Chiebot-Mirror
/
FFmpeg
mirror of https://github.com/FFmpeg/FFmpeg.git
8
0
Fork 0
Code Issues Projects Releases Wiki Activity

avutil/md5: fix misaligned reads

Browse Source 
This makes ubsan happy and also considerably increases performance on
big endian systems.

Tested on an IBM POWER7 3.55 GHz

Before:

2.24user 0.14system 0:02.39elapsed 99%CPU (0avgtext+0avgdata 2624maxresident)k
2.26user 0.11system 0:02.38elapsed 99%CPU (0avgtext+0avgdata 2688maxresident)k
2.23user 0.15system 0:02.38elapsed 99%CPU (0avgtext+0avgdata 2624maxresident)k
2.25user 0.12system 0:02.38elapsed 100%CPU (0avgtext+0avgdata 2624maxresident)k
2.20user 0.15system 0:02.36elapsed 99%CPU (0avgtext+0avgdata 2624maxresident)k

After:

1.86user 0.13system 0:02.00elapsed 99%CPU (0avgtext+0avgdata 2624maxresident)k
1.89user 0.11system 0:02.01elapsed 99%CPU (0avgtext+0avgdata 2624maxresident)k
1.85user 0.14system 0:02.00elapsed 99%CPU (0avgtext+0avgdata 2624maxresident)k
1.84user 0.15system 0:01.99elapsed 99%CPU (0avgtext+0avgdata 2624maxresident)k
1.89user 0.13system 0:02.02elapsed 99%CPU (0avgtext+0avgdata 2688maxresident)k

Tested-by: Nicolas George <george@nsup.org>
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Sean McGovern <gseanmcg@gmail.com>
pull/272/head
James Almer 8 years ago committed by Sean McGovern
parent 61cec5adaa
commit a876958d0f
1 changed files with 8 additions and 12 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 20
      libavutil/md5.c

20
libavutil/md5.c
Unescape View File

@ -83,15 +83,15 @@ static const uint32_t T[64] = { // T[i]= fabs(sin(i+1)<<32)
a += T[i]; \
\
if (i < 32) { \
if (i < 16) \
a += (d ^ (b & (c ^ d))) + X[i & 15]; \
else \
a += (c ^ (d & (c ^ b))) + X[(1 + 5 * i) & 15]; \
if (i < 16) \
a += (d ^ (b & (c ^ d))) + AV_RL32(X + (i & 15)); \
else \
a += (c ^ (d & (c ^ b))) + AV_RL32(X + ((1 + 5 * i) & 15)); \
} else { \
if (i < 48) \
a += (b ^ c ^ d) + X[(5 + 3 * i) & 15]; \
a += (b ^ c ^ d) + AV_RL32(X + ((5 + 3 * i) & 15)); \
else \
a += (c ^ (b | ~d)) + X[(7 * i) & 15]; \
a += (c ^ (b | ~d)) + AV_RL32(X + ((7 * i) & 15)); \
} \
a = b + (a << t | a >> (32 - t)); \
} while (0)
@ -99,18 +99,14 @@ static const uint32_t T[64] = { // T[i]= fabs(sin(i+1)<<32)
static void body(uint32_t ABCD[4], uint32_t X[16])
{
int t;
int i av_unused;
unsigned int a = ABCD[3];
unsigned int b = ABCD[2];
unsigned int c = ABCD[1];
unsigned int d = ABCD[0];
#if HAVE_BIGENDIAN
for (i = 0; i < 16; i++)
X[i] = av_bswap32(X[i]);
#endif
#if CONFIG_SMALL
int i;
for (i = 0; i < 64; i++) {
CORE(i, a, b, c, d);
t = d;

Write Preview
Loading…
Cancel
Save