[Mesa-stable] Notes from the stable-branch maintainer

[Carl Worth]

First off, what happened to February? Or, in other words, I apologize
for slacking off so badly with the 10.0 branch lately.

I've just made a first pass over the pending commits, and I've
cherry-picked an impressive set (about 52 patches). I'll be pushing that
out after some testing.

But there are a number of patches that haven't made the cut in my first
pass, (the authors of which are all CCed here). Some of these simply
failed to cherry-pick cleanly. I'll be taking a look at these and I will
either merge it myself (if trivial) or else ask the author for help.

But there's also a number of proposed patches that fail to meet the
stable branch criteria. Some of the problems I have seen:

  * Patch is not a fix, (eg., moves code without functional change)

  * Patch is not obviously a fix, (commit message is mostly empty, no
    description of bug fix, no mention of bugzilla)

  * Patch has not obviously been reviewed (no Reviewed-by,
    Signed-off-by, nor Tested-by from anyone but the author)

  * Patch is too large

  * Patch depends on too many other patches

  * Patch introduces a new failure (such as an assert), which is not a
    bug fix appropriate for the stable branch

I'll follow up directly with the authors of each patch (or patch series)
with my specific determination. That will likely be as a reply to the
patch as originally sent to mesa-dev or mesa-stable. So look for those.

Or, if you'd like to act proactively, I've included the list of patches
not-yet-applied to 10.0 below. So if you want to take a look at these,
that could help. Here is some of the information you could give me that
would be particularly helpful:

  * Updated patch that cherry-picks cleanly to the 10.0 branch

  * An updated commit message that provides additional information that
    the original commit message is missing, (such as justification that
    a patch is a fix, or additional Reviewed-by tags).

  * Your agreement that a patch is inappropriate (for any of the reasons
    above).

So that's either a quick note to tell me to drop a patch (or series of
patches) or else a new version of the patch with merged content or a
better commit message.

But like I said, if I don't hear anything in advance, I'll be emailing
specific patch authors with my own determination of the status of each
patch .

Thanks,

-Carl

c7373b7 nv50: rework primid logic
364bdd2 nv50: fix layerid to be the fp input number rather than vp output number
f76c7ad nv50: only over-allocate by a page for code
882e98e nvc0: handle TGSI_SEMANTIC_LAYER
2e9ee44 nv50/ir/ra: some register spilling fixes
d34d5fd gallium: add geometry shader output limits
025d99c glsl: Do not vectorize vector array dereferences.
01e6371 radeon: reverse DBG_NO_HYPERZ logic
2e0c908 svga: split / update svga3d header files
0247113 svga: update dumping code with new GBS commands, etc
d993ada svga: update svga_winsys interface for GBS
823fbfd svga: add new GBS commands
31dfefc svga: add svga_have_gb_objects/dma() functions
2f1fc8d svga: update constant buffer code for GBS
f84c830 svga: update shader code for GBS
c1e60a6 svga: add helpers for tracking rendering to textures
d0c22a6 svga: track which textures are rendered to
f8bbd82 svga: adjust adjustment for point coordinates
6476bcb svga: remove a couple unneeded assertions
e0a6fb0 svga: add new helper functions for GBS buffers
72b0e95 svga: update buffer code for GBS
3d1fd6d svga: update texture code for GBS
c9e9b18 pipebuffer, winsys: Add a size match parameter to the cached buffer manager
8af358d gallium/pipebuffer: Add a cache buffer manager bypass mask
59e7c59 gallium/util: Add flush/map debug utility code
fe6a854 svga/winsys: implement GBS support
141e39a svga/winsys: Propagate surface shared information to the winsys
e4a5a9f gallium/pipebuffer: change pb_cache_manager_create() size_factor to float
cf0172d r600g,radeonsi: Consolidate logic for short-circuiting flushes
1f4bfb8 nv50/ir/ra: fix SpillCodeInserter::offsetSlot usage
ffde483 i965/vec4: Add a brw->gen >= 6 assertion in three-source emitters.
2fdea48 i965: Stop lowering ir_triop_lrp.
f5e681f winsys/svga: Avoid calling drm getparam for max surface size on older kernels
4bd7f1d glsl: Don't vectorize horizontal expressions.
7189fce i965: Allocate vec4_visitor's uniform_size and uniform_vector_size arrays dynamically.
8ba1570 i915: Only allow 8 vertex texture units
590920f i965: Widen sampler key bitfields for 32 samplers
aff7c5e mesa: don't call ctx->Driver.ClearBufferSubData() if size==0
dff3ecc radeonsi: move translate_colorswap to common code
ac35ded r600g: port color buffer format conversion from radeonsi
1337da5 r600g: implement edge flags
f34d75d c11/threads: Fix nano to milisecond conversion.
e8d8503 mapi/u_thread: Use GetCurrentThreadId
a61d859 c11/threads: Don't implement thrd_current on Windows.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/mesa-stable/attachments/20140304/24559be8/attachment.pgp>