patches/gcc/3.3.3/pr15089-fix.patch
author "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Sat Feb 24 11:00:05 2007 +0000 (2007-02-24)
changeset 1 eeea35fbf182
permissions -rw-r--r--
Add the full crosstool-NG sources to the new repository of its own.
You might just say: 'Yeah! crosstool-NG's got its own repo!".
Unfortunately, that's because the previous repo got damaged beyond repair and I had no backup.
That means I'm putting backups in place in the afternoon.
That also means we've lost history... :-(
yann@1
     1
See http://gcc.gnu.org/PR15089
yann@1
     2
"In some cases the specified register for a local variable meant to be
yann@1
     3
used with inline assembly code is not respected.  This breaks code 
yann@1
     4
relying on this feature to implement non-standard calling convension."
yann@1
     5
yann@1
     6
This fix is probably needed to properly implement linux syscalls in some architectures.
yann@1
     7
yann@1
     8
First added to crosstool by Jamie Hicks, see
yann@1
     9
http://www.handhelds.org/hypermail/kernel-discuss/current/0066.html
yann@1
    10
yann@1
    11
Index: gcc/loop.c
yann@1
    12
===================================================================
yann@1
    13
RCS file: /cvs/gcc/gcc/gcc/loop.c,v
yann@1
    14
retrieving revision 1.488.2.3
yann@1
    15
diff -u -r1.488.2.3 loop.c
yann@1
    16
--- gcc-old/gcc/loop.c	14 Feb 2004 14:46:03 -0000	1.488.2.3
yann@1
    17
+++ gcc-new/gcc/loop.c	28 Apr 2004 22:02:53 -0000
yann@1
    18
@@ -929,6 +929,7 @@
yann@1
    19
 			  || (! (GET_CODE (SET_SRC (set)) == REG
yann@1
    20
 				 && (REGNO (SET_SRC (set))
yann@1
    21
 				     < FIRST_PSEUDO_REGISTER))))
yann@1
    22
+		      && regno >= FIRST_PSEUDO_REGISTER 
yann@1
    23
 		      /* This test is not redundant; SET_SRC (set) might be
yann@1
    24
 			 a call-clobbered register and the life of REGNO
yann@1
    25
 			 might span a call.  */