AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
128,885 Commits 1 Branch 0 Tags
4fb2eddfdf9adafde2e6f94de23202ee44123d59
Commit Graph

30028 Commits

Author SHA1 Message Date
Alyssa Rosenzweig 8249e2b9a2 panfrost: Explicitly handle nr_cbufs=0 case 
Avoids some indirection around colour buffer count.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6373>
 2020-08-20 18:15:01 +02:00
Alyssa Rosenzweig dc7fbe114b panfrost: Drop implicit blend pooling 
We dropped blend shader pools a long time ago for various reasons, but
the indirection required remained. Oops.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6373>
 2020-08-20 18:15:00 +02:00
Alyssa Rosenzweig d8deb1eb6a panfrost: Share tiler_heap across batches/contexts 
There's only one tiler, so this is safe. (The blob does the same
optimization.) This avoids allocating multiple heaps for multiple
batches, which wastes memory and CPU time.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6373>
 2020-08-20 18:15:00 +02:00
Alyssa Rosenzweig b41692caf4 panfrost: Avoid minimum stack allocations 
If stack isn't used, don't allocate it - simple as that.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6373>
 2020-08-20 18:15:00 +02:00
Alyssa Rosenzweig 17c617cdb7 panfrost: Introduce invisible pool 
Whereas the main batch->pool is CPU read/write, the new
batch->invisible_pool is not. This enables GPU-internal structures that
the CPU must allocate from a pool dynamically but does not read,
corresponding to the BO_INVISIBLE create flag.

The use case is speeding up varying allocation by skipping the
CPU-side mmap/munmap.

We simultaneously half the pool's minimal allocation to avoid negatively
affecting memory usage.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6373>
 2020-08-20 18:15:00 +02:00
Alyssa Rosenzweig 40c0d7a13d panfrost: Pre-allocate memory for pool 
Seeing random BO creates in the middle of command stream generation is
annoying, let's preallocate memory so CPU profiling is less random.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6373>
 2020-08-20 18:15:00 +02:00
Kenneth Graunke 3fed1c75ef iris: Fix headerless sampler messages in compute shaders with preemption 
We were failing to set the "Headerless Message for Preemptable Contexts"
bit in SAMPLER_MODE in the compute context.  Other drivers use a single
hardware context, so setting it on the render engine was sufficient to
flip it in both pipelines.  But iris uses a separate hardware context
for compute, so we were only getting these set for the render context.

Thanks to Jason Ekstrand for catching this bug.

Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6380>
 2020-08-20 14:57:40 +00:00
Nanley Chery 9d503b36ca iris: Drop buffer support in resource_from_handle 
The callers don't seem to pass targets of PIPE_BUFFER. Stop nesting an
if-else block by dropping support for this target.

Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6311>
 2020-08-19 22:30:39 +00:00
Nanley Chery c54bb23967 iris: Add and use iris_resource_configure_main 
Reduce duplication of ISL surface creation code for main surfaces. This
is similar to iris_resource_configure_aux.

Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6311>
 2020-08-19 22:30:39 +00:00
Nanley Chery 51e42e2183 iris: Move size/offset calculations out of configure_aux 
Not all callers want the result of the offset and size calculations.

Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6311>
 2020-08-19 22:30:39 +00:00
Nanley Chery 8b2fc91959 iris: Drop old comment on clear color BO allocation 
Iris hasn't used a clear color BO on gen9 since commit 1cd13ccee7.

Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6311>
 2020-08-19 22:30:39 +00:00
Nanley Chery fd3c20674e iris: Drop unused resource allocation optimization 
Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6311>
 2020-08-19 22:30:39 +00:00
Nanley Chery 2555321a10 iris: Drop iris_resource_alloc_separate_aux 
When this function was called, the res->mod_info->aux_usage parameter
was ISL_AUX_USAGE_NONE. As a result, no aux buffer was ever allocated.

Reviewed-by: Anuj Phogat <anuj.phogat@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6311>
 2020-08-19 22:30:39 +00:00
Karol Herbst 5ef9c110d5 nv50/ir/nir: support load_work_dim 
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6367>
 2020-08-19 18:11:42 +00:00
Karol Herbst 080bee33da nvc0: handle nr being 0 in nvc0_set_global_bindings 
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6367>
 2020-08-19 18:11:42 +00:00
Karol Herbst b283fb335b nv50/ir/nir: fix global_atomic_comp_swap 
Fixes: 20d0ae464c ("nv50/ir: implement global atomics and handle it for nir")
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6367>
 2020-08-19 18:11:42 +00:00
Karol Herbst 38e904f929 nv50/ir/nir: assert on unknown alu ops 
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6367>
 2020-08-19 18:11:42 +00:00
Karol Herbst ada8c76cca nve4: fix uploading unaligned sized input buffers 
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Reviewed-by: Pierre Moreau <dev@pmoreau.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6367>
 2020-08-19 18:11:42 +00:00
Rob Clark 7e3a788bd3 freedreno/a6xx: fix occlusion query with more than one tile 
We need to emit epilogue after each tile, not just after the last tile.

Fixes: 13fc03f4c0 ("freedreno/a6xx: Avoid stalling for occlusion queries")
Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6376>
 2020-08-19 14:03:42 +00:00
Rob Clark 5550bc0423 freedreno/a6xx: only generate streamout for draw pass shader 
Previously we were doing this both for draw and binning pass.. resulting
that the stateobj contained first the incorrect state (based on binning
pass shader with varyings DCEd) followed by the correct state.

Also, in the streamout case we should link binning pass VS against the
real FS, rather than the dummy FS, so that OUTLOC's match (since same
streamout stateobj is used for both draw and binning pass).

Signed-off-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6376>
 2020-08-19 14:03:42 +00:00
Alyssa Rosenzweig 415eb43fd5 panfrost: Fix shared memory size computation 
Based on core count. Also, avoid some of the more complex programming
and stick to powers-of-two for now.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6321>
 2020-08-19 12:08:51 +00:00
Alyssa Rosenzweig 39bf1fb322 panfrost: Fix WRITES_GLOBAL bit 
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Fixes: dce7722ef8 ("panfrost: Handle writes_memory correctly")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6321>
 2020-08-19 12:08:51 +00:00
Eduardo Lima Mitev a457256e2d freedreno: Destroy syncobj too when destroying fence 
This is a missing part of freedreno EXT_semaphore support,
recently merged as part of series adding EXT_external_objects. This hunk was
actually lost due to a mistake on my side while doing a rebase.

Reviewed-by: Hyunjun Ko <zzoon@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6383>
 2020-08-19 09:11:21 +00:00
Alyssa Rosenzweig 1cf7a3968f panfrost: Hoist assert from bind to create 
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6327>
 2020-08-19 08:24:37 +00:00
Alyssa Rosenzweig 3b129dd335 panfrost: Remove useless comment 
Less than useless, actually - it's not even factually correct.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6327>
 2020-08-19 08:24:37 +00:00
Alyssa Rosenzweig a2e4159029 panfrost: Hoist add_fbo_bo call 
Move it to batch initialization, to take it out of the clear/draw path.
This should make sense - as soon as the batch is created, we need to
hold the refference. Functionally there shouldn't be unused batches
floating around anyway.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6327>
 2020-08-19 08:24:37 +00:00
Alyssa Rosenzweig 433fa39c2e panfrost: Drop QUADS primitive convert 
Even Bifrost retains QUADS support.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6327>
 2020-08-19 08:24:37 +00:00
Alyssa Rosenzweig 4b21c8b15f panfrost: Drop panfrost_invalidate_frame 
Only used to initialize active_queries once at context initialization,
then unneeded.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6327>
 2020-08-19 08:24:37 +00:00
Alyssa Rosenzweig 1362371a96 panfrost: Drop ZSA null checks in draws 
Likewise.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6327>
 2020-08-19 08:24:37 +00:00
Alyssa Rosenzweig ec35159fba panfrost: Drop rasterizer null checks in draw calls 
rasterizer must be non-null when draw_vbo is called (indeed, we assert
this there), so anything in the per-draw graphics path can safely assume
we have a rasterizer object bound.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6327>
 2020-08-19 08:24:37 +00:00
Alyssa Rosenzweig e46d4fe8bc panfrost: Decontextualize rasterizer 
Tightens the scope.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6327>
 2020-08-19 08:24:36 +00:00
Dave Airlie eb787773b2 llvmpipe/nir: lower some more packing at nir level. 
Fixes:
dEQP-VK.glsl.builtin.function.pack_unpack.*

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6339>
 2020-08-19 10:16:49 +10:00
Dave Airlie 95d611e036 llvmpipe: lower mul 2x32_64 
Fixes:
dEQP-VK.glsl.builtin.function.integer.umulextended*
dEQP-VK.glsl.builtin.function.integer.imulextended*

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6339>
 2020-08-19 10:16:46 +10:00
Dave Airlie 25ba200985 llvmpipe: lower cs local index from id 
Fixes:
dEQP-VK.glsl.atomic_operations.*_shared

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6339>
 2020-08-19 10:16:43 +10:00
Dave Airlie e616223024 gallivm/nir: fixup compact TCS variable storage. 
This fixes a lot of tessellation shaders, since tess factors
get emitted with spir-v now.

Fixes a bunch of:
dEQP-VK.tessellation*
along with 7000 others.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6339>
 2020-08-19 10:16:36 +10:00
Dave Airlie fe9667e336 llvmpipe: lower uadd_carry/usub_borrow. 
Fixes:
dEQP-VK.glsl.builtin.function.integer.uaddcarry.*
dEQP-VK.glsl.builtin.function.integer.usubborrow.*

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6339>
 2020-08-19 10:16:31 +10:00
Dave Airlie e137bbb87f llvmpipe/cs: update compute counters not fragment shader. 
This was updating the wrong counters.

Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Fixes: a6f6ca37c8 ("llvmpipe: add initial shader create/bind/destroy variants framework.")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6339>
 2020-08-19 10:16:14 +10:00
Hyunjun Ko e0e9712a4d freedreno: support GL_EXT_semaphore 
Signed-off-by: Hyunjun Ko <zzoon@igalia.com>
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4565>
 2020-08-18 20:40:40 +00:00
Eduardo Lima Mitev f6187aa1c3 freedreno: Enable GL_EXT_memory_object and GL_EXT_memory_object_fd 
Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4565>
 2020-08-18 20:40:40 +00:00
Eduardo Lima Mitev e3c39e5052 freedreno: Implement pipe screen's get_device/driver_uuid() 
v2: Don't need to prefix included uuid headers path with
    'freedreno'. (Rob Clark)

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4565>
 2020-08-18 20:40:40 +00:00
Eduardo Lima Mitev 6bef126c62 freedreno: implement pipe screen's resource_from_memobj 
v2: Use 'prsc' and 'rsc', 'pmemobj' and 'memobj' for consistency with
    rest of the code. (Rob Clark)

v3: - Use the existing flag PIPE_BIND_LINEAR instead (Marek Olšák)
    - Assert that the resource is not intended for scanout (Rob Clark)
    - Use the fd_resource_allocate_and_resolve() helper (Rob Clark)
    - Check that bo's resolved size fit into memobj's bo size (Rob Clark)

v4: Don't steal memobj's bo, but share it instead by getting a new
ref. (Rob Clark)

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4565>
 2020-08-18 20:40:40 +00:00
Eduardo Lima Mitev 536ec9d7f5 freedreno: Refactor fd_resource_create_with_modifiers() into a helper 
The helper just allocates and resolves layout, but does not deal with
scanout buffers nor allocation of the actual bo.

The resolved bo size is returned as an output argument.

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4565>
 2020-08-18 20:40:40 +00:00
Eduardo Lima Mitev 0eecd3d684 freedreno: Implement memory object create/destroy for GL_EXT_memory_object 
v2: Add implementation of fd_memobj_destroy() virtual func, which was newly
added.

v3: The memobj bo must be non-NULL and destroyed as part of memobj
destruction (instead of its reference being stolen). (Rob Clark)

Reviewed-by: Rob Clark <robdclark@chromium.org>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4565>
 2020-08-18 20:40:40 +00:00
Dave Airlie f9e773b920 llvmpipe: enable GL 4.5 
This passes conformance on both the master and 4.6.0 (with patches)
branches. I'll be submitting results shortly from the 20.2.x branch
this lands in.

Cc: "20.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6044>
 2020-08-17 20:12:40 +00:00
Dave Airlie 2e40b2c155 llvmpipe/blit: for 32-bit unorm depth blits just copy 32-bit 
Don't have the shader do a 32-bit unorn to float conversion,
the values don't remain accurate then.

Fixes:
GTF-GL45.gtf30.GL3Tests.framebuffer_blit.framebuffer_blit_functionality_multisampled_to_singlesampled_blit
for 32-bit unorm depth buffers.

Cc: "20.2" <mesa-stable@lists.freedesktop.org>
Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6044>
 2020-08-17 20:12:40 +00:00
Karol Herbst d79c946fb3 nv50/ir/nir: rework indirect function_temp handling 
Signed-off-by: Karol Herbst <kherbst@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6338>
 2020-08-17 12:05:45 +00:00
Alyssa Rosenzweig ec58cda5da panfrost: Use packs for varying buffers 
In addition to better aesthetics, this automatically fixed multiple
instances of accidental CPU readback of GPU buffers -- in a hot path,
too!

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>
 2020-08-17 11:43:52 +00:00
Alyssa Rosenzweig 7ef205dabd panfrost: Use MALI_ATTRIBUTE_LENGTH 
Removes an unwanted reference to mali_attr_meta.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>
 2020-08-17 11:43:52 +00:00
Alyssa Rosenzweig 59fa26902d panfrost: Use pack for general varying 
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>
 2020-08-17 11:43:52 +00:00
Alyssa Rosenzweig 6c850637f5 panfrost: Use pack for XFB varying 
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Reviewed-by: Tomeu Vizoso <tomeu.vizoso@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6326>
 2020-08-17 11:43:52 +00:00