From: Serge Martin <edb+pig...@sigluy.net>
---
tests/cl/api/get-mem-object-info.c | 253 +++--
1 file changed, 213 insertions(+), 40 deletions(-)
diff --git a/tests/cl/api/get-mem-object-info.c
b/tests/cl/api/get-mem-object-info.c
index c24b51d..b68262c
.
+ */
+
clReleaseMemObject(matrix_buffer)
EdB
+ return result;
+}
___
Piglit mailing list
Piglit@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/piglit
OpenCL objects are just pointers in the end, using random pointer value
is likely to make an application crash.
Also, all others tests use NULL as an invalid object.
See also Francisco Jerez comments:
http://lists.freedesktop.org/archives/mesa-dev/2014-November/070520.html
--- /dev/null
+++ b/tests/cl/api/enqueue-migrate-mem-objects.c
@@ -0,0 +1,227 @@
+/*
+ * Copyright © 2015 Serge Martin (EdB) edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the Software
@@
+/*
+ * Copyright © 2015 Serge Martin (EdB) edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the Software),
+ * to deal in the Software without restriction, including without limitation
-arg-info.c
@@ -0,0 +1,274 @@
+/*
+ * Copyright © 2014 EdB edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the Software),
+ * to deal in the Software without restriction, including
mode 100644
index 000..a77c422
--- /dev/null
+++ b/tests/cl/api/enqueue-fill-buffer.c
@@ -0,0 +1,270 @@
+/*
+ * Copyright © 2015 Serge Martin (EdB) edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated
@@
+/*
+ * Copyright © 2015 Serge Martin (EdB) edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the Software),
+ * to deal in the Software without restriction, including without
EdB edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the Software),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify
from that buffer object to host memory,
+ * finally compare the read result with reference matrix.
+*/
You mix spaces and tabs here.
Other than that : Reviewed by EdB edb+pig...@sigluy.net
EdB
+ test_write_rect(env-context-command_queues[0], matrix_buffer, CL_TRUE
Hello
Thanks for the patch, some comments bellow.
On Friday 10 July 2015 20:17:52 Meng Mengmeng wrote:
It's a simple function test for clEnqueueReadBufferRect and
clEnqueueWriteBufferRect. ---
tests/cl.py | 2 +
tests/cl/api/CMakeLists.cl.txt
Ping
On Sunday 14 June 2015 17:02:26 EdB wrote:
this query is only valid for a custom device or a build-in kernel
---
tests/cl/api/get-kernel-work-group-info.c | 18 --
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/tests/cl/api/get-kernel-work-group-info.c
this query is only valid for a custom device or a build-in kernel
---
tests/cl/api/get-kernel-work-group-info.c | 18 --
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/tests/cl/api/get-kernel-work-group-info.c
b/tests/cl/api/get-kernel-work-group-info.c
index
++;
+ }
+ }
Usually after checking for correct execution with valid input, piglit cl tests
test for correct behavior and return code with invalid parameters.
EdB
+ return PIGLIT_PASS;
+}
___
Piglit mailing list
Piglit@lists.freedesktop.org
http
On Monday 18 May 2015 14:06:02 Marek Olšák wrote:
From: Marek Olšák marek.ol...@amd.com
Reviewed-by: EdB e...@sigluy.net
---
tests/cl/interop/CMakeLists.cl.txt | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/cl/interop/CMakeLists.cl.txt
b/tests/cl/interop
How do I do ?
I make clean and even git clean -ifx and still got the error.
I see not rule to clean generated test, but I may have miss it.
On Tuesday 21 April 2015 10:58:59 Jan Vesely wrote:
On Tue, 2015-04-21 at 16:47 +0200, EdB wrote:
Piglit no loger work for me with commit
: /home/edb/projects/mesa-opencl/piglit-
git/bin/cl-program-tester /home/edb/projects/mesa-opencl/piglit-
git/generated_tests/cl/builtin/math/builtin-float-mix-1.0.generated.cl
The new test has this command:/home/edb/projects/mesa-opencl/piglit-
git/bin/cl-program-tester /home/edb/projects
Le 2015-04-21 17:27, Jan Vesely a écrit :
On Tue, 2015-04-21 at 17:18 +0200, EdB wrote:
How do I do ?
I make clean and even git clean -ifx and still got the error.
I see not rule to clean generated test, but I may have miss it.
yeah, for some reason tests in generated_tests/cl/builtin
against AMD catalyst. The test failed on
CL_KERNEL_ARG_INFO_NOT_AVAILABLE because kernel info were always available
On Sun, 2014-12-28 at 14:23 +0100, EdB wrote:
---
tests/cl/api/CMakeLists.cl.txt | 1 +
tests/cl/api/get-kernel-arg-info.c | 229
On Thursday 05 February 2015 13:49:37 Jan Vesely wrote:
On Tue, 2015-01-06 at 21:23 +0100, EdB wrote:
Ping
pushed
thanks
Le 2014-12-06 23:56, Jan Vesely a écrit :
On Sat, 2014-12-06 at 12:17 +0100, EdB wrote:
Hello
The bug is Verified.Fixed.
I don't have commit access
Ping
Le 2014-12-28 14:23, EdB a écrit :
---
tests/cl/api/CMakeLists.cl.txt | 1 +
tests/cl/api/get-kernel-arg-info.c | 229
+
tests/util/piglit-util-cl-enum.c | 11 ++
tests/util/piglit-util-cl-enum.h | 1 +
4 files changed, 242 insertions
I don't have commit acces. Can someone push it?
Le 2014-12-19 20:41, Tom Stellard a écrit :
On Fri, Dec 19, 2014 at 02:39:06PM +0100, EdB wrote:
v2:
Use piglit_cl_get_program_build_info instead of
piglit_cl_get_program_info,
I was expected it to fail so I didn't paid attention.
Remove
Ping
Le 2014-12-06 23:56, Jan Vesely a écrit :
On Sat, 2014-12-06 at 12:17 +0100, EdB wrote:
Hello
The bug is Verified.Fixed.
I don't have commit access. Can someone push it?
you can add
Reviewed-by: Jan Vesely jan.ves...@rutgers.edu
but I can't push changes
On Thursday 27 November 2014
get-kernel-work-group-info.c)
# Events
diff --git a/tests/cl/api/get-kernel-arg-info.c
b/tests/cl/api/get-kernel-arg-info.c
new file mode 100644
index 000..7e6c8df
--- /dev/null
+++ b/tests/cl/api/get-kernel-arg-info.c
@@ -0,0 +1,229 @@
+/*
+ * Copyright © 2014 EdB edb+pig...@sigluy.net
-info get-program-build-info.c)
diff --git a/tests/cl/api/link-program.c b/tests/cl/api/link-program.c
new file mode 100644
index 000..610aa8f
--- /dev/null
+++ b/tests/cl/api/link-program.c
@@ -0,0 +1,386 @@
+/*
+ * Copyright © 2014 EdB edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted
-program.c
new file mode 100644
index 000..e98f428
--- /dev/null
+++ b/tests/cl/api/link-program.c
@@ -0,0 +1,394 @@
+/*
+ * Copyright © 2014 EdB edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation
Hello
The bug is Verified.Fixed.
I don't have commit access. Can someone push it?
On Thursday 27 November 2014 12:38:52 EdB wrote:
This fix https://bugs.freedesktop.org/show_bug.cgi?id=86361
---
tests/cl/api/compile-program.c | 6 ++
1 file changed, 6 insertions(+)
diff --git
By OpenCL 1.2 spec, -cl-std= should only be CL1.1 or CL1.2
We were checking against the device version instead of the desired version
Report by: yan.w...@linux.intel.com
---
tests/util/piglit-framework-cl-program.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git
On Tuesday, December 02, 2014 08:17:46 PM Vinson Lee wrote:
Signed-off-by: Vinson Lee v...@freedesktop.org
Reviewed-by: EdB edb+pig...@sigluy.net
___
Piglit mailing list
Piglit@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/piglit
, CL_SUCCESS)) {
+ fprintf(stderr,
+ Failed (error code: %s): Create kernel.\n,
+ piglit_cl_get_error_name(errNo));
+ piglit_merge_result(result, PIGLIT_FAIL);
PIGLIT_WARN ?
- EdB
+ }
+ free
---
tests/cl/api/get-kernel-work-group-info.c | 159 +-
1 file changed, 132 insertions(+), 27 deletions(-)
diff --git a/tests/cl/api/get-kernel-work-group-info.c
b/tests/cl/api/get-kernel-work-group-info.c
index a38cdf0..ea899ca 100644
---
This fix https://bugs.freedesktop.org/show_bug.cgi?id=86361
---
tests/cl/api/compile-program.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/tests/cl/api/compile-program.c b/tests/cl/api/compile-program.c
index d280f98..54adecc 100644
--- a/tests/cl/api/compile-program.c
+++
because -cl-strict-aliasing options argument to
clBuildProgram is no longer supported in OpenCL 1.1.
---
or may be we should just test this in api/build-program.c
---
tests/cl/program/build/optimization-options-cl10.cl | 13 +
tests/cl/program/build/optimization-options-cl11+.cl |
---
tests/cl/api/enqueue-map-buffer.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/cl/api/enqueue-map-buffer.c
b/tests/cl/api/enqueue-map-buffer.c
index f101d5c..214ef7f 100644
--- a/tests/cl/api/enqueue-map-buffer.c
+++ b/tests/cl/api/enqueue-map-buffer.c
@@ -31,7
change existing tests in order to add CL_MEM_HOST_* checks
The fisrt two patches are standalone fixes
EdB (3):
cl: fix doxygen filename
cl: fix test name
cl: add CL_MEM_HOST_* releated tests
tests/cl/api/create-buffer.c | 4 +-
tests/cl/api/enqueue-map-buffer.c| 113
---
tests/cl/api/create-buffer.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/cl/api/create-buffer.c b/tests/cl/api/create-buffer.c
index b42dcba..1c4c4e9 100644
--- a/tests/cl/api/create-buffer.c
+++ b/tests/cl/api/create-buffer.c
@@ -22,7 +22,7 @@
*/
/**
-
---
tests/cl/api/enqueue-map-buffer.c| 111 +++
tests/cl/api/enqueue-read_write-buffer.c | 49 +-
2 files changed, 158 insertions(+), 2 deletions(-)
diff --git a/tests/cl/api/enqueue-map-buffer.c
b/tests/cl/api/enqueue-map-buffer.c
index
get-program-build-info.c)
diff --git a/tests/cl/api/compile-program.c b/tests/cl/api/compile-program.c
new file mode 100644
index 000..d280f98
--- /dev/null
+++ b/tests/cl/api/compile-program.c
@@ -0,0 +1,366 @@
+/*
+ * Copyright © 2014 EdB edb+pig...@sigluy.net
+ *
+ * Permission is hereby
If there is only one device associed with the kernel and device arg is NULL
you don't have to trigger CL_INVALID_DEVICE
v2:
Test for CL_SUCCESS instead of disabling it (Jan Vesely)
---
tests/cl/api/get-kernel-work-group-info.c | 28 +---
1 file changed, 21 insertions(+),
/tests/cl/api/compile-program.c
@@ -0,0 +1,345 @@
+/*
+ * Copyright © 2014 EdB edb+pig...@sigluy.net
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the Software),
+ * to deal in the Software without
On Tuesday 12 August 2014 17:25:37 Jan Vesely wrote:
On Sun, 2014-08-10 at 16:46 +0200, EdB wrote:
From: EdB e...@sigluy.net
If there is only one device associed with the kernel and device arg is
NULL
you don't have to trigger CL_INVALID_DEVICE
---
tests/cl/api/get-kernel-work
From: EdB e...@sigluy.net
If there is only one device associed with the kernel and device arg is NULL
you don't have to trigger CL_INVALID_DEVICE
---
tests/cl/api/get-kernel-work-group-info.c | 31 +++
1 file changed, 19 insertions(+), 12 deletions(-)
diff --git
42 matches
Mail list logo