firefox/Bug-1610814-Fix-NEON-compil...

37 lines
1.4 KiB
Diff

From: Andrew Osmond <aosmond@mozilla.com>
Date: Wed, 22 Jan 2020 15:19:20 +0000
Subject: Bug 1610814 - Fix NEON compile error with gcc and RGB unpacking.
r=lsalzman
This patch makes us use the correct intrinsic for loading a uint8x16
register. It is not entirely clear why clang accepts this without
complaint but beyond the types, it should be equivalent.
Differential Revision: https://phabricator.services.mozilla.com/D60667
---
gfx/2d/SwizzleNEON.cpp | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/gfx/2d/SwizzleNEON.cpp b/gfx/2d/SwizzleNEON.cpp
index 6b34f76cab1..63f211ff294 100644
--- a/gfx/2d/SwizzleNEON.cpp
+++ b/gfx/2d/SwizzleNEON.cpp
@@ -412,7 +412,7 @@ void UnpackRowRGB24_NEON(const uint8_t* aSrc, uint8_t* aDst, int32_t aLength) {
src -= 4 * 3;
dst -= 4 * 4;
while (src >= aSrc) {
- uint8x16_t px = vld1q_u16(reinterpret_cast<const uint16_t*>(src));
+ uint8x16_t px = vld1q_u8(src);
// G2R2B1G1 R1B0G0R0 -> X1R1G1B1 X0R0G0B0
uint8x8_t pxlo = vtbl1_u8(vget_low_u8(px), masklo);
// B3G3R3B2 G2R2B1G1 -> X3R3G3B3 X2R2G2B2
@@ -420,7 +420,7 @@ void UnpackRowRGB24_NEON(const uint8_t* aSrc, uint8_t* aDst, int32_t aLength) {
vtbl1_u8(vext_u8(vget_low_u8(px), vget_high_u8(px), 4), maskhi);
px = vcombine_u8(pxlo, pxhi);
px = vorrq_u8(px, alpha);
- vst1q_u16(reinterpret_cast<uint16_t*>(dst), px);
+ vst1q_u8(dst, px);
src -= 4 * 3;
dst -= 4 * 4;
}