AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

pan/bi: Remove conversion lowering

Browse Source 
Inherited from Midgard, it's easier to just do at NIR->BIR time now that
we have a builder that doesn't suck. Impeded certain optimizations.

Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10065>
This commit is contained in:
Alyssa Rosenzweig
2020-12-31 22:15:55 -05:00
committed by Marge Bot
parent 266de379c2
commit d01628d24d
1 changed files with 3 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/panfrost/bifrost/bifrost_nir_algebraic.py
+3 -41
View File
@@ -50,51 +50,13 @@ for sz in ('8', '16', '32'):
((t + 'max', 'a@' + sz, 'b@' + sz), ('b' + sz + 'csel', (t + 'lt' + sz, b, a), a, b))
]
# Midgard is able to type convert down by only one "step" per instruction; if
# NIR wants more than one step, we need to break up into multiple instructions
converts = []
for op in ('u2u', 'i2i', 'f2f', 'i2f', 'u2f', 'f2i', 'f2u'):
srcsz_max = 64
dstsz_max = 64
# 8 bit float doesn't exist
srcsz_min = 8 if op[0] != 'f' else 16
dstsz_min = 8 if op[2] != 'f' else 16
dstsz = dstsz_min
# Iterate over all possible destination and source sizes
while dstsz <= dstsz_max:
srcsz = srcsz_min
while srcsz <= srcsz_max:
# Size converter lowering is only needed if src and dst sizes are
# spaced by a factor > 2.
# Type converter lowering is needed as soon as src_size != dst_size
if srcsz != dstsz and ((srcsz * 2 != dstsz and srcsz != dstsz * 2) or op[0] != op[2]):
cursz = srcsz
rule = a
# When converting down we first do the type conversion followed
# by one or more size conversions. When converting up, we do
# the type conversion at the end. This way we don't have to
# deal with the fact that f2f8 doesn't exists.
sizeconvop = op[0] + '2' + op[0] if srcsz < dstsz else op[2] + '2' + op[2]
if srcsz > dstsz and op[0] != op[2]:
rule = (op + str(int(cursz)), rule)
while cursz != dstsz:
cursz = cursz / 2 if dstsz < srcsz else cursz * 2
rule = (sizeconvop + str(int(cursz)), rule)
if srcsz < dstsz and op[0] != op[2]:
rule = (op + str(int(cursz)), rule)
converts += [((op + str(int(dstsz)), 'a@' + str(int(srcsz))), rule)]
srcsz *= 2
dstsz *= 2
# Bifrost doesn't have fp16 for a lot of special ops
SPECIAL = ['fexp2', 'flog2', 'fsin', 'fcos']
for op in SPECIAL:
converts += [((op + '@16', a), ('f2f16', (op, ('f2f32', a))))]
algebraic_late += [((op + '@16', a), ('f2f16', (op, ('f2f32', a))))]
converts += [(('f2b32', a), ('fneu32', a, 0.0)),
algebraic_late += [(('f2b32', a), ('fneu32', a, 0.0)),
(('i2b32', a), ('ine32', a, 0)),
(('b2i32', a), ('iand', 'a@32', 1))]
@@ -112,7 +74,7 @@ def run():
print('#include "bifrost_nir.h"')
print(nir_algebraic.AlgebraicPass("bifrost_nir_lower_algebraic_late",
algebraic_late + converts).render())
algebraic_late).render())
if __name__ == '__main__':
main()