On Tue, 2017-12-12 at 14:12 -0600, Segher Boessenkool wrote:
> That looks better :-) Okay for trunk, thanks!
As we discussed on IRC before christmas, I've simplified this to use
TARGET_EFFICIENT_UNALIGNED_VSX instead of checking explicitly for P8/P9
processors. Has the same effect but is cleaner/clearer. Committed in
256112.
Aaron
Index: gcc/config/rs6000/rs6000-string.c
===================================================================
--- gcc/config/rs6000/rs6000-string.c (revision 256110)
+++ gcc/config/rs6000/rs6000-string.c (working copy)
@@ -73,7 +73,7 @@
When optimize_size, avoid any significant code bloat; calling
memset is about 4 instructions, so allow for one instruction to
load zero and three to do clearing. */
- if (TARGET_ALTIVEC && align >= 128)
+ if (TARGET_ALTIVEC && (align >= 128 || TARGET_EFFICIENT_UNALIGNED_VSX))
clear_step = 16;
else if (TARGET_POWERPC64 && (align >= 64 || !STRICT_ALIGNMENT))
clear_step = 8;
@@ -90,7 +90,7 @@
machine_mode mode = BLKmode;
rtx dest;
- if (bytes >= 16 && TARGET_ALTIVEC && align >= 128)
+ if (bytes >= 16 && TARGET_ALTIVEC && (align >= 128 ||
TARGET_EFFICIENT_UNALIGNED_VSX))
{
clear_bytes = 16;
mode = V4SImode;
@@ -1260,7 +1260,7 @@
/* Altivec first, since it will be faster than a string move
when it applies, and usually not significantly larger. */
- if (TARGET_ALTIVEC && bytes >= 16 && align >= 128)
+ if (TARGET_ALTIVEC && bytes >= 16 && (TARGET_EFFICIENT_UNALIGNED_VSX ||
align >= 128))
{
move_bytes = 16;
mode = V4SImode;
--
Aaron Sawdey, Ph.D. acsaw...@linux.vnet.ibm.com
050-2/C113 (507) 253-7520 home: 507/263-0782
IBM Linux Technology Center - PPC Toolchain
