Builder ffmpeg-solaris10-sparc Build #12766

Results:

Failed shell_2 shell_3 shell_4 shell_5

SourceStamp:

BuildSlave:

Reason:

The SingleBranchScheduler scheduler named 'schedule-ffmpeg-solaris10-sparc' triggered this build

Steps and Logfiles:

1. git update ( 10 secs )
   1. stdio
2. shell 'gsed -i ...' ( 0 secs )
   1. stdio
3. shell_1 'gsed -i ...' ( 0 secs )
   1. stdio
4. shell_2 'gsed -i ...' failed ( 0 secs )
   1. stdio
5. shell_3 './configure --samples="../../../ffmpeg/fate-suite" ...' failed ( 7 secs )
   1. stdio
   2. config.log
6. shell_4 'gmake fate-rsync' failed ( 0 secs )
   1. stdio
7. shell_5 '../../../ffmpeg/fate.sh ../../../ffmpeg/fate_config.sh' failed ( 3 secs )
   1. stdio
   2. configure.log
   3. compile.log
   4. test.log

Build Properties:

Forced Build Properties:

Responsible Users:

1. Russell Greene

   russellgreene8ohnoyoudont@gmail.com

Timing:

All Changes:

:

1. Change #250560

   Category	ffmpeg
   Changed by	Russell Greene <russellgreene8ohnoyoudont@gmail.com>
   Changed at	Sun 30 Nov 2025 22:47:12
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	3beaa2d70f99e6ec28c0120385ccd2abc922be25

   Comments

   hwcontext_vulkan: remove VK_HOST_IMAGE_COPY_MEMCPY flag
   Reading the spec for what this flag means, it copies the data verbatim, including any swizzling/tiling, this has two issues
   
   1. the format may not be what ffmpeg expects elsewhere, as it is expecing normal pitch linear host memeory in `swf`
   2. the size of the copied data may not match the size of buffer provided, causing heap buffer overflow
   
   It seems like addition of this flag is an oversight as it seems to be for caching/backups of image data, just to be used with copying back to the GPU with the MEMCPY flag, which is *not* how its used in ffmpeg.
   
   Additionally, set memoryRowLength as if it isn't set, it assumes pitch = width_in_bytes, which I don't think is necessarily the case

   Changed files

   • libavutil/hwcontext_vulkan.c