AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
120,222 Commits 1 Branch 0 Tags
ccfe5f9d288fa36e85ac1140f2eae8429decaae6
Commit Graph

120222 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Qiang Yu ccfe5f9d28 lima: delay add plb buffer to submit when flush 
Prepare for multi submit in which case plb buffer is known
only when flush.

Keep the write back buffer update which is needed for FB
dirty track.

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3755>
 2020-02-17 02:54:15 +00:00
Qiang Yu 92387ca236 lima: pass array as parameter to PLBU and VS command macros 
Don't assume the ctx parameter, prepare for moving PLBU and
VS arrary from lima_context to lima_submit and adding new
plbu_cmd_head array.

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3755>
 2020-02-17 02:54:15 +00:00
Qiang Yu c3bbe4f7f8 lima: remove lima_ctx_buff_va submit flags (v2) 
We don't have any shared lima_ctx_buff for both GP and PP,
so no need to have these flags.

v2:
still add submit in lima_ctx_buff_va because some bo need
to exist cross flush, so not every submit will call
lima_ctx_buff_alloc.

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3755>
 2020-02-17 02:54:15 +00:00
Qiang Yu 9f924c795b lima: always add texture bo to submit 
No matter texture desc change, we need to add texture to submit.
Otherwise texture may be freed before submit finish.

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3755>
 2020-02-17 02:54:15 +00:00
Qiang Yu 3c4ff27250 lima: use util_copy_framebuffer_state 
Use this helper to replace self written code.

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3755>
 2020-02-17 02:54:15 +00:00
Qiang Yu c8b53d8020 lima: remove definition of lima_is_scanout 
There is no implementation of this function.

Reviewed-by: Vasily Khoruzhick <anarsoul@gmail.com>
Signed-off-by: Qiang Yu <yuq825@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3755>
 2020-02-17 02:54:15 +00:00
Alyssa Rosenzweig 0c4a70b64d pan/decode: Remove extraneous newline 
pandecode_log already does this.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:50 -05:00
Alyssa Rosenzweig 8ab0bf1f93 pan/midgard: Use fprintf instead of printf for constants 
I was wondering where those constants disappeared to :-)

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Fixes: 968f36d1fc ("pan/midgard: Support disassembling to a file")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:50 -05:00
Alyssa Rosenzweig 6af14d3685 pan/midgard: Don't crash with constants on unknown ops 
Just use a dummy name instead.. we can't know a priori what type an
unknown op will consume, but we don't want to dereference a null
pointer.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Fixes: 24360966ab ("panfrost/midgard: Prettify embedded constant
prints")

Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:50 -05:00
Alyssa Rosenzweig 5c06ecd2c6 pan/midgard: Identify stack barrier flag 
In case thread local storage is used.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:49 -05:00
Alyssa Rosenzweig d3747fb1eb pan/midgard: Set xyzx swizzle for load_compute_arg 
Probably harmless but the w component doesn't appear valid so let's
match the blob... one less bit to be nervous about.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:48 -05:00
Alyssa Rosenzweig f0ee55ad2a pan/midgard: Infer tags entirely 
We're so close, again marking off a few edge cases is enough to allow us
to omit this data entirely. Woot!

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:48 -05:00
Alyssa Rosenzweig 57a84278fd pan/midgard: Imply next tags 
As long as we can disambiguate a few edge cases, we can imply next tags
entirely which cleans up the disassembly a fair bit (though not as much
as implying tags entirely would -- we'll get there!)

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:47 -05:00
Alyssa Rosenzweig 453c64663c pan/midgard: Overhaul tag handling 
We unify disparate metadata about tags into a single structure to ensure
information is not left out.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:47 -05:00
Alyssa Rosenzweig 9168e7a65d pan/midgard: Improve barrier disassembly 
Just move some state from unknowns to actual keywords.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:47 -05:00
Alyssa Rosenzweig d208212f80 pan/midgard: Use dummy tag for empty shaders 
Fixes INSTR_INVALID_ENC in dEQP-GLES31.functional.compute.basic.empty

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:47 -05:00
Alyssa Rosenzweig b2cab6b6db pan/midgard: Fix 32/64 mixed swizzle packing 
Occurs in SSBO address computation.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:47 -05:00
Alyssa Rosenzweig a55a2e02a5 pan/midgard: Allow jumping out of a shader 
This comes up as a `return;` instruction in a compute shader. We need to
use the special tag 1 to signify "break". Fixes numerous
INSTR_INVALID_ENC faults in dEQP-GLES31.functional.compute.basic.*

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:47 -05:00
Alyssa Rosenzweig 3f59098d1a pan/midgard: Implement barriers 
Barriers execute on the texture pipeline on Midgard, so let's
tentatively handle barrier() as conservatively as possible (forcing
memory barriers of both buffers and shared memory). Implementation isn't
quite there yet -- it doesn't look at interactions of adjacent barriers
like it's supposed to -- but the core is there.

Fixes dEQP-GLES31.functional.compute.basic.ssbo_local_barrier_single_invocation

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:47 -05:00
Alyssa Rosenzweig 4f0b928921 pan/midgard: Fix swizzles harder 
Just for disassembly for now~

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:47 -05:00
Alyssa Rosenzweig fbe1fd3de0 pan/midgard: Fix missing prefixes 
I was wondering where those were going... :)

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Fixes: c1952779d6 ("pan/decode: Dump to a file")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Alyssa Rosenzweig 521406a069 pan/midgard: Track pressure when scheduling ld/st 
Fixes RA failure in
dEQP-GLES31.functional.shaders.builtin_functions.common.modf.* (which
uses multiple indirect SSBO writes)

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Alyssa Rosenzweig 9603126b74 panfrost: Allocate RAM backing of shared memory 
Unlike other GPUs, Mali does not have dedicated shared memory for
compute workloads. Instead, we allocate shared memory (backed to RAM),
and the general memory access functions have modes to access shared
memory (essentially, think of these modes as adding this allocates base
+ workgroupid * stride  in harder). So let's allocate enough memory
based on the shared_size parameter and supply it.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Alyssa Rosenzweig 50138abb5a panfrost: Rename unknown2_8 to padding 
It's zero everywhere.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Alyssa Rosenzweig 6d9ee3e65a panfrost: Rename bifrost_framebuffer->mali_framebuffer 
(And bifrost_fb_extra to mali_framebuffer_extra, bifrost_render_target
to mali_render_target)

These structures are the norm on midgard t760+, drop the bifrost names,
it's silly... unrelated to the rest of the series but while I'm messing
with pandecode and cleaning up bifrost abstractions, might as well.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Alyssa Rosenzweig 6dc105555b panfrost: Unify bifrost_scratchpad with mali_shared_memory 
It looks like these are the same structure, so this allows us to reuse
mali_shared_memory across architectures, and dispels with the
Bifrost-specific mystery of the scratchpads... nothing so mysterious
after all, just stack.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Alyssa Rosenzweig 254f40fd53 panfrost: Identify mali_shared_memory structure 
This small structure is used to configure shared memory and stack for
compute shaders, and is also present at the beginning of framebuffer
descriptors. Let's factor it out.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Alyssa Rosenzweig 418ca5dc1a panfrost: Ensure compute shader_meta is zeroed 
In theory the hardware doesn't care but it'll make for easier traces.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Alyssa Rosenzweig 058faf5a4b panfrost: Update comment about magic number relating to barriers 
It's a complicated story. But from what I can tell, in GL compute
without barriers, the blob is able to redistribute the workgroups in
various ways (that are not yet understood), whereas with barriers it
cannot redistribute anything, which accounts for erratic workgroup
packing without barriers.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3835>
 2020-02-16 09:16:46 -05:00
Dave Airlie 8f5a252d35 ci: bump debian image and change llvm deps to 8 
v3: remove version in a few places (Michel)

Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3805>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3805>
 2020-02-15 04:15:00 +00:00
Dave Airlie e7375e1795 gallivm/s390: fix pass init order on s390 with llvm 8 (v2) 
llvm 8 has some missing pass dependencies, fix the s390 case
as well.

v2: add ARM also (Michel)

Reviewed-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3805>
 2020-02-15 04:15:00 +00:00
Kenneth Graunke a603822b2f iris: Trim "../../src/gallium/drivers/iris/" out of debug dump filenames 
Easier to read.

v2: Also trim "/iris/" (Jordan Justen)

Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3830>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3830>
 2020-02-15 00:55:55 +00:00
Kenneth Graunke 96f247d1b3 iris: Dump frame markers with INTEL_DEBUG=submit 
Now you can see which batches go with which frames.

Acked-by: Jason Ekstrand <jason@jlekstrand.net>
Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3830>
 2020-02-15 00:55:55 +00:00
Marek Olšák e395ce03e9 gallium/cso_hash: remove another layer of pointer indirection 
Convert this:

    struct cso_hash {
       union {
          struct cso_hash_data *d;
          struct cso_node      *e;
       } data;
    };

to this:

    struct cso_hash {
       struct cso_hash_data data;
       struct cso_node      *end;
    };

1) data is not a pointer anymore.
2) "end" points to "data" and acts as the end of the linked list.
3) This code is still crazy.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Tested-by: Marge Bot <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:28 -05:00
Marek Olšák e0bb7b87e2 gallium/cso_hash: cosmetic changes, no behavior changes 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:28 -05:00
Marek Olšák 789ed29d59 gallium/cso_hash: remove always constant variable nodeSize 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:28 -05:00
Marek Olšák a8bbf10540 gallium/cso_hash: make cso_hash declared within structures instead of alloc'd 
This removes one level of indirection.

Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:28 -05:00
Marek Olšák f8594a06e4 gallium/cso_hash: inline a bunch of functions 
I'm probably not getting anything out of this, but it's harmless.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák cf86f522b2 gallium/u_vbuf: adjust the heuristic for unrolling indices 
This improves performance in the first subtest of Viewperf11/Catia by 10%.

Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák 55d8baa285 gallium/u_upload_mgr: don't do align twice in the u_upload_alloc fast path 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák 19c18d532e gallium/u_upload_mgr: reduce dereferences by adding buffer_size 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák 909a2d0ed3 st/mesa: simplify releasing the current attrib buffer 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák 6954efce23 st/mesa: make st_setup_current static 
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák e3617fd00b st/mesa: change some loops from while to do..while in st_atom_array.c 
Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák fd6636ebc0 st/mesa: simplify determination whether a draw needs min/max index 
Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák 1d93372802 st/mesa: simplify determination whether a draw has user vertex buffers 
Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák 61e4c582e0 st/mesa: always inline the code setting non-64bit vertex elements 
Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák 3c98dccd40 mesa: remove unused _mesa_draw_indirect 
All drivers that expose ARB_draw_indirect also set the driver callback.

Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Marek Olšák e6448f993b mesa: translate into gallium vertex formats in mesa/main 
Reviewed-by: Mathias Fröhlich <mathias.froehlich@web.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/merge_requests/3829>
 2020-02-14 18:16:27 -05:00
Francisco Jerez 8d3b86e34a intel/fs/gen7+: Implement discard/demote for SIMD32 programs. 
At this point this simply involves fixing the initialization of the
sample mask flag register to take the right dispatch mask from the
thread payload, and fixing sample_mask_reg() to return f1.1 for the
second half of a SIMD32 thread.  This improves Manhattan 3.1
performance by 2.4%±0.31% (N>40) on my ICL with SIMD32 enabled
relative to falling back to SIMD16 for the shaders that use discard.

Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
 2020-02-14 14:31:49 -08:00