| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
| |
These fields are of little importance as they're used to recognize
instructions.
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We used to print invalid data when the last field was being clamped to
32bits due to Dword Length of the whole instruction. Here is an
example where the decoder read part of the next instruction instead of
stopping at the 32bit limit:
0x000ce0b4: 0x10000002: MI_STORE_DATA_IMM
0x000ce0b4: 0x10000002 : Dword 0
DWord Length: 2
Store Qword: 0
Use Global GTT: false
0x000ce0b8: 0x00045010 : Dword 1
Core Mode Enable: 0
Address: 0x00045010
0x000ce0bc: 0x00000000 : Dword 2
0x000ce0c0: 0x00000000 : Dword 3
Immediate Data: 8791026489807077376
With this change we have the proper value :
0x000ce0b4: 0x10000002: MI_STORE_DATA_IMM (4 Dwords)
0x000ce0b4: 0x10000002 : Dword 0
DWord Length: 2
Store Qword: 0
Use Global GTT: false
0x000ce0b8: 0x00045010 : Dword 1
Core Mode Enable: 0
Address: 0x00045010
0x000ce0bc: 0x00000000 : Dword 2
0x000ce0c0: 0x00000000 : Dword 3
Immediate Data: 0
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Due to the new way we handle fields, we need *not* to forget the first
field when decoding instructions. The issue was that the advance
function was called first and skipped the first field.
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
| |
This should be inside the function that actually decodes fields.
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
| |
Making the next change more readable.
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For example, we were skipping Dword 3 in this PIPE_CONTROL :
0x000ce130: 0x7a000004: PIPE_CONTROL
DWord Length: 4
0x000ce134: 0x00000010 : Dword 1
Flush LLC: false
Destination Address Type: 0 (PPGTT)
LRI Post Sync Operation: 0 (No LRI Operation)
Store Data Index: 0
Command Streamer Stall Enable: false
Global Snapshot Count Reset: false
TLB Invalidate: false
Generic Media State Clear: false
Post Sync Operation: 0 (No Write)
Depth Stall Enable: false
Render Target Cache Flush Enable: false
Instruction Cache Invalidate Enable: false
Texture Cache Invalidation Enable: false
Indirect State Pointers Disable: false
Notify Enable: false
Pipe Control Flush Enable: false
DC Flush Enable: false
VF Cache Invalidation Enable: true
Constant Cache Invalidation Enable: false
State Cache Invalidation Enable: false
Stall At Pixel Scoreboard: false
Depth Cache Flush Enable: false
0x000ce138: 0x00000000 : Dword 2
Address: 0x00000000
0x000ce140: 0x00000000 : Dword 4
Immediate Data: 0
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
The xml files don't always have fields in order. This might confuse
our parsing of the commands. Let's have the fields in order. To do
this, the easiest way it to use a linked list. It also helps a bit
with the iterator.
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
|
|
| |
v2:
* Return unsigned instead of size_t. (Ken)
Signed-off-by: Jordan Justen <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Jordan Justen <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
|
| |
The caller can now use brw_stage_prog_data::program_size which is set
by the brw_compile_* functions.
Cc: Jason Ekstrand <[email protected]>
Signed-off-by: Jordan Justen <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
This will be used by the on disk shader cache.
v2:
* Set in brw_compile_* rather than brw_codegen_*. (Jason)
Signed-off-by: Timothy Arceri <[email protected]>
[[email protected]: Only add to brw_stage_prog_data]
Signed-off-by: Jordan Justen <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
|
| |
CannonLake additionally supports R11G11B10_FLOAT and four 10-10-10-2
formats with CCS_E. None of these formats fit within the current
blorp_copy framework so disable them until support is added.
Signed-off-by: Nanley Chery <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Groups containing fields smaller than a DWord were not being decoded
correctly. For example:
<group count="32" start="32" size="4">
<field name="Vertex Element Enables" start="0" end="3" type="uint"/>
</group>
gen_field_iterator_next would properly walk over each element of the
array, incrementing group_iter, and calling iter_group_offset_bits()
to advance to the proper DWord. However, the code to print the actual
values only considered iter->field->start/end, which are 0 and 3 in the
above example. So it would always fetch bits 3:0 of the current DWord
when printing values, instead of advancing to each element of the array,
printing bits 0-3, 4-7, 8-11, and so on.
To fix this, we add new iter->start/end tracking, which properly
advances for each instance of a group's field.
Caught by Matt Turner while working on 3DSTATE_VF_COMPONENT_PACKING,
with a patch to convert it to use an array of bitfields (the example
above).
This also fixes the decoding of 3DSTATE_SBE's "Attribute Active
Component Format" fields.
Reviewed-by: Jordan Justen <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Jordan Justen <[email protected]>
|
|
|
|
|
|
|
|
| |
Fixes: 2c873060d3578c7004c0 "i965: Delete unused
brw_vs_prog_data::nr_attributes field."
Cc: Kenneth Graunke <[email protected]>
Signed-off-by: Eric Engestrom <[email protected]>
Reviewed-by: Eduardo Lima Mitev <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
NIR does not have these instructions. TGSI and Mesa IR both implement
them using < and >=, repsectively. Removing them deletes a bunch of
code and means I don't have to add code to the SPIR-V generator for
them.
v2: Rebase on 2+ years of change... and fix a major bug added in the
rebase.
text data bss dec hex filename
8255291 268856 294072 8818219 868e2b 32-bit i965_dri.so before
8254235 268856 294072 8817163 868a0b 32-bit i965_dri.so after
7815339 345592 420592 8581523 82f193 64-bit i965_dri.so before
7813995 345560 420592 8580147 82ec33 64-bit i965_dri.so after
Signed-off-by: Ian Romanick <[email protected]>
Reviewed-by: Nicolai Hähnle <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Patch uses mem_ctx for allocation to ensure param array gets freed
later.
==6164== 48 bytes in 1 blocks are definitely lost in loss record 61 of 193
==6164== at 0x4C2EB6B: malloc (vg_replace_malloc.c:299)
==6164== by 0x12E31C6C: ralloc_size (ralloc.c:121)
==6164== by 0x130189F1: fs_visitor::assign_constant_locations() (brw_fs.cpp:2095)
==6164== by 0x13022D32: fs_visitor::optimize() (brw_fs.cpp:5715)
==6164== by 0x13024D5A: fs_visitor::run_fs(bool, bool) (brw_fs.cpp:6229)
==6164== by 0x1302549A: brw_compile_fs (brw_fs.cpp:6570)
==6164== by 0x130C4B07: blorp_compile_fs (blorp.c:194)
==6164== by 0x130D384B: blorp_params_get_clear_kernel (blorp_clear.c:79)
==6164== by 0x130D3C56: blorp_fast_clear (blorp_clear.c:332)
==6164== by 0x12EFA439: do_single_blorp_clear (brw_blorp.c:1261)
==6164== by 0x12EFC4AF: brw_blorp_clear_color (brw_blorp.c:1326)
==6164== by 0x12EFF72B: brw_clear (brw_clear.c:297)
Fixes: 8d90e28839 ("intel/compiler: Allocate pull_param in assign_constant_locations")
Signed-off-by: Tapani Pälli <[email protected]>
Reviewed-by: Lionel Landwerlin <[email protected]>
Cc: [email protected]
|
|
|
|
|
|
| |
This has been unused since we switched to nir_lower_wpos_ytransform.
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fixes intermittent GPU hangs on Broxton with an Intel internal
test case.
There are plenty of similar fragment shaders in piglit that do
not use any varyings and any uniforms. According to the
documentation special timing is needed between pipeline stages.
Apparently we just don't hit that with piglit. Even with the
failing test case one doesn't always get the hang.
Moreover, according to the error states the hang happens
significantly later than the execution of the problematic shader.
There are multiple render cycles (primitive submissions) in between.
I've also seen error states where the ACTHD points outside the
batch. Almost as if the hardware writes somewhere that gets used
later on. That would also explain why piglit doesn't suffer from
this - most tests kick off one render cycle and any corruption
is left unseen.
v2 (Ken): Instead of enabling push constants, enable one of the
inputs (PSIZ).
v3 (Ken, Jason): Use LAYER instead making vulkan emit_3dstate_sbe()
happy.
Cc: "17.3 17.2" <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
Signed-off-by: Topi Pohjolainen <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Asserting slot >= 2 made sense when the URB read offset was always 1
(pair of slots). Commit 566a0c43f0b9fbf5106161471dd5061c7275f761 made
it possible to read from the VUE header in slot 0, by adjusting the
offset to be 0. So, this assert is now bogus. Use the one from GL.
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Commit 566a0c43f0b9fbf5106161471dd5061c7275f761 started setting the
3DSTATE_SBE bit to override these values with the one calculated there.
So, they're dead. Stop setting them.
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
| |
Reviewed-by: Matt Turner <[email protected]>
|
|
|
|
|
|
| |
Signed-off-by: Kevin Rogovin <[email protected]>
Reviewed-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
|
|
|
| |
The disassembler does not (and should not) be modifying the data.
Signed-off-by: Kevin Rogovin <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
|
|
|
|
| |
Reviewed-by: Lionel Landwerlin <[email protected]>
|
|
|
|
| |
Reviewed-by: Lionel Landwerlin <[email protected]>
|
|
|
|
|
|
|
|
|
|
| |
We currently have a bug where nir_lower_system_values gets called before
nir_lower_var_copies so it will miss any system value uses which come
from a copy_var intrinsic. Moving it to after brw_preprocess_nir fixes
this problem.
Reviewed-by: Lionel Landwerlin <[email protected]>
Cc: [email protected]
|
|
|
|
|
|
| |
We already handle it in brw_preprocess_nir
Reviewed-by: Lionel Landwerlin <[email protected]>
|
|
|
|
| |
Reviewed-by: Lionel Landwerlin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
| |
The PRM says "The execution size must be 1." In 73137997e23ff6c11, the
execution size was set to 1 when it should have been BRW_EXECUTE_1
(which maps to 0). Later, in dc2d3a7f5c217a7cee9, JMPI was used for
line AA on gen6 and earlier and we started manually stomping the
exeution size to BRW_EXECUTE_1 in the generator. This commit fixes the
original bug and makes brw_JMPI just do the right thing.
Reviewed-by: Matt Turner <[email protected]>
Fixes: 73137997e23ff6c1145d036315d1a9ad96651281
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Returns the brw_type for a given ssa.bit_size, and a reference type.
So if bit_size is 64, and the reference type is BRW_REGISTER_TYPE_F,
it returns BRW_REGISTER_TYPE_DF. The same applies if bit_size is 32
and reference type is BRW_REGISTER_TYPE_HF it returns BRW_REGISTER_TYPE_F
v2 (Jason Ekstrand):
- Use better unreachable() messages
- Add Q types
Signed-off-by: Jose Maria Casanova Crespo <[email protected]>
Signed-off-by: Alejandro Piñeiro <[email protected]
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
| |
Reviewed-by: Lionel Landwerlin <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In order to implement the ballot intrinsic, we do a MOV from flag
register to some GRF. If that GRF is used in a SEL, cmod propagation
helpfully changes it into a MOV from the flag register with a cmod.
This is perfectly valid but when lower_simd_width comes along, it simply
splits into two instructions which both have conditional modifiers.
This is a problem since we're reading the flag register. This commit
makes us check whether or not flags_written() overlaps with the flag
values that we are reading via the instruction source and, if we have
any interference, will force us to emit a copy of the source.
Reviewed-by: Matt Turner <[email protected]>
Cc: [email protected]
|
|
|
|
|
|
|
| |
Signed-off-by: Jordan Justen <[email protected]>
Reviewed-by: Timothy Arceri <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
| |
Also needed in freedreno/ir3.
Signed-off-by: Rob Clark <[email protected]>
Reviewed-by: Eric Engestrom <[email protected]>
Reviewed-by: Dylan Baker <[email protected]>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
gcc is throwing this warning in my meson build:
../src/intel/compiler/brw_eu_validate.c:50:11: warning
argument 1 null where non-null expected [-Wnonnull]
return memmem(haystack.str, haystack.len,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
needle.str, needle.len) != NULL;
~~~~~~~~~~~~~~~~~~~~~~~
The first check for CONTAINS has a NULL error_msg.str and 0 len. The
glibc implementation will exit without looking at any haystack bytes if
haystack.len < needle.len, so this was safe, but silence the warning
anyway by guarding against implementation variablility.
Fixes: 122ef3799d56 ("i965: Only insert error message if not already present")
Reviewed-by: Matt Turner <[email protected]>
|
|
|
|
|
|
|
| |
v2: Warn that support is still in alpha (Jason)
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
| |
This workaround isn't listed on Gen10.
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Kenneth Graunke <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
| |
Signed-off-by: Lionel Landwerlin <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Align1 mode offers some nice features over align16, like access to more
data types and the ability to use a 16-bit immediate. This patch does
not start using any new features. It just emits ternary instructions in
align1 mode.
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
| |
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
| |
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
| |
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
| |
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
| |
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
| |
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
|
|
| |
Put hw_ in the name so that it's clear these are the hardware encodings.
Similar to commit 9fb832332868 ("i965: Rename brw_inst's functions that
access the register type")
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
| |
Reviewed-by: Scott D Phillips <[email protected]>
|
|
|
|
|
|
|
|
| |
The instruction word contains SubRegNum[4:2] so it's in units of dwords
(hence the * 4 to get it in terms of bytes). Before this patch, the
subreg would have been wrong for DF arguments.
Reviewed-by: Scott D Phillips <[email protected]>
|