wrapper/libavc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

armv8: Fix ime_calculate_sad3_prog_av8

Browse source 
Previously, this used x5 (which is a stride parameter)
as destination pointer, and overwrote the destination in x6
with a counter, and looped to a label in another function.

In practice, this function is never used, which is why this
hasn't been noticed before.

Change-Id: Iab87a2960c36b7dc4de4216b1d4bbbe4e9f03578
This commit is contained in:
Martin Storsjo 2016-09-06 11:17:01 +03:00
parent d91f49ad65
commit e17ec6c736
1 changed files with 4 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

8
encoder/armv8/ime_distortion_metrics_av8.s
View file

@ -476,7 +476,7 @@ ime_calculate_sad3_prog_av8:
push_v_regs
sxtw x4, w4
sxtw x5, w5
mov x6, #16
mov x7, #16
movi v29.8h, #0
movi v30.8h, #0
movi v31.8h, #0
@ -507,15 +507,15 @@ core_loop_ime_calculate_sad3_prog_av8:
uabal v31.8h, v6.8b, v7.8b
uabal2 v31.8h, v6.16b, v7.16b
subs x6, x6, #1
bne core_loop_ime_calculate_sad2_prog_av8
subs x7, x7, #1
bne core_loop_ime_calculate_sad3_prog_av8
addp v30.8h, v30.8h, v31.8h
uaddlp v30.4s, v30.8h
addp v30.2s, v30.2s, v30.2s
shl v30.2s, v30.2s, #1
st1 {v30.2s}, [x5]
st1 {v30.2s}, [x6]
pop_v_regs
ret