powerpc: Remove UP only lazy floating point and vector optimisations

Browse source

The UP only lazy floating point and vector optimisations were written
back when SMP was not common, and neither glibc nor gcc used vector
instructions. Now SMP is very common, glibc aggressively uses vector
instructions and gcc autovectorises.

We want to add new optimisations that apply to both UP and SMP, but
in preparation for that remove these UP only optimisations.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

...

This commit is contained in:

Anton Blanchard 2015-10-29 11:43:57 +11:00 • committed by Michael Ellerman

parent 68bfa962bf

commit af1bbc3dd3

9 changed files with 1 additions and 304 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

6

arch/powerpc/include/asm/processor.h

View file

8

arch/powerpc/include/asm/switch_to.h

View file

35

arch/powerpc/kernel/fpu.S

View file

32

arch/powerpc/kernel/head_fsl_booke.S

View file

7

arch/powerpc/kernel/idle_power7.S

View file

113

arch/powerpc/kernel/process.c

View file

18

arch/powerpc/kernel/signal_32.c

View file

18

arch/powerpc/kernel/signal_64.c

View file

68

arch/powerpc/kernel/vector.S

View file