at 01:37:03AM +0200, Marek Olšák wrote:
From: Marek Olšák marek.ol...@amd.com
Reviewed-by: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/radeonsi/si_blit.c| 2 +-
src/gallium/drivers/radeonsi/si_descriptors.c | 12 +-
src/gallium
growing the pool faster, and we can also save
64KB of memory that were allocated for the 'shadow', even if they
weren't used.
Reviewed-by: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/r600/compute_memory_pool.c | 61
++
1 file changed, 43 insertions
On Thu, Jul 03, 2014 at 10:33:41PM -0500, Darius Goad wrote:
Hello. I'm trying to get my feet wet with Mesa, and I was wondering
what some good tasks for me would be. Thanks again.
What hardware do you have? What are you interested in working on?
-Tom
- Darius Goad
On Fri, Jul 04, 2014 at 12:28:05PM +0200, Francisco Jerez wrote:
Tom Stellard t...@stellard.net writes:
On Fri, Jul 04, 2014 at 12:28:20AM +0200, Francisco Jerez wrote:
Tom Stellard t...@stellard.net writes:
On Thu, Jul 03, 2014 at 01:12:07AM +0200, Francisco Jerez wrote:
Tom
On Fri, Jul 04, 2014 at 05:25:42PM +0200, Francisco Jerez wrote:
Tom Stellard t...@stellard.net writes:
On Fri, Jul 04, 2014 at 12:28:05PM +0200, Francisco Jerez wrote:
Tom Stellard t...@stellard.net writes:
On Fri, Jul 04, 2014 at 12:28:20AM +0200, Francisco Jerez wrote:
Tom
On Thu, Jul 03, 2014 at 06:26:04PM +0200, Marek Olšák wrote:
From: Marek Olšák marek.ol...@amd.com
Reviewed-by: Tom Stellard thomas.stell...@amd.com
It was conflicting with def TEX_SHADOW_ARRAY, which also handles them.
---
lib/Target/R600/R600Instructions.td | 2 +-
1 file changed, 1
On Thu, Jul 03, 2014 at 01:12:07AM +0200, Francisco Jerez wrote:
Tom Stellard t...@stellard.net writes:
On Thu, Jun 26, 2014 at 04:15:39PM +0200, Francisco Jerez wrote:
Tom Stellard thomas.stell...@amd.com writes:
v2:
- Report correct values
!work_group_size_hint, i32 4, i32 1, i32 1}
Attempting to get the kernel without the correct number of attributes led
to memory corruption and luxrays crashing out.
Fixes the cl/program/execute/attributes.cl piglit test.
Thanks for tracking this down.
Reviewed-by: Tom Stellard thomas.stell
On Thu, Jul 03, 2014 at 10:56:24AM -0400, Tom Stellard wrote:
On Wed, Jul 02, 2014 at 04:34:24PM -0500, Aaron Watry wrote:
Previously, we were assuming that kernel metadata nodes only had 1 operand.
Kernels which have attributes can have more than 1, e.g.:
!0 = metadata !{void (i32
On Thu, Jul 03, 2014 at 11:59:00AM -0400, Alex Deucher wrote:
On Thu, Jul 3, 2014 at 11:46 AM, Aaron Watry awa...@gmail.com wrote:
On Thu, Jul 3, 2014 at 9:56 AM, Tom Stellard t...@stellard.net wrote:
On Wed, Jul 02, 2014 at 04:34:24PM -0500, Aaron Watry wrote:
Previously, we were assuming
.
If you enable image support, I believe that it is still going to fail
due to mismatches/oddness with the number of supported pixel formats.
--Aaron
Reviewed-by: Tom Stellard thomas.stell...@amd.com
Signed-off-by: Aaron Watry awa...@gmail.com
CC: Tom Stellard thomas.stell
On Fri, Jul 04, 2014 at 12:28:20AM +0200, Francisco Jerez wrote:
Tom Stellard t...@stellard.net writes:
On Thu, Jul 03, 2014 at 01:12:07AM +0200, Francisco Jerez wrote:
Tom Stellard t...@stellard.net writes:
On Thu, Jun 26, 2014 at 04:15:39PM +0200, Francisco Jerez wrote:
Tom
was released last week, so
you don't have to wait until LLVM 3.5.
-Tom
Marek
On Fri, Jun 27, 2014 at 5:26 PM, Tom Stellard t...@stellard.net wrote:
On Tue, Jun 17, 2014 at 01:51:10AM +0200, Marek Olšák wrote:
Since LLVM 3.5 will be released in August and my radeon patches adding
Which LLVM
On Wed, Jul 02, 2014 at 06:29:25PM +0200, Laurent Carlier wrote:
Le mercredi 25 juin 2014, 11:58:47 Michel Dänzer a écrit :
On 25.06.2014 09:15, Tom Stellard wrote:
https://bugs.freedesktop.org/show_bug.cgi?id=80015
CC: 10.1 10.2 mesa-sta...@lists.freedesktop.org
---
src
On Thu, Jun 26, 2014 at 04:15:39PM +0200, Francisco Jerez wrote:
Tom Stellard thomas.stell...@amd.com writes:
v2:
- Report correct values for CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE and
CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE.
- Only define cl_khr_fp64 if the extension is supported
v2:
- Report correct values for CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE
and CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE.
- Only define cl_khr_fp64 if the extension is supported.
- Remove trailing space from extension string.
- Rename device query function from cl_khr_fp86() to
---
src/gallium/state_trackers/clover/api/device.cpp | 3 +--
src/gallium/state_trackers/clover/core/device.cpp | 6 ++
src/gallium/state_trackers/clover/core/device.hpp | 1 +
3 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/src/gallium/state_trackers/clover/api/device.cpp
On Tue, Jun 17, 2014 at 01:51:10AM +0200, Marek Olšák wrote:
Since LLVM 3.5 will be released in August and my radeon patches adding
Which LLVM patches are required for ARB_draw_indirect?
-Tom
ARB_draw_indirect depend on it, I will commit ARB_draw_indirect
support for Gallium with softpipe
v2:
- Report correct values for CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE and
CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE.
- Only define cl_khr_fp64 if the extension is supported.
- Remove trailing space from extension string.
- Rename device query function from cl_khr_fp86() to has_doubles().
From: Francisco Jerez curroje...@riseup.net
---
src/gallium/state_trackers/clover/api/kernel.cpp | 4 +++-
src/gallium/state_trackers/clover/util/compat.hpp | 8
2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/src/gallium/state_trackers/clover/api/kernel.cpp
This is for reporting whether or not double precision floating-point
operations are supported.
Reviewed-by: Francisco Jerez curroje...@riseup.net
---
src/gallium/auxiliary/gallivm/lp_bld_limits.h | 2 ++
src/gallium/auxiliary/tgsi/tgsi_exec.h| 2 ++
src/gallium/docs/source/screen.rst
From: Matt Arsenault arse...@gmail.com
v2:
-Fix indentation
---
src/gallium/state_trackers/clover/api/device.cpp | 11 +--
src/gallium/state_trackers/clover/core/device.cpp | 24 +++
src/gallium/state_trackers/clover/core/device.hpp | 3 +++
3 files changed, 36
From: Matt Arsenault arse...@gmail.com
If there were only warnings, they would not be added to the log.
Also fixes valgrind use after free errors.
v2:
- Use compat::string.
---
src/gallium/state_trackers/clover/core/compiler.hpp | 3 ++-
src/gallium/state_trackers/clover/core/error.hpp
caused
This fixes crashes and assertion failures in two gegl tests.
Signed-off-by: Jan Vesely jan.ves...@rutgers.edu
CC: Bruno Jimenez brunoji...@gmail.com
CC: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/r600/compute_memory_pool.c | 3 ++-
1 file
https://bugs.freedesktop.org/show_bug.cgi?id=80015
CC: 10.1 10.2 mesa-sta...@lists.freedesktop.org
---
src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/drivers/radeon/radeon_setup_tgsi_llvm.c
On Sat, Jun 21, 2014 at 06:33:17PM +0200, Francisco Jerez wrote:
Tom Stellard thomas.stell...@amd.com writes:
From: Matt Arsenault arse...@gmail.com
If there were only warnings, they would not be added to the log.
Also fixes valgrind use after free errors.
---
src/gallium
On Sun, Jun 22, 2014 at 04:05:49PM +0200, Francisco Jerez wrote:
Bruno Jimenez brunoji...@gmail.com writes:
On Sat, 2014-06-21 at 17:39 +0200, Francisco Jerez wrote:
[...]
The implementation of PIPE_TRANSFER_MAP_DIRECTLY introduced in PATCH 10
has somewhat worrying semantics: A mapping
On Wed, Jun 18, 2014 at 05:01:50PM +0200, Bruno Jiménez wrote:
Hi,
This is my second attempt to fix the mapping bug adding all the
suggestions that Tom Stellard sent, and, so far, it seems that
it is resolved.
This series changes completely how OpenCL buffers are handled
by the r600g
On Fri, Jun 20, 2014 at 07:14:43PM -0500, Aaron Watry wrote:
Signed-off-by: Aaron Watry awa...@gmail.com
Reviewed-by: Tom Stellard thomas.stell...@amd.com
---
src/gallium/auxiliary/gallivm/lp_bld_debug.cpp | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/gallium
On Thu, Jun 19, 2014 at 03:53:42PM +0900, Michel Dänzer wrote:
From: Michel Dänzer michel.daen...@amd.com
I just pushed this patch to fix the regressions. We can update the
other intrinsics in a follow on patch.
We also need to change RSQ to use llvm.AMDGPU.rsq.clamped once
On Thu, Jun 19, 2014 at 10:21:32AM -0400, Jan Vesely wrote:
The important part is the change of the condition to = 0. Otherwise the loop
gets stuck never actually growing the pool.
The change in the aux-need calculation guarantees max 2 iterations, and
avoids wasting memory in case a smaller
On Thu, Jun 19, 2014 at 11:22:28AM -0400, Jan Vesely wrote:
On Thu, 2014-06-19 at 17:12 +0200, Bruno Jimenez wrote:
Hi,
To which failure are you refering? Could you please send me a
test/program that I can try to track this down?
well, the compute_memory_finalize_pending() function
, so that it prints an error message when it fails to load a
shared object.
Other than that, it's hard to review a series like this, but I'll
give it an:
Acked-by: Tom Stellard thomas.stell...@amd.com
Since I like the general approach.
-Tom
In a nutshell:
- Convert one target per patch
On Wed, Jun 18, 2014 at 09:54:28AM +1000, Dave Airlie wrote:
On 18 June 2014 05:08, Roland Scheidegger srol...@vmware.com wrote:
This looks ok to me though since tgsi currently doesn't have any double
opcodes (well the docs have them...) it doesn't really apply to most
drivers (at least I
We were printing messages for all diagnostic types, which was
spamming the console for some OpenCL programs.
---
src/gallium/drivers/radeon/radeon_llvm_emit.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/src/gallium/drivers/radeon/radeon_llvm_emit.c
that
was by replacing the intrinsic with an SDNode.
-Tom
Marek
On Mon, Jun 16, 2014 at 5:45 PM, Tom Stellard t...@stellard.net wrote:
On Thu, Jun 12, 2014 at 02:11:10AM +0200, Marek Olšák wrote:
From: Marek Olšák marek.ol...@amd.com
This adds a new type of intrinsic and SDNode: SampleRaw.
All fields
This is for reporting whether or not double precision floating-point
operations are supported.
---
src/gallium/auxiliary/gallivm/lp_bld_limits.h | 2 ++
src/gallium/auxiliary/tgsi/tgsi_exec.h| 2 ++
src/gallium/docs/source/screen.rst| 2 ++
From: Matt Arsenault arse...@gmail.com
---
src/gallium/state_trackers/clover/api/device.cpp | 11 +--
src/gallium/state_trackers/clover/core/device.cpp | 24 +++
src/gallium/state_trackers/clover/core/device.hpp | 3 +++
3 files changed, 36 insertions(+), 2
---
src/gallium/state_trackers/clover/api/device.cpp | 4 +++-
src/gallium/state_trackers/clover/core/device.cpp | 6 ++
src/gallium/state_trackers/clover/core/device.hpp | 1 +
3 files changed, 10 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/clover/api/device.cpp
An LLVMContext should only be accessed by a single and using the global
context was causing crashes in multi-threaded environments. Now we use
a separate context for each compile.
---
src/gallium/state_trackers/clover/llvm/invocation.cpp | 15 +--
1 file changed, 9 insertions(+), 6
From: Matt Arsenault arse...@gmail.com
If there were only warnings, they would not be added to the log.
Also fixes valgrind use after free errors.
---
src/gallium/state_trackers/clover/core/compiler.hpp | 3 ++-
src/gallium/state_trackers/clover/core/error.hpp | 2 +-
On Fri, Jun 13, 2014 at 10:35:30PM +0200, Bruno Jiménez wrote:
This patch changes completely the way buffers are added to the
compute_memory_pool. Before this, whenever we were going to
map a buffer or write to or read from it, it would get placed
into the pool. Now, every unallocated buffer
On Fri, Jun 13, 2014 at 10:35:31PM +0200, Bruno Jiménez wrote:
These statuses will help track whether the items are mapped
or if they should be promoted to or demoted from the pool
---
src/gallium/drivers/r600/compute_memory_pool.h | 7 ++-
src/gallium/drivers/r600/evergreen_compute.c
On Fri, Jun 13, 2014 at 10:35:32PM +0200, Bruno Jiménez wrote:
Now we will have a list with the items that are in the pool
(item_list) and the items that are outside it (unallocated_list)
Reviewed-by: Tom Stellard thomas.stell...@amd.com
---
src/gallium/drivers/r600/compute_memory_pool.c
On Sun, Jun 15, 2014 at 01:08:14PM +0200, Francisco Jerez wrote:
Tom Stellard thomas.stell...@amd.com writes:
We were serializing the binaries once when clGetProgramInfo was called
with CL_PROGRAM_BINARY_SIZES and then again when it was called with
CL_PROGRAM_BINARIES. This was slowing
401 - 500 of 1623 matches
Mail list logo