[Beignet] [PATCH V2] Add -cl-fast-relaxed-math into incompatible opts and fix the PreprocessorOptions bug

From: Junyan He Signed-off-by: Junyan He --- backend/src/backend/program.cpp |7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/backend/src/backend/program.cpp b/backend/src/backend/program.cpp index 3ab1bc2..728fa5f 100644 --- a/backend/src/backend/program.cpp +++ b/b

Re: [Beignet] utests results for IVB2

There are several failures in atan, atan2, atanpi, cos, cospi, erf, erfc, nextafter, pow, pown, powr, rootn, sin, sinpi, tan, tanpi and tgamma. Hi Simon, These are all precision issues. we are fixing them, and most of the patches are already in the mail list for review. Thanks for your attenti

Re: [Beignet] Piglit results for IVB2

Hi Simon, Thanks for the reporting. Logic operators are not applied on float data type according to OpenCL 1.1 spec. However in 1.2 spec, they are applied to float type. So we don't think this is a real issue in this the OpenCL 1.1 implementation. Thanks Zou Nan

Re: [Beignet] opencl-1.2 branch head compile error and warnings

opencl-1.2 branch HEAD is compiled fine, but when I ran utest I saw this warning : libEGL warning: failed to create a pipe screen for i965 2014/1/9 Zhigang Gong > This is a pure llvm related issue. The error messages you gave should be > simply from the > following command: > > llvm-as -o ocl

[Beignet] Piglit results for IVB2

Hi, Piglit doesn't seem to have any real issues. Full results at http://www.psi5.com/~geier/beignet/ The failure in the "double" tests can probably be explained by the fact that I haven't hidden the cl_khr_fp64 extension in my build, so these can be ignored. The only remaining Piglit failure i

Re: [Beignet] [PATCH] build: use auto libdir

Hi, On 14.01.2014 18:51, Igor Gnatenko wrote: > From 4a47aac87719b0b56565cd257cac0868ae574cd4 Mon Sep 17 00:00:00 2001 > From: Igor Gnatenko > Date: Tue, 14 Jan 2014 21:22:42 +0400 > Subject: [PATCH] build: use auto libdir That looks immensely useful for distribution maintainers. Not sure about

Re: [Beignet] Libdir

Hi, > I'm not sure, but why we installing libs to prefix/lib but not libdir? > I.e. on x86_64 I have installed beignet in /usr/lib, but not in /usr/lib64. /usr/lib64 is dead and should never be used. In principle, it would be good to install to /usr/lib/x86_64-linux-gnu, but that is actually the

[Beignet] [PATCH] build: use auto libdir

>From 4a47aac87719b0b56565cd257cac0868ae574cd4 Mon Sep 17 00:00:00 2001 From: Igor Gnatenko Date: Tue, 14 Jan 2014 21:22:42 +0400 Subject: [PATCH] build: use auto libdir Signed-off-by: Igor Gnatenko --- CMakeLists.txt | 3 +++ backend/src/CMakeLists.txt | 10 +- src/CMakeLi

[Beignet] Libdir

Hi, I'm not sure, but why we installing libs to prefix/lib but not libdir? I.e. on x86_64 I have installed beignet in /usr/lib, but not in /usr/lib64. ___ Beignet mailing list Beignet@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/be

[Beignet] [PATCH] Add the device id for haswell GT.

From: Junyan He Signed-off-by: Junyan He --- src/cl_device_data.h | 70 +++- src/cl_device_id.c | 100 ++- 2 files changed, 145 insertions(+), 25 deletions(-) diff --git a/src/cl_device_data.h b/src/cl_device_da

Re: [Beignet] [Patch v2 1/4] CL: back port ICD support to 1.1 branch.

I tested the patchset, with this patchset, it is easy to run opencv's cl program. Thanks! Thanks! Ruiling -Original Message- From: beignet-bounces+ruiling.song=intel@lists.freedesktop.org [mailto:beignet-bounces+ruiling.song=intel@lists.freedesktop.org] On Behalf Of Zhigang Gong