MIPS: math-emu: Optimise NaN handling in comparisons
authorMaciej W. Rozycki <macro@linux-mips.org>
Fri, 3 Apr 2015 22:25:38 +0000 (23:25 +0100)
committerRalf Baechle <ralf@linux-mips.org>
Tue, 7 Apr 2015 23:09:35 +0000 (01:09 +0200)
commitc9a1084516e35ff4f1d5b83e77530ed019ca364b
treede1cfa5b17d0e2db63616994f42fb12de8b207c6
parentd5afa7e905544a3d9e2bb29d9cafebc8e544c978
MIPS: math-emu: Optimise NaN handling in comparisons

We have the input operands already classified in `ieee754sp_cmp' and
`ieee754dp_cmp' comparison operations, so use the class obtained to tell
NaNs and numbers apart rather than classifying inputs again for this
purpose, reducing the size of code by 24 and 40 instructions or 96 and
160 bytes respectively.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/9689/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
arch/mips/math-emu/dp_cmp.c
arch/mips/math-emu/ieee754dp.c
arch/mips/math-emu/ieee754int.h
arch/mips/math-emu/ieee754sp.c
arch/mips/math-emu/sp_cmp.c