mesa: Changes to 'drm-gem'

Eric Anholt anholt at kemper.freedesktop.org
Fri May 23 05:07:45 UTC 2008 

The branch, drm-gem has been created
        Commits at: http://cgit.freedesktop.org/mesa/mesa/log/?h=drm-gem

        at  6cefae5354fb3015c5a14677071871613faa9c3a (commit)

- Commits -----------------------------------------------
commit 6cefae5354fb3015c5a14677071871613faa9c3a
Author: Eric Anholt <eric at anholt.net>
Date:   Mon May 19 15:42:00 2008 -0700

    Add back a mostly-correct glFinish for GEM and fake.
    
    The right solution would probably be keeping a list of regions which have been
    rendered to.

commit 76286bc76c5ea2217378809a9dcab6794aae7b5e
Author: Keith Packard <keithp at keithp.com>
Date:   Thu May 22 10:44:47 2008 -0700

    [intel-gem] Make sure set_domain is called often enough.
    
    The write_domain needs to be set after any batch buffer uses an object,
    track when that happens in the new 'cpu_domain_set' field.

commit 8b49cc104dd556218fc769178b96f4a8a428d057
Author: Keith Packard <keithp at keithp.com>
Date:   Sat May 17 23:34:47 2008 -0700

    [intel-gem] Don't calloc reloc buffers
    
    Only a few relocations are typically used, so don't clear the
    whole thing.

commit cba90d4a778ab5202f2f7547ad7ffeb84216d07b
Author: Eric Anholt <eric at anholt.net>
Date:   Tue May 13 11:30:18 2008 -0700

    [GEM] Actually include the presumed offset in initial relocations.
    
    This avoids kernel relocations for most batchbuffer relocs.

commit 145523ba3acb95a9ff390430a9e0a3fa958cae1b
Author: Keith Packard <keithp at keithp.com>
Date:   Sun May 11 00:16:25 2008 -0700

    [intel] update GEM api. Add bo_subdata and bo_get_subdata driver hooks.
    
    Track DRM GEM name changes.
    Add driver hooks for bo_subdata and bo_get_subdata so that GEM can use pread
    and pwrite.

commit 0cb006c1fdb75e1fe282120cc5455a4e8c59b1a7
Author: Keith Packard <keithp at keithp.com>
Date:   Thu May 8 10:38:55 2008 -0700

    [intel-gem] move domains to relocations. add set_domain to bo_map.
    
    Fix the kernel API to place the read/write domain information in the
    relocation instead of the buffer.

commit fda5687241f4ce5cab3bf2eac437b52d4b37dd10
Author: Keith Packard <keithp at keithp.com>
Date:   Thu May 8 10:37:23 2008 -0700

    [intel] intel_batchbuffer_flush using uninit 'used' to check for buffer empty
    
    Make sure 'used' tracks the right value through the whole function.
    Also, use GLint for intel_batchbuffer_space in case we do bad things
    in the future.

commit 68a916183276f727948e73fe752ebf36ef9f8ba9
Merge: c75b2d74d88e9926cbd10aa1eee4aeba93ba4850 ead798eb103e4cfe801704bc15eb4fe8df078fa8
Author: Keith Packard <keithp at keithp.com>
Date:   Wed May 7 20:20:26 2008 -0700

    Merge commit 'anholt/drm-gem' into drm-gem

commit ead798eb103e4cfe801704bc15eb4fe8df078fa8
Author: Eric Anholt <eric at anholt.net>
Date:   Wed May 7 14:01:03 2008 -0700

    GEM: Remove already-disabled PIPE_CONTROL command.
    
    This existed to get the icache flushed.  However, GEM handles this for us
    now for sure, and we had disabled it prematurely anyway.

commit ab50ddaa9173ae108833db0edb209045788efc41
Author: Eric Anholt <eric at anholt.net>
Date:   Wed May 7 13:51:29 2008 -0700

    GEM: Make dri_emit_reloc take GEM domain flags instead of TTM flags.
    
    The GEM flags are much more descriptive for what we need.  Since this makes
    bufmgr_fake rather device-specific, move it to the intel common directory.
    We've wanted to do device-specific stuff to it before.

commit 8b2a7f08bc446deef497f2a0d3b54d9b70bdaf9c
Author: Eric Anholt <eric at anholt.net>
Date:   Wed May 7 10:01:14 2008 -0700

    GEM: Don't emit an extra MI_FLUSH in the batch since GEM handles it.

commit c75b2d74d88e9926cbd10aa1eee4aeba93ba4850
Merge: a2ec8570aeb838700fa97b8c5ba6d9d383e5606e 42d4f89264f193e8beae7ba975df3507b81b6da0
Author: Keith Packard <keithp at keithp.com>
Date:   Tue May 6 22:07:58 2008 -0700

    Merge commit 'anholt/drm-gem' into drm-gem

commit a2ec8570aeb838700fa97b8c5ba6d9d383e5606e
Author: Keith Packard <keithp at keithp.com>
Date:   Tue May 6 22:06:41 2008 -0700

    [intel-GEM] partial support for memory domains.
    
    Doesn't deal with local modifications yet (need new kernel set_domain ioctl
    for that to work). Also, guesses what domains are affected based on the
    read/write bits set in the flags. Works for 915, probably not so much for
    965.

commit 42d4f89264f193e8beae7ba975df3507b81b6da0
Author: Eric Anholt <eric at anholt.net>
Date:   Tue May 6 15:25:51 2008 -0700

    GEM: Fix previous commit to avoid asserting when we run into reserved space.
    
    These are the dwords that the reserved space is for.

commit 96f52f089f42b4bca8fa5fb573c687d233851126
Author: Eric Anholt <eric at anholt.net>
Date:   Tue May 6 13:57:08 2008 -0700

    GEM: Don't emit an extra MI_FLUSH in the batch since GEM handles it.

commit 537bbe6dec780f6f85838fe7e6036579c509f8a6
Author: Keith Packard <keithp at keithp.com>
Date:   Tue May 6 10:51:08 2008 -0700

    [intel-GEM] Add tiling support to swrast.
    
    Accessing tiled surfaces without using the fence registers requires that
    software deal with the address swizzling itself.

commit df4b49c2cedde60c02f869977ee426f280b2985b
Author: Keith Packard <keithp at keithp.com>
Date:   Mon May 5 22:08:05 2008 -0700

    Dump buffer tiled status from intelPrintSAREA

commit e9a2a67745d46509928263f0556f5c0a4211b94f
Author: Eric Anholt <eric at anholt.net>
Date:   Mon May 5 15:45:15 2008 -0700

    GEM: Allocate the right number of relocs, avoiding heap smashing.

commit 1decab06d15f0dead0a544dbed2f10041caac844
Author: Eric Anholt <eric at anholt.net>
Date:   Mon May 5 15:44:49 2008 -0700

    GEM: Include target buffer handle in relocation debug.

commit 01d1a292bf53ab949cf2075f18986b58fa468a61
Author: Eric Anholt <eric at anholt.net>
Date:   Mon May 5 14:20:18 2008 -0700

    GEM: Set validate index to keep the same buffer from being duped on the list.

commit be59d52ca0c0a5b93963297d596972fccb792b69
Author: Eric Anholt <eric at anholt.net>
Date:   Mon May 5 14:15:40 2008 -0700

    Print GEM handles instead of BO pointers in debugging.
    
    small integers are much prettier, and let me correlate to DRM debug output.

commit 5290ec4756eb33ec27e06bb68d64c33472276ac3
Author: Eric Anholt <eric at anholt.net>
Date:   Mon May 5 13:45:03 2008 -0700

    Initialize bufmgr_gem->validate_array[i].offset.
    
    This is just cosmetic, to produce less scary values when the ioctl fails and
    doesn't return values there.

commit 1f810b85b1e9393c8e606d2f28250cbb19cf916b
Author: Eric Anholt <eric at anholt.net>
Date:   Mon May 5 13:40:50 2008 -0700

    Make intel_{batch,exec}_ioctl return an error code so we can recover better.

commit 87ccc03736166db9ef85f3eee3723b82f395d3cf
Author: Keith Packard <keithp at keithp.com>
Date:   Mon May 5 10:46:27 2008 -0700

    Add intel_bufmgr_gem.c to i915

commit 367b1e35dc1dbeda65709b0ab4f7983d0c7a6cc2
Author: Keith Packard <keithp at keithp.com>
Date:   Mon May 5 10:45:30 2008 -0700

    Temporarily disable intel pixel ops on i915 for GEM
    
    Instead of attempting to fix these for GEM, just disable until GEM is
    working.

commit 81ec0545c93d57f72cff5099c6a34f04e9257a38
Author: Eric Anholt <eric at anholt.net>
Date:   Fri May 2 18:25:00 2008 -0700

    Don't forget to set handle of shared buffers.
    
    (And fix a nearby whitespace nit)

commit 3d19a095cda30ac8abdbe26cd3b664a4b97c899b
Author: Eric Anholt <eric at anholt.net>
Date:   Fri May 2 18:24:13 2008 -0700

    Fix GEM execbuf ioctl argument.

commit 7349f218b47b21595a13103aaa45ddbfdc14dd13
Author: Eric Anholt <eric at anholt.net>
Date:   Fri May 2 17:13:45 2008 -0700

    Fix to use GEM execbuf instead of TTM.

commit ef33e76cebed39551aabce397d165d3990ba517c
Author: Eric Anholt <eric at anholt.net>
Date:   Fri May 2 17:00:47 2008 -0700

    Minor fixups to get GEM to the point of execbuf ioctl.

commit eb10cdc838fc31ea2cf59f556f6f7d8b072f5bae
Author: Eric Anholt <eric at anholt.net>
Date:   Fri May 2 14:11:19 2008 -0700

    [intel] Fix build for GEM.  TTM is now disabled, and fencing is gone.
    
    Fencing was used in two places: ensuring that we didn't get too many frames
    ahead of ourselves, and glFinish.  glFinish will be satisfied by waiting on
    buffers like we would do for CPU access on them.  The "don't get too far ahead"
    is now the responsibility of the execution manager (kernel).

commit 89bba44e969f15bf20da6d700c493237b095a588
Author: Eric Anholt <eric at anholt.net>
Date:   Fri May 2 11:52:52 2008 -0700

    Add intel_bufmgr_gem for new graphics execution manager.

---------------------------------

More information about the mesa-commit mailing list