Hi Lvzhihong,
we usually track problems like these in Bugzilla so they do not get
forgotten. I made a report for you, please CC yourself there and any
details you can think of:
https://bugs.freedesktop.org/show_bug.cgi?id=104266
Regards,
Vedran
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
26.10.2017 u 01:09, Jan Vesely je napisao/la:
> On Tue, 2017-10-24 at 15:32 +0200, Vedran Miletić wrote:
>> On 10/23/2017 05:24 AM, Jan Vesely wrote:
>>> From: Jan Vesely
>>>
>>> v2: use a more generic compat function
>>>
>>> Bugzill
v2: remove/inline compat stuff
Reviewed-by: Francisco Jerez
---
.../state_trackers/clover/llvm/codegen/native.cpp | 8 +-
src/gallium/state_trackers/clover/llvm/compat.hpp | 109 +
.../state_trackers/clover/llvm/invocation.cpp | 22 +++--
.../state_trackers/clover/l
lang::LangAS::Offset;
> const clang::InputKind ik_opencl = clang::IK_OpenCL;
> #endif
>
>
Thanks for improving the patch. Future-proof thinking: what if the value
of clang::LangAS::Default changes from 0 to some other constant?
Other than that, this patch is:
Reviewed-by: Vedran Miletić
Regards,
Vedran
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
---
src/gallium/state_trackers/clover/llvm/codegen/common.cpp | 5 +++--
src/gallium/state_trackers/clover/llvm/compat.hpp | 2 +-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/src/gallium/state_trackers/clover/llvm/codegen/common.cpp
b/src/gallium/state_trackers/clover/l
---
src/gallium/state_trackers/clover/llvm/compat.hpp | 59 +-
.../state_trackers/clover/llvm/invocation.cpp | 2 -
.../state_trackers/clover/llvm/metadata.hpp| 30 +--
3 files changed, 2 insertions(+), 89 deletions(-)
diff --git a/src/gallium/state_trac
th this in mind, there a handful of dead code that we could remove.
> That will come with later commits.
>
> Note: this drops the LLVM 3.6 build from the Travis build. LLVM 3.9 (and
> later) are already covered in there.
>
> Cc: Vedran Miletić
> Cc: Jan Vesely
> Cc: Aaron Wat
" side, but given that all drivers that Clover works with
already require LLVM 3.9, I'd say yes, bump the requirement (this allows
us to get rid of three quarter of #if HAVE_LLVM version checks).
Regards,
Vedran
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
LVM >= 0x0600
> + const auto default_code_model = ::llvm::None;
> +#else
> + const auto default_code_model = ::llvm::CodeModel::Default;
> +#endif
> +
> #if HAVE_LLVM >= 0x0309
> const auto default_reloc_model = ::llvm::None;
> #else
>
This patch en
-181,41 +200,41 @@ disk_cache_create(const char *gpu_name, const char
> *timestamp,
> if (local == NULL)
>goto fail;
>
> /* At user request, disable shader cache entirely. */
> if (getenv("MESA_GLSL_CACHE_DISABLE"))
>goto fail;
>
>
+ mgr()->deregisterEHFrames(Addr, LoadAddr, Size);
> + }
> +#endif
>virtual void *getPointerToNamedFunction(const std::string &Name,
>bool AbortOnFailure=true) {
> return mgr()->get
mention of the compiler and say something along the lines of:
"C++11 requires space between < and :: as <: is an alternate spelling
for [ (see standard documents ... and ...)."
I see you caught all the occurrences.
Reviewed-by: Vedran Miletić
--
Vedran Miletić
vedran.miletic.net
Can you find it in Freedesktop Bugzilla?
Regards,
Vedran
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
t
website, not to mention easier to maintain.
As for the theme, I suggest Guzzle [1] instead; RTD is very pretty, but
it's also fairly common, as is Bootswatch. Guzzle is pretty and not so
commonly used, as far as my anecdotal experience goes.
Regards,
Vedran
[1] https://github.com/guzzle/guzzl
AttributeSet got renamed to AttributeList.
Signed-off-by: Vedran Miletić
---
src/amd/common/ac_llvm_helper.cpp | 10 --
src/gallium/auxiliary/gallivm/lp_bld_misc.cpp | 8 +++-
2 files changed, 15 insertions(+), 3 deletions(-)
diff --git a/src/amd/common
> that's a good question, so I did a bit of digging. for EG global
> buffers are shadowed in global buffer memory pool, so mapping uses
> memory pool's pipe_resource. I'm still not 100% sure where exactly
> unmapping the global pool accesses the original buffer's p
swr/rasterizer/jitter/streamout_jit.cpp
> +++ b/src/gallium/drivers/swr/rasterizer/jitter/streamout_jit.cpp
> @@ -285,7 +285,7 @@ struct StreamOutJit : public Builder
> IRB()->SetInsertPoint(entry);
>
> // arguments
> -auto argitr = soFunc->getArgum
il/build_id.h| 5 +++
> src/util/format_r11g11b10f.h | 5 +++
> src/util/register_allocate.h | 6 ++-
> src/util/strndup.c | 47
> --
> sr
VICE_TYPE_DISCRETE_GPU,
Works with Dota 2, Talos and Doom over Wine. Mesa version appears nicely
in Doom. This patch and the previous one is
Tested-by: Vedran Miletić
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@list
OpenMM holds events created during runtime in the queue and destroys
them at the end, which causes a stack overflow in nontrivial examples.
This patch forces flushing of the event queue and destruction of
events which are CL_COMPLETE.
I am pretty sure this isn't the nicest solution and probably no
On 03/09/2017 12:21 AM, Jan Vesely wrote:
> v2: buffers are created with one reference.
> Signed-off-by: Jan Vesely
> ---
> Vedran, can you confirm that this one still fixes the problem? Seems OK on EG.
>
Unfortunately it does not, valgrind reports Invalid read here.
Regards
It does; this approach was actually mentioned on IRC as an option.
Good work.
Reviewed-by: Vedran Miletić
On 03/08/2017 02:30 AM, Jan Vesely wrote:
> Signed-off-by: Jan Vesely
> ---
> Does this patch work as well?
>
> src/gallium/state_trackers/clover/core/re
Piglit's tests/cl/custom/buffer-flags.c test in-0x0-out-0x8 subtest
reports Invalid read of size 8 when ran under Valgrind due to pipe
resource being destroyed prior to the mapping of the said resource.
This patch amends the destructor to clear the mappings first.
Signed-off-by: Vedran Mi
it beyond
printing non-errors.
2) Do not print debug_message() (including si_shader_dump_stats()) via
pfn_notify.
3) Introduce separate error_message() for pfn_notify instead of
debug_message().
Regards,
Vedran
[1]
https://www.khronos.org/registry/OpenCL/sdk/1.1/docs/man/xhtml/clCreateCo
cc_cache_line_size = 64; /* TC L2 line size on GCN */
>
> -ws->check_vm = strstr(debug_get_option("R600_DEBUG", ""), "check_vm") !=
> NULL;
> +if ((strstr(debug_get_option("R600_DEBUG", ""),
uot; warnings in your browser for a bit...
>
> Eventually, the pages on www.mesa3d.org <http://www.mesa3d.org> should
> get automatically updated from git check-ins to the docs/ directory.
>
> -Brian
>
Great! We also got https in the process, very nice.
--
Vedran Miletić
vedran.m
struct si_shader *current = state->current;
> struct si_shader *iter, *shader = NULL;
>
> - if (unlikely(sscreen->b.chip_class & DBG_NO_OPT_VARIANT)) {
> + if (unlikely(sscreen->b.debug_flags & DBG_NO_OPT_VARIANT)) {
> memset(&key->opt,
On 09/19/2016 08:39 PM, Vedran Miletić wrote:
> On 09/07/2016 06:52 PM, Vedran Miletić wrote:
>> LLVM and Mesa both define the DEBUG macro in incompatible ways. As a
>> general practice, we should avoid using such generic names when it is
>> possible to do so.
>>
test.
v2: fix inconsistent code formatting
Signed-off-by: Vedran Miletić
Reviewed-by: Francisco Jerez
---
src/gallium/state_trackers/clover/llvm/invocation.cpp | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/clover/llvm/invocation.cpp
b/src/ga
On 12/20/2016 11:59 AM, Jan Vesely wrote:
> On Fri, 2016-12-16 at 13:43 -0800, Francisco Jerez wrote:
>> Vedran Miletić writes:
>>
>>> Current implementation of event handling keeps an extra reference to
>>> the hardware event, in addition to the reference retu
test.
Signed-off-by: Vedran Miletić
---
src/gallium/state_trackers/clover/llvm/invocation.cpp | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/clover/llvm/invocation.cpp
b/src/gallium/state_trackers/clover/llvm/invocation.cpp
index 675cf19..29
lease-event test.
Signed-off-by: Vedran Miletić
---
src/gallium/state_trackers/clover/api/event.cpp | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/clover/api/event.cpp
b/src/gallium/state_trackers/clover/api/event.cpp
index 5d1a0e5..74bc4d9 1
ing a different doxygen config for usage with sphinx and I'm
> wondering if we should try and preserve what's already there.
>
> --Jason
>
Somewhat related effort was already discussed several months ago:
https://lists.freedesktop.org/archives/mesa-dev/2016-September/128182.html
V.
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
Nice cleanup.
Reviewed-by: Vedran Miletić
On 12/13/2016 04:58 AM, Dave Airlie wrote:
> From: Dave Airlie
>
> Signed-off-by: Dave Airlie
> ---
> src/amd/vulkan/radv_private.h | 6 --
> 1 file changed, 6 deletions(-)
>
> diff --git a/src/amd/vulkan/radv_pr
: simplify using templates and lambda
Signed-off-by: Vedran Miletić
Reviewed-by[v2]: Jan Vesely
Tested-by[v4]: Vinson Lee
Tested-by[v4]: Pierre Moreau
Reviewed-by: Francisco Jerez
---
.../state_trackers/clover/llvm/codegen/bitcode.cpp | 9 +++--
src/gallium/state_trackers/clover/llvm
The commit 8e430ff8b060b4e8e922bae24b3c57837da6ea77 support for LLVM
3.9 and older versionsin Clover. This patch restores it and refactors
the support using Clover compatibility layer for LLVM.
Signed-off-by: Vedran Miletić
---
.../state_trackers/clover/llvm/codegen/bitcode.cpp | 9
The commit 8e430ff8b060b4e8e922bae24b3c57837da6ea77 support for LLVM
3.9 and older versionsin Clover. This patch restores it and refactors
the support using Clover compatibility layer for LLVM.
Signed-off-by: Vedran Miletić
---
.../state_trackers/clover/llvm/codegen/bitcode.cpp | 9
On 11/16/2016 11:21 PM, Vinson Lee wrote:
> On Wed, Nov 16, 2016 at 10:10 AM, Jan Vesely wrote:
>> On Wed, 2016-11-16 at 12:29 +0100, Vedran Miletić wrote:
>>> The commit 8e430ff8b060b4e8e922bae24b3c57837da6ea77 support for LLVM
>>> 3.9 and older versionsin Clover.
The commit 8e430ff8b060b4e8e922bae24b3c57837da6ea77 support for LLVM
3.9 and older versionsin Clover. This patch restores it and refactors
the support using Clover compatibility layer for LLVM.
Signed-off-by: Vedran Miletić
---
.../state_trackers/clover/llvm/codegen/bitcode.cpp | 9
The commit 8e430ff8b060b4e8e922bae24b3c57837da6ea77 broke 3.9 and older
versions of LLVM for Clover. This patch restores it and refactors the
support using Clover compatibility layer for LLVM.
Signed-off-by: Vedran Miletić
---
.../state_trackers/clover/llvm/codegen/bitcode.cpp | 9 ++-
src
On 11/14/2016 04:49 PM, Tom Stellard wrote:
> On Mon, Nov 14, 2016 at 01:44:18PM +0100, Dieter Nützel wrote:
>> Tested-by: Dieter Nützel
>>
>> Thanks Vedran!
>>
>
> Pushed, thanks!
>
> -Tom
>
Thank you!
Ved
---
src/gallium/state_trackers/clover/llvm/codegen/bitcode.cpp | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/src/gallium/state_trackers/clover/llvm/codegen/bitcode.cpp
b/src/gallium/state_trackers/clover/llvm/codegen/bitcode.cpp
index 8e89a49..5dcc4f8 100644
--- a
od.takeError()) {
+ std::string msg;
+ ::llvm::handleAllErrors(std::move(err), [&](::llvm::ErrorInfoBase
&EIB) {
+ msg = EIB.message();
+ fail(r_log, error(CL_INVALID_PROGRAM), msg.c_str());
+ });
+ }
return std::unique_ptr<::llvm::Module>(std::move(*mod));
}
On 10/30/2016 08:16 PM, Francisco Jerez wrote:
> Vedran Miletić writes:
>
>> OpenCL apps can quote arguments they pass to the OpenCL compiler, most
>> commonly include paths containing spaces.
>>
>> If the Clang OpenCL compiler was called via a shell, the shell wou
On 10/25/2016 04:15 AM, Michel Dänzer wrote:
> On 25/10/16 11:01 AM, Michel Dänzer wrote:
>> On 25/10/16 09:17 AM, Vedran Miletić wrote:
>>> On 10/25/2016 01:41 AM, Vinson Lee wrote:
>>>> Fix SCons build. drmDevicePtr is not available until libdrm 2.4.65.
>>>
('DRM', ['libdrm >= 2.4.65'])
if env['x11']:
env.Append(CPPPATH = env['X11_CPPPATH'])
That's a large bump and 2.4.65 is just over a year old. It would be nice
to support a couple more versions before.
Is there an acceptable workaroun
27; ') {
std::vector ss;
std::istringstream iss(s);
std::string t;
- while (getline(iss, t, ' '))
+ while (getline(iss, t, sep))
ss.push_back(t);
return ss;
diff --git a/src/gallium/state_trackers/clover/util/vers
On 10/13/2016 07:45 PM, Vedran Miletić wrote:
> On 10/07/2016 12:05 PM, Serge Martin wrote:
>> On Thursday 06 October 2016 16:26:21 Vedran Miletić wrote:
>>> PACKAGE_VERSION #ifdef MESA_GIT_SHA1
>>> " (" MESA_GIT_SHA1 ")
On 10/07/2016 12:05 PM, Serge Martin wrote:
> On Thursday 06 October 2016 16:26:21 Vedran Miletić wrote:
>> CLOVER_CL_VERSION_OVERRIDE allows overriding default OpenCL version
>> supported by Clover, analogous to MESA_GL_VERSION_OVERRIDE for OpenGL.
>> CLOVER_CL_C_VERSION_OVER
On 10/06/2016 07:15 PM, Jan Vesely wrote:
> On Thu, 2016-10-06 at 16:26 +0200, Vedran Miletić wrote:
>> CLOVER_CL_VERSION_OVERRIDE allows overriding default OpenCL version
>> supported by Clover, analogous to MESA_GL_VERSION_OVERRIDE for
>> OpenGL.
>> CLOVER_CL_C_VERSIO
On 10/06/2016 04:26 PM, Vedran Miletić wrote:
> CLOVER_CL_VERSION_OVERRIDE allows overriding default OpenCL version
> supported by Clover, analogous to MESA_GL_VERSION_OVERRIDE for OpenGL.
> CLOVER_CL_C_VERSION_OVERRIDE allows overridng default OpenCL C version.
> ---
> do
://reviews.llvm.org/D20168 or an equivalent patch.
Signed-off-by: Vedran Miletić
---
src/gallium/state_trackers/clover/llvm/codegen/common.cpp | 3 +++
1 file changed, 3 insertions(+)
diff --git a/src/gallium/state_trackers/clover/llvm/codegen/common.cpp
b/src/gallium/state_trackers/clover/llvm/codegen
CLOVER_CL_VERSION_OVERRIDE allows overriding default OpenCL version
supported by Clover, analogous to MESA_GL_VERSION_OVERRIDE for OpenGL.
CLOVER_CL_C_VERSION_OVERRIDE allows overridng default OpenCL C version.
---
docs/envvars.html | 12
src/galliu
const char *p_name) {
>> + return GetExtensionFunctionAddressForPlatform(d_platform, p_name);
>> +}
>> +
>> CLOVER_ICD_API cl_int
>> clIcdGetPlatformIDsKHR(cl_uint num_entries, cl_platform_id *rd_platforms,
>> cl_uint *rnum_platforms) {
>
> ___
> mesa-dev mailing list
> mesa-dev@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
>
Reviewed-by: Vedran Miletić
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
m/invocation.cpp | 2 +-
> .../state_trackers/clover/tgsi/compiler.cpp| 2 +-
> 11 files changed, 61 insertions(+), 5 deletions(-)
>
Nice work.
The series is
Reviewed-by: Vedran Miletić
--
Vedran Miletić
vedran.miletic.net
__
ernel
> + /// argument given by \p arg.
> + ///
> + inline uint64_t
> + get_uint_argument_metadata(const ::llvm::Function &f,
> + const ::llvm::Argument &arg,
> + const std::string &name) {
comment
Signed-off-by: Vedran Miletić
---
src/gallium/state_trackers/clover/llvm/util.hpp | 39 ++---
1 file changed, 35 insertions(+), 4 deletions(-)
diff --git a/src/gallium/state_trackers/clover/llvm/util.hpp
b/src/gallium/state_trackers/clover/llvm/util.hpp
index 8db6f20
On 09/28/2016 04:22 PM, Edward O'Callaghan wrote:
> Just drop the Rb as soon as you revise any patch please. Obviously there
> was a bunch of stuff that happened and discussion..
>
Indeed, v1 and v6 are quite different. Will do in the future.
Regards,
Vedran
--
typo
* use CLOVER_EXTRA_COMPILER_OPTIONS in link stage
v6:
* separate in CLOVER_EXTRA_{BUILD,COMPILE,LINK}_OPTIONS
* append options in cl{Build,Compile,Link}Program
Signed-off-by: Vedran Miletić
Reviewed-by[v1]: Edward O'Callaghan
---
docs/envvars.html
Signed-off-by: Vedran Miletić
Reviewed-by[v1]: Edward O'Callaghan
---
docs/envvars.html | 13 +
src/gallium/state_trackers/clover/llvm/invocation.cpp | 11 ---
2 files changed, 21 insertions(+), 3 deletions(-)
diff --git a
On 09/07/2016 06:52 PM, Vedran Miletić wrote:
> LLVM and Mesa both define the DEBUG macro in incompatible ways. As a
> general practice, we should avoid using such generic names when it is
> possible to do so.
>
> This patch renames all occurrences of the DEBUG macro to MESA_DEBUG
OpenCL apps can quote arguments they pass to the OpenCL compiler, most
commonly include paths containing spaces.
If the Clang OpenCL compiler was called via a shell, the shell would
split the arguments with respect to to quotes and then remove quotes
before passing the arguments to the compiler. S
On 09/19/2016 07:08 PM, Vedran Miletić wrote:
> On 07/28/2016 07:52 AM, Francisco Jerez wrote:
>> Emil Velikov writes:
>>
>>> On 6 June 2016 at 00:02, Vedran Miletić wrote:
>>>> On 06/04/2016 04:18 AM, Francisco Jerez wrote:
>>>>>
>>&g
On 07/28/2016 07:52 AM, Francisco Jerez wrote:
> Emil Velikov writes:
>
>> On 6 June 2016 at 00:02, Vedran Miletić wrote:
>>> On 06/04/2016 04:18 AM, Francisco Jerez wrote:
>>>>
>>>> Serge Martin writes:
>>>>
>>>>> From:
till doesn't have it in their stable package (and it's not
> even in test in Bodhi). I'll try to compile my own LLVM as Edward advised.
>
Fedora 24 is unlikely to get LLVM 3.9 in updates. If packaged, it will
come to Fedora 25 or newer.
Good choice.
--
Vedran Miletić
vedran.m
On 09/07/2016 07:58 PM, Nicholas Bishop wrote:
> Bump, any interest in this?
>
In my experience rST is much nicer to write than pure HTML, and I like
how the output looks. Definitive yes on the rST instead of HTML, no
strong opinion on readthedocs.org vs self-hosted.
Regards,
Vedran
--
CLOVER_EXTRA_COMPILER_OPTIONS
* use debug_get_option
* append to linker options as well
v3: code cleanups
v4: separate CLOVER_EXTRA_LINKER_OPTIONS options
Signed-off-by: Vedran Miletić
Reviewed-by[v1]: Edward O'Callaghan
---
docs/envvars.html | 12
On 09/07/2016 04:14 PM, Jan Vesely wrote:
> On Wed, 2016-09-07 at 13:43 +0200, Vedran Miletić wrote:
>> Options specified via the CLOVER_EXTRA_COMPILER_OPTIONS shell
>> variable
>> are appended to the compiler and linker options specified by the
>> OpenCL
&g
-off-by: Vedran Miletić
Reviewed-by[v1]: Edward O'Callaghan
---
docs/envvars.html | 9 +
src/gallium/state_trackers/clover/llvm/invocation.cpp | 10 +++---
2 files changed, 16 insertions(+), 3 deletions(-)
diff --git a/docs/envvars.html b
On 08/31/2016 12:11 PM, Vedran Miletić wrote:
> Options specified via the CLOVER_EXTRA_COMPILER_OPTIONS shell variable
> are appended to the compiler and linker options specified by the OpenCL
> program (if any).
>
> v2:
> * rename to CLOVER_EXTRA_COMPILER_OPTIONS
> *
On 08/28/2016 04:42 PM, Niels Ole Salscheider wrote:
> This is required by the OpenCL standard.
>
> Signed-off-by: Niels Ole Salscheider
Reviewed-by: Vedran Miletić
Good catch. Do we miss more defines from [1]?
Regards,
Vedran
[1]
https://www.khronos.org/registry/cl/sdk/1.1/docs/
Options specified via the CLOVER_EXTRA_COMPILER_OPTIONS shell variable
are appended to the compiler and linker options specified by the OpenCL
program (if any).
v2:
* rename to CLOVER_EXTRA_COMPILER_OPTIONS
* use debug_get_option
* append to linker options as well
Signed-off-by: Vedran
On 08/30/2016 09:27 AM, Serge Martin wrote:
> On Tuesday 30 August 2016 01:20:55 Vedran Miletić wrote:
>> Options specified via the CLOVER_COMPILER_OPTIONS shell variable are
>> appended to the compiler options specified by the OpenCL program (if
>> any).
>>
>&
Options specified via the CLOVER_COMPILER_OPTIONS shell variable are
appended to the compiler options specified by the OpenCL program (if
any).
Signed-off-by: Vedran Miletić
---
docs/envvars.html | 2 ++
src/gallium/state_trackers/clover/llvm/invocation.cpp
On 08/29/2016 09:30 PM, Jan Vesely wrote:
On Mon, 2016-08-29 at 20:13 +0200, Vedran Miletić wrote:
Clover presently reports 32 as CL_DEVICE_ADDRESS_BITS, which is not
correct for AMD SI and newer chip generations. This patch introduces
the PIPE_COMPUTE_CAP_ADDRESS_BITS pipe capability queried
older than EG will not use this capability).
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=97513
Signed-off-by: Vedran Miletić
---
src/gallium/drivers/radeon/r600_pipe_common.c | 9 +
src/gallium/include/pipe/p_defines.h | 3 ++-
src/gallium/state_trackers/clover/api
>
> (anyways, it's just for non-autoconf builds)
>
>
If there aren't too many I would keep them.
Regards,
Vedran
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
On 07/26/2016 10:29 PM, Kenneth Graunke wrote:
> On Tuesday, July 26, 2016 5:41:29 PM PDT Vedran Miletić wrote:
>> On 07/15/2016 07:27 AM, Edward O'Callaghan wrote:
>>> Reviewed-by: Edward O'Callaghan
>>>
>>
>> Anything I can do to get this push
On 07/15/2016 07:27 AM, Edward O'Callaghan wrote:
Reviewed-by: Edward O'Callaghan
Anything I can do to get this pushed?
Regards,
Vedran
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop
On 07/22/2016 03:37 AM, Rob Clark wrote:
On Thu, Jul 21, 2016 at 9:35 PM, Rob Clark wrote:
On Thu, Jul 21, 2016 at 1:48 PM, Vedran Miletić wrote:
LLVM and Mesa both define the DEBUG macro in incompatible ways. As a
general practice, we should avoid using such generic names when it is
;
+#include "llvm/compat.hpp"
+#include "llvm/invocation.hpp"
+#include "llvm/metadata.hpp"
+#include "llvm/util.hpp"
+#include "util/algorithm.hpp"
+
+
using namespace clover;
using namespace clover::llvm;
The series is:
Reviewed-by: Vedran Mi
Few files were missing the copyright headers. This patch adds correct
copyright headers according to the commit author and date.
Signed-off-by: Vedran Miletić
---
src/gallium/state_trackers/nine/nine_dump.c | 21 +
src/gallium/state_trackers/nine/nine_dump.h | 21
Signed-off-by: Vedran Miletić
---
docs/xlibdriver.html| 2 +-
src/gallium/docs/source/drivers/openswr/faq.rst | 4 ++--
src/gallium/include/pipe/p_defines.h| 2 +-
src/mesa/main/version.c | 2 +-
4 files changed, 5 insertions(+), 5
OpenCL apps can quote arguments they pass to the OpenCL compiler, most
commonly include paths containing spaces.
If the Clang OpenCL compiler was called via a shell, the shell would
split the arguments with respect to to quotes and then remove quotes
before passing the arguments to the compiler. S
o any general feedback is also welcome.
You've got some more feedback in my previous reply it doesn't look like
you've taken into account yet, it would be nice to see those addressed
before we get your patch merged.
Indeed. Is [1] what you had in mind?
Regards,
Vedran
[1] htt
k
in the game; in the first run there is a very long stop (approx 1
second) the moment the character picks up the green sigil (and a few
other points, I would have to check exactly), second run is all smooth.
Regards,
Vedran
--
Vedran Miletić
vedran.mileti
06.06.2016 u 12:24, Vedran Miletić je napisao/la:
On 06/06/2016 02:04 AM, Francisco Jerez wrote:
Vedran Miletić writes:
Aside from working just like NVIDIA and AMD proprietary stacks, no.
However, what we do right now is most certainly broken. Consider the
following argument
-I"/fo
) to stop using generic macro names, but I don't
think that's on anybody's todo list.
it was just a suggestion, feel free to ignore it.
Jan
I love cleaning stuff like that and can provide a patch changing DEBUG
to e.g. MESA_DEBUG, if that would solve the problem on Mesa's
Reviewed-by: Vedran Miletić
On 07/01/2016 04:18 PM, Marek Olšák wrote:
From: Marek Olšák
v2: - squashed the patches
- use INT_MAX
- clamp max_const_buffer_size
- check the DRM version in radeon
---
src/gallium/drivers/r600/r600_pipe.c | 4 ++--
src/gallium/drivers
On 07/01/2016 02:27 PM, Marek Olšák wrote:
On Fri, Jul 1, 2016 at 1:45 PM, Vedran Miletić wrote:
On 07/01/2016 11:11 AM, Marek Olšák wrote:
On Fri, Jul 1, 2016 at 10:54 AM, Marek Olšák wrote:
On Fri, Jul 1, 2016 at 2:52 AM, Vedran Miletić
wrote:
On 07/01/2016 01:29 AM, Marek Olšák
On 07/01/2016 11:11 AM, Marek Olšák wrote:
On Fri, Jul 1, 2016 at 10:54 AM, Marek Olšák wrote:
On Fri, Jul 1, 2016 at 2:52 AM, Vedran Miletić wrote:
On 07/01/2016 01:29 AM, Marek Olšák wrote:
From: Marek Olšák
also fix max_global_size to take a maximum of {vram_size, gart_size}
---
src
;info.gart_size);
The radeon driver in kernels older than 3.17 can't allocate BOs larger
than 256MB.
Indeed. You should ensure DRM >= 2.40.
Regards,
Vedran
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@list
On 07/01/2016 02:52 AM, Vedran Miletić wrote:
Had something similar in the works, Bas did as well, but this approach
is cleaner.
With these changes, in si_pipe.c and r600_pipe.c, you should not return
max_const_buffer_size anymore, since it can exceed int limits, but
instead something like
ead something like
MAX2(0x7ff, max_const_buffer_size)
Haven't tried running memory-heavy OpenCL programs, but clinfo other
than const buffer size looks good.
--
Vedran Miletić
vedran.miletic.net
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.o
On 06/22/2016 02:04 PM, Emil Velikov wrote:
From: Emil Velikov
Considering how hard/annoying it was for many peoples' workflow to
properly generate the macro, it will be demoted to conditionally
available with follow-up commits.
Cc: mesa-sta...@lists.freedesktop.org
Cc: Vedran Mileti
eonsi: boolean -> bool, TRUE -> true, FALSE -> false
radeonsi: make si_is_format_supported static
Please review.
I presume the next series will address boolean-> bool in si_state.c.
This series is
Reviewed-by: Vedran Miletić
Marek
__
On 06/21/2016 03:13 AM, Michel Dänzer wrote:
On 21.06.2016 08:17, Vedran Miletić wrote:
setLangDefaults() now requires PreprocessorOptions as an argument.
---
src/gallium/state_trackers/clover/llvm/invocation.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium
setLangDefaults() now requires PreprocessorOptions as an argument.
---
src/gallium/state_trackers/clover/llvm/invocation.cpp | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/gallium/state_trackers/clover/llvm/invocation.cpp
b/src/gallium/state_trackers/clover/llvm/invocatio
Change MESA into Mesa in CL_PLATFORM_VERSION and CL_DEVICE_VERSION. For
both, always append git version suffix from git_sha1.h.
v5: move semicolon to same line as MESA_GIT_SHA1.
v4: drop #ifdef guards.
v3: add missing include.
v2: change CL_DEVICE_VERSION as well.
Reviewed-by: Francisco Jerez
--
On 06/06/2016 02:04 AM, Francisco Jerez wrote:
Vedran Miletić writes:
Aside from working just like NVIDIA and AMD proprietary stacks, no.
However, what we do right now is most certainly broken. Consider the
following argument
-I"/foo bar/baz"
which is going to be sent to Cl
1 - 100 of 136 matches
Mail list logo
