diff options
| author | Siarhei Siamashka <siarhei.siamashka@gmail.com> | 2012-12-03 03:01:21 +0200 |
|---|---|---|
| committer | Siarhei Siamashka <siarhei.siamashka@gmail.com> | 2012-12-18 20:45:57 +0200 |
| commit | f9a41703b2d46c988b9e4e378d27396f718006ae (patch) | |
| tree | 9b02b72c21711625798c1826f70545f9840b145c | |
| parent | 3922e90c400fca3ac43dc77b8dd0c0591e7e4fbc (diff) | |
Faster conversion from a8r8g8b8 to r5g6b5 in C code
This change reduces 3 shifts, 3 ANDs and 2 ORs (total 8 arithmetic
operations) to 3 shifts, 2 ANDs and 2 ORs (total 7 arithmetic
operations).
We get garbage in the high 16 bits of the result, which might need
to be cleared when casting to uint16_t (it would bring us back to
total 8 arithmetic operations). However in the case if the result
of a8r8g8b8->r5g6b5 conversion is immediately stored to memory, no
extra instructions for clearing these garbage bits are needed.
This allows the a8r8g8b8->r5g6b5 conversion code to be compiled
into 4 instructions for ARM instead of 5 (assuming a good optimizing
compiler), which has no pipeline stalls on ARM11 as an additional
bonus.
The change in benchmark results for 'lowlevel-blt-bench src_8888_0565'
with PIXMAN_DISABLE="arm-simd arm-neon mips-dspr2 mmx sse2" and pixman
compiled by gcc-4.7.2:
MIPS 74K 480MHz : 40.44 MPix/s -> 40.13 MPix/s
ARM11 700MHz : 50.28 MPix/s -> 62.85 MPix/s
ARM Cortex-A8 1000MHz : 124.38 MPix/s -> 141.85 MPix/s
ARM Cortex-A15 1700MHz : 281.07 MPix/s -> 303.29 MPix/s
Intel Core i7 2800MHz : 515.92 MPix/s -> 531.16 MPix/s
The same trick was used in xomap (X server for Nokia N800/N810):
http://repository.maemo.org/pool/diablo/free/x/xorg-server/
xorg-server_1.3.99.0~git20070321-0osso20083801.tar.gz
| -rw-r--r-- | pixman/pixman-private.h | 10 |
1 files changed, 7 insertions, 3 deletions
diff --git a/pixman/pixman-private.h b/pixman/pixman-private.h index ce71bbd..ea447aa 100644 --- a/pixman/pixman-private.h +++ b/pixman/pixman-private.h @@ -887,9 +887,13 @@ pixman_list_move_to_front (pixman_list_t *list, pixman_link_t *link) static force_inline uint16_t convert_8888_to_0565 (uint32_t s) { - return ((((s) >> 3) & 0x001f) | - (((s) >> 5) & 0x07e0) | - (((s) >> 8) & 0xf800)); + /* The following code can be compiled into just 4 instructions on ARM */ + uint32_t a, b; + a = (s >> 3) & 0x1F001F; + b = s & 0xFC00; + a |= a >> 5; + a |= b >> 5; + return (uint16_t)a; } static force_inline uint32_t |