mirror of https://git.ffmpeg.org/ffmpeg.git
softfloat: decrease MIN_EXP to cover full float range
floats are not necessarily normalized, so a normalized softfloat needs
MIN_EXP lowered by 23 to cover that range.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
(cherry picked from commit 2d6f46d801
)
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
This commit is contained in:
parent
e70caba384
commit
5d1502d4b6
|
@ -27,7 +27,7 @@
|
|||
#include "avassert.h"
|
||||
#include "softfloat_tables.h"
|
||||
|
||||
#define MIN_EXP -126
|
||||
#define MIN_EXP -149
|
||||
#define MAX_EXP 126
|
||||
#define ONE_BITS 29
|
||||
|
||||
|
|
Loading…
Reference in New Issue