[vta, trunk?] enable bootstrap with BUILD_CONFIG=bootstrap-O3

[vta, trunk?] enable bootstrap with BUILD_CONFIG=bootstrap-O3

[Alexandre Oliva]

[-- Attachment #1: Type: text/plain, Size: 343 bytes --]

Bootstrapping on x86-linux-gnu (not sure 32 or 64 any more, probably
both) fails because of -Werror on these two files, because functions get
inlined and expose paths in which the compiler can't determine that
variables or their fields are initialized before they're referenced.

I'm installing this in the branch.  Is this ok for the trunk?


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gcc-uninit-O3.patch --]
[-- Type: text/x-patch, Size: 1207 bytes --]

for  gcc/ChangeLog.vta
from  Alexandre Oliva  <aoliva@redhat.com>

	* tree-ssa-structalias.c (handle_lhs_call): Initialize tmp.
	* fortran/module.c (mio_f2k_derived): Initialize cur.

Index: gcc/fortran/module.c
===================================================================
--- gcc/fortran/module.c.orig	2009-05-12 03:45:55.000000000 -0300
+++ gcc/fortran/module.c	2009-05-15 23:43:21.000000000 -0300
@@ -3340,7 +3340,7 @@ mio_f2k_derived (gfc_namespace *f2k)
       f2k->finalizers = NULL;
       while (peek_atom () != ATOM_RPAREN)
 	{
-	  gfc_finalizer *cur;
+	  gfc_finalizer *cur = NULL;
 	  mio_finalizer (&cur);
 	  cur->next = f2k->finalizers;
 	  f2k->finalizers = cur;
Index: gcc/tree-ssa-structalias.c
===================================================================
--- gcc/tree-ssa-structalias.c.orig	2009-05-12 03:48:03.000000000 -0300
+++ gcc/tree-ssa-structalias.c	2009-05-15 23:42:50.000000000 -0300
@@ -3425,7 +3425,7 @@ handle_lhs_call (tree lhs, int flags, VE
 static void
 handle_const_call (gimple stmt, VEC(ce_s, heap) **results)
 {
-  struct constraint_expr rhsc, tmpc;
+  struct constraint_expr rhsc, tmpc = { NULL, 0, 0 };
   tree tmpvar = NULL_TREE;
   unsigned int k;
 

[-- Attachment #3: Type: text/plain, Size: 258 bytes --]



-- 
Alexandre Oliva, freedom fighter    http://FSFLA.org/~lxoliva/
You must be the change you wish to see in the world. -- Gandhi
Be Free! -- http://FSFLA.org/   FSF Latin America board member
Free Software Evangelist      Red Hat Brazil Compiler Engineer

Re: [vta, trunk?] enable bootstrap with BUILD_CONFIG=bootstrap-O3

[Alexandre Oliva]

[-- Attachment #1: Type: text/plain, Size: 495 bytes --]

On May 18, 2009, Alexandre Oliva <aoliva@redhat.com> wrote:

> Bootstrapping on x86-linux-gnu (not sure 32 or 64 any more, probably
> both) fails because of -Werror on these two files, because functions get
> inlined and expose paths in which the compiler can't determine that
> variables or their fields are initialized before they're referenced.

Erhm...  I jumped the gun on this one, the patch I posted apparently
hadn't got into my last test cycle.

Here's a correct version of the patch.


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gcc-uninit-O3.patch --]
[-- Type: text/x-patch, Size: 1209 bytes --]

for  gcc/ChangeLog.vta
from  Alexandre Oliva  <aoliva@redhat.com>

	* tree-ssa-structalias.c (handle_lhs_call): Initialize tmp.
	* fortran/module.c (mio_f2k_derived): Initialize cur.

Index: gcc/fortran/module.c
===================================================================
--- gcc/fortran/module.c.orig	2009-05-12 03:45:55.000000000 -0300
+++ gcc/fortran/module.c	2009-05-15 23:43:21.000000000 -0300
@@ -3340,7 +3340,7 @@ mio_f2k_derived (gfc_namespace *f2k)
       f2k->finalizers = NULL;
       while (peek_atom () != ATOM_RPAREN)
 	{
-	  gfc_finalizer *cur;
+	  gfc_finalizer *cur = NULL;
 	  mio_finalizer (&cur);
 	  cur->next = f2k->finalizers;
 	  f2k->finalizers = cur;
Index: gcc/tree-ssa-structalias.c
===================================================================
--- gcc/tree-ssa-structalias.c.orig	2009-05-12 03:48:03.000000000 -0300
+++ gcc/tree-ssa-structalias.c	2009-05-15 23:42:50.000000000 -0300
@@ -3425,7 +3425,7 @@ handle_lhs_call (tree lhs, int flags, VE
 static void
 handle_const_call (gimple stmt, VEC(ce_s, heap) **results)
 {
-  struct constraint_expr rhsc, tmpc;
+  struct constraint_expr rhsc, tmpc = { SCALAR, 0, 0 };
   tree tmpvar = NULL_TREE;
   unsigned int k;
 

[-- Attachment #3: Type: text/plain, Size: 258 bytes --]



-- 
Alexandre Oliva, freedom fighter    http://FSFLA.org/~lxoliva/
You must be the change you wish to see in the world. -- Gandhi
Be Free! -- http://FSFLA.org/   FSF Latin America board member
Free Software Evangelist      Red Hat Brazil Compiler Engineer

Re: [vta, trunk?] enable bootstrap with BUILD_CONFIG=bootstrap-O3

[Alexandre Oliva]

[-- Attachment #1: Type: text/plain, Size: 588 bytes --]

On May 19, 2009, Alexandre Oliva <aoliva@redhat.com> wrote:

> On May 18, 2009, Alexandre Oliva <aoliva@redhat.com> wrote:
>> Bootstrapping on x86-linux-gnu (not sure 32 or 64 any more, probably
>> both) fails because of -Werror on these two files, because functions get
>> inlined and expose paths in which the compiler can't determine that
>> variables or their fields are initialized before they're referenced.

tree-ssa-structalias.c was already fixed, but fortran/module.c still
needs a patch for BUILD_CONFIG=bootstrap-O3 AKA BOOT_CFLAGS='-O3 -g' on
x86_64 or i686.

Ok for trunk?


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gcc-uninit-O3.patch --]
[-- Type: text/x-patch, Size: 620 bytes --]

for  gcc/fortran/ChangeLog
from  Alexandre Oliva  <aoliva@redhat.com>

	* module.c (mio_f2k_derived): Initialize cur.

Index: gcc/fortran/module.c
===================================================================
--- gcc/fortran/module.c.orig	2009-05-12 03:45:55.000000000 -0300
+++ gcc/fortran/module.c	2009-05-15 23:43:21.000000000 -0300
@@ -3340,7 +3340,7 @@ mio_f2k_derived (gfc_namespace *f2k)
       f2k->finalizers = NULL;
       while (peek_atom () != ATOM_RPAREN)
 	{
-	  gfc_finalizer *cur;
+	  gfc_finalizer *cur = NULL;
 	  mio_finalizer (&cur);
 	  cur->next = f2k->finalizers;
 	  f2k->finalizers = cur;

[-- Attachment #3: Type: text/plain, Size: 257 bytes --]


-- 
Alexandre Oliva, freedom fighter    http://FSFLA.org/~lxoliva/
You must be the change you wish to see in the world. -- Gandhi
Be Free! -- http://FSFLA.org/   FSF Latin America board member
Free Software Evangelist      Red Hat Brazil Compiler Engineer

Re: [vta, trunk?] enable bootstrap with BUILD_CONFIG=bootstrap-O3

[Richard Henderson]

Alexandre Oliva wrote:
> tree-ssa-structalias.c was already fixed, but fortran/module.c still
> needs a patch for BUILD_CONFIG=bootstrap-O3 AKA BOOT_CFLAGS='-O3 -g' on
> x86_64 or i686.

Ok.


r~