> -Original Message-
> From: Richard Guenther [mailto:richard.guent...@gmail.com]
> Sent: Saturday, September 24, 2011 6:12 PM
> To: Jiangning Liu
> Cc: Mike Stump; gcc-patches@gcc.gnu.org; r...@cebitec.uni-bielefeld.de
> Subject: Re: [PATCH, testsuite] Add loop unrolling command line optio
Dinar Temirbulatov writes:
> I found typo in the patch instead of checking *set_after != 0 it was
> set_after != 0, here is corrected version of patch. I retested the
> patch without typo on mipsel-unknown-linux-gnu with no new
> regressions.
But my point was that *set_after should always be non
Fat pointer types are the device used in GNAT to represent unconstrained array
types and pointers to such array types, i.e. arrays whose bounds depend on the
object. They are "fat" because they contain two pointers, one that points to
the array itself and the other to a structure containing the
In Ada, access types (the equivalent of pointer types) are checked before being
dereferenced. You can eliminate these checks, or more precisely replace them
with checks on assignments, by putting a null-exclusion marker on them. As a
result, the compiler can guarantee that some access values a
Hi!
A couple of unreviewed patches:
tree-ssa-strlen optimization:
[1/2] http://gcc.gnu.org/ml/gcc-patches/2011-09/msg00890.html
[2/2] http://gcc.gnu.org/ml/gcc-patches/2011-09/msg00891.html
optimize all ones vectors in simplify-rtx.c (and i386 expansion):
http://gcc.gnu.org/ml/gcc-patches/
It turned out that the front-end wasn't fully using the information about the
null exclusion either in some cases. This eliminates redundant access checks
in the callee for subprograms with IN or IN/OUT parameters of an null-exclusion
access subtype, which were introduced in Ada 2005. The comp
Tested on i586-suse-linux, applied on the mainline.
2011-09-26 Eric Botcazou
* gcc-interface/trans.c (assoc_to_constructor): Minor tweaks.
* gcc-interface/utils2.c (build_simple_component_ref): Fix formatting
issues. Use COMPLETE_TYPE_P in assertion. Also set TREE_RE
Some TREE_CHAINs have surreptitiously crept back...
Tested on i586-suse-linux, applied on the mainline.
2011-09-26 Eric Botcazou
* gcc-interface/gigi.h (create_subprog_decl): Replace TREE_CHAIN with
DECL_CHAIN in comment.
* gcc-interface/trans.c (gigi): Likewise.
This happens when the array has an alignment clause, because we fail to look up
the array type within the padded type it is wrapped up in.
Tested on i586-suse-linux, applied on the mainline.
2011-09-26 Eric Botcazou
* gcc-interface/utils.c (maybe_unconstrained_array): Declare TYPE l
Jakub Jelinek writes:
> optimize all ones vectors in simplify-rtx.c (and i386 expansion):
> http://gcc.gnu.org/ml/gcc-patches/2011-09/msg01364.html
I wonder whether we should have a CONSTM1_RTX(MODE). It seems
inconsistent to have vector 0s and 1s, but only have integer -1s.
As far as simplif
Hi,
I notice the following description is different from how spu & m32c use it.
In internal manual:
bool TARGET_ADDR_SPACE_SUBSET_P (addr space t superset, [Target Hook]
addr space t subset)
Define this to return whether the subset named address space is contained
within the
superset named add
It may arise on platforms with conditional execution because of an awkward CFG,
but it ultimately comes from a discrepancy in the way we translate allocation
expressions for unconstrained array types in gigi.
Tested on i586-suse-linux, applied on the mainline.
2011-09-26 Eric Botcazou
On 9 September 2011 18:04, Ulrich Weigand wrote:
>
> In theory, LEGITIMIZE_RELOAD_ADDRESS could attempt to handle them by
> substituting the equivalent constant and then reloading the result.
> However, this might need additional steps (pushing to the constant pool,
> reloading the constant pool a
> Jan Hubicka writes:
> >
> > + Link-time optimization improvements:
> > +
> > + Improved scalability and reduced memory usage. Link time
> > optimization
> > + of Firefox now require 3GB of RAM on 64bit system, while over 8GB
> > was needed
> > + previously. L
This is a fallout of the recent gigi changes: we can now set MEM_NOTRAP_P on
the memory accesses more often in Ada. This uncovered a problem in the ifcvt
pass, which is hoisting a MEM_NOTRAP_P load before a test guarding it. This
is a known pattern and the fix is to use may_trap_or_fault_p to
Ian Lance Taylor writes:
> Rainer Orth writes:
>
>> Unfortunately, the
>>
>> /vol/gcc/src/hg/trunk/local/libgo/syscalls/exec.go:11:14: error: imported
>> and not used: unsafe
>>
>> error remains. I've no idea what triggers it.
>
> Bother. Neither do I. I just looked through the code and it a
> Ugh, yeah. I suppose PTA assigned a HEAP var as pointed-to object for the
> original pointer, even if the transformed stmt
>
> orig_ptr_1 = &a;
>
> has the points-to information preserved for orig_ptr_1 further propagation
> of &a will make accesses through orig_ptr_1 have different alias
> pro
> I tried to implement the approach you describe above in attached patch.
Thanks a lot, this indeed fixes the problem!
> Currently testing on x86_64.
Please also install the testcase I posted in the other message in conjunction
with the fix. Thanks in advance.
--
Eric Botcazou
On 09/26/2011 02:34 AM, Charles Wilson wrote:
I note one other comment in the referenced bugzilla thread, where
Paolo mentions that the ABI will break for C++1x (libstdc++7?) anyway,
and that in the "new" string implementation this issue will be not
actually be an issue. Is that 2009 statement
This patch is fix PR38644, a 3-year-old bug.
>From the discussions in mail list and bugzilla, I think the middle end fix
is a common view. Although there are stills some gaps on how to fix it in
middle end, I think this patch at least moves the problem from back-end to
middle-end, which makes GCC
On Sun, 25 Sep 2011, Tom de Vries wrote:
> On 09/25/2011 10:57 AM, Richard Guenther wrote:
> > On Sat, Sep 24, 2011 at 5:29 PM, Eric Botcazou
> > wrote:
> >>> This is an updated version of the patch. I have 2 new patches and an
> >>> updated testcase which I will sent out individually.
> >>>
> >
On Mon, Sep 26, 2011 at 9:08 AM, Jiangning Liu wrote:
>> -Original Message-
>> From: Richard Guenther [mailto:richard.guent...@gmail.com]
>> Sent: Saturday, September 24, 2011 6:12 PM
>> To: Jiangning Liu
>> Cc: Mike Stump; gcc-patches@gcc.gnu.org; r...@cebitec.uni-bielefeld.de
>> Subject:
Arnaud Charlet writes:
>> Mainline Ada bootstrap is currently broken on Solaris 8 and 9/x86 with
>> Sun as:
>>
>> Assembler: a-strunb.adb
>> "/var/tmp//cc6ZxWWa.s", line 2395 : Syntax error
>> "/var/tmp//cc6ZxWWa.s", line 2591 : Syntax error
>> make[7]: *** [a-strunb.o] Error 1
>>
>>
This patch is housekeeping to clean up avr.c:adjust_insn_length i.e. replace
digging in RTXes by using insn attribute "adjust_len".
There's nothing special about it, it's just mechanical change.
As alternative "yes" is no more needed, I removed it and set the default to
"no": No insn needs length
On Mon, Sep 26, 2011 at 11:20 AM, Bingfeng Mei wrote:
> Hi,
> I notice the following description is different from how spu & m32c use it.
>
> In internal manual:
>
> bool TARGET_ADDR_SPACE_SUBSET_P (addr space t superset, [Target Hook]
> addr space t subset)
> Define this to return whether the sub
2011/9/26 Georg-Johann Lay :
> This patch is housekeeping to clean up avr.c:adjust_insn_length i.e. replace
> digging in RTXes by using insn attribute "adjust_len".
>
> There's nothing special about it, it's just mechanical change.
>
> As alternative "yes" is no more needed, I removed it and set th
Hi,
in private email, Daniel pointed out that, post my recent fix in this
area, instantiation is not handled yet. Thus I prepared the very simple
patch below. Should we provide a more specific error message?
Tested x86_64-linux.
Thanks,
Paolo.
/cp
2011-09-26 Paolo Carl
Richard,
Here are updated patches. tm.texi.in doesn't need any change as the
description of the hook is correct, just the order of parameters needs
change.
Thanks,
Bingfeng
2011-09-26 Bingfeng Mei
* doc/tm.texi: Correct documentation for TARGET_ADDR_SPACE_SUBSET_P.
* target.d
On Sep 23, 2011, at 1:58 AM, Bruce Korb wrote:
> On 09/22/11 08:00, Tristan Gingold wrote:
>> Hi,
>>
>> this patch adds some fixes in fixincludes specific to VMS.
>> Until now we were using a manually modified subset of the VMS headers, but
>> using fix includes is the right way.
>>
>> Ok for
Hi Robert,
> 2011/8/17 Rainer Orth :
>> Hi Robert,
>>
>>> My patch still applies cleanly to current HEAD, has this migration
>>> happened already? If not, what's the current ETA? I'll have almost
>>> no spare time after this week, I'd like to sort this out before/during
>>> the weekend if possib
Hi Richard,
gcc/
* config/iq2000/iq2000.md: Use match_test rather than eq/ne symbol_ref
throughout file.
Approved - please apply.
Cheers
Nick
Hi Richard,
gcc/
* config/m32r/m32r.md: Use match_test rather than eq/ne symbol_ref
throughout file.
Approved - please apply.
Cheers
Nick
On Mon, Sep 26, 2011 at 1:59 PM, Bingfeng Mei wrote:
> Richard,
> Here are updated patches. tm.texi.in doesn't need any change as the
> description of the hook is correct, just the order of parameters needs
> change.
Ok.
Thanks,
Richard.
> Thanks,
> Bingfeng
>
>
>
> 2011-09-26 Bingfeng Mei
>
Hi,
as analyzed in the audit trail, it's pretty sure that this check in
get_bc_label is not serving any purpose anymore, because such
diagnostics is always (eg, also with -fsyntax-only, double checked)
produced by the parser. Thus I tested the below on x86_64-linux.
Ok for mainline?
Thanks,
On 09/26/2011 12:29 PM, Richard Guenther wrote:
> On Sun, 25 Sep 2011, Tom de Vries wrote:
>
>> On 09/25/2011 10:57 AM, Richard Guenther wrote:
>>> On Sat, Sep 24, 2011 at 5:29 PM, Eric Botcazou
>>> wrote:
> This is an updated version of the patch. I have 2 new patches and an
> updated t
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2011-09-26 Richard Guenther
PR tree-optimization/50472
* gimple-fold.c (fold_const_aggregate_ref_1): Do not fold
volatile references.
* gcc.dg/torture/pr50472.c: New testcase.
Index: gcc/
On 09/25/2011 02:39 PM, Fabien Chêne wrote:
Great, I would be glad to see it applied to 4.6. If you do that, you
may also want to adjust the last paragraph in the release note of 4.6.
Done, thanks for the suggestion.
Jason
This is a minor tweak to do eq/ne comparisons one instruction shorter in the
case where the compare target is unused after the comparison:
For 1:
DEC R0
OR R0,R1
OR R0,R2
OR R0,R3
For -1:
AND R0,R3
AND R0,R2
AND R0,R1
COM R0
The text peephole casesi+2 used 0x where -1 is the right
OK.
Jason
OK.
Jason
This patch fixes an issue that limit opportunities for shrink-wrapping
on PowerPC. The rs6000 REG_ALLOC_ORDER chooses r0 as the very first
gpr to use in code, with r11 also having high priority. This means it
is quite likely that r0 or r11 is live on the edge chosen for
shrink-wrapping. That's u
On 09/25/2011 05:06 PM, Fabien Chêne wrote:
+ else if ((using_decl = strip_using_decl (member)) != member)
+ /* If it is a using decl, use its underlying decl. */
+ type_decl = strip_using_decl (type_decl);
- if (DECL_NAME (field) == name
+ if (DECL_NAME (decl) == name
On 09/25/2011 11:57 PM, Jason Merrill wrote:
Ville pointed out that my earlier patch failed to support 'this' in
NSDMIs. So this patch implements that.
...and this one fixes 'this' in templates.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit 7893474d771c6449e05f8c85b3e9e418b9c72a57
Aut
This patch increases opportunities for shrink-wrapping. With this
applied, http://gcc.gnu.org/ml/gcc-patches/2011-09/msg00754.html and
http://gcc.gnu.org/ml/gcc-patches/2011-03/msg01499.html plus the fix
mentioned in http://gcc.gnu.org/ml/gcc-patches/2011-09/msg01016.html,
along with my powerpc ch
On Mon, Sep 19, 2011 at 9:54 AM, Ira Rosen wrote:
> Hi,
>
> This patch adds a support of widening shift left. The following
> pattern is detected:
>
> type a_t;
> TYPE a_T, res_T;
>
> a_t = ;
> a_T = (TYPE) a_t;
> res_T = a_T << CONST;
>
> ('TYPE' is at least 2 times bigger than 'type', and CONST
Ramana Radhakrishnan wrote:
> On 9 September 2011 18:04, Ulrich Weigand wrote:
> >
> > In theory, LEGITIMIZE_RELOAD_ADDRESS could attempt to handle them by
> > substituting the equivalent constant and then reloading the result.
> > However, this might need additional steps (pushing to the constant
On Wed, Sep 7, 2011 at 5:06 PM, Joseph S. Myers wrote:
> This looks like it has the same issue with maybe needing to use
> TYPE_MAIN_VARIANT in type comparisons as the shuffle patch.
I don't think so, we move qualifiers to the vector type from the element type
in make_vector_type and the tests on
On Thu, Sep 15, 2011 at 7:00 PM, Jakub Jelinek wrote:
> Hi!
>
> Here is an updated version of the support patch for the tree-ssa-strlen.c
> optimization. The only change is the propagation of gimple_block
> to the new call from the old call, otherwise Wobjsize-1.c testcase
> fails when we optimiz
2011/9/26 Georg-Johann Lay :
> This is a minor tweak to do eq/ne comparisons one instruction shorter in the
> case where the compare target is unused after the comparison:
>
> For 1:
> DEC R0
> OR R0,R1
> OR R0,R2
> OR R0,R3
>
> For -1:
> AND R0,R3
> AND R0,R2
> AND R0,R1
> COM R0
>
> Th
On Wed, Sep 14, 2011 at 10:38 PM, Richard Henderson wrote:
> On 09/13/2011 04:27 PM, Michael Meissner wrote:
>> This patch touches a bunch of files, but most of the changes are fairly
>> mechanical. What this does is change the infrastructure of how machine
>> independent builtin functions are ha
Hi!
As discussed in the PR, we currently don't vectorize
#include
std::valarray
f1 (std::valarray a, std::valarray b, std::valarray c, int z)
{
int i;
for (i = 0; i < z; i++)
{
a[i] = b[i] + c[i];
a[i] += b[i] * c[i];
}
return a;
}
void
f2 (std::valarray &__restrict a,
Ping 3 ? I'd like to get this in before stage1 ends... There's
RX-specific code that goes with it that can't go in until the core
functionality is approved.
http://gcc.gnu.org/ml/gcc-patches/2011-07/msg01889.html
http://gcc.gnu.org/ml/gcc-patches/2011-07/msg02555.html
http://gcc.gnu.org/ml/gcc-
On Mon, Sep 26, 2011 at 10:18:43AM +0100, Richard Sandiford wrote:
> Jakub Jelinek writes:
> > optimize all ones vectors in simplify-rtx.c (and i386 expansion):
> > http://gcc.gnu.org/ml/gcc-patches/2011-09/msg01364.html
>
> I wonder whether we should have a CONSTM1_RTX(MODE). It seems
> incon
On Mon, 26 Sep 2011, Jakub Jelinek wrote:
> Hi!
>
> As discussed in the PR, we currently don't vectorize
> #include
>
> std::valarray
> f1 (std::valarray a, std::valarray b, std::valarray c, int z)
> {
> int i;
> for (i = 0; i < z; i++)
> {
> a[i] = b[i] + c[i];
> a[i] += b[
> > ld -r is now supported with LTO
>
> Thanks, forgot about this one. This also needs support at linker side, right?
Only if you include assembler or non LTO code.
Without that it should work with any linker.
> Do you know minimal GNU ld/Gold versions that works fine?
I believe the assembler
On Mon, Sep 26, 2011 at 4:25 PM, Richard Guenther
wrote:
> On Wed, Sep 7, 2011 at 5:06 PM, Joseph S. Myers
> wrote:
>> This looks like it has the same issue with maybe needing to use
>> TYPE_MAIN_VARIANT in type comparisons as the shuffle patch.
>
> I don't think so, we move qualifiers to the ve
build_user_type_conversion_1 complains about ambiguity if
LOOKUP_COMPLAIN is set, so we need to avoid setting that flag when we
call the function from reference_binding. It seems to make sense to
mask away all inappropriate lookup flags at the top of
implicit_conversion rather than doing it fu
This patch consists intrinsics to properly create the bases and
direct_bases of a class in the correct order (including multiple nested
ambiguous virtual and non-virtual classes) for N2965
(http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2009/n2965.html).
This allows you to create type trait
Hi!
Adding Joseph and Jason to CC.
On Mon, Sep 26, 2011 at 04:56:20PM +0200, Richard Guenther wrote:
> Let's see what kind of fallout we get ;) For example, if the
> following is valid C code I expect we will vectorize the second
> loop (disambiguating p[i] and q[i]) bogously:
>
> void foo (int
On Mon, 26 Sep 2011, Jakub Jelinek wrote:
> Hi!
>
> Adding Joseph and Jason to CC.
>
> On Mon, Sep 26, 2011 at 04:56:20PM +0200, Richard Guenther wrote:
> > Let's see what kind of fallout we get ;) For example, if the
> > following is valid C code I expect we will vectorize the second
> > loop
On Mon, Sep 19, 2011 at 5:52 PM, Doug Evans wrote:
> Hi.
>
> This patch adds two functions, timeval_add and timeval_sub,
> to libiberty. GDB has use for them in a few places and since
> they're general purpose I wish to check them into libiberty.
>
> Ok to check in?
>
> 2011-09-19 Doug Evans
>
On Tue, Sep 20, 2011 at 11:23 AM, Doug Evans wrote:
> Hi.
>
> Part of the abstraction of an argv is a count of the number elements.
> This patch adds countargv which I have use for in gdb.
>
> Ok to check in?
>
> 2011-09-20 Doug Evans
>
> include/
> * libiberty.h (countargv): Decl
This patchlet represents the impact of insn ashrqi3 on CC by means of attribute
"cc" instead of hard-coding it in notice_update_cc.
Testsuite passes fine. Moreover, tested against code like
char c;
void func_1 (char a)
{
a = a >> 7;
if (a)
c = a;
}
that triggered PR39633.
Ok?
On Mon, Sep 26, 2011 at 05:50:40PM +0200, Jakub Jelinek wrote:
> On Mon, Sep 26, 2011 at 04:56:20PM +0200, Richard Guenther wrote:
> > Let's see what kind of fallout we get ;) For example, if the
> > following is valid C code I expect we will vectorize the second
> > loop (disambiguating p[i] and
On 9/23/2011 6:03 PM, Sriraman Tallam wrote:
This patch adds a new linker plugin to re-order functions.
This is great stuff. We were experimenting with using the coverage
files to generate an ordering for --section-ordering-file, but this
might be even better, will have to experiment with it
On 26 September 2011 15:24, Ulrich Weigand wrote:
> Ramana Radhakrishnan wrote:
>> On 9 September 2011 18:04, Ulrich Weigand wrote:
>> >
>> > In theory, LEGITIMIZE_RELOAD_ADDRESS could attempt to handle them by
>> > substituting the equivalent constant and then reloading the result.
>> > However,
Paolo,
> Ok with a couple of changes:
>
>> gcc:
>> * gthr-posix.h, gthr-single.h, gthr.h: Move to ../libgcc.
>
> Move gthr-posix.h to config/
I had left it in libgcc directly since the file, like gthr-single.h, is
included in a couple of others, and is pretty generic. The updated
patch
On Mon, Sep 26, 2011 at 9:22 AM, Nathan Froyd wrote:
> On 9/23/2011 6:03 PM, Sriraman Tallam wrote:
>>
>> This patch adds a new linker plugin to re-order functions.
>
> This is great stuff. We were experimenting with using the coverage files to
> generate an ordering for --section-ordering-file,
On 09/26/2011 11:50 AM, Mike Spertus wrote:
This patch consists intrinsics to properly create the bases and direct_bases of
a class
Looks pretty good. Some comments:
#define GCC_TREE_H
-
#include "hashtab.h"
I don't see any reason to remove this blank line.
+ if (TREE_CODE (par
On Mon, Sep 26, 2011 at 06:01:45PM +0200, Jakub Jelinek wrote:
> In the above function with this e.g. a[4] object is modified during
> the extecution of the foo block B, thus the 6.7.3.1/4 rules should apply.
> And within that block a[4] is accessed through lvalues whose address
> is p1 based as we
On Fri, Sep 23, 2011 at 10:34 AM, Jan Hubicka wrote:
> Hi,
> this patch extends inliner's predicate infrastructure to track optimization
> oppurtunities that depends on properties of call site parameters that are
> not readilly available (and do not belong to) jump functions.
>
> For this new inli
My patch for DR 1328 started comparing rvaluedness_matches_p for all
reference binding conversions, not just those where the target type
differs in rvaluedness. In this testcase, the rvaluedness of the
initializer differs: when calling the first function the initializer is
a pointer constant,
On Sep 25, 2011, at 4:01 AM, Iain Sandoe wrote:
> Following Eric's fix to the stack-check output for rs6000/Darwin, I think we
> can enable this function for the port.
>
> bootstrapped/tested (including Java, Ada), on *-darwin9, x86-64-darwin10.
>
> OK for trunk?
Ok.
2011/9/26 Georg-Johann Lay :
> This patchlet represents the impact of insn ashrqi3 on CC by means of
> attribute
> "cc" instead of hard-coding it in notice_update_cc.
>
> Testsuite passes fine. Moreover, tested against code like
>
> char c;
>
> void func_1 (char a)
> {
> a = a >> 7;
> if (a)
> >>> This patch is OK, with or without the testsuite additions Jakub
> >>> mentions.
> >>
> >> Thanks, I added the new tests, and tweaked the clone suffix parsing to
> >> allow the leading "_".
> >>
> >> Committed as r179132.
> >
> > Should I backport this patch to gcc-4_6-branch?
>
> Since you of
http://codereview.appspot.com/5090041/diff/1/gcc/cp/pph-streamer-out.c
File gcc/cp/pph-streamer-out.c (right):
http://codereview.appspot.com/5090041/diff/1/gcc/cp/pph-streamer-out.c#newcode1924
gcc/cp/pph-streamer-out.c:1924: tree enclosing_namespace )
1922 void
1923 pph_write_namespace_tree (
Hi,
this patch implements slim LTO. It is updated version of Andi's patch. It is
done by terminating IPA optimization after analyzing and outputting LTO and in
compile_file skipping stuff that outputs assembly.
After some consideration I neded up with -ffat-lto-objects and
-fno-fat-lto-objects.
This patch from Chung-Lin Tang fixes PR50496.
I post it under his name (hope this is OK). The original patch can be
found here:
https://code.launchpad.net/~cltang/gcc-linaro/lp-748138-cfgrtl-fix-4_5/+merge/60742/+preview-diff/+files/preview.diff
https://bugs.launchpad.net/ubuntu/+source/gcc-4.5/+b
From: David Miller
Date: Sat, 24 Sep 2011 17:30:57 -0400 (EDT)
> From: Hans-Peter Nilsson
> Date: Sat, 24 Sep 2011 17:15:06 -0400 (EDT)
>
>> One more: please consider adding a
>> if (TARGET_VIS) builtin_define ("__VIS__=something") so I as a
>> user theoretically wouldn't *have* to autoconfisc
On 09/26/2011 10:21 AM, Dodji Seketeli wrote:
> > + /* The cpp_reader the macro comes from. */
> > + cpp_reader *pfile;
>
> This seems to only be used to decide whether or not to increment
> location_ptr. Rather than base that decision on going all the way
> back to the CPP_OPTION, le
Rainer Orth writes:
> the problem turned out to be obvious once you ignore most of the error
> message ;-( The spurious import is in sysinfo.go, which unconditionally
> imports the package, but only uses in some cases.
Ah. This patch fixes the location of the error message. Bootstrapped
and r
On Fri, Sep 17, 2010 at 5:57 AM, Martin Jambor wrote:
> Hi,
>
> this is a followup to my previous build_ref_for_offset patch. If
> build_user_friendly_ref_for_offset does not succeed when building an
> artificial SRA access, we use build_ref_for_model which produces an
> expression we cannot use
Hi all,
I just committed as obvious two small patches for two recently
reported accepts-invalid PRs (after regtesting):
http://gcc.gnu.org/viewcvs?view=revision&revision=179213
Cheers,
Janus
Jason Merrill writes:
> On 09/21/2011 02:32 PM, Dodji Seketeli wrote:
>> FWIW, I'd like the LRK_MACRO_PARM_REPLACEMENT name (or its
>> replacement. ha ha) to hint at the fact that it really has to do with
>> a token that is an /argument/ for a function-like macro.
>
> I disagree; arguments are t
I foolishly tried to use an unspec to keep the compiler from
eliminating stores to %gsr, when the proper thing to do is to
mark it in global_regs[] instead.
Committed to trunk.
gcc/
* config/sparc/sparc.c (sparc_conditional_register_usage): When VIS
is enabled, mark %gsr as glob
On 09/26/2011 10:10 PM, Janus Weil wrote:
Hi all,
I just committed as obvious two small patches for two recently
reported accepts-invalid PRs (after regtesting):
http://gcc.gnu.org/viewcvs?view=revision&revision=179213
Thanks !
As far as the fortran/interface.c change is concerned, why isn'
On Mon, Sep 26, 2011 at 10:28:14PM +0200, Toon Moene wrote:
> On 09/26/2011 10:10 PM, Janus Weil wrote:
>
> >Hi all,
> >
> >I just committed as obvious two small patches for two recently
> >reported accepts-invalid PRs (after regtesting):
> >
> >http://gcc.gnu.org/viewcvs?view=revision&revision=17
Hi,
>> >I just committed as obvious two small patches for two recently
>> >reported accepts-invalid PRs (after regtesting):
>> >
>> >http://gcc.gnu.org/viewcvs?view=revision&revision=179213
>>
>> Thanks !
>>
>> As far as the fortran/interface.c change is concerned, why isn't a full
>> TKR (Type/Ki
On 09/26/2011 04:21 PM, Dodji Seketeli wrote:
Jason Merrill writes:
On 09/21/2011 02:32 PM, Dodji Seketeli wrote:
FWIW, I'd like the LRK_MACRO_PARM_REPLACEMENT name (or its
replacement. ha ha) to hint at the fact that it really has to do with
a token that is an /argument/ for a function-like
Hi,
Here is a simple patch to add SECTION_EXCLUDE to the list of
section flag macros. Is this ok for trunk?
Thanks,
-Sri.
* output.h (SECTION_EXCLUDE): New macro.
* varasm.c (default_elf_asm_named_section): Check for
SECTION_EXCLUDE.
Index: gcc/varasm.c
=
Committed to trunk.
gcc/testsuite/
* gcc.target/sparc/rdgsr.c: New test.
* gcc.target/sparc/edge.c: New test.
* gcc.target/sparc/fcmp.c: New test.
---
gcc/testsuite/ChangeLog|3 ++
gcc/testsuite/gcc.target/sparc/edge.c | 39 +++
On Mon, Sep 26, 2011 at 3:26 AM, Jiangning Liu wrote:
> This patch is fix PR38644, a 3-year-old bug.
>
> From the discussions in mail list and bugzilla, I think the middle end fix
> is a common view. Although there are stills some gaps on how to fix it in
> middle end, I think this patch at least
On 09/26/2011 07:55 AM, Jakub Jelinek wrote:
> * rtl.h (const_tiny_rtx): Change into array of 4 x MAX_MACHINE_MODE
> from 3 x MAX_MACHINE_MODE.
> (CONSTM1_RTX): Define.
> * emit-rtl.c (const_tiny_rtx): Change into array of 4 x MAX_MACHINE_MODE
> from 3 x MAX_MACHINE_
Rainer Orth writes:
> the problem turned out to be obvious once you ignore most of the error
> message ;-( The spurious import is in sysinfo.go, which unconditionally
> imports the package, but only uses in some cases.
And this patch should fix the Solaris problem. Bootstrapped and ran Go
test
The gcc source uses several constructs that GDB does not understand.
This patch corrects some of them. It affects only compilers built
with ENABLE_TREE_CHECKING, and hence release compilers are unaffected.
In particular, I change the implementation of CHECK macros using
__extension__ into macros
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 08/18/11 15:59, Richard Henderson wrote:
> On 08/17/2011 12:21 AM, Richard Guenther wrote:
>> The patch itself looks sensible, though I am surprised ifcvt
>> doesn't run in cfglayout mode (so you have to use reg notes to
>> find probabilities ...)
>
On Mon, Sep 26, 2011 at 11:22:54PM +0930, Alan Modra wrote:
> Two regressions appeared due to a problem in the shrink-wrap code.
These two.
+FAIL: g++.dg/torture/pr46111.C -O1 (internal compiler error)
+FAIL: gcc.dg/autopar/pr46099.c (internal compiler error)
Both "internal compiler error: in m
Rainer Orth writes:
> Apart from the (already fixed) go.* dejagnu errors, I noticed another
> problem: the libgo net test now fails on Solaris, but only with Sun ld:
>
> Undefined first referenced
> symbol in file
> getaddrinfo
On 09/27/11 00:32, Alan Modra wrote:
> On Mon, Sep 26, 2011 at 11:22:54PM +0930, Alan Modra wrote:
>> Two regressions appeared due to a problem in the shrink-wrap code.
>
> These two.
> +FAIL: g++.dg/torture/pr46111.C -O1 (internal compiler error)
> +FAIL: gcc.dg/autopar/pr46099.c (internal comp
Hi,
the below implements what submitter requested - ie, separate bindings
with semicolons instead of commas, already used in many other places.
Seems a pretty straightforward change to me. Tested x86_64-linux.
Ok for mainline?
Thanks,
Paolo.
/
/cp
2011-09-26 Paolo Carl
1 - 100 of 128 matches
Mail list logo
