Your message dated Fri, 07 Dec 2012 10:02:43 +0000
with message-id <[email protected]>
and subject line Bug#695136: fixed in khronos-opencl-headers 1.2-2012.11.30-1
has caused the Debian Bug report #695136,
regarding opencl-headers: Include AMD extensions in opencl-headers
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)


-- 
695136: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=695136
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: opencl-headers
Version: 1.2-2012.04.18a-1
Severity: normal

  Hi,

  Members of the Khronos group take care toguether to not use the same ID, in
particular for extensions. Released Khronos headers include these vendor
extensions (for example CL_DEVICE_PROFILING_TIMER_OFFSET_AMD in cl_ext.h).
  However, these extensions are not always included (or they are but too late).

  In particular, AMD has some extensions that are not present in the current
headers (even the last released one that are not packaged yet). This leads
to difficulties for people that want to compile OpenCL programs for AMD
plate-form using opencl-headers (see #695112). Current packaging of AMD OpenCL
(in non-free) do not repackage AMD headers. Instead, they depend on
opencl-headers.

  We have several possibilities:
[A] include AMD (and other verndors) extensions into opencl-headers
  AMD SDK packaging can still depend on opencl-headers
[B] let opencl-headers be the official, unmodified Khronos headers
  This will imply that AML headers must be packaged:
  [B1] at the same place (with Replace/Conflict)
  [B2] with other name
  [B3] into another directory (with the same name)

B1 should be avoided if possible (will lead to complexity when other
  vendors also have extensions)
B2 imply that user program sources must be modified (to include the
  new header names such as "amd-opencl.h")
B3 imply that compiler must use a correct -I option

  Knowing the OpenCL (with ICD) have been designed to allow multiple
runtime OpenCL implementations at run-time (with fundamental problems
from my point of view) and that the Khronos group try hard to avoid header
conflict, I would push for the [A] solution.

  If you agree, I prepared a patch (probably to add with quilt into
the debian package, in order to clearly separate it from the plain
Khronos headers) that apply to the last Khonos headers (ie not the
one in the current package, see #695124) and that adds the missing
AMD extensions.

  Regards,
    Vincent

-- System Information:
Debian Release: wheezy/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable'), (1, 
'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386 armel mipsel

Kernel: Linux 3.2.0-4-amd64 (SMP w/8 CPU cores)
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

-- no debconf information
diff -ru a/cl_ext.h b/cl_ext.h
--- a/cl_ext.h	2012-11-30 12:49:46.000000000 +0100
+++ b/cl_ext.h	2012-05-14 09:44:34.000000000 +0200
@@ -40,6 +40,9 @@
	#include <CL/cl.h>
 #endif
 
+/* cl_khr_fp64 extension - no extension #define since it has no functions  */
+#define CL_DEVICE_DOUBLE_FP_CONFIG                  0x1032
+
 /* cl_khr_fp16 extension - no extension #define since it has no functions  */
 #define CL_DEVICE_HALF_FP_CONFIG                    0x1033
 
@@ -169,11 +130,46 @@
 #define CL_DEVICE_KERNEL_EXEC_TIMEOUT_NV            0x4005
 #define CL_DEVICE_INTEGRATED_MEMORY_NV              0x4006
 
+/*********************************
+* cl_amd_device_memory_flags *
+*********************************/
+#define cl_amd_device_memory_flags 1
+
+#define CL_MEM_USE_PERSISTENT_MEM_AMD       (1 << 6)        // Alloc from GPU's CPU visible heap
+
+/* cl_device_info */
+#define CL_DEVICE_MAX_ATOMIC_COUNTERS_EXT           0x4032
 
 /*********************************
 * cl_amd_device_attribute_query *
 *********************************/
 #define CL_DEVICE_PROFILING_TIMER_OFFSET_AMD        0x4036
+#define CL_DEVICE_TOPOLOGY_AMD                      0x4037
+#define CL_DEVICE_BOARD_NAME_AMD                    0x4038
+#define CL_DEVICE_GLOBAL_FREE_MEMORY_AMD            0x4039
+#define CL_DEVICE_SIMD_PER_COMPUTE_UNIT_AMD         0x4040
+#define CL_DEVICE_SIMD_WIDTH_AMD                    0x4041
+#define CL_DEVICE_SIMD_INSTRUCTION_WIDTH_AMD        0x4042
+#define CL_DEVICE_WAVEFRONT_WIDTH_AMD               0x4043
+#define CL_DEVICE_GLOBAL_MEM_CHANNELS_AMD           0x4044
+#define CL_DEVICE_GLOBAL_MEM_CHANNEL_BANKS_AMD      0x4045
+#define CL_DEVICE_GLOBAL_MEM_CHANNEL_BANK_WIDTH_AMD 0x4046
+#define CL_DEVICE_LOCAL_MEM_SIZE_PER_COMPUTE_UNIT_AMD   0x4047
+#define CL_DEVICE_LOCAL_MEM_BANKS_AMD               0x4048
+
+typedef union
+{
+    struct { cl_uint type; cl_uint data[5]; } raw;
+    struct { cl_uint type; cl_char unused[17]; cl_char bus; cl_char device; cl_char function; } pcie;
+} cl_device_topology_amd;
+
+#define CL_DEVICE_TOPOLOGY_TYPE_PCIE_AMD            1
+
+
+/**************************
+* cl_amd_offline_devices *
+**************************/
+#define CL_CONTEXT_OFFLINE_DEVICES_AMD              0x403F
 
 #ifdef CL_VERSION_1_1
    /***********************************
@@ -239,7 +235,11 @@
     #define CL_PARTITION_BY_COUNTS_LIST_END_EXT         ((cl_device_partition_property_ext) 0)
     #define CL_PARTITION_BY_NAMES_LIST_END_EXT          ((cl_device_partition_property_ext) 0 - 1)
 
-
+    /* cl_ext_atomic_counters_32 and cl_ext_atomic_counters_64 extensions 
+     * no extension #define since they have no functions                                              
+     */
+    #define CL_DEVICE_MAX_ATOMIC_COUNTERS_EXT           0x4032
+    
 
 #endif /* CL_VERSION_1_1 */
 
diff -ru a/cl.h b/cl.h
--- a/cl.h	2012-11-30 01:49:11.000000000 +0100
+++ b/cl.h	2012-05-14 09:44:34.000000000 +0200
@@ -1152,6 +1147,15 @@
                              const cl_event *  /* event_wait_list */,
                              cl_event *        /* event */) CL_API_SUFFIX__VERSION_1_2;
 
+extern CL_API_ENTRY cl_int CL_API_CALL
+clSetPrintfCallback(cl_context          /* context */,
+                    void (CL_CALLBACK * /* pfn_notify */)(cl_context /* program */, 
+                                                          cl_uint /*printf_data_len */, 
+                                                          char * /* printf_data_ptr */, 
+                                                          void * /* user_data */),
+                    void *              /* user_data */) CL_API_SUFFIX__VERSION_1_2;
+
+
 
 /* Extension function access
  *

--- End Message ---
--- Begin Message ---
Source: khronos-opencl-headers
Source-Version: 1.2-2012.11.30-1

We believe that the bug you reported is fixed in the latest version of
khronos-opencl-headers, which is due to be installed in the Debian FTP archive.

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to [email protected],
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Andreas Beckmann <[email protected]> (supplier of updated 
khronos-opencl-headers package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing [email protected])


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Format: 1.8
Date: Fri, 07 Dec 2012 10:42:50 +0100
Source: khronos-opencl-headers
Binary: opencl-headers
Architecture: source all
Version: 1.2-2012.11.30-1
Distribution: unstable
Urgency: low
Maintainer: Debian NVIDIA Maintainers <[email protected]>
Changed-By: Andreas Beckmann <[email protected]>
Description: 
 opencl-headers - OpenCL (Open Computing Language) header files
Closes: 684444 695124 695136
Changes: 
 khronos-opencl-headers (1.2-2012.11.30-1) unstable; urgency=low
 .
   * New upstream release.  (Closes: #695124, #684444)
   * Run some trivial compile tests during package build.
   * CL_EXT_SUFFIX__VERSION_1_2.diff: New, fix typo in macro name.
   * clSetPrintfCallback.patch: New, add missing declaration.
   * AMD_extensions.patch: New, add a few extensions from AMD that are not yet
     in the headers published by khronos. Thanks Vincent Danjean!
     (Closes: #695136)
Checksums-Sha1: 
 a346325ea39e36c360df4cdb6334fd9118d20b42 2201 
khronos-opencl-headers_1.2-2012.11.30-1.dsc
 736ec1e97c6a806801a468e22406b91118285f94 54153 
khronos-opencl-headers_1.2-2012.11.30.orig.tar.gz
 2043c2392687a01e2221589c7baa7bd48a4a1b21 5664 
khronos-opencl-headers_1.2-2012.11.30-1.debian.tar.gz
 b46809cf78be09ffb0bf713e81d1ac36a33b5df3 58264 
opencl-headers_1.2-2012.11.30-1_all.deb
Checksums-Sha256: 
 4a8a59da9f9978e22d9fa594c59c7f55a97c57f98014fa6794e31d619377e888 2201 
khronos-opencl-headers_1.2-2012.11.30-1.dsc
 76f53c2010b9200a8da0ddc4f5ea19fef5d84a4925d6430f736260d4990b77ab 54153 
khronos-opencl-headers_1.2-2012.11.30.orig.tar.gz
 6ec670e23931ae0fa7888eeb0ac522813b6fc8aa591ca1f208e9063f6037f47d 5664 
khronos-opencl-headers_1.2-2012.11.30-1.debian.tar.gz
 1617d5f88acf4eaeaf60f2a0362b3417e217d381218549341b887e6c550b7674 58264 
opencl-headers_1.2-2012.11.30-1_all.deb
Files: 
 6eb7b0de447649d130ce1be5ef29bac6 2201 libdevel extra 
khronos-opencl-headers_1.2-2012.11.30-1.dsc
 6edd745918cec1b31483b2c7e1308a03 54153 libdevel extra 
khronos-opencl-headers_1.2-2012.11.30.orig.tar.gz
 854ea68ebda70df0e2b210d384015c6d 5664 libdevel extra 
khronos-opencl-headers_1.2-2012.11.30-1.debian.tar.gz
 c72e3bd6c36575984952e1017cadf3b4 58264 libdevel extra 
opencl-headers_1.2-2012.11.30-1_all.deb

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
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=it7h
-----END PGP SIGNATURE-----

--- End Message ---

Reply via email to