On Tue, Apr 2, 2013 at 7:43 AM, Wei Mi w...@google.com wrote:
I attached the patch.4 based on r197308. r197308 changes shift-and
type truncation from define_insn_and_split to define_insn. patch.4
changes ix86_rtx_costs for shift-and type rtx to get the correct cost
for the result after the
Hi Marcus,
Thanks for reviewing the patch and your comments.
I'm not sure how good the coverage is from these test cases
The shift instructions are not generated with the test case since
multiply patterns are generated for the same. Its the same case with
other add and sub instructions which
Thanks, here is a version taking into account all your comments, and which
still passes bootstrap+testsuite on x86_64-linux-gnu. I am not completely
sure if there is a point checking !side_effects_p (op1) after rtx_equal_p
(op0, op1), but I am still doing it as it seems safe.
It's also done
On Tue, 2 Apr 2013, Eric Botcazou wrote:
Thanks, here is a version taking into account all your comments, and which
still passes bootstrap+testsuite on x86_64-linux-gnu. I am not completely
sure if there is a point checking !side_effects_p (op1) after rtx_equal_p
(op0, op1), but I am still
By the way, shouldn't this be:
unsigned HOST_WIDE_INT mask = ((unsigned HOST_WIDE_INT) 2 (n_elts - 1)) -
1;
so it doesn't cause undefined behavior for V64QI?
You're right, but I think that we'd rather write:
if (n_elts == HOST_BITS_PER_WIDE_INT)
mask = -1;
else
mask =
Well, the goal was to have arbitrary canonicalization for all cases where
precise rules are not already provided. In (a8)|(b24), there is no
obvious reason why lshift should have higher or lower priority than
rshift, but we don't want to have to write patterns with both orders in
the target
On Sun, Mar 31, 2013 at 10:01:21PM +0200, Manuel López-Ibáñez wrote:
The following patch implements an option that allows printing the
diagnostics markers (error:, warning:, note:) in different
colors (red, magenta, and green, respectively).
Thanks for working on that.
I followed the
Using set_block_for_insn instead of using BLOCK_FOR_INSN is the proper
way.
Yes, but BLOCK_FOR_INSN as accessor around INSN_BASIC_BLOCK is ugly and a bit
misleading. Either keep BLOCK_FOR_INSN or make the full change (the number of
occurrences of BLOCK_FOR_INSN in the back-ends is
On Fri, Mar 29, 2013 at 1:05 PM, Steven Bosscher stevenb@gmail.com wrote:
Hello,
It looks like there are places in the middle end that use remove_insn
on insns that are not actually emitted. This breaks the assert I added
in df_insn_delete. The patch disables the assert for now. The
On Fri, Mar 29, 2013 at 6:02 PM, Gabriel Dos Reis g...@axiomatics.org wrote:
Paolo Carlini paolo.carl...@oracle.com writes:
| Hi,
|
| On 03/29/2013 04:59 PM, Gabriel Dos Reis wrote:
| This patch introduces the predicate VAR_P and use it in place of direct
|
| TREE_CODE (t) ==
On Tue, Apr 2, 2013 at 4:14 AM, Jakub Jelinek ja...@redhat.com wrote:
Yeah, IMHO we definitely want to support GCC_COLORS env var or similar, with
same syntax as e.g. GREP_COLORS, but with different names of the (two
letter?) color names.
Agreed.
This patch only allows two options
On Tue, Apr 2, 2013 at 4:26 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Fri, Mar 29, 2013 at 6:02 PM, Gabriel Dos Reis g...@axiomatics.org wrote:
Paolo Carlini paolo.carl...@oracle.com writes:
| Hi,
|
| On 03/29/2013 04:59 PM, Gabriel Dos Reis wrote:
| This patch introduces the
On Tue, Apr 2, 2013 at 11:17 AM, Eric Botcazou wrote:
Using set_block_for_insn instead of using BLOCK_FOR_INSN is the proper
way.
Yes, but BLOCK_FOR_INSN as accessor around INSN_BASIC_BLOCK is ugly and a bit
misleading.
True, I don't like that bit very much myself, either. But I expected
On Tue, Apr 02, 2013 at 04:26:37AM -0500, Gabriel Dos Reis wrote:
IMO we also want that autodetection and default to auto.
I disagree.
For this release, I think the default should be off. We need more
I think a year is plenty of time to agree on the default color scheme
(and yes, the
On Sun, Mar 31, 2013 at 7:20 PM, Marc Glisse marc.gli...@inria.fr wrote:
On Sun, 31 Mar 2013, Andrew Pinski wrote:
On Sun, Mar 31, 2013 at 6:31 AM, Marc Glisse marc.gli...@inria.fr wrote:
Hello,
this adds constant folding of VEC_COND_EXPR at the tree level by
forwarding
to the
On Sun, Mar 31, 2013 at 7:51 PM, Kenneth Zadeck
zad...@naturalbridge.com wrote:
richard,
I was able to add everything except for the checking asserts.While I
think that this is a reasonable idea, it is difficult to add that to a
function that is defined in hwint.h because of circular
On Mon, Apr 1, 2013 at 12:19 AM, Lawrence Crowl cr...@googlers.com wrote:
Remove unused symbols related to ivtype_map. This map does not appear to
exist and I see no evidence of its removal in the ChangeLog.
Tested on x86_64.
Okay for trunk?
Ok.
Thanks,
Richard.
Index: gcc/ChangeLog
Hi,
On 04/02/2013 11:30 AM, Gabriel Dos Reis wrote:
[]
Sorry for chiming in late - but VAR_P loses the fact that we are checking
for a decl ... VAR_DECL_P would be more like following existing practice
(otherwise we can shorten VAR_OR_FUNCTION_DECL_P to
VAR_OR_FUNCTION_P for example). As for
True, I don't like that bit very much myself, either. But I expected
more resistance for the full change ;-)
Possibly. You might really want to wait until the end of the week. :-)
Right. OK if I call it get_block_for_insn() and make the replacements?
get_block_for_insn will break
Hi,
According to Vladimir Makarov's analysis, the root cause of PR target/54338
is that ALL_REGS doesn't contain IWMMXT_GR_REGS in REG_CLASS_CONTENTS.
It seems there is no reason to exclude the IWMMXT_GR_REGS from ALL_REGS as
IWMMXT_GR_REGS are the real registers.
This patch simply makes
On 04/02/13 10:40, Xinyu Qi wrote:
Hi,
According to Vladimir Makarov's analysis, the root cause of PR target/54338
is that ALL_REGS doesn't contain IWMMXT_GR_REGS in REG_CLASS_CONTENTS.
It seems there is no reason to exclude the IWMMXT_GR_REGS from ALL_REGS as
IWMMXT_GR_REGS are the real
On Tue, Apr 2, 2013 at 4:44 AM, Paolo Carlini paolo.carl...@oracle.com wrote:
Hi,
On 04/02/2013 11:30 AM, Gabriel Dos Reis wrote:
[]
Sorry for chiming in late - but VAR_P loses the fact that we are checking
for a decl ... VAR_DECL_P would be more like following existing practice
Ping.
On 21 March 2013 14:58, Zhenqiang Chen zhenqiang.c...@linaro.org wrote:
Hi,
When shrink-wrap is enabled, the returns from simple-return path and
normal return path can be merged. The code is like:
tst ...
/ \
| push ...
| ...
| pop ...
\ /
bx
On Thu, Mar 21, 2013 at 7:03 AM, Zhenqiang Chen
zhenqiang.c...@linaro.org wrote:
Hi,
The patch is to enable shrink-wrap for TARGET_ARM and TARGET_THUMB2.
Bootstrapped and no make check regression.
All previous Linaro shrink-wrap bugs (http://goo.gl/6fGg5) are verified.
Is it OK?
The tests
On Mon, Apr 1, 2013 at 7:07 PM, Pavel Chupin pavel.v.chu...@gmail.com wrote:
On Android pthread is integrated into libc.
Attached patch fixes configures for this case by trying to build test
without -pthread -lpthread.
2013-04-01 Pavel Chupin pavel.v.chu...@intel.com
Fix libatomic
Hello,
This idiom: if (GET_CODE (body) == ASM_INPUT || asm_noperands (body)
= 0) appears in multiple places. There's even one place where the
idiom above is used in reverse (making the GET_CODE... check
redundant). A few more places to the equivalent by checking
extract_asm_operands != NULL.
It
On Thu, Mar 28, 2013 at 03:55:38PM +0100, Paolo Carlini wrote:
.. oops, the patch I attached has a typo. This is the right one.
This regressed
FAIL: obj-c++.dg/try-catch-13.mm -fgnu-runtime (test for errors, line 12)
Can you please adjust that testcase too?
Jakub
On Thu, Mar 21, 2013 at 6:09 PM, Kyrylo Tkachov kyrylo.tkac...@arm.com wrote:
Hi all,
This patch adds a splitter variant of the minmax_arithsi pattern for when
the operator
is non-commutative (MINUS) and the ordering of the operands is not
canonical.
That is, it will trigger for:
#define
On 04/02/2013 12:03 PM, Jakub Jelinek wrote:
On Thu, Mar 28, 2013 at 03:55:38PM +0100, Paolo Carlini wrote:
.. oops, the patch I attached has a typo. This is the right one.
This regressed
FAIL: obj-c++.dg/try-catch-13.mm -fgnu-runtime (test for errors, line 12)
Can you please adjust that
Hello!
.. oops, the patch I attached has a typo. This is the right one.
This regressed
FAIL: obj-c++.dg/try-catch-13.mm -fgnu-runtime (test for errors, line 12)
Can you please adjust that testcase too?
Tested patch is at [1].
[1] http://gcc.gnu.org/ml/gcc-patches/2013-04/msg00013.html
On Tue, 2 Apr 2013, Eric Botcazou wrote:
I am not sure about adding just a few rules. If I just say that lshift is
stronger than rshift, the relation is not an order (transitive) anymore.
Why? Can't you give them precedences in commutative_operand_precedence that
preserve the transitivity?
This idiom: if (GET_CODE (body) == ASM_INPUT || asm_noperands (body)
= 0) appears in multiple places. There's even one place where the
idiom above is used in reverse (making the GET_CODE... check
redundant). A few more places to the equivalent by checking
extract_asm_operands != NULL.
I
Ping (second try)
Sorry if you received it twice, it seems that my gmail account
switched in text/html mode :(
Many thanks,
Yvan
On 2 April 2013 11:21, Yvan Roux yvan.r...@linaro.org wrote:
Ping
On 17 March 2013 21:34, Yvan Roux yvan.r...@linaro.org wrote:
Hi,
this is a backport from
I can, but then I am giving lshift higher priority than every other
operation, not just rshift. And if I want to give (vec_select x 0) a
higher precedence than (vec_select x 1) but lower than (vec_select
(vec_concat a b) 1), the weights may become complicated, whereas the
comparison function
Hi!
Jason's http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34949#c12
patch attempts to emit a clobber for *this at the end of the destructor,
so that we can DSE stores into the object when the object is dead, but
so far only VAR_DECLs have been allowed on the lhs of gimple_clobber_p
stmts.
This
Hi!
The patch I've just posted wasn't enough, because stmt_kills_ref_p_1
only did something if base == ref-base, but in the case of the dtors
base and ref-base are often MEM_REFs, which aren't equal, but they
just operand_equal_p. And, for MEM_REFs, we don't even need to require
that the two
Hi!
As discussed in the PR, when __builtin_constant_p is used in the first
argument of __builtin_choose_expr, as we require a constant value
immediately, we'd better force folding of __builtin_constant_p immediately,
as if we don't, and it isn't 1 known immediately, we won't end up with a
Hi,
This patch adds support for the Cortex-A53. Tested with a simple build
of arm-none-eabi for no regressions on a version before the builds broke
with PR56809. We have been carrying this internally for quite some time
now - so I'm ok with it's stability.
The default tuning for
On Tue, 2 Apr 2013, Jakub Jelinek wrote:
Hi!
Jason's http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34949#c12
patch attempts to emit a clobber for *this at the end of the destructor,
so that we can DSE stores into the object when the object is dead, but
so far only VAR_DECLs have been allowed
Hi!
On the (undefined behavior) testcase below, we end up with
then_bb ending with __builtin_unreachable () at the tree level, therefore
no successor at the RTL level, and else_bb being EXIT_BLOCK_PTR (i.e.
conditional return before a bb with undefined behavior at the end).
Trying to optimize
On Tue, 2 Apr 2013, Jakub Jelinek wrote:
Hi!
The patch I've just posted wasn't enough, because stmt_kills_ref_p_1
only did something if base == ref-base, but in the case of the dtors
base and ref-base are often MEM_REFs, which aren't equal, but they
just operand_equal_p. And, for
This backports a fix to fix PR56768.
Bootstrapped on the 4.7 branch, testing in progress.
Richard.
2013-04-02 Richard Biener rguent...@suse.de
PR middle-end/56768
Backport from mainline
2012-05-16 Richard Guenther rguent...@suse.de
* tree-inline.c
On Tue, Apr 02, 2013 at 01:44:46PM +0200, Richard Biener wrote:
@@ -412,7 +412,8 @@ get_mem_ref_of_assignment (const gimple
{
gcc_assert (gimple_assign_single_p (assignment));
- if (gimple_store_p (assignment))
+ if (gimple_store_p (assignment)
+ !gimple_clobber_p
This patch is for GCC 4.8 and 4.9.
libcpp/ChangeLog
2013-04-02 Sebastian Huber sebastian.hu...@embedded-brains.de
* configure.ac: Require 64-bit int for arm*-*-rtems*.
* configure: Regenerate.
---
libcpp/configure|1 +
libcpp/configure.ac |1 +
2 files changed, 2
On Tue, 2 Apr 2013, Jakub Jelinek wrote:
On Tue, Apr 02, 2013 at 01:44:46PM +0200, Richard Biener wrote:
@@ -412,7 +412,8 @@ get_mem_ref_of_assignment (const gimple
{
gcc_assert (gimple_assign_single_p (assignment));
- if (gimple_store_p (assignment))
+ if
On Mon, Apr 1, 2013 at 5:52 PM, Marc Glisse marc.gli...@inria.fr wrote:
On Sat, 30 Mar 2013, Marc Glisse wrote:
* tree-flow-inline.h (get_addr_base_and_unit_offset_1): Handle
BIT_FIELD_REF.
I wrote a safer version of this for PR52436:
That variant is ok - please test and
On Sat, Mar 30, 2013 at 5:14 PM, Marc Glisse marc.gli...@inria.fr wrote:
On Tue, 29 Jan 2013, Richard Biener wrote:
So yes, handling BIT_FIELD_REF in the vectorizer looks like the correct
way to do - but mind that you should constrain the BIT_FIELD_REFs you
allow (I suppose in the end that's
On Tue, 2 Apr 2013, Jakub Jelinek wrote:
Hi!
On the (undefined behavior) testcase below, we end up with
then_bb ending with __builtin_unreachable () at the tree level, therefore
no successor at the RTL level, and else_bb being EXIT_BLOCK_PTR (i.e.
conditional return before a bb with
could someone please review and commit if it is OK?
I don't have commit access.
Regards,
Pitchumani
On 4/1/2013 4:30 PM, Pitchumani Sivanupandi wrote:
Fix test case sra-13.c that assumed int is always 4 bytes.
Regards,
Pitchumani
2013-04-01 Pitchumani Sivanupandi pitchuman...@atmel.com
On Wed, Mar 27, 2013 at 01:56:48PM +0400, Kirill Yukhin wrote:
Otherwise OK.
Thanks,
Hi, chacked into trunk: http://gcc.gnu.org/ml/gcc-cvs/2013-03/msg00785.html
This leads to:
../../gcc/config/t-linux-android:22: warning: overriding recipe for target
`linux-android.o'
Ping ^2
- Original Message -
From: Kai Tietz ktiet...@googlemail.com
To: GCC Patches gcc-patches@gcc.gnu.org
Cc: Richard Henderson r...@redhat.com
Sent: Thursday, March 28, 2013 2:35:15 PM
Subject: Re: [patch i386 windows]: Fix PR/52790 also required for workig
upcoming cygwin x64
On Tue, Apr 02, 2013 at 02:57:13PM +0200, Richard Biener wrote:
On Tue, 2 Apr 2013, Jakub Jelinek wrote:
On the (undefined behavior) testcase below, we end up with
then_bb ending with __builtin_unreachable () at the tree level, therefore
no successor at the RTL level, and else_bb being
Hi,
I think this is my last obvious clean-up for now. I'm going to commit it
when testing on x86_64-linux completes.
Thanks,
Paolo.
//
2013-04-02 Paolo Carlini paolo.carl...@oracle.com
* friend.c (do_friend): Use COMPLETE_OR_OPEN_TYPE_P.
* pt.c
On Tue, 2 Apr 2013, Jakub Jelinek wrote:
On Tue, Apr 02, 2013 at 02:57:13PM +0200, Richard Biener wrote:
On Tue, 2 Apr 2013, Jakub Jelinek wrote:
On the (undefined behavior) testcase below, we end up with
then_bb ending with __builtin_unreachable () at the tree level, therefore
no
Richard,
did everything that you asked here. bootstrapped and regtested on
x86-64. ok to commit?
kenny
On 04/02/2013 05:38 AM, Richard Biener wrote:
On Sun, Mar 31, 2013 at 7:51 PM, Kenneth Zadeck
zad...@naturalbridge.com wrote:
richard,
I was able to add everything except for the
Yep.. we missed that:
t-linux-android was added here:
# Add Android userspace support to Linux targets.
case $target in
*linux*)
tm_p_file=${tm_p_file} linux-protos.h
tmake_file=${tmake_file} t-linux-android
tm_file=$tm_file linux-android.h
On Tue, Apr 02, 2013 at 06:24:06PM +0400, Alexander Ivchenko wrote:
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,7 @@
+2013-04-02 Alexander Ivchenko alexander.ivche...@intel.com
+
+ * config.gcc (arm*-*-linux-*): Remove duplicate t-linux-android.
+
2013-04-02 Richard Biener
Hi,
is it ok?
Yes.
Checked into trunk: http://gcc.gnu.org/ml/gcc-cvs/2013-04/msg00066.html
Thanks, K
LAST_UPDATED: Mon Apr 1 18:01:32 UTC 2013 (revision 197313)
Native configuration is powerpc-ibm-aix7.1.0.0
=== g++ tests ===
Running target unix
FAIL: tmpdir-g++.dg-struct-layout-1/t024
cp_compat_x_tst.o-cp_compat_y_tst.o execute
FAIL: tmpdir-g++.dg-struct-layout-1/t026
On Tue, Apr 2, 2013 at 8:34 AM, Paolo Carlini paolo.carl...@oracle.com wrote:
Hi,
I think this is my last obvious clean-up for now. I'm going to commit it
when testing on x86_64-linux completes.
thanks.
-- Gaby
On Tue, Apr 2, 2013 at 3:49 PM, Kenneth Zadeck zad...@naturalbridge.com wrote:
Richard,
did everything that you asked here. bootstrapped and regtested on x86-64.
ok to commit?
diff --git a/gcc/hwint.c b/gcc/hwint.c
index 330b42c..7e5b85c 100644
--- a/gcc/hwint.c
+++ b/gcc/hwint.c
@@ -204,3
On Wed, Feb 27, 2013 at 2:59 AM, Kenneth Zadeck
zad...@naturalbridge.com wrote:
This patch contains a large number of the changes requested by Richi. It
does not contain any of the changes that he requested to abstract the
storage layer. That suggestion appears to be quite unworkable.
I of
This patch is for GCC 4.6 and 4.7.
libcpp/ChangeLog
2013-04-02 Sebastian Huber sebastian.hu...@embedded-brains.de
PR target/56771
* configure.ac: Require 64-bit int for arm*-*-rtems[.0-9]*.
* configure: Regenerate.
---
libcpp/configure|1 +
libcpp/configure.ac
This patch is for GCC 4.8 and 4.9.
v2: Fix ChangeLog.
libcpp/ChangeLog
2013-04-02 Sebastian Huber sebastian.hu...@embedded-brains.de
PR target/56771
* configure.ac: Require 64-bit int for arm*-*-rtems*.
* configure: Regenerate.
---
libcpp/configure|1 +
On 15 March 2013 08:55, Jakub Jelinek ja...@redhat.com wrote:
On Thu, Feb 28, 2013 at 08:32:02AM -0700, Tom Tromey wrote:
2013-02-27 Tom Tromey tro...@redhat.com
* libsupc++/unwind-cxx.h: Include sys/sdt.h if detected.
(PROBE2): New macro.
* libsupc++/eh_throw.cc
On 04/02/2013 04:54 AM, Eric Botcazou wrote:
I wonder if those issues might in most cases be bugs in the back-ends
(optimizations missed depending on the order), that the canonicalization
would make more noticable (and thus easier to fix).
There are certainly such issues in the back-ends (and
On Tue, 2 Apr 2013, Jonathan Wakely wrote:
On 15 March 2013 08:55, Jakub Jelinek ja...@redhat.com wrote:
On Thu, Feb 28, 2013 at 08:32:02AM -0700, Tom Tromey wrote:
2013-02-27 Tom Tromey tro...@redhat.com
* libsupc++/unwind-cxx.h: Include sys/sdt.h if detected.
(PROBE2): New
From: Ramana Radhakrishnan [mailto:ramana@googlemail.com]
Sent: 02 April 2013 11:10
To: Kyrylo Tkachov
Cc: gcc-patches@gcc.gnu.org; Richard Earnshaw; Ramana Radhakrishnan
Subject: Re: [PATCH][ARM] minmax_arithsi for non-canonical operand
order with MINUS operator
On Thu, Mar 21, 2013
Hi Richard,
I've now updated this patch. How does this version look?
Catherine
2013-04-02 Catherine Moore c...@codesourcery.com
* config/mips/micromips.md (jraddiusp): New pattern.
* config/mips/mips.c (mips_expand_epilogue): Use the JRADDIUSP
instruction if possible.
This patch updates the C_LOC checking fixes for array. In particular:
Fortran 2003 allows:
(1) have interoperable type and type parameters and be
(a) a variable that has the TARGET attribute and is interoperable,
(b) an allocated allocatable variable that has the TARGET attribute
and
On Tue, Apr 2, 2013 at 5:00 AM, Jeff Law wrote:
On 04/01/2013 01:47 PM, Eric Botcazou wrote:
I'm not sure what this is supposed to do. How is pat == target ever
going to be true when ANY_RETURN_P (pat) is true? Isn't target supposed
to be a CODE_LABEL or NULL? What am I missing?
The
On 2013-03-22 02:27, Kai Tietz wrote:
Hi,
this patch fixes PR/52790 and supports for x64 Windows targets the use
of large and medium code-model.
This feature is required for upcoming new cygwin x64 target, which
uses full 48-bit available address-space of x64 Windows.
The cygwin-target depends
Hello!
2013-04-02 Uros Bizjak ubiz...@gmail.com
* config/i386/i386.md (*testqi_ext_3): Merge with *testqi_ext_3_rex64
using SWI48 mode attribute.
Tested on x86_64-pc-linux-gnu {,-m32} and committed to mainline SVN.
Uros.
Index: config/i386/i386.md
On Apr 2, 2013, at 5:59 AM, Pitchumani Sivanupandi pitchuman...@atmel.com
wrote:
could someone please review
Ok.
and commit if it is OK?
Committed revision 197366.
Yes, I agree that you win the challenge that it can be done.What you
have always failed to address is why anyone would want to do this. Or
how this would at all be desirable.But I completely agree that from
a purely abstract point of view you can add a storage model.
Now here is why
On Tue, Apr 02, 2013 at 02:41:07PM +0200, Richard Biener wrote:
I've just come along 56787 and decided that not collecting
CLOBBERs during data-reference gathering would be wrong.
Thus changing gimple_store_p would be wrong, too. So in the
end the above hunk looks ok and we shouldn't change
On Tue, Apr 02, 2013 at 01:52:16PM +0200, Richard Biener wrote:
in all relevant cases operand zero of the MEM_REFs are SSA names
and thus can be compared for equality using == (exception are
integer constant pointers where the type could be theoretically
different).
...
Here is what I've
Build and regtested on x86-64-gnu-linux.
OK for the trunk?
Tobias
2013-04-02 Tobias Burnus bur...@net-b.de
PR fortran/56810
* io/list_read.c (check_type): Fix kind checking for COMPLEX.
2013-04-02 Tobias Burnus bur...@net-b.de
PR fortran/56810
* gfortran.dg/read_repeat_2.f90: New.
this time for sure.
kenny
On 04/02/2013 10:54 AM, Richard Biener wrote:
On Tue, Apr 2, 2013 at 3:49 PM, Kenneth Zadeck zad...@naturalbridge.com wrote:
Richard,
did everything that you asked here. bootstrapped and regtested on x86-64.
ok to commit?
diff --git a/gcc/hwint.c b/gcc/hwint.c
First round of results for gcc 4.8.x.
-tgc
Testresults for 4.8.0
arm-unknown-linux-gnueabi
hppa2.0w-hp-hpux11.00
hppa2.0w-hp-hpux11.11
hppa64-hp-hpux11.11
i386-apple-darwin10.8.0
i386-pc-solaris2.9
i386-pc-solaris2.10
i386-pc-solaris2.11
mipsel-unknown-linux-gnu
On Tue, Apr 2, 2013 at 12:34 PM, Eric Botcazou wrote:
This idiom: if (GET_CODE (body) == ASM_INPUT || asm_noperands (body)
= 0) appears in multiple places. There's even one place where the
idiom above is used in reverse (making the GET_CODE... check
redundant). A few more places to the
On 2 April 2013 16:39, Marc Glisse wrote:
On Tue, 2 Apr 2013, Jonathan Wakely wrote:
Should we update the prerequisites documentation to say that if
Systemtap is installed it needs to be at least version X?
I thought you were going to suggest enhancing the configure test so it fails
on old
On 4/2/13, Richard Biener richard.guent...@gmail.com wrote:
On Mon, Apr 1, 2013 at 12:19 AM, Lawrence Crowl cr...@googlers.com wrote:
Remove unused symbols related to ivtype_map. This map does not appear to
exist and I see no evidence of its removal in the ChangeLog.
Tested on x86_64.
Okay
Now that Jakub has checked in support for MEM_REF clobbers, we can use
that to let the front end tell the back end that nothing in an object is
interesting after the destructor is complete. This should allow us to
optimize away assignments to vtable pointers if the destructor doesn't
use
On Tue, 2 Apr 2013, Jakub Jelinek wrote:
The following patch folds it there immediately too. Bootstrapped/regtested
on x86_64-linux and i686-linux, ok for trunk?
OK.
BTW, in the PR I've also mentioned what has been reported to the Red Hat
bugzilla about __builtin_constant_p in array bounds
On Tue, 2 Apr 2013, Ramana Radhakrishnan wrote:
2013-04-02 Ian Caulfield ian.caulfi...@arm.com
Ramana Radhakrishnan ramana.radhakrish...@arm.com
* config/arm/arm-arches.def (armv8-a): Default to cortex-a53.
* config/arm/t-arm (MD_INCLUDES): Depend on
This patch implements
http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2010/n3189.htm
The function pointer holding the current new handler was exposed as an
extern variable, which is unnecessary now there's an accessor for it.
Atomic operations are used to set and retrieve the handlers to avoid
Hi,
When I look into code of auto-inc-dec.c in GCC, I found this typo in rtl.h,
as
#if (defined (HAVE_PRE_INCREMENT) || defined (HAVE_PRE_DECREMENT) \
|| defined (HAVE_POST_INCREMENT) || defined (HAVE_POST_DECREMENT) \
|| defined (HAVE_PRE_MODIFY_DISP) || defined (HAVE_PRE_MODIFY_DISP) \
-Original Message-
From: gcc-patches-ow...@gcc.gnu.org [mailto:gcc-patches-ow...@gcc.gnu.org]
On
Behalf Of Bin Cheng
Sent: Wednesday, April 03, 2013 9:38 AM
To: gcc-patches@gcc.gnu.org
Subject: [PATCH GCC]Fix typo in definition of macro AUTO_INC_DEC in rtl.h
Hi,
When I look
I remember thinking that I'd need to deal with mangling for
ref-qualifiers, but somehow it slipped my mind. This patch implements
it. Having to mess with re-adding the ref-qualifier after
TYPE_MAIN_VARIANT makes me more determined to deal with giving
ref-qualified types their own
90 matches
Mail list logo