The patch looks good to me. But I am not sure if anybody else has different
comment.
Thanks!
Ruiling
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
Midhun Kodiyath
Sent: Monday, August 10, 2015 11:35 PM
To: beignet@lists.freedesktop.org
Luo Xionghu
Best Regards
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
junyan...@inbox.com
Sent: Wednesday, August 12, 2015 4:49 PM
To: beignet@lists.freedesktop.org
Cc: Junyan He
Subject: [Beignet] [PATCH 02/15] Backend: Add
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
Guo Yejun
Sent: Wednesday, August 12, 2015 9:10 AM
To: beignet@lists.freedesktop.org
Cc: Guo, Yejun
Subject: [Beignet] [PATCH] fix issue when build against llvm3.3
llvm 3.3 has a
LGTM, thanks.
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
Midhun Kodiyath
Sent: Monday, August 10, 2015 11:35 PM
To: beignet@lists.freedesktop.org
Cc: Kodiyath, Midhunchandra
Subject: [Beignet] [PATCH] Set proper Vendor ID
Device ID and
Luo Xionghu
Best Regards
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
Zhigang Gong
Sent: Thursday, August 13, 2015 10:24 AM
To: beignet@lists.freedesktop.org
Cc: Gong, Zhigang
Subject: [Beignet] [PATCH 4/4] GBE: a potential bug in
Pushed, thanks.
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
Guo, Yejun
Sent: Thursday, August 13, 2015 12:23
To: Kodiyath, Midhunchandra; beignet@lists.freedesktop.org
Cc: Kodiyath, Midhunchandra
Subject: Re: [Beignet] [PATCH] Set
Yes exactly why I have added that piece of code there. So that every time the
program is built, it will clear the contents of that file and then on it will
append the ASM dump into the file.
Regards
Manasi
-Original Message-
From: Song, Ruiling
Sent: Wednesday, August 12, 2015 6:27 PM
--- a/backend/src/backend/gen_context.cpp
+++ b/backend/src/backend/gen_context.cpp
@@ -2299,6 +2299,14 @@ namespace gbe
if (OCL_OUTPUT_ASM)
outputAssembly(stdout, genKernel);
+if (this-asmFileName) {
+ FILE *asmDumpStream = fopen(this-asmFileName, a);
+ if
We should treat it as a 2D image as image 1d buffer may be
exceed the 1D image size restrication.
Signed-off-by: Zhigang Gong zhigang.g...@intel.com
---
backend/src/libocl/src/ocl_image.cl | 14 +-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git
We need to test large image 1d buffer read and write testing.
Signed-off-by: Zhigang Gong zhigang.g...@intel.com
---
kernels/image_1D_buffer.cl | 11 ++-
utests/image_1D_buffer.cpp | 73 ++
2 files changed, 32 insertions(+), 52 deletions(-)
diff
ENDIF should be treated as barrier-like instruction
in instruction scheduling.
Signed-off-by: Zhigang Gong zhigang.g...@intel.com
---
backend/src/backend/gen_insn_scheduling.cpp | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/backend/src/backend/gen_insn_scheduling.cpp
llvm 3.3 has a different constructure of llvm::raw_fd_ostream
V2: refine the code
Signed-off-by: Guo Yejun yejun@intel.com
---
backend/src/backend/program.cpp | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/backend/src/backend/program.cpp
LGTM, thanks.
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
Guo Yejun
Sent: Monday, August 10, 2015 08:33
To: beignet@lists.freedesktop.org
Cc: Guo, Yejun
Subject: [Beignet] [PATCH V2] generate sub_group_id inside kernel instead
of
LGTM, thanks.
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
xionghu@intel.com
Sent: Wednesday, August 5, 2015 13:13
To: beignet@lists.freedesktop.org
Cc: Luo, Xionghu
Subject: [Beignet] [PATCH] GBE/IR: add collectInsnNum to collect
From: Junyan He junyan...@linux.intel.com
The profiling support is enabled by this patch set.
The profiling information is as following:
-- Log 0 --
| fix functions id: 7 simd: 16 kernel id:0 |
| thread id: 0 EU id: 1
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Junyan He junyan...@linux.intel.com
---
backend/src/ir/unit.cpp |5 -
backend/src/ir/unit.hpp |5 +
2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/backend/src/ir/unit.cpp b/backend/src/ir/unit.cpp
index
From: Junyan He junyan...@linux.intel.com
We will maintain a real clock to record the real execute time
of the orginal code. We do not want to introduce overhead
because of adding the profiling instructions, so every time
we enter the proliling instructions block, we will calculate the
real time
From: Junyan He junyan...@linux.intel.com
When user enables profiling feature, we need to insert
extra instructions to record and store the timestamps.
By now, the function pass will just insert the requred
instructions at the head of first 20 blocks. Later, we
will support to insert timestamps
From: Junyan He junyan...@linux.intel.com
Add four timestamp reigsters and one pointer register
into curbe. The four timestamp reigsters will hold
all the infomation of profiling timestamps, includes
20 uint timestamps for each point, 1 ulong prolog holding
the start time and and 1 ulong epilog
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Junyan He junyan...@linux.intel.com
---
backend/src/llvm/llvm_gen_backend.cpp | 45
backend/src/llvm/llvm_gen_backend.hpp |3 ++
backend/src/llvm/llvm_gen_ocl_function.hxx |5
3 files
llvm 3.3 has a different constructure of llvm::raw_fd_ostream
Signed-off-by: Guo Yejun yejun@intel.com
---
backend/src/backend/program.cpp | 11 ++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/backend/src/backend/program.cpp b/backend/src/backend/program.cpp
index
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Junyan He junyan...@linux.intel.com
---
backend/src/backend/gen_context.hpp|3 +++
backend/src/backend/gen_program.cpp|6 +-
backend/src/backend/gen_program.hpp|2 +-
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Junyan He junyan...@linux.intel.com
---
src/cl_command_queue.c |8 +++
src/cl_command_queue.h |2 ++
src/cl_command_queue_gen7.c | 38 ++
src/cl_driver.h | 16 +
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Junyan He junyan...@linux.intel.com
---
backend/src/backend/gen_context.cpp | 156 +++
1 file changed, 156 insertions(+)
diff --git a/backend/src/backend/gen_context.cpp
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Bai Yannan yannan@intel.com
---
backend/src/backend/program.cpp | 26 +-
backend/src/backend/program.h | 11 +++
backend/src/backend/program.hpp | 22 ++
3 files changed, 58
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Junyan He junyan...@linux.intel.com
---
backend/src/gbe_bin_interpreter.cpp |4
src/cl_command_queue_gen7.c |1 -
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/backend/src/gbe_bin_interpreter.cpp
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Junyan He junyan...@linux.intel.com
---
backend/src/backend/gen_context.cpp | 17 +
1 file changed, 17 insertions(+)
diff --git a/backend/src/backend/gen_context.cpp
b/backend/src/backend/gen_context.cpp
index
From: Junyan He junyan...@linux.intel.com
Signed-off-by: Junyan He junyan...@linux.intel.com
---
backend/src/backend/gen_insn_scheduling.cpp |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/backend/src/backend/gen_insn_scheduling.cpp
From: Laura Ekstrand laura.d.ekstr...@intel.com
This unifies the printing of the Gen Assembly to C file I/O, eliminating the
use of iostream. This is so that a single stream can be passed into the
function to receive all of the output.
Signed-off-by: Manasi Navare manasi.d.nav...@intel.com
From: Laura Ekstrand laura.d.ekstr...@intel.com
Part of the plumbing that passes the ASM file name from the compiler options
level down to the emitCode level so that the assembly can be written to that
file
Signed-off-by: Manasi Navare manasi.d.nav...@intel.com
Signed-off-by: Laura Ekstrand
From: Laura Ekstrand laura.d.ekstr...@intel.com
Part of the plumbing that passes the ASM file name from the compiler options
level down to the emitCode level so that the assembly can be written to that
file.
Signed-off-by: Manasi Navare manasi.d.nav...@intel.com
Signed-off-by: Laura Ekstrand
From: Laura Ekstrand laura.d.ekstr...@intel.com
Part of the plumbing that passes the ASM file name from the compiler options
level down to the emitCode level so that the assembly can be written to that
file.
Signed-off-by: Manasi Navare manasi.d.nav...@intel.com
Signed-off-by: Laura Ekstrand
From: Laura Ekstrand laura.d.ekstr...@intel.com
This prepares the compiler to dump assembly to an arbitrary file.
Signed-off-by: Manasi Navare manasi.d.nav...@intel.com
Signed-off-by: Laura Ekstrand laura.d.ekstr...@intel.com
---
backend/src/backend/gen_context.cpp | 62
Open the file specified for the ASM dump and write the assembly to it.
Signed-off-by: Manasi Navare manasi.d.nav...@intel.com
Signed-off-by: Laura Ekstrand laura.d.ekstr...@intel.com
---
backend/src/backend/gen_context.cpp | 8
backend/src/backend/program.cpp | 3 +++
2 files
From: Laura Ekstrand laura.d.ekstr...@intel.com
Signed-off-by: Manasi Navare manasi.d.nav...@intel.com
Signed-off-by: Laura Ekstrand laura.d.ekstr...@intel.com
---
backend/src/backend/program.cpp | 15 +++
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git
This is a revised patch set to enable the Gen ASM dump featuree
These patches are rebased on the latest Beignet codebase
It includes te changes requested by Ruiling Song.
Now the ASM dump file is cleared everytime a kernel/program is built
and then the ASM dump for each built kernel is appended to
36 matches
Mail list logo