Re: [Mesa-dev] [PATCH] st/clover: Define __OPENCL_VERSION__ on the device side

On Wednesday, 31 August 2016, 15:53:05 CEST, Serge Martin wrote: > On Wednesday 31 August 2016 12:39:23 Vedran Miletić wrote: > > On 08/28/2016 04:42 PM, Niels Ole Salscheider wrote: > > > This is required by the OpenCL standard. > > > > > > Signed-off-by: Niel

[Mesa-dev] [PATCH] st/clover: Define __OPENCL_VERSION__ on the device side

This is required by the OpenCL standard. Signed-off-by: Niels Ole Salscheider --- src/gallium/state_trackers/clover/llvm/invocation.cpp | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/gallium/state_trackers/clover/llvm/invocation.cpp b/src/gallium/state_trackers/clover/llvm

[Mesa-dev] [PATCH] winsys/radeon: Do not deinit the pb cache if it was not initialized

This fixes a crash in pb_cache_release_all_buffers. Signed-off-by: Niels Ole Salscheider --- src/gallium/winsys/radeon/drm/radeon_drm_winsys.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/gallium/winsys/radeon/drm/radeon_drm_winsys.c b/src/gallium/winsys/radeon

Re: [Mesa-dev] [PATCH] configure: Link against all LLVM targets when building clover

On Saturday 24 January 2015, 18:24:16, Jan Vesely wrote: > On Sat, 2015-01-24 at 22:49 +0100, Niels Ole Salscheider wrote: > > Since 8e7df519bd8556591794b2de08a833a67e34d526, we initialise all targets > > in clover. This fixes bug 85380. > > > > v2: Mention co

[Mesa-dev] [PATCH] configure: Link against all LLVM targets when building clover

Since 8e7df519bd8556591794b2de08a833a67e34d526, we initialise all targets in clover. This fixes bug 85380. v2: Mention correct bug in commit message Signed-off-by: Niels Ole Salscheider --- configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configure.ac b

Re: [Mesa-dev] [PATCH] configure: Link against all LLVM targets when building clover

On Thursday 22 January 2015, 13:46:14, Jan Vesely wrote: > On Thu, 2015-01-22 at 16:45 +, Emil Velikov wrote: > > On 15/01/15 21:38, Tom Stellard wrote: > > > On Thu, Jan 15, 2015 at 07:25:56PM +0100, Niels Ole Salscheider wrote: > > >> Since 8e7df519bd8556

[Mesa-dev] [PATCH] configure: Link against all LLVM targets when building clover

Since 8e7df519bd8556591794b2de08a833a67e34d526, we initialise all targets in clover. This fixes bug 85189. Signed-off-by: Niels Ole Salscheider --- configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configure.ac b/configure.ac index c72fe92..1761c32 100644 --- a

Re: [Mesa-dev] [PATCH v2] r600: Use DMA transfers in r600_copy_global_buffer

, it is used from clEnqueueCopyBuffer - and it does not regress any piglit tests for me. Can someone with write access please push this? Thanks, Ole > On Mon, 2014-09-08 at 20:10 +0200, Niels Ole Salscheider wrote: > > v2: Do not demote items that are already in th

Re: [Mesa-dev] [PATCH 1/4] radeonsi/compute: directly emit CONTEXT_CONTROL

sktop.org > > http://lists.freedesktop.org/mailman/listinfo/mesa-dev > > ___ > mesa-dev mailing list > mesa-dev@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/mesa-dev >From 9714d3ee55ee0ddb0bcf63934b552df641b866a2

Re: [Mesa-dev] [PATCH] r600: Use DMA transfers in r600_copy_global_buffer

On Tuesday 09 September 2014, 11:40:49, Bruno Jimenez wrote: > On Mon, 2014-09-08 at 18:30 +0200, Niels Ole Salscheider wrote: > > On Monday 08 September 2014, 15:19:15, Bruno Jimenez wrote: > > > Hi, > > > > > > I'm not sure if this will work. Imagine th

[Mesa-dev] [PATCH v2] r600: Use DMA transfers in r600_copy_global_buffer

v2: Do not demote items that are already in the pool Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/r600/evergreen_compute.h | 1 + src/gallium/drivers/r600/r600_blit.c | 59 2 files changed, 43 insertions(+), 17 deletions(-) diff --git a/src

Re: [Mesa-dev] [PATCH] r600: Use DMA transfers in r600_copy_global_buffer

to directly access the pool bo if the item is already in it. > - Bruno > > On Sun, 2014-09-07 at 18:32 +0200, Niels Ole Salscheider wrote: > > Signed-off-by: Niels Ole Salscheider > > --- > > > > src/gallium/drivers/r600/evergreen_compute.c | 27 -

[Mesa-dev] [PATCH] r600: Use DMA transfers in r600_copy_global_buffer

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/r600/evergreen_compute.c | 27 --- src/gallium/drivers/r600/evergreen_compute.h | 1 + src/gallium/drivers/r600/r600_blit.c | 40 3 files changed, 41 insertions(+), 27 deletions

[Mesa-dev] [PATCH] gallium/radeon: Do not use u_upload_mgr for buffer downloads

RAM) does not offer any performance improvements for buffer downloads. Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeon/r600_buffer_common.c | 20 1 file changed, 8 insertions(+), 12 deletions(-) diff --git a/src/gallium/drivers/radeon/r600_buffer_common.c b/

[Mesa-dev] [PATCH] gallium/radeon: Set gpu_address to 0 if r600_virtual_address is false

Without this patch I get the following during DMA transfers: [drm:radeon_cs_ib_chunk] *ERROR* Invalid command stream ! radeon :01:00.0: CP DMA dst buffer too small (21475829792 4096) This is a fixup for e878e154cdfd4dbb5474f776e0a6d86fcb983098. Signed-off-by: Niels Ole Salscheider --- src

Re: [Mesa-dev] [PATCH 0/2] radeon: Use the DMA engine for buffer downloads

get any performance improvements. Would it now be OK to change u_upload_mgr or do you have a better proposal? Ole > Marek > > On Mon, Mar 3, 2014 at 9:29 PM, Niels Ole Salscheider > > wrote: > > Using the DMA engine for buffer downloads vastly improves performance. > >

Re: [Mesa-dev] [PATCH 1/2] clover: Call end_query before getting timestamp result v2

Reviewed-by: Niels Ole Salscheider On Wednesday 16 July 2014, 17:37:48, Tom Stellard wrote: > 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/ti

Re: [Mesa-dev] [PATCH 1/2] clover: Call end_query before getting timestamp result

On Wednesday 16 July 2014, 16:49:08, Tom Stellard wrote: > 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. I think you should call end_query in the constructor right after the call to create_quer

[Mesa-dev] [PATCH v2] egl/gallium: Set defines for supported APIs when using automake

This fixes automake builds which are broken since b52a530ce2aada1967bc8fefa83ab53e6a737dae. v2: This patch also adds the FEATURE_* defines back to targets/egl-static for Android and Scons that have been removed in the mentioned commit. Signed-off-by: Niels Ole Salscheider Bugzilla: https

Re: [Mesa-dev] [PATCH] egl/gallium: Set defines for supported APIs when using automake

On Tuesday 10 June 2014, 16:18:56, Emil Velikov wrote: > On 10/06/14 15:17, Niels Ole Salscheider wrote: > > This fixes automake builds which are broken since > > b52a530ce2aada1967bc8fefa83ab53e6a737dae. > > Not sure what I was smoking with the above mentioned patch. > S

[Mesa-dev] [PATCH] egl/gallium: Set defines for supported APIs when using automake

This fixes automake builds which are broken since b52a530ce2aada1967bc8fefa83ab53e6a737dae. Signed-off-by: Niels Ole Salscheider --- src/gallium/state_trackers/egl/Makefile.am | 20 1 file changed, 20 insertions(+) diff --git a/src/gallium/state_trackers/egl/Makefile.am b

Re: [Mesa-dev] [PATCH 2/2] radeonsi: Implement DMA blit

;. > > 5) Flushing the gfx CS is not required, because r600_context_bo_reloc > flushes it for you. > > Please see also my latest DMA patches for r600g. > > Thanks. > > Marek > > On Thu, Mar 13, 2014 at 8:45 AM, Niels Ole Salscheider > > wrote: > >

[Mesa-dev] [PATCH 1/2] radeon: Move r600_need_dma_space to common code

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/r600/evergreen_hw_context.c | 2 +- src/gallium/drivers/r600/evergreen_state.c | 2 +- src/gallium/drivers/r600/r600_hw_context.c | 12 +--- src/gallium/drivers/r600/r600_pipe.h| 1 - src/gallium

[Mesa-dev] [PATCH 2/2] radeonsi: Implement DMA blit

re if they should be renamed. Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeonsi/Makefile.sources | 1 + src/gallium/drivers/radeonsi/si_dma.c | 352 ++ src/gallium/drivers/radeonsi/si_pipe.h| 9 + src/gallium/drivers/radeonsi/si_state.c

[Mesa-dev] [PATCH 2/2] radeonsi: flush the dma ring in si_flush_from_st

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeonsi/si_pipe.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/src/gallium/drivers/radeonsi/si_pipe.c b/src/gallium/drivers/radeonsi/si_pipe.c index 827e9fe..401bf6a 100644 --- a/src/gallium/drivers/radeonsi/si_pipe.c

[Mesa-dev] [PATCH 1/2] radeon: Move DMA ring creation to common code

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/r600/r600_pipe.c | 30 - src/gallium/drivers/r600/r600_pipe.h | 1 - src/gallium/drivers/radeon/r600_pipe_common.c | 32 +++ src/gallium/drivers/radeon

[Mesa-dev] [PATCH 1/2] radeonsi: Add DMA ring

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeonsi/si_hw_context.c | 3 +++ src/gallium/drivers/radeonsi/si_pipe.c | 22 ++ 2 files changed, 25 insertions(+) diff --git a/src/gallium/drivers/radeonsi/si_hw_context.c b/src/gallium/drivers/radeonsi

[Mesa-dev] [PATCH 2/2] radeonsi: Implement DMA blit

re if they should be renamed. Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeonsi/si_hw_context.c | 65 +++ src/gallium/drivers/radeonsi/si_pipe.h | 7 + src/gallium/drivers/radeonsi/si_state.c | 265 ++- src/gallium/drivers/rade

Re: [Mesa-dev] [PATCH 1/3] r600g, radeonsi: use a fallback in dma_copy instead of failing

You are right, r600-r700 require dword alignment while linear copies can be byte aligned on EG+. Apart from that, patch 1 and 2 look good to me... Ole ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo

Re: [Mesa-dev] [PATCH 1/3] r600g, radeonsi: use a fallback in dma_copy instead of failing

On Sunday 09 March 2014, 02:24:51, Marek Olšák wrote: > From: Marek Olšák > > --- > src/gallium/drivers/r600/evergreen_state.c | 37 +--- > src/gallium/drivers/r600/r600_state.c | 41 ++--- > src/gallium/drivers/radeon/r600_buffer_common.c | 58 > ++

Re: [Mesa-dev] [PATCH v2] radeon: Use upload manager for buffer downloads

s yet. > > Other than those two, the patch looks good. > > CP DMA != async DMA (dma_copy). CP DMA is actually a feature of the > graphics ring. > > Marek > > On Tue, Mar 4, 2014 at 6:23 PM, Niels Ole Salscheider > > wrote: > > Using DMA for

[Mesa-dev] [PATCH v3] radeon: Use upload manager for buffer downloads

Using DMA for reads is much faster. Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeon/r600_buffer_common.c | 74 +++-- 1 file changed, 56 insertions(+), 18 deletions(-) diff --git a/src/gallium/drivers/radeon/r600_buffer_common.c b/src/gallium/drivers

Re: [Mesa-dev] [PATCH 0/2] radeon: Use the DMA engine for buffer downloads

> Could you please do this without changing u_upload_mgr? You can still > use u_upload_alloc to allocate buffer memory in the driver and the map > buffer read/write flags are not important with persistent coherent > buffer mappings anyway. I have sent an updated patch to the list. Ole ___

[Mesa-dev] [PATCH v2] radeon: Use upload manager for buffer downloads

Using DMA for reads is much faster. Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeon/r600_buffer_common.c | 78 +++-- 1 file changed, 60 insertions(+), 18 deletions(-) diff --git a/src/gallium/drivers/radeon/r600_buffer_common.c b/src/gallium/drivers

[Mesa-dev] [PATCH 0/2] radeon: Use the DMA engine for buffer downloads

613.49 MB/s(DD) 1841.07 MB/s(DH) Judging by these numbers, it might even make sense to use the DMA engine for larger buffer downloads... Niels Ole Salscheider (2): util/u_upload_mgr: Allow to also use it for downloads radeon: Use transfer manager for buffer downloads src/g

[Mesa-dev] [PATCH 2/2] radeon: Use transfer manager for buffer downloads

Using DMA for reads is much faster. Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeon/r600_buffer_common.c | 78 +++-- src/gallium/drivers/radeon/r600_pipe_common.c | 11 src/gallium/drivers/radeon/r600_pipe_common.h | 1 + 3 files changed, 72

[Mesa-dev] [PATCH 1/2] util/u_upload_mgr: Allow to also use it for downloads

Signed-off-by: Niels Ole Salscheider --- src/gallium/auxiliary/hud/hud_context.c | 3 +- src/gallium/auxiliary/util/u_blitter.c| 3 +- src/gallium/auxiliary/util/u_upload_mgr.c | 49 +++ src/gallium/auxiliary/util/u_upload_mgr.h | 13 +-- src

[Mesa-dev] [PATCH] r600: compute memory pool size is given in dw

Multiply the dw value by 4 in order to map the complete buffer. Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/r600/compute_memory_pool.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/gallium/drivers/r600/compute_memory_pool.c b/src/gallium/drivers

[Mesa-dev] [PATCH] winsys/radeon: remove superfluous distinction of cases

Signed-off-by: Niels Ole Salscheider --- src/gallium/winsys/radeon/drm/radeon_drm_cs.c | 20 +--- 1 file changed, 5 insertions(+), 15 deletions(-) diff --git a/src/gallium/winsys/radeon/drm/radeon_drm_cs.c b/src/gallium/winsys/radeon/drm/radeon_drm_cs.c index acb12b2..d8ad297

Re: [Mesa-dev] [PATCH] clover: Calculate the optimal work group size when local_size is NULL

Hi Tom, this has been on my todo list for quite a while. Your patch looks good to me, but in my experience a block with approximately the same size for each dimension gives slightly better performance in many cases when compared to one where one dimension is significantly larger. Maybe you coul

Re: [Mesa-dev] [PATCH 2/2] st/clover: Always flush the queue when waiting on an hard_event

Am Donnerstag, 3. Oktober 2013, 11:08:26 schrieb Francisco Jerez: > Niels Ole Salscheider writes: > >> Do you have any example of a real world application that relies on this? > >> Or at least some reasonable use case? > > > > The problem is that the queue is on

Re: [Mesa-dev] [PATCH 2/2] st/clover: Always flush the queue when waiting on an hard_event

> Do you have any example of a real world application that relies on this? > Or at least some reasonable use case? The problem is that the queue is only cleared from already signalled events when we flush it. And we might not do this if the user only calls clWaitForEvents once the corresponding

Re: [Mesa-dev] [PATCH 2/2] st/clover: Always flush the queue when waiting on an hard_event

> I don't think this is right, with this patch we remove *all* events from > the command queue, signalled or not, every time the command queue is > flushed. You are right, I got the logic wrong here (see also http://lists.freedesktop.org/archives/mesa-dev/2013-September/044363.html). The problem

Re: [Mesa-dev] [PATCH 1/2] st/clover: Clear the complete queue

Ping ___ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev

[Mesa-dev] [PATCH 1/2] st/clover: Clear the complete queue

Events that are already signalled can be removed from the queue, too. Signed-off-by: Niels Ole Salscheider --- src/gallium/state_trackers/clover/core/queue.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/state_trackers/clover/core/queue.cpp b/src/gallium

Re: [Mesa-dev] [PATCH 3/6] st/clover: Unreference fences as early as possible

While unreferencing fences as early as possible is not a bad idea, this patch hides the underlying problem. That is, events are never deleted from the queued_events list of the queue object if their fences are signalled before the queue is flushed. I will send a patch that fixes the problem shor

[Mesa-dev] [PATCH 2/2] st/clover: Always flush the queue when waiting on an hard_event

ist of the queue object. Signed-off-by: Niels Ole Salscheider --- src/gallium/state_trackers/clover/core/event.cpp | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/gallium/state_trackers/clover/core/event.cpp b/src/gallium/state_trackers/clover/core/event.cpp index cbb97b

[Mesa-dev] [PATCH] radeonsi: Do not suspend timer queries

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeonsi/r600.h| 1 + src/gallium/drivers/radeonsi/r600_hw_context.c | 28 ++ src/gallium/drivers/radeonsi/r600_query.c | 7 +-- src/gallium/drivers/radeonsi/radeonsi_pipe.c | 2 +- src

[Mesa-dev] [PATCH 2/2] R600/SI: FMA is faster than fmul and fadd for f64

--- lib/Target/R600/SIISelLowering.cpp | 18 ++ lib/Target/R600/SIISelLowering.h | 1 + test/CodeGen/R600/fmuladd.ll | 31 +++ 3 Dateien geändert, 50 Zeilen hinzugefügt(+) create mode 100644 test/CodeGen/R600/fmuladd.ll diff --git a/lib/Target

[Mesa-dev] [PATCH 1/2] R600/SI: Add FMA pattern

--- lib/Target/R600/SIInstructions.td | 8 ++-- test/CodeGen/R600/fma.ll | 31 +++ 2 Dateien geändert, 37 Zeilen hinzugefügt(+), 2 Zeilen entfernt(-) create mode 100644 test/CodeGen/R600/fma.ll diff --git a/lib/Target/R600/SIInstructions.td b/lib/Target

[Mesa-dev] [PATCH 1/6] st/clover: Profiling support

Signed-off-by: Niels Ole Salscheider --- src/gallium/state_trackers/clover/api/event.cpp | 26 - src/gallium/state_trackers/clover/core/event.cpp | 116 --- src/gallium/state_trackers/clover/core/event.hpp | 18 +++- 3 Dateien geändert, 142 Zeilen hinzugefügt(+), 18

[Mesa-dev] [PATCH 6/6] radeonsi: Handle additional PIPE_COMPUTE_CAP_*

This patch adds support for: PIPE_COMPUTE_CAP_MAX_INPUT_SIZE PIPE_COMPUTE_CAP_MAX_LOCAL_SIZE Return the values reported by the closed source driver for now. Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeonsi/radeonsi_pipe.c | 15 ++- 1 Datei geändert, 14 Zeilen

[Mesa-dev] [PATCH 2/6] st/clover: Add event to deps even if it has been triggered

The command is submitted once the event has been triggered, but it might not have completed yet. Therefore, we have to add it to deps in order to wait on it. Signed-off-by: Niels Ole Salscheider --- src/gallium/state_trackers/clover/core/event.cpp | 2 +- 1 Datei geändert, 1 Zeile hinzugefügt

[Mesa-dev] [PATCH 4/6] radeonsi: Implement PIPE_QUERY_TIMESTAMP

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeonsi/r600.h| 1 + src/gallium/drivers/radeonsi/r600_hw_context.c | 31 ++ src/gallium/drivers/radeonsi/r600_query.c | 14 +++- src/gallium/drivers/radeonsi/radeonsi_pipe.c | 2

[Mesa-dev] [PATCH 5/6] radeonsi: copy r600_get_timestamp

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/radeonsi/radeonsi_pipe.c | 9 + 1 Datei geändert, 9 Zeilen hinzugefügt(+) diff --git a/src/gallium/drivers/radeonsi/radeonsi_pipe.c b/src/gallium/drivers/radeonsi/radeonsi_pipe.c index 3ba8232..7ae5598 100644 --- a/src

[Mesa-dev] [PATCH 3/6] st/clover: Unreference fences as early as possible

This makes sure that there are not too many concurrent fences. Also, simplify status handling by keeping track of the current state. Signed-off-by: Niels Ole Salscheider --- src/gallium/state_trackers/clover/core/event.cpp | 29 +++- src/gallium/state_trackers/clover/core

[Mesa-dev] [PATCH 2/2] R600/SI: Implement fp32<->fp64 conversions

--- lib/Target/R600/SIISelLowering.cpp | 3 +++ lib/Target/R600/SIInstructions.td | 8 ++-- test/CodeGen/R600/fpext.ll | 9 + test/CodeGen/R600/fptrunc.ll | 9 + 4 Dateien geändert, 27 Zeilen hinzugefügt(+), 2 Zeilen entfernt(-) create mode 100644 test/CodeGen/R

[Mesa-dev] [PATCH 1/2] R600/SI: Implement sint<->fp64 conversions

--- lib/Target/R600/SIInstrInfo.td| 6 ++ lib/Target/R600/SIInstructions.td | 8 ++-- test/CodeGen/R600/fp64_to_sint.ll | 9 + test/CodeGen/R600/sint_to_fp64.ll | 9 + 4 Dateien geändert, 30 Zeilen hinzugefügt(+), 2 Zeilen entfernt(-) create mode 100644 test/CodeGen/R6

[Mesa-dev] [PATCH] clover: Fix linkage of libOpenCL

Clover needs the option component of llvm. Signed-off-by: Niels Ole Salscheider --- configure.ac | 4 1 Datei geändert, 4 Zeilen hinzugefügt(+) diff --git a/configure.ac b/configure.ac index 62d06e0..0dcd2a5 100644 --- a/configure.ac +++ b/configure.ac @@ -1617,6 +1617,10 @@ if test &q

Re: [Mesa-dev] R600/SI: Initial double precision support for Radeon SI

alar move in patch #2 by a vector move since there is probably no point in having a floating point value in a scalar register. Kind regards, Ole>From 4224b314cf2d97cdf2ac99564d6155fa04fbb971 Mon Sep 17 00:00:00 2001 From: Niels Ole Salscheider Date: Sat, 1 Jun 2013 16:48:56 +0200 Subject

[Mesa-dev] R600/SI: Initial double precision support for Radeon SI

double values. Regards, Ole>From 4224b314cf2d97cdf2ac99564d6155fa04fbb971 Mon Sep 17 00:00:00 2001 From: Niels Ole Salscheider Date: Sat, 1 Jun 2013 16:48:56 +0200 Subject: [PATCH 1/6] R600/SI: Add initial double precision support for SI --- lib/Target/R600/AMDGPUISelLowering.cpp |

[Mesa-dev] [PATCH] st/clover: Allow double precision operations

Pass "cl_khr_fp64" preprocessor definition to clang Signed-off-by: Niels Ole Salscheider --- src/gallium/state_trackers/clover/llvm/invocation.cpp | 1 + 1 Datei geändert, 1 Zeile hinzugefügt(+) diff --git a/src/gallium/state_trackers/clover/llvm/invocation.cpp b/src/gallium/stat

Re: [Mesa-dev] [PATCH 5/5] radeonsi/compute: Upload work group, work item size in input buffer

Am Freitag, 24. Mai 2013, 14:07:29 schrieb Tom Stellard: > From: Tom Stellard > > --- > src/gallium/drivers/radeonsi/radeonsi_compute.c | 38 > ++--- 1 file changed, 27 insertions(+), 11 deletions(-) > > diff --git a/src/gallium/drivers/radeonsi/radeonsi_compute.c > b/src/gal

[Mesa-dev] [PATCH] r600g: fixup for MSAA texture support checking

Signed-off-by: Niels Ole Salscheider --- src/gallium/drivers/r600/r600_shader.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/gallium/drivers/r600/r600_shader.c b/src/gallium/drivers/r600/r600_shader.c index 4e5af70..4d74db0 100644 --- a/src/gallium/drivers/r600

Re: [Mesa-dev] [PATCH] gallium/opencl: Fix out-of-tree build

Am Dienstag, 9. April 2013, 11:17:39 schrieb Michel Dänzer: > From: Michel Dänzer > > > Signed-off-by: Michel Dänzer > --- > src/gallium/targets/opencl/Makefile.am | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/src/gallium/targets/opencl/Makefile.am > b/src/galli

[Mesa-dev] [PATCH] clover: Fix linkage of libOpenCL

Clover needs the irreader component of llvm v2: Check for irreader component irreader is only available with LLVM 3.3 >= 177971 Signed-off-by: Niels Ole Salscheider --- configure.ac | 4 1 file changed, 4 insertions(+) diff --git a/configure.ac b/configure.ac index 81d4a3f..fea5

[Mesa-dev] [PATCH] clover: Fix linkage of libOpenCL

Clover needs the irreader component of llvm --- configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configure.ac b/configure.ac index 81d4a3f..bfba1b3 100644 --- a/configure.ac +++ b/configure.ac @@ -1649,7 +1649,7 @@ if test "x$enable_gallium_llvm" = xyes; then

Re: [Mesa-dev] [PATCH] pipe-loader: Fix out of source build

Am Sonntag, 24. Februar 2013, 15:02:33 schrieb Matt Turner: > On Sun, Feb 24, 2013 at 2:00 PM, Niels Ole Salscheider > > wrote: > > Signed-off-by: Niels Ole Salscheider > > --- > > > > src/gallium/targets/opencl/Makefile.am | 4 ++-- > > 1 fil

[Mesa-dev] [PATCH] pipe-loader: Fix out of source build

Signed-off-by: Niels Ole Salscheider --- src/gallium/targets/opencl/Makefile.am | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/gallium/targets/opencl/Makefile.am b/src/gallium/targets/opencl/Makefile.am index c5c3003..709112f 100644 --- a/src/gallium/targets/opencl

[Mesa-dev] [PATCH] st/mesa: index can be negative in the PROGRAM_CONSTANT case

--- src/mesa/state_tracker/st_glsl_to_tgsi.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp b/src/mesa/state_tracker/st_glsl_to_tgsi.cpp index 39717b6..9f58312 100644 --- a/src/mesa/state_tracker/st_glsl_to_tgsi.cpp +++ b/src/mesa/

[Mesa-dev] Commit 683ef52e19576f6e1263bc7d25fc9475c519eade / Assertion `type != WRITE_COLOR' failed.

Hello, when I try to run any OpenGL program like glxgears I get the following (radeon x700, gallium, mesa git): r300_state_derived.c:214:r300_rs_col_write: Assertion `type != WRITE_COLOR' failed. This assertion was added by Marek Olšák in commit 683ef52e19576f6e1263bc7d25fc9475c519eade. Unfortun

Re: [Mesa-dev] [PATCH] dri2GetGlxDrawableFromXDrawableId may return NULL

Hello, > > Since dri2GetGlxDrawableFromXDrawableId may return NULL we > > should only dereference the returned pointer if it is not NULL. > > It shouldn't return NULL... when does this happen? During shutdown? > We only get the DRI2 events for drawables we've created a DRI2 > drawable for, which

[Mesa-dev] [PATCH] check if glx_displays->dpy == dpy and update glx_display

Hello, in src/glx/glxext.c line 146 following dpy should be removed from the glx_displays list. But it does not handle the case when dpy is the first item in glx_displays or when glx_displays is empty. This should be fixed by the attached patch. Signed-of-by: Niels Ole Salscheider Kind

Re: [Mesa-dev] [PATCH] dri2GetGlxDrawableFromXDrawableId may return NULL

And finally some further functions I missed... All three patches are Signed-off-by: Niels Ole Salscheider Kind regards Ole On Wednesday 28 July 2010 10:32:06 Niels Ole Salscheider wrote: > Hello, > > there's another place where the returned pointer is not checked in > sr

Re: [Mesa-dev] [PATCH] dri2GetGlxDrawableFromXDrawableId may return NULL

Hello, there's another place where the returned pointer is not checked in src/glx/dri2.c (see attachment). This might fix bug 29148. Kind regards Ole On Wednesday 28 July 2010 09:46:42 Niels Ole Salscheider wrote: > Hello, > > I have written a small patch that fixes a cras

[Mesa-dev] [PATCH] dri2GetGlxDrawableFromXDrawableId may return NULL

From: Niels Ole Salscheider Date: Wed, 28 Jul 2010 09:15:22 +0200 Subject: [PATCH] dri2GetGlxDrawableFromXDrawableId may return NULL --- src/glx/dri2_glx.c | 37 + 1 files changed, 21 insertions(+), 16 deletions(-) diff --git a/src/glx/dri2_glx.c b/src/glx