diff options
author | Tobias Klausmann <[email protected]> | 2017-08-13 03:02:39 +0200 |
---|---|---|
committer | Ilia Mirkin <[email protected]> | 2017-08-31 22:58:06 -0400 |
commit | 1c4e6d7ca83578caf5212f7a484538cb1b4ae2a3 (patch) | |
tree | 6d0cf9c9ccd3f5559691c25cd2a6c5ab356793d5 /src/gallium/README.portability | |
parent | b672c3833b7ea51c09a6bcbed86b8fb6789448f1 (diff) |
nvc0/ir: propagate immediates to CALL input MOVs
On using builtin functions we have to move the input to registers $0 and $1, if
one of the input value is an immediate, we fail to propagate the immediate:
...
mov u32 $r477 0x00000003 (0)
...
mov u32 $r0 %r473 (0)
mov u32 $r1 $r477 (0)
call abs BUILTIN:0 (0)
mov u32 %r495 $r1 (0)
...
With this patch the immediate is propagated, potentially causing the first MOV
to be superfluous, which we'd remove in that case:
...
mov u32 $r0 %r473 (0)
mov u32 $r1 0x00000003 (0)
call abs BUILTIN:0 (0)
mov u32 %r495 $r1 (0)
...
Shaderdb stats:
total instructions in shared programs : 4893460 -> 4893324 (-0.00%)
total gprs used in shared programs : 582972 -> 582881 (-0.02%)
total local used in shared programs : 17960 -> 17960 (0.00%)
local gpr inst bytes
helped 0 91 112 112
hurt 0 0 0 0
v2:
implement some changes proposed by imirkin, the manual deletion of the dead
mov is necessary after ea22ac23e0 ("nvc0/ir: unlink values pre- and post-call
to division function") as the potentially dead mov is unlinked properly,
causing later passes to not notice the mov op at all and thus not cleaning it
up. That makes up a big chunk of the regression the above commit caused.
Keep the deletion of the op where it is, deleting it later unnecessarily blows
up size of the change.
Signed-off-by: Tobias Klausmann <[email protected]>
Reviewed-by: Ilia Mirkin <[email protected]>
Diffstat (limited to 'src/gallium/README.portability')
0 files changed, 0 insertions, 0 deletions