[Bug 105950] radeonsi: OpenCL not working correctly on a big endian machine
https://bugs.freedesktop.org/show_bug.cgi?id=105950 Timothy Arceri changed: What|Removed |Added Resolution|--- |FIXED Status|NEW |RESOLVED --- Comment #3 from Timothy Arceri --- These patches were committed. Closing as fixed. -- You are receiving this mail because: You are the assignee for the bug.___ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 105950] radeonsi: OpenCL not working correctly on a big endian machine
https://bugs.freedesktop.org/show_bug.cgi?id=105950 --- Comment #2 from Bas Vermeulen--- Created attachment 138698 --> https://bugs.freedesktop.org/attachment.cgi?id=138698=edit Patch for dispatch_packet endianness fix -- You are receiving this mail because: You are the assignee for the bug.___ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 105950] radeonsi: OpenCL not working correctly on a big endian machine
https://bugs.freedesktop.org/show_bug.cgi?id=105950 --- Comment #1 from Bas Vermeulen--- Created attachment 138697 --> https://bugs.freedesktop.org/attachment.cgi?id=138697=edit Patch for si_vgt_param_key endian fix -- You are receiving this mail because: You are the assignee for the bug.___ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel
[Bug 105950] radeonsi: OpenCL not working correctly on a big endian machine
https://bugs.freedesktop.org/show_bug.cgi?id=105950 Bug ID: 105950 Summary: radeonsi: OpenCL not working correctly on a big endian machine Product: Mesa Version: 17.3 Hardware: PowerPC OS: Linux (All) Status: NEW Severity: normal Priority: medium Component: Drivers/Gallium/radeonsi Assignee: dri-devel@lists.freedesktop.org Reporter: b...@daedalean.ai QA Contact: dri-devel@lists.freedesktop.org Running any OpenCL program on a big endian machine using a radeonsi card fails. - union si_vgt_param_key uses a bitfield shared with a 32 bit value. This causes the index to be wrong on a big endian machine, and crashes any OpenCL program with a segfault. - si_setup_user_sgprs_co_v2 in si_compute.c fills a dispatch_packet struct with values which are uploaded to the GPU. To ensure this works correctly, these values need to be converted to little endian. I'll send patches to the mesa-dev list for both these issues. Bas Vermeulen -- You are receiving this mail because: You are the assignee for the bug.___ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel