Changes:
- disallow NGG culling for GS, fast launch for tess using template args
(GS can't do NGG culling, tess can't do fast launch)
- skip checking current_rast_prim with tessellation
(bake the condition into ngg_cull_vert_threshold)
- use only 1 vertex count threshold for enabling NGG shader culling
to simplify it. I think it doesn't have a big impact. The threshold
computation depends on more parameters than just fast launch.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8434>
This decreases the release libgallium_dri.so size without debug symbols
by 16384 bytes. The CPU time spent in si_emit_draw_packets decreased
from 4.5% to 4.1% in viewperf13/catia/plane01.
The previous code did:
cs->current.buf[cs->current.cdw++] = ...;
cs->current.buf[cs->current.cdw++] = ...;
cs->current.buf[cs->current.cdw++] = ...;
cs->current.buf[cs->current.cdw++] = ...;
The new code does:
unsigned num = cs->current.cdw;
uint32_t *buf = cs->current.buf;
buf[num++] = ...;
buf[num++] = ...;
buf[num++] = ...;
buf[num++] = ...;
cs->current.cdw = num;
The code is the same (radeon_emit is redefined as a macro) except that
all set and emit functions must be surrounded by radeon_begin(cs) and
radeon_end().
radeon_packets_added() returns whether there has been any new packets added
since radeon_begin.
radeon_end_update_context_roll(sctx) sets sctx->context_roll = true
if there has been any new packets added since radeon_begin.
For now, the "cs" parameter is intentionally unused in radeon_emit and
radeon_emit_array.
Reviewed-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8653>
It's incorrect because si_get_vs_state returns gs_copy_shader for legacy
GS. It was harmless, but let's use si_get_vs, which is simpler.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8548>
It didn't do anything useful. GS doesn't use the other user SGPRs.
If we decrease the number of user SGPRs we declare for the GS prolog,
we can remove gfx9_prev_is_vs.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8344>
LLVM expects that exec != 0 when entering loops and generates this code
that becomes an infinite loop if exec == 0:
BB5_1:
vcc_lo = (inverted terminating condition)
s_and_b32 vcc_lo, exec_lo, vcc_lo
s_cbranch_vccnz BB5_3 // jump if vcc != 0 (break statement)
// ... loop body ...
s_branch BB5_1
BB5_3:
For non-monolithic VS before TCS, VS before GS, and TES before GS,
we set exec = (thread enabledmask), which sets 0 for HS-only and GS-only
waves, causing the infinite loop condition above.
Fix it as follows:
- set exec = ~0 at the beginning
- wrap the whole shader (LS and ES) in a conditional block, so that HS-only
and GS-only waves jump over it and never enter such a loop
The TES before GS hang can be reproduced by gfxbench:
testfw_app --gfx egl -w 1920 -h 1080 --gl_api gles -t gl_tess
Fixes: 68d6d097f1 - radeonsi/gfx9: add GFX9 and VEGA10 enums
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8344>
Enable vrs2x2 coarse shading if flat shading as per
idea and guidance given by Marek.
is_flat_shading variable in struct si_shader_info is set
based on the data from gather_intrinsic_info() function
and struct si_state_rasterizer. If is_flat_shading_variable
is set, then in function si_emit_db_render_state() vrs2x2
shading is enabled in hardware.
v2: Fix review comments from Pierre-Eric. Code optimizations.
v3: Fix indentation style issue.
v4: Fix review comments from Marek. Fixed logical issue pointed
by Marek where info->is_flat_shading variable can be corrupted
and other code cleanup.
v5: Make the code compact as suggested by Pierre-Eric.
v6: Fix new review comments from Marek.
v7: use info->uses_interp_color variable fix from Marek.
v8: Fix coding style comment from Marek.
v9: Add uses_fbfetch_output check as suggested by Marek.
Signed-off-by: Yogesh Mohan Marimuthu <yogesh.mohanmarimuthu@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8161>
Fixes:
- Sample shading now uses per-sample interpolation for colors if colors
are the only inputs. (this is the only case that was broken)
Optimizations:
- BC_OPTIMIZE (barycentric optimization) is now enabled with MSAA if colors
are qualified with both center and centroid. (BC_OPTIMIZE means that
the hardware skips initializing centroid (i,j) if they are equal to
center (i,j))
- If MSAA is disabled and at least 2 out of (center, centroid, sample) are
used by all inputs now including colors, center is forced for all inputs.
- If INTERP_MODE_COLOR is not used and the legacy GL shade model is flat,
the shader variant for flat shading is not generated.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8225>
The problem was that the shader constants were based on the framebuffer
sample count and ignored the multisample enable state and the line/polygon
smoothing state, which uses MSAA rasterization that only sets SampleMaskIn
to get the coverage for alpha-blended smoothing (the PS epilog computes
the alpha channel from SampleMaskIn and blending generates the AA results).
- This is a complete rework that adds a new state for NGG cull constants.
- It fixes the same thing for the prim discard compute shader.
- It documents how VS_STATE.SMALL_PRIM_PRECISION is encoded.
It fixes blue corruption in Unigine Heaven with MSAA and Medium details
or better.
Fixes: 7648060dc0 - radeonsi: enable NGG culling by default on gfx10.3 dGPUs
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8022>
It's straightforward except that the amdgpu winsys had to be cleaned up
to allow this.
radeon_cmdbuf is inlined and optionally the winsys can save the pointer
to it. radeon_cmdbuf::priv points to the winsys cs structure.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7907>
I think that reducing the CU mask to 1 disabled CU per SA broke the WGP mode
on VanGogh, causing a hang. To be sure, disable it on all chips.
Fixes: 9538b9a68e - radeonsi: add support for Sienna Cichlid
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7721>
with a sequence like this:
glClear(STENCIL)
glBeginTransformFeedback()
...
glEndTransformFeedback()
glClear(STENCIL)
The second clear sometimes may produce an unexpected result.
Calling si_flush_gfx_cs() when doing ngg -> legacy transition seems to be a
valid workaround (both for the synthetic reproducer and the real Blender bug).
Using flush flags or events (BOTTOM_OF_PIPE_TS, RESET_TO_LOWEST_VGT) didn't help.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/2941
Cc: mesa-stable
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7750>
It can only be done if a TCS input is accessed without indirect indexing and
with gl_InvocationID as the vertex index, and the number of VS and TCS threads
is the same.
This eliminates LDS stores and loads for VS->TCS IO, reducing shader lifetime
and LDS traffic.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7623>
Instead of:
if (VS) {
VS;
}
if (TCS) {
TCS;
}
Do this if the number of threads is the same in VS and TCS:
exec = enabled_threads;
VS;
TCS;
Skipping declare_vb_descriptor_input_sgprs is needed to match the VS return
values.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7623>
This improves performance for uber shaders.
It must be enabled using the new driconf option.
The driver compiles the specialized shaders in another thread without stalls,
same as all other optimizations.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/7057>
Add a vertex count threshold into si_shader_selector to simplify
the draw_vbo code.
The new option is supposed to be used in 00-mesa-defaults.conf and should be
tweaked for best performance unlike the AMD_DEBUG experimental options.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6948>
tess_rings must be encrypted when used in a secure job so this commit
introduces a tess_rings_tmz resource.
The cs_preamble_state doesn't contain the tess_rings address anymore since
it can change. The tess_rings related registers go in a separate preamble.
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6049>
Tag allocations as driver internal.
Some of these allocations will need to be doubled to handle TMZ (one secure bo,
one normal bo) but these allocations shouldn't switch the winsys in "the app
is using TMZ".
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6049>
Fixed-func shaders can contain the output, because their generator
doesn't consider the current primitive type into account.
Acked-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/6620>
