Hello list,

The candidate for the Mesa 17.0.5 is now available. Currently we have:
 - 38 queued
 - 7 nominated (outstanding)
 - and 2 rejected patch(es)


The current queue contains fixes in several fronts.

We have a fix for nouveau regarding the instructions emission with
GF100's ISA encoding.

We have several fixes for Intel drivers.  The Vulkan driver has some
corrections for flusing the cache (VF and texture) while setting up a
null surface state, which is cheaper than calculating when one is
needed and corrections for the handling of the VK_ATTACHMENT_UNUSED. It
also includes a patch to disable CCS on BDW input attachments since it
is not supported by the sampler.  i965 includes a change in spilling's
cost heuristic that improves performance in Unigine Heaven, and another
fix in the register coalesce optimization.

In Mesa core we include corrections in the API validation, including a
set of patches for glMultiDrawArrays handling and a fix for the
validation of the sampler types.  Additionally, NIR includes a
correction when using ARB_shader_clock.

On integration side, there are several patches to fix configure and
build errors.

radv comes with a fix to properly report the timestampPeriod.

Finally, Gallivm brings a fix to avoid a hangs when apps call exit and
we also include a fix in the state tracker to invalidate the readpix
cache.

Take a look at section "Mesa stable queue" for more information.


Testing reports/general approval
--------------------------------

Any testing reports (or general approval of the state of the branch)
will be greatly appreciated.

The plan is to have 17.0.5 this Friday (28th of April), around or
shortly after 22:00 GMT.

If you have any questions or suggestions - be that about the current
patch queue or otherwise, please go ahead.


Trivial merge conflicts
-----------------------

commit c09f729d939e8f01ed26045156a2595e3f4b3cf0
Author: Jason Ekstrand <jason.ekstr...@intel.com>

    anv/blorp: Flush the texture cache in UpdateBuffer

    (cherry picked from commit 939337e49fb8a6e16f83e4e030743f91b1b05cc5)

    Squashed with commit:

    (cherry picked from commit d71efbe5f2a0ff934b8e9eeb96cd680a83bc0259)

commit f26047871f5a56522f803e26dc7639edf20c7365
Author: Jason Ekstrand <jason.ekstr...@intel.com>

    anv/cmd_buffer: Flush the VF cache at the top of all primaries

    (cherry picked from commit e1f6fb8021bcdd407fae77917502891b9239f53c)

commit af58e0dbe6daa9f5de20f0d2fdd3c776ffebc704
Author: Jason Ekstrand <jason.ekstr...@intel.com>

    anv/cmd_buffer: Use the null surface state for ATTACHMENT_UNUSED

    (cherry picked from commit 21d2ca72d88e00ddf37c7e1030472e184a038195)

commit 255d689dcd09a8c3737b11d7491aeab3547a2525
Author: Jason Ekstrand <jason.ekstr...@intel.com>

    anv/blorp: Properly handle VK_ATTACHMENT_UNUSED

    (cherry picked from commit 220974b38dfcd557f4a6bc723e4b5d15add39f84)

commit d066714fbb1d2e1271c077a3749a61d8d9ed07ad
Author: Nicolai Hähnle <nicolai.haeh...@amd.com>

    mesa: move glMultiDrawArrays to vbo and fix error handling

    (cherry picked from commit 42d5465b9ba85b4918b9e6fb57994720e3c8a80b)

commit a7c86ab750bc7e712068ac32ce83ad81c0b21bf8
Author: Francisco Jerez <curroje...@riseup.net>

    intel/fs: Use regs_written() in spilling cost heuristic for improved 
accuracy.

    (cherry picked from commit ecc19e12dca95d2571d3761dea6dec24b061013c)

commit 4c2356f13cb199dad474d38fecd424d200e4cffd
Author: Francisco Jerez <curroje...@riseup.net>

    intel/fs: Take into account amount of data read in spilling cost heuristic.

    (cherry picked from commit 58324389be7bc7c5e10093b9cc0a8efa9b4c93a9)

commit 313f48f0300f3e07c026265fc692279a245e1616
Author: Nanley Chery <nanley.g.ch...@intel.com>

    anv/cmd_buffer: Disable CCS on BDW input attachments

    (cherry picked from commit d9d793696bf54e970491302605a1efd0aa182d1b)

commit df53e5844b383abfd1b2ece00d941150141008f9
Author: Rob Clark <robdcl...@gmail.com>

    util/queue: don't hang at exit

    (cherry picked from commit 6fb7935dedc87ffd767a2999f402ce1a46d18cce)

commit b7916134a0102218d6b4ec0d482bfd0e3aeeba43
Author: Timothy Arceri <tarc...@itsqueeze.com>

    mesa: validate sampler type across the whole program

    (cherry picked from commit d682f8aa8e0edd166166f87fcd774dd2d57b4180)


Cheers,
Andres


Mesa stable queue
-----------------

Nominated (7)
=============

Marek Olšák (7):
      radeonsi/gfx9: fix texture buffer objects and image buffers with IDXEN==0
      radeonsi/gfx9: fix most things wrong with shader images
      radeonsi/gfx9: fix 1D array shader images
      radeonsi/gfx9: add a workaround for viewing a slice of 3D as a 2D image
      radeonsi/gfx9: set MAX_PRIMGRP_IN_WAVE in the correct register
      radeonsi/gfx9: don't set deprecated field PARTIAL_ES_WAVE_ON
      radeonsi: adjust ESGS ring buffer size computation on VI

Queued (38)
===========

Andres Gomez (15):
      cherry-ignore: Add the pci_id into the shader cache UUID
      cherry-ignore: fix crash if ctx torn down with no rendering
      cherry-ignore: Fix typos.
      cherry-ignore: Revert "etnaviv: Cannot render to rb-swapped formats"
      cherry-ignore: Revert "i965/fs: Don't emit SEL instructions for 
type-converting MOVs."
      cherry-ignore: fix typo in a2b10g10r10 fast clear calculation
      cherry-ignore: remove unused anv_dispatch_table dtable
      cherry-ignore: remove unused radv_dispatch_table dtable
      cherry-ignore: make radv_resolve_entrypoint static
      cherry-ignore: vulkan: add support for libmesa_vulkan_util
      cherry-ignore: r600: fix libmesa_amd_common dependency
      cherry-ignore: remove dead brw_new_shader() declaration
      cherry-ignore: remove i965_symbols_test reference from .gitignore
      cherry-ignore: automake: ensure that the destination directory is created
      cherry-ignore: provide required gem stubs for the tests

Boyan Ding (2):
      nvc0/ir: Properly handle a "split form" of predicate destination
      nir: Destination component count of shader_clock intrinsic is 2

Emil Velikov (4):
      winsys/sw/dri: don't use GNU void pointer arithmetic
      st/clover: add space between < and ::
      configure.ac: check require_basic_egl only if egl enabled
      st/mesa: automake: honour the vdpau header install location

Francisco Jerez (2):
      intel/fs: Use regs_written() in spilling cost heuristic for improved 
accuracy.
      intel/fs: Take into account amount of data read in spilling cost 
heuristic.

Grazvydas Ignotas (1):
      radv: report timestampPeriod correctly

Jason Ekstrand (5):
      anv/blorp: Flush the texture cache in UpdateBuffer
      anv/cmd_buffer: Flush the VF cache at the top of all primaries
      anv/cmd_buffer: Always set up a null surface state
      anv/cmd_buffer: Use the null surface state for ATTACHMENT_UNUSED
      anv/blorp: Properly handle VK_ATTACHMENT_UNUSED

Kenneth Graunke (1):
      i965/vec4: Avoid reswizzling MACH instructions in opt_register_coalesce().

Marek Olšák (1):
      st/mesa: invalidate the readpix cache in st_indirect_draw_vbo

Nanley Chery (1):
      anv/cmd_buffer: Disable CCS on BDW input attachments

Nicolai Hähnle (4):
      mesa: fix remaining xfb prims check for GLES with multiple instances
      mesa: extract need_xfb_remaining_prims_check
      mesa: move glMultiDrawArrays to vbo and fix error handling
      vbo: fix gl_DrawID handling in glMultiDrawArrays

Rob Clark (1):
      util/queue: don't hang at exit

Timothy Arceri (1):
      mesa: validate sampler type across the whole program

Rejected (2)
============

Jason Ekstrand (1):
      anv: Add the pci_id into the shader cache UUID

The commit depends on earlier commit 656e30b6860 and some other
dependencies which did not land in branch, We can come up with a
different solution if we actually see that the problem that was
addressed by the solution in master gets important.

Vinson Lee (1):
      configure.ac: Fix typos.

The commit depends on earlier commit 7748c3f5eb1 which did not land in
branch and the part that could be merged is not really providing
anything relevant to stable.

_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev

Reply via email to