Re: [Beignet] [PATCH] generate MOV instruction at selection stage when do simd_shuffle with imm value.

Pint for review, thanks. -Original Message- From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of Guo, Yejun Sent: Tuesday, September 08, 2015 4:08 PM To: beignet@lists.freedesktop.org Subject: Re: [Beignet] [PATCH] generate MOV instruction at selection stage when do

[Beignet] [PATCH 3/3] add utest runtime_use_host_ptr_image

Signed-off-by: Guo Yejun --- kernels/runtime_use_host_ptr_image.cl | 10 + utests/CMakeLists.txt | 1 + utests/runtime_use_host_ptr_image.cpp | 72 +++ 3 files changed, 83 insertions(+) create mode 100644 kernels/runtime_use_host_ptr_image.cl

[Beignet] [PATCH 2/3] enable USE_HOST_PTR for cl image with userptr to avoid extra copying

the pointer must be 64 byte aligned, and only when w,h equals to its aligned value, otherwise, roll back to the old method with extra copying. Signed-off-by: Guo Yejun --- src/cl_command_queue.c | 7 ++- src/cl_enqueue.c | 10 ++ src/cl_mem.c | 45 +++

[Beignet] [PATCH 1/3] refine code to separate the usage of data and image2d_from_buffer

currently, 'void* data' has two meanings: the pointer from application, and the buffer to create image2d from. It makes the code a little complex when supporting userptr for image. So, add a new function parameter to separate the two meanings. Signed-off-by: Guo Yejun --- src/cl_mem.c| 61 ++

Re: [Beignet] [PATCH V2] GBE: Implement liveness dump.

LGTM. On Thu, Sep 24, 2015 at 03:47:30PM +0800, Ruiling Song wrote: > v2: > remove old printf debug code. > Signed-off-by: Ruiling Song > --- > backend/src/ir/liveness.cpp | 58 > - > 1 file changed, 20 insertions(+), 38 deletions(-) > > diff --git a

[Beignet] [Patch v2] GBE: refine longjmp checking.

v2: simplify the logic in function.hpp. Let the user to prepare correct start and end point. Fix the incorrect start/end point for one forward jump and one backward jump case. Signed-off-by: Zhigang Gong --- backend/src/backend/gen_insn_selection.cpp | 17 +++-- backend/src/ir/functi

Re: [Beignet] [PATCH 2/5] GBE: refine longjmp checking.

On Thu, Sep 24, 2015 at 07:05:38AM +, Yang, Rong R wrote: > One comment. > > > -Original Message- > > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > > Zhigang Gong > > Sent: Monday, September 14, 2015 14:20 > > To: beignet@lists.freedesktop.org > > Cc: Gong

Re: [Beignet] [PATCH V2 3/3] add local copy propagation optimization for each basic block

> > > > + > > > > + void SelBasicBlockOptimizer::changeInsideReplaceInfos(const > > > > + SelectionInstruction& insn, GenRegister& var) { > > > > +for (ReplaceInfo* info : replaceInfos) { > > > > + if (info->intermedia.reg() == var.reg()) { > > A new comment here is that the above loop

[Beignet] [PATCH 1/4] Fix gpgpu node related bug

This should be a typo, we should wait for the gpgpu and create node only if the batch buffer is busy. Signed-off-by: Pan Xiuli --- src/intel/intel_gpgpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/intel/intel_gpgpu.c b/src/intel/intel_gpgpu.c index 901bd98..60d318a

[Beignet] [PATCH 3/4] Fix a event leak in create context

We get an event out of NDRangeKernel, and we don't release it. As an gpgpu event it can also make drm buffer leak, to avoid potenial error we just release it.w Signed-off-by: Pan Xiuli --- src/cl_device_id.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/cl_device_id.c b/src/cl_device_i

[Beignet] [PATCH 2/4] runtime: refine the last_event in queue to a list

Refine the evetn struct to make last_event become a list to store all uncompelete and update them every queue flush. This can make sure all events created in the runtime have a chance for updating status and upcoming callback functions and delete. We will also need not to worry about other memory l

[Beignet] [PATCH 4/4] utests: event should be released

This is also gpgpu event, which can cause leakes. Just release it. Signed-off-by: Pan Xiuli --- utests/profiling_exec.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/utests/profiling_exec.cpp b/utests/profiling_exec.cpp index afa55ba..4232772 100644 --- a/utests/profiling_exec.cpp +++ b/

Re: [Beignet] [PATCH V2 3/3] add local copy propagation optimization for each basic block

On Thu, Sep 24, 2015 at 06:51:12AM +, Guo, Yejun wrote: > > > -Original Message- > From: Zhigang Gong [mailto:zhigang.g...@linux.intel.com] > Sent: Thursday, September 24, 2015 1:12 PM > To: Guo, Yejun > Cc: beignet@lists.freedesktop.org > Subject: Re: [Beignet] [PATCH V2 3/3] add lo

[Beignet] [PATCH V2] GBE: Implement liveness dump.

v2: remove old printf debug code. Signed-off-by: Ruiling Song --- backend/src/ir/liveness.cpp | 58 - 1 file changed, 20 insertions(+), 38 deletions(-) diff --git a/backend/src/ir/liveness.cpp b/backend/src/ir/liveness.cpp index c5a6374..9f456a3 100644

Re: [Beignet] [PATCH] GBE: Implement liveness dump.

Yes, I agree with you. When I was drafting the patch, I think the old printf code may be still useful for debugging. But I take a detailed look, the dump to ostream can replace the old debug printf. I will remove old code in V2. Thanks! Ruiling > -Original Message- > From: Zhigang Gong [

Re: [Beignet] [PATCH 2/5] GBE: refine longjmp checking.

One comment. > -Original Message- > From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of > Zhigang Gong > Sent: Monday, September 14, 2015 14:20 > To: beignet@lists.freedesktop.org > Cc: Gong, Zhigang > Subject: [Beignet] [PATCH 2/5] GBE: refine longjmp checking. > >