summaryrefslogtreecommitdiff
path: root/toolchain/gcc/3.4.0/820-no-mips-empic-relocs.patch
blob: c126b787d94f26a34cc4f1706b5a9e355d29d2e1 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
From: cgd at broadcom dot com
To: gcc-patches at gcc dot gnu dot org
Cc: mark at codesourcery dot com
Date: 13 Jun 2004 22:51:30 -0700
Subject: [trunk + 3.4-branch RFA] don't use empic relocs for mips-linuxeh

This patch changes mips-linux to avoid using embedded-pic relocs for
its eh data.  (Support for generating these for new code is removed in
current binutils srcs.)

Relating to this, previously, mips-linux and mips64-linux would use
different representations for their EH data (even for mips64-linux o32
abi), due to the mips64-linux n32/64 BFDs not supporting the
embedded-pic relocs.  This was a bug.

For more explanation, see the thread of the URL quoted in the comment
in linux.h.


Tested the same w/ sources of about a week ago for c/c++ for
mips-linux (native) before/after.  Also verified .o compatibility
before/after just to be sure.

I'd like this approved for the branch as well, so 3.4.1 will work
nicely w/ the next major binutils release.


thanks,

chris

2004-06-13  Chris Demetriou  <cgd@broadcom.com>

	* config/mips/linux.h (ASM_PREFERRED_EH_DATA_FORMAT): Redefine
	to return DW_EH_PE_absptr.
	* config/mips/linux64.h (ASM_PREFERRED_EH_DATA_FORMAT): Remove.

Index: config/mips/linux.h
===================================================================
RCS file: /cvs/gcc/gcc/gcc/config/mips/linux.h,v
retrieving revision 1.77
diff -u -p -r1.77 linux.h
--- gcc/gcc/config/mips/linux.h	19 Feb 2004 22:07:51 -0000	1.77
+++ gcc/gcc/config/mips/linux.h	14 Jun 2004 05:49:51 -0000
@@ -170,10 +170,11 @@ Boston, MA 02111-1307, USA.  */
 #undef FUNCTION_NAME_ALREADY_DECLARED
 #define FUNCTION_NAME_ALREADY_DECLARED 1
 
-#define ASM_PREFERRED_EH_DATA_FORMAT(CODE, GLOBAL)       		\
-  (flag_pic								\
-    ? ((GLOBAL) ? DW_EH_PE_indirect : 0) | DW_EH_PE_pcrel | DW_EH_PE_sdata4\
-   : DW_EH_PE_absptr)
+/* If possible, we should attempt to use GP-relative relocs for this
+   (see <a  href="http://sources.redhat.com/ml/binutils/2004-05/msg00227.html">http://sources.redhat.com/ml/binutils/2004-05/msg00227.html</a>).
+   However, until that is implement, this just uses standard, absolute
+   references.  */
+#define ASM_PREFERRED_EH_DATA_FORMAT(CODE, GLOBAL)	DW_EH_PE_absptr
 
 /* The glibc _mcount stub will save $v0 for us.  Don't mess with saving
    it, since ASM_OUTPUT_REG_PUSH/ASM_OUTPUT_REG_POP do not work in the
Index: config/mips/linux64.h
===================================================================
RCS file: /cvs/gcc/gcc/gcc/config/mips/linux64.h,v
retrieving revision 1.7
diff -u -p -r1.7 linux64.h
--- gcc/gcc/config/mips/linux64.h	18 Feb 2004 15:05:10 -0000	1.7
+++ gcc/gcc/config/mips/linux64.h	14 Jun 2004 05:49:51 -0000
@@ -81,14 +81,6 @@ Boston, MA 02111-1307, USA.  */
    specification.  The SGI/MIPS ABI defines it to be the same as PTR_SIZE.  */
 #define DWARF_OFFSET_SIZE PTR_SIZE
 
-#undef ASM_PREFERRED_EH_DATA_FORMAT
-#if 0  /* We can't use relative addressing modes on NEWABI :-(  */
-#define ASM_PREFERRED_EH_DATA_FORMAT(CODE, GLOBAL)       		\
-   (flag_pic								\
-    ? ((GLOBAL) ? DW_EH_PE_indirect : 0) | DW_EH_PE_sdata4		\
-    : DW_EH_PE_absptr)
-#endif
-
 /* GNU/Linux doesn't use the same floating-point format that IRIX uses
    for long double.  There's no need to override this here, since
    ieee_quad_format is the default, but let's put this here to make