the legacyMode means what kind of address mode to choose.
when legacyMode is true, we need to do complex bti analysis.
dynamicBti and staticBti are most for platforms before BDW.
And stateless is for platform BDW+
v2:
only do analyzePointerOrigin() under legacyMode.
v3:
fix conflict with maste
Thanks your comments, I'll send v2.
-Original Message-
From: Song, Ruiling
Sent: Wednesday, November 04, 2015 3:17 PM
To: Meng, Mengmeng; beignet@lists.freedesktop.org
Cc: Meng, Mengmeng
Subject: RE: [Beignet] [PATCH 1/2] add benckmark for copy data from buffer to
buffer
> -Origin
> -Original Message-
> From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
> Meng Mengmeng
> Sent: Monday, November 2, 2015 2:29 PM
> To: beignet@lists.freedesktop.org
> Cc: Meng, Mengmeng
> Subject: [Beignet] [PATCH 1/2] add benckmark for copy data from buffer to
>
From: Luo Xionghu
include case compile_spir_binary and build_spir_binary.
Signed-off-by: Luo Xionghu
---
utests/get_cl_info.cpp | 99 ++
1 file changed, 99 insertions(+)
diff --git a/utests/get_cl_info.cpp b/utests/get_cl_info.cpp
index 3109c8e.
From: Luo Xionghu
allow user to pass "-dump-spir-binary=[file_name]" to generate the spir
binary to the file.
Signed-off-by: Luo Xionghu
---
backend/src/backend/program.cpp | 27 ++-
1 file changed, 22 insertions(+), 5 deletions(-)
diff --git a/backend/src/backend/prog
Currently it will reject the creation of image based on host_ptr
if the height is not aligned. In fact this is wrong. And we can't
pass the aligned height in image_desc. Otherwise the sampler doesn't
work as expected because of padding height. It is enough if the image_row_pitch
follows the aligned
Pushed.
> -Original Message-
> From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
> Guo, Yejun
> Sent: Thursday, October 29, 2015 15:05
> To: Luo, Xionghu; beignet@lists.freedesktop.org
> Cc: Luo, Xionghu
> Subject: Re: [Beignet] [patch V2] set the pitch of image fro
> -Original Message-
> From: Pan, Xiuli
> Sent: Tuesday, November 3, 2015 17:23
> To: Yang, Rong R; beignet@lists.freedesktop.org
> Subject: RE: [Beignet] [PATCH 2/2] Backend: add debugwait function
>
> 1. I will look into the side effect and the scheduling. And the backend part
> is
>
LGTM, Thanks for your help!
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
Rebecca N. Palmer
Sent: Wednesday, November 4, 2015 6:19 AM
To: beignet@lists.freedesktop.org
Subject: Re: [Beignet] [PATCH v3] GBE: Don't read past end of printf forma
When p == end (the null terminator byte), don't try to read p + 1:
as this is outside the string, it might be a '%' from a different
object (causing __parse_printf_state(end + 2, end, ...) to be called,
which will fail), or an invalid address.
Signed-off-by: Rebecca Palmer
---
backend/src/llvm/l
> -Original Message-
> From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
> junyan...@inbox.com
> Sent: Wednesday, September 9, 2015 8:01
> To: beignet@lists.freedesktop.org
> Subject: [Beignet] [PATCH 17/19] Backend: Add ADD_ and SUB_ timestamps
> help functions.
>
> -Original Message-
> From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
> junyan...@inbox.com
> Sent: Wednesday, September 9, 2015 8:01
> To: beignet@lists.freedesktop.org
> Subject: [Beignet] [PATCH 19/19] Backend: Implement
> StoreProfilingInstruction in GenCont
> -Original Message-
> From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
> junyan...@inbox.com
> Sent: Wednesday, September 9, 2015 8:01
> To: beignet@lists.freedesktop.org
> Subject: [Beignet] [PATCH 11/19] Backend: Add profilingProlog function for
> GenContext.
>
> -Original Message-
> From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
> junyan...@inbox.com
> Sent: Wednesday, September 9, 2015 8:01
> To: beignet@lists.freedesktop.org
> Subject: [Beignet] [PATCH 08/19] Backend: Add IVAR OCL_PROFILING_LOG
> to control profilin
One comment.
> -Original Message-
> From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
> junyan...@inbox.com
> Sent: Wednesday, September 9, 2015 8:01
> To: beignet@lists.freedesktop.org
> Subject: [Beignet] [PATCH 03/19] Backend: Add ProfilingInserter and a new
> fu
The patch LGTM, but has some reorder warning when compiler. And the patch
conflict with "GBE: Fix unaligned load/store issues.", don't include float
support. Can you send a new version.
/home/yr/maintain/beignet/backend/src/llvm/llvm_gen_backend.cpp:416:31:
warning: ‘gbe::MemoryInstHelper::unit
The patch has been merged, and have sent a patch update the README.
I will add the LLVM 3.7 information after have the LLVM 3.7 performance date.
> -Original Message-
> From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
> Zou, Nanhai
> Sent: Monday, November 2, 2015
After fix clang conversion corner case, The pass rate of LLVM 3.6 is same as
3.5.
Signed-off-by: Yang Rong
---
docs/Beignet.mdwn | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/docs/Beignet.mdwn b/docs/Beignet.mdwn
index 9a2b516..f763b6f 100644
--- a/docs/Beignet.mdwn
1. I will look into the side effect and the scheduling. And the backend part
is used part of the existed barrier wait inst, may need some change.
2.The sel.WAIT(1); is not for the execwidth, it is used for select the
different
reg used for host-thread from the barrier host-host reg.
-Origin
Sorry about that, it seems that my email client did something for it.
It worked.
How about the format-patch? :)
-Original Message-
From: Beignet [mailto:beignet-boun...@lists.freedesktop.org] On Behalf Of
Rebecca N. Palmer
Sent: Tuesday, November 3, 2015 3:46 PM
To: beignet@lists.freedes
20 matches
Mail list logo
