[Bug rtl-optimization/50696] [x32] Unnecessary lea

2018-03-11 Thread egallager at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=50696 Eric Gallager changed: What|Removed |Added Keywords||patch CC|

[Bug rtl-optimization/50696] [x32] Unnecessary lea

2011-10-14 Thread hjl.tools at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50696 H.J. Lu changed: What|Removed |Added URL||http://gcc.gnu.org/ml/gcc-p |

[Bug rtl-optimization/50696] [x32] Unnecessary lea

2011-10-14 Thread bonzini at gnu dot org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50696 Paolo Bonzini changed: What|Removed |Added CC||bonzini at gnu dot org --- Comment #6 fro

[Bug rtl-optimization/50696] [x32] Unnecessary lea

2011-10-11 Thread hjl.tools at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50696 --- Comment #5 from H.J. Lu 2011-10-11 23:29:05 UTC --- This patch changes combine not to generate: (plus:DI (subreg:DI (mult:SI (reg/v:SI 85 [ i ]) (const_int 4 [0x4])) 0) (subreg:DI (reg:SI 100) 0)) and changes const_32bit_mas

[Bug rtl-optimization/50696] [x32] Unnecessary lea

2011-10-11 Thread hjl.tools at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50696 --- Comment #4 from H.J. Lu 2011-10-11 22:13:47 UTC --- const_32bit_mask is incorrect since combine may optimize VAL in ADDR & VAL from 0x to 0xfffc. Even if we take this into account, we can't decompose (plus:DI (subreg:DI (mult:SI

[Bug rtl-optimization/50696] [x32] Unnecessary lea

2011-10-11 Thread hjl.tools at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50696 --- Comment #3 from H.J. Lu 2011-10-11 20:11:59 UTC --- Does this patch diff --git a/gcc/combine.c b/gcc/combine.c index 6c3b17c..52259b6 100644 --- a/gcc/combine.c +++ b/gcc/combine.c @@ -5078,6 +5078,22 @@ subst (rtx x, rtx from, rtx to, int i

[Bug rtl-optimization/50696] [x32] Unnecessary lea

2011-10-11 Thread hjl.tools at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50696 H.J. Lu changed: What|Removed |Added Component|target |rtl-optimization --- Comment #2 from H.J. Lu 2