On Fri, Oct 5, 2012 at 8:43 PM, Andrew Pinski
wrote:
> On Sun, Aug 19, 2012 at 10:13 AM, Richard Sandiford
> wrote:
>> Andrew Pinski writes:
>>> Right now we only produce ins when a zero_extract is used on the
>>> right hand side. We can do better by adding some patterns which
>>> combine for
On 10/05/2012 07:09 PM, Marc Glisse wrote:
[LR]SHIFT_EXPR are special and also accept to have a vector as first
argument and a scalar as second argument (but not the reverse).
Fair enough.
These 2 lines are in a switch in the case where scalar_to_vector
returned stv_firstarg, meaning that the
On Sun, Aug 19, 2012 at 10:13 AM, Richard Sandiford
wrote:
> Andrew Pinski writes:
>> Right now we only produce ins when a zero_extract is used on the
>> right hand side. We can do better by adding some patterns which
>> combine for the ins instruction. This patch adds those patterns and a
>>
Oleg Endo wrote:
> The attached patch is the next step that adds the thread pointer
> builtins. The GBR address mode stuff will follow afterwards separately.
> Tested on rev 192142 with 'make all' and
> 'make -k check-gcc RUNTESTFLAGS="sh.exp --target_board=sh-sim
> \{-m2/-ml,-m2/-mb,-m2a/-mb,-m4
The following patch fixes a bug found on x86-64 in a big program with
some options combination.
The patch was successfully bootstrapped on x86/x86-64. The patch
changes the code for about 30% of SPEC2000 tests. But these changes are
quite small (in code size and performance point of view)
On 12-10-05 5:53 PM, Steven Bosscher wrote:
On Thu, Oct 4, 2012 at 2:59 PM, Steven Bosscher wrote:
On Thu, Oct 4, 2012 at 1:30 PM, Richard Guenther
wrote:
Isn't _REVERSE vs. non-_RESERVE still kind-of "random" order?
Not at this stage. For cfglayout mode I would answer yes, but IRA/LRA
opera
Hi,
This patch fixes PR54826. When lowering the gimple, the block for call
arg also need to be reset.
Bootstrapped and passed gcc regression test on x86.
Okay for trunk?
Thanks,
Dehao
2012-10-05 Dehao Chen
* gimple-low.c (lower_stmt): Set the block for call args.
Index: gcc/gimple-
On 10/06/2012 02:33 AM, Joe Seymour wrote:
I'm seeing tr2/headers/all.cc fail in the libstdc++ testsuite:
In file included from
src/gcc-mainline/libstdc++-v3/testsuite/tr2/headers/all.cc:22:0:
/scratch/jseymour/mainline/i686-pc-linux-gnu/install/opt/codesourcery/include/c++/4.8.0/tr2/dynamic_bit
The 32-bit DImode patterns for and and friends are not split on parisc.
Based on inspection of the assembly code generated for gcc.dg/lower-subreg-1.c,
it better if we let lower subreg split the DImode objects on parisc and
remove the current DImode patterns.
Tested on hppa-unknown-linux-gnu and c
I'm seeing tr2/headers/all.cc fail in the libstdc++ testsuite:
In file included from
src/gcc-mainline/libstdc++-v3/testsuite/tr2/headers/all.cc:22:0:
/scratch/jseymour/mainline/i686-pc-linux-gnu/install/opt/codesourcery/include/c++/4.8.0/tr2/dynamic_bitset:42:27:
fatal error: cxxabi_forced.h: No s
On Oct 5, 2012, at 6:05 PM, Cary Coutant wrote:
>> There certainly is a fair amount of code in dwarf2read.c in gdb to handle
>> DW_AT_declaration and do things differently for declarations.
>>
>> Should I rework this patch to use that mechanism instead? If so, how? If
>> the class is marked
> -Original Message-
> From: Georg-Johann Lay
> Sent: Friday, October 05, 2012 8:30 AM
> To: gcc-patches@gcc.gnu.org
> Cc: Denis Chertykov; Weddington, Eric
> Subject: [Patch,avr]: Fix PR54815
>
> avr-gcc compiles code like
>
> void f (int, int);
>
> void f_or (int x)
> {
> f (x, x
On Fri, 5 Oct 2012, Kenneth Zadeck wrote:
> +# define HOST_HALF_WIDE_INT_PRINT "h"
This may cause problems on hosts not supporting %hd (MinGW?), and there's
no real need for using "h" here given the promotion of short to int; you
can just use "" (rather than e.g. needing special handling in xm-
On Fri, Oct 5, 2012 at 12:13 PM, Andrew Pinski wrote:
> On Fri, Oct 5, 2012 at 7:32 AM, Pavel Chupin wrote:
>> I can't configure libstdc++ separately. To reproduce:
>>
>> mkdir BUILD
>> cd BUILD
>> ../libstdc++-v3/configure
>>
>> Error:
>> make: *** No rule to make target
>> `/users/pvchupin/andr
On Fri, Oct 5, 2012 at 3:50 PM, Jason Merrill wrote:
> On 10/05/2012 05:57 PM, Sriraman Tallam wrote:
>>
>> In general, the dispatcher is always necessary since it is not known
>> what function version will be called at compile time. This is true
>> whether it is a direct or an indirect call.
>
>
On Sat, 15 Sep 2012, Ian Lance Taylor wrote:
>> 2012-09-02 Mark Kettenis
>>
>> * config.host (*-*-openbsd*): Add t-eh-dw2-dip to tmake_file.
>> * unwind-dw2-fde-dip.c: Don't include on OpenBSD.
>> (USE_PT_GNU_EH_FRAME): Define for OpenBSD.
>> (ElfW): Likewise.
>
On Sat, 6 Oct 2012, Mark Kettenis wrote:
> Fortunately OpenBSD releases are completely predictable: two releases
> per year, and the version number gets increased by 0.1 every release.
> And it is highly unlikely that this policy will ever be changed. So
> since 5.2 will be released on November 1s
On Fri, Oct 5, 2012 at 6:53 PM, Diego Novillo wrote:
> Harshit, why didn't you propose this patch for trunk? Why should we
> make it a google-local patch?
In the meantime, let's put it in the google branches. Please make
sure that you ping the upstream patch. It will need more testing than
jus
On Fri, 5 Oct 2012, Jason Merrill wrote:
On 09/21/2012 02:32 PM, Marc Glisse wrote:
+ gcc_assert (TREE_CODE (type0) == VECTOR_TYPE
+ || TREE_CODE (type1) == VECTOR_TYPE);
+ switch (code)
+{
+ case RSHIFT_EXPR:
+ case LSHIFT_EXPR:
+ if (TREE_CODE (type0) == INTE
Harshit, why didn't you propose this patch for trunk? Why should we
make it a google-local patch?
Diego.
On Fri, Sep 28, 2012 at 5:24 AM, Harshit Chopra wrote:
> commit fc3a55ccec9bc770c79f8a221f5abd397befc8f6
> Author: Harshit Chopra
> Date: Thu Sep 20 17:49:59 2012 -0700
>
> Instead o
Harshit, why didn't you propose this patch for trunk? Why should we
make it a google-local patch?
Diego.
On Fri, Sep 28, 2012 at 5:24 AM, Harshit Chopra wrote:
> commit fc3a55ccec9bc770c79f8a221f5abd397befc8f6
> Author: Harshit Chopra
> Date: Thu Sep 20 17:49:59 2012 -0700
>
> Instead o
On Wed, Oct 3, 2012 at 6:46 PM, Lawrence Crowl wrote:
> On 10/2/12, Richard Guenther wrote:
>> You are changing a hashtable used by fold checking, did you test
>> with fold checking enabled?
>
> One small thinko fixed. Patch passes tests.
>
>> The cfg.c, dse.c and hash-table.h parts are ok for
On 10/05/2012 05:57 PM, Sriraman Tallam wrote:
In general, the dispatcher is always necessary since it is not known
what function version will be called at compile time. This is true
whether it is a direct or an indirect call.
So you want to compile with lowest common denominator flags and then
This patch adds machinery to genmodes.c so that largest possible sizes
of various data structures can be determined at gcc build time. These
functions create 3 symbols that are available in insn-modes.h:
MAX_BITSIZE_MODE_INT - the bitsize of the largest int.
MAX_BITSIZE_MODE_PARTIAL_INT - the b
> Date: Fri, 5 Oct 2012 23:41:28 +0200 (CEST)
> From: Gerald Pfeifer
>
> On Sat, 15 Sep 2012, Mark Kettenis wrote:
> > Index: config.gcc
> > ===
> > --- config.gcc (revision 191120)
> > +++ config.gcc (working copy)
> > @@
On Fri, Oct 5, 2012 at 6:08 PM, Lawrence Crowl wrote:
>> For many people the time to compile (almost) empty file is very
>> important, we are already bad about that right now, initializing
>> too much stuff dynamically is going to make it worse.
>
> So far, we are looking at dynamic initializatio
>
>>
>> > +
>> > + default_edge->count = default_count;
>> > + if (count)
>> > +{
>> > + edge e;
>> > + edge_iterator ei;
>> > + FOR_EACH_EDGE (e, ei, stmt_bb->succs)
>> > +e->probability = e->count * REG_BR_PROB_BASE / count;
>> > +}
>>
>> Hmm, this updates origina
this patch adds two groups of things to hwint.h that are needed for
wide-int.[ch].
A new data type, the HOST_HALF_WIDE_INT (and all of it related
macros). This type is defined to be exactly 1/2 the width of a
HOST_WIDE_INT. This is used by the multiplication and division routines
in wide-i
On 10/5/12, Jakub Jelinek wrote:
> On Fri, Oct 05, 2012 at 01:59:18PM -0700, Lawrence Crowl wrote:
> > With the constructor, you don't have to remember and you don't
> > have to type more. If you have a variable, you know that it is
> > properly initialized.
>
> But we really don't want hundreds
> There certainly is a fair amount of code in dwarf2read.c in gdb to handle
> DW_AT_declaration and do things differently for declarations.
>
> Should I rework this patch to use that mechanism instead? If so, how? If
> the class is marked only by prune_unused_types_mark visiting it as a parent,
On Fri, Oct 5, 2012 at 11:50 PM, Lawrence Crowl wrote:
> If no one cares about these time reports, then I will gladly stop
> spending the effort to make them.
It's not that no-one cases, I think, but the mathematics don't have to
be so complicated. Just showing or saying there's no significant
co
On Fri, Oct 5, 2012 at 10:43 AM, Jason Merrill wrote:
> On 08/24/2012 08:34 PM, Sriraman Tallam wrote:
>>
>> + /* If the address of a multiversioned function dispatcher is taken,
>> + generate the body to dispatch the right function at run-time. This
>>
>> + is needed as the address can
On Thu, Oct 4, 2012 at 2:59 PM, Steven Bosscher wrote:
> On Thu, Oct 4, 2012 at 1:30 PM, Richard Guenther
> wrote:
>> Isn't _REVERSE vs. non-_RESERVE still kind-of "random" order?
>
> Not at this stage. For cfglayout mode I would answer yes, but IRA/LRA
> operates in cfgrtl mode, so the sequence
On 10/5/12, Diego Novillo wrote:
> On Oct 5, 2012 Richard Guenther wrote:
> > Sorry, that wasn't intended. I question these numbers because
> > unless you bootstrap say 100 times the noise in bootstrap
> > speed is way too high to make such claims. Of course critical
> > information is missing:
On 09/21/2012 02:32 PM, Marc Glisse wrote:
+ gcc_assert (TREE_CODE (type0) == VECTOR_TYPE
+ || TREE_CODE (type1) == VECTOR_TYPE);
+ switch (code)
+{
+ case RSHIFT_EXPR:
+ case LSHIFT_EXPR:
+ if (TREE_CODE (type0) == INTEGER_TYPE
+ && TREE_CODE (TREE_TYP
On Fri, 2012-10-05 at 21:55 +0900, Kaz Kojima wrote:
> Oleg Endo wrote:
> > Do you mean something like the attached patch as a preparation step?
> > (checked with 'make all')
>
> Yes. The patch is OK with removing the first line of the ChangeLog
> entry for PR number.
Done.
The attached patch i
On Sat, 15 Sep 2012, Mark Kettenis wrote:
> Index: config.gcc
> ===
> --- config.gcc(revision 191120)
> +++ config.gcc(working copy)
> @@ -708,6 +708,11 @@
> *-*-openbsd2.*|*-*-openbsd3.[012])
>tm_defines="
> This fixes PR54811 for me - still Ada LTO bootstrap fails for me
> with cgraph verification ICEs.
Thanks for the fix! Yes, I also get the cgraph verification ICEs now, but
that's kind of a progress if you ask me. :-) I'll look into them once the
dust has settled on Jan's recent changes in th
On Oct 5, 2012, at 2:43 PM, Cary Coutant wrote:
>>> It seems to me that there are cases where we just want to emit the
>>> class for the context info (like a namespace, which doesn't have to be
>>> complete everywhere). Is there a way to tell the debugger that this
>>> class declaration is incomp
More testsuite changes, tracking GSOC work.
tested x86/linux
-benjamin2012-10-05 Benjamin Kosnik
* testsuite/28_regex/algorithms/match: Rename to...
* testsuite/28_regex/algorithms/regex_match: ...this.
* testsuite/28_regex/basic_regex/regex.cc: Rename to...
* testsuite/28_regex/basic_re
On Fri, Oct 05, 2012 at 01:59:18PM -0700, Lawrence Crowl wrote:
> With the constructor, you don't have to remember and you don't
> have to type more. If you have a variable, you know that it is
> properly initialized.
But we really don't want hundreds or thousands of dynamic constructors
for glob
The attached change fixes a reload failure encountered testing another
change. It occurred for a constant that didn't match the "J" constraint.
The change makes the HImode pattern the same as the corresponding SImode
pattern.
Tested on hppa-unknown-linux-gn, hppa2.0w-hp-hpux11.11 and
hppa64-hp-hp
thanks. That will be helpful.
David
On Fri, Oct 5, 2012 at 2:09 PM, Dehao Chen wrote:
> Sure, I'll add a detailed documentation in a gcc wiki page.
>
> Dehao
>
> On Fri, Oct 5, 2012 at 2:01 PM, Xinliang David Li wrote:
>> Dehao, the file auto-profile.c has some high level description of
>> aFD
Could you please review the patch at
http://www.mail-archive.com/gcc-patches@gcc.gnu.org/msg40961.html?
On 10/3/12, Lawrence Crowl wrote:
> Sorry, one more time with the right file contents.
>
> On 10/2/12, Richard Guenther wrote:
>> You are changing a hashtable used by fold checking, did you te
On 10/5/12, Richard Guenther wrote:
> On Thu, 4 Oct 2012, Lawrence Crowl wrote:
>> On 10/4/12, Richard Guenther wrote:
>> > On Tue, 2 Oct 2012, Lawrence Crowl wrote:
>> >> On 10/2/12, Richard Guenther wrote:
>> >> > On Mon, 1 Oct 2012, Lawrence Crowl wrote:
>> >> > > Change more non-GTY hash tab
On Fri, Sep 14, 2012 at 2:26 PM, Richard Guenther wrote:
> If you can figure out a better name for the function we should
> probably move it to cfganal.c
It looks like my previous e-mail about this appears to have gone got
somehow, so retry:
Your my_rev_post_order_compute is simply inverted_post
Ping
http://gcc.gnu.org/ml/gcc-patches/2012-09/msg01557.html
On Fri, 21 Sep 2012, Marc Glisse wrote:
Hello,
this patch adds mixed scalar-vector operations to the C++ front-end. It also
adds a few operators to the C front-end (comparisons in particular). This
patch is mostly an import from t
On Wed, Oct 03, 2012 at 09:42:05PM -0400, David Edelsohn wrote:
> @@ -1115,7 +1118,8 @@ static const struct attribute_spec rs600
>{ NULL,0, 0, false, false, false, NULL, false }
> };
>
> -#ifndef MASK_STRICT_ALIGN
> +#ifndef OPTION_MASK_STRICT_ALIGN
> +#define OPTION_MASK_STRICT_ALI
After commiting the last SH changes updates for 4.8 I was kindly
informed that the br tags were left open. I've committed the attached
fix as obvious.
Cheers,
Oleg
? www_4_8_sh_changes_2_close_br.patch
Index: htdocs/gcc-4.8/changes.html
On Fri, Oct 5, 2012 at 7:32 AM, Pavel Chupin wrote:
> I can't configure libstdc++ separately. To reproduce:
>
> mkdir BUILD
> cd BUILD
> ../libstdc++-v3/configure
>
> Error:
> make: *** No rule to make target
> `/users/pvchupin/android/toolchain/gcc/gcc-4.8/BUILD/../libgcc/gthr-default.h',
> neede
On Oct 5, 2012, at 2:43 PM, Cary Coutant wrote:
>>> It seems to me that there are cases where we just want to emit the
>>> class for the context info (like a namespace, which doesn't have to be
>>> complete everywhere). Is there a way to tell the debugger that this
>>> class declaration is incomp
I've rewritten my rs6000 infrastructure patch to allow more target switches, so
that it does not change the flags currently in the target_flags. This should
reduce the number of changes due to TARGET_ being called OPTION_, and
MASK_ being called OPTION_MASK_. In addition, since it might be
possib
>> It seems to me that there are cases where we just want to emit the
>> class for the context info (like a namespace, which doesn't have to be
>> complete everywhere). Is there a way to tell the debugger that this
>> class declaration is incomplete and that it should look elsewhere for
>> a full d
On 10/04/2012 08:26 PM, Cary Coutant wrote:
It seems to me that there are cases where we just want to emit the
class for the context info (like a namespace, which doesn't have to be
complete everywhere). Is there a way to tell the debugger that this
class declaration is incomplete and that it sho
On 08/24/2012 08:34 PM, Sriraman Tallam wrote:
+ /* For function versions, their parms and types match
+but they are not duplicates. Record function versions
+as and when they are found. */
+ if (TREE_CODE (fn) == FUNCTION_DECL
+ && TREE_CODE
On Wed, Oct 03, 2012 at 09:42:05PM -0400, David Edelsohn wrote:
> @@ -1115,7 +1118,8 @@ static const struct attribute_spec rs600
>{ NULL,0, 0, false, false, false, NULL, false }
> };
>
> -#ifndef MASK_STRICT_ALIGN
> +#ifndef OPTION_MASK_STRICT_ALIGN
> +#define OPTION_MASK_STRICT_ALI
On Thu, Oct 04, 2012 at 06:33:33PM +0200, Gunther Nikl wrote:
> Michael Meissner schrieb:
> > On Tue, Oct 02, 2012 at 10:13:25AM +0200, Gunther Nikl wrote:
> >> Michael Meissner wrote:
> >>> Segher Boessenkool asked me on IRC to break out the fix in the last
> >>> change.
> >>> This patch is just
On 10/05/2012 01:43 PM, Jason Merrill wrote:
On 08/24/2012 08:34 PM, Sriraman Tallam wrote:
+ /* If the address of a multiversioned function dispatcher is taken,
+ generate the body to dispatch the right function at run-time. This
+ is needed as the address can be used to do an indirec
On 10/05/2012 01:29 PM, Richard Sandiford wrote:
+/* This is the maximal size of the buffer needed for dump. */
>>+const int MAX = (MAX_BITSIZE_MODE_ANY_INT / 4
>>++ MAX_BITSIZE_MODE_ANY_INT / HOST_BITS_PER_WIDE_INT + 32);
>>
>>I'd prefer a target macro for this, not anything ba
> So given the comments, is this patch now ok to commit?
Yes, this is OK. Thanks for doing the extra testing! (I also ran a
quick test with -fdebug-types-section, just to make sure.)
-cary
On 08/24/2012 08:34 PM, Sriraman Tallam wrote:
+ /* If the address of a multiversioned function dispatcher is taken,
+ generate the body to dispatch the right function at run-time. This
+ is needed as the address can be used to do an indirect call. */
It seems to me that you don't ne
Hi,
remove some cruft noticed by Marc (and myself). Sanity checked
x86_64-linux, committed to mainline.
Thanks,
Paolo.
//
2012-10-05 Paolo Carlini
* include/c_global/cstdlib: Remove redundant pasto code protected
by __GXX_EXPERIMENTAL_CXX0X__.
* inc
On 10/05/2012 04:29 AM, Richard Guenther wrote:
Or if we have the extra indirection via a reference anyway, we could
have a pointer TLS variable (NULL initialized) that on the first access
will trap where in a trap handler we could then perform initialization
and setup of that pointer.
Interest
Kenneth Zadeck writes:
>> Or do you want to optimize encoding like for CONST_INT (and unlike
>> CONST_DOUBLE)? I doubt the above packs nicely into rtx_def?
>> A general issue of it though - we waste 32bits on 64bit hosts in
>> rtx_def between the bits and the union. Perfect place for num_elem
>>
On 10/05/2012 04:41 AM, Jakub Jelinek wrote:
Unfortunately, that penalty is not only for thread_local vars with
ctors/dtors. There is some penalty even for using
extern thread_local int i;
int foo (void)
{
return i;
}
(as compared to extern __thread int i;), because we have to at least check
On Oct 5, 2012, at 11:34 AM,
wrote:
>
> On Oct 5, 2012, at 4:16 AM, Jakub Jelinek wrote:
>
>> On Thu, Oct 04, 2012 at 05:26:11PM -0700, Cary Coutant wrote:
Index: gcc/testsuite/g++.dg/debug/dwarf2/localclass1.C
===
>>>
Hi Jason, Sri has addressed the comments you had on FE part. Can you
take a look if it is ok? Stage-1 is going to be closed soon, and we
hope to get this major feature in 4.8.
thanks,
David
On Tue, Sep 18, 2012 at 9:29 AM, Sriraman Tallam wrote:
> Ping.
>
> On Fri, Aug 24, 2012 at 5:34 PM,
On Fri, 5 Oct 2012, Iain Buclaw wrote:
> Would the best approach be to move all handle_* functions and any
> helper functions into a new source file that can be shared between
> frontends, and define two new frontend hooks,
> LANG_HOOK_ATTRIBUTE_TABLE and LANG_HOOK_FORMAT_ATTRIBUTE_TABLE ?
I don'
richard s,
there are two comments that i deferred to you. that have the word
richard in them,
richi,
thank, i will start doing this now.
kenny
On 10/05/2012 09:49 AM, Richard Guenther wrote:
On Fri, Oct 5, 2012 at 2:39 PM, Richard Guenther
wrote:
Ok, I see where you are going. Let me loo
As already discussed, this patch removes the -mshort-calls command option from
avr-gcc.
Ok to apply?
If the change is on order, changes to wwwdocs will follow, i.e. deprecate the
option in 4.7 and tell it is removed in the 4.8 caveats.
Johann
* doc/invoke.texi (AVR Options): Remove -ms
Hi,
This patch refactors the initialisation code for the Advanced
SIMD builtins under the AArch64 target. The patch has been
regression tested on aarch64-none-elf.
OK for aarch64-branch?
(If yes, someone will have to commit this for me as I do not
have commit rights)
Thanks,
James Greenhalgh
On Oct 5, 2012, at 4:16 AM, Jakub Jelinek wrote:
> On Thu, Oct 04, 2012 at 05:26:11PM -0700, Cary Coutant wrote:
>>> Index: gcc/testsuite/g++.dg/debug/dwarf2/localclass1.C
>>> ===
>>> --- gcc/testsuite/g++.dg/debug/dwarf2/localclass1
Hi,
I'm adding the testcase and closing the PR. Tested x86_64-linux.
Thanks,
Paolo.
/
2012-10-05 Paolo Carlini
PR c++/50893
* g++.dg/cpp0x/defaulted38.C: New.
Index: g++.dg/cpp0x/defaulted38.C
==
[I am still a little confused, sorry for the long email...]
On Tue, 2 Oct 2012, Richard Guenther wrote:
+ if (TREE_CODE (op0) == VECTOR_CST && TREE_CODE (op1) == VECTOR_CST)
+{
+ int count = VECTOR_CST_NELTS (op0);
+ tree *elts = XALLOCAVEC (tree, count);
+ gcc_assert (TREE
Kenneth Zadeck writes:
> i will convert ppc if that is what it takes. david's office is 4 isles
> away and mike has a lot of experience on ppc also. (this is unless
> richard is willing to do mips one afternoon.)
'Fraid MIPS is very boring as far as this is concerned. MIPS sets
need_64bit_h
On 10/05/2012 10:36 AM, Richard Guenther wrote:
On Fri, Oct 5, 2012 at 4:14 PM, Richard Sandiford
wrote:
Richard Guenther writes:
On Fri, Oct 5, 2012 at 3:18 PM, Richard Sandiford
wrote:
Richard Sandiford writes:
How is CONST_WIDE_INT variable size?
It's just the usual trailing variable
On Fri, Oct 5, 2012 at 4:14 PM, Richard Sandiford
wrote:
> Richard Guenther writes:
>> On Fri, Oct 5, 2012 at 3:18 PM, Richard Sandiford
>> wrote:
>>> Richard Sandiford writes:
>>> How is CONST_WIDE_INT variable size?
>>
>> It's just the usual trailing variable-length array thing.
>
I can't configure libstdc++ separately. To reproduce:
mkdir BUILD
cd BUILD
../libstdc++-v3/configure
Error:
make: *** No rule to make target
`/users/pvchupin/android/toolchain/gcc/gcc-4.8/BUILD/../libgcc/gthr-default.h',
needed by `bits/gthr-default.h'. Stop.
See fix attached.
Ok for trunk and
This patch fixes the handling of type invariants when they are specified on
the completion of a private type, and when public expression functions return
a type with invariants.
Running the following:
gnatmake -q -gnat12a main
main
must yield:
OK
---
with Ada.Assertions; use Ada.Asser
This patch changes the default overflow checking mode to suppressed
in the case of targets that support backend divide and overflow
checking as indicated by the appropriate flags in Targparm being
set. There are no such targets currently, so this change has no
effect and no test is needed.
Tested
avr-gcc compiles code like
void f (int, int);
void f_or (int x)
{
f (x, x | 42);
}
to
f_or:
ldi r22,lo8(42) ; 15*movhi/5[length = 2]
ldi r23,0
or r22,r24 ; 6iorhi3/1[length = 2]
or r23,r25
rjmp f
but this is more efficient:
f_or:
movw
The code for statically computing bounds for division and exponent was
incorrect in some cases, possibly leading to wrong results when -gnato2 or
-gnato3 was used. The code for mod and rem was not using optimal bounds, with
possible minor consequences on efficiency of the generated code with -gnato
This patch fixes a subtle race condition where the master task and the child
task may free the child task concurrently. Found by code reading.
Tested on x86_64-pc-linux-gnu, committed on trunk
2012-10-05 Arnaud Charlet
* s-tassta.adb: Update comments.
(Vulnerable_Complete_Mas
This patch ensures that Deallocate_Subpool is invoked on each subpool when the
owner pool_with_subpools is finalized.
-- Source --
-- gc_spool.ads
with System.Storage_Pools.Subpools;
private with System.Storage_Elements;
package GC_SPool is
use System;
use Syste
On Fri, Oct 05, 2012 at 03:59:55PM +0200, Richard Guenther wrote:
> I don't think we want to rely on that ... so just keep the push/pop_cfun.
Ok, so this is what I'm retesting (basically just comments added and the two
lines (subcode and set) swapped:
2012-10-05 Jakub Jelinek
PR debug
Richard Guenther writes:
> On Fri, Oct 5, 2012 at 3:18 PM, Richard Sandiford
> wrote:
>> Richard Sandiford writes:
>> How is CONST_WIDE_INT variable size?
>
> It's just the usual trailing variable-length array thing.
Good. Do you get rid of CONST_DOUBLE (for integers) at t
On Fri, Oct 5, 2012 at 2:43 PM, Diego Novillo wrote:
> Because (...) there has been so much
> negative pressure on our work, that we sometimes try to find some
> benefit when reality may provide neutral results.
When people say "your work sucks", they probably don't mean to apply
negative pressure
On Fri, Oct 5, 2012 at 2:49 PM, Jakub Jelinek wrote:
> On Fri, Oct 05, 2012 at 02:20:13PM +0200, Richard Guenther wrote:
>> The following could use a comment on what you are doing ...
>
> Will add something.
>
>> > + if (args_to_skip)
>> > +for (parm = DECL_ARGUMENTS (current_function_decl),
On Fri, Oct 5, 2012 at 2:43 PM, Diego Novillo wrote:
> Because (...) there has been so much
> negative pressure on our work, that we sometimes try to find some
> benefit when reality may provide neutral results.
When people say "your work sucks", they probably don't mean to apply
negative pressure
On Fri, Oct 5, 2012 at 3:18 PM, Richard Sandiford
wrote:
> Richard Sandiford writes:
> How is CONST_WIDE_INT variable size?
It's just the usual trailing variable-length array thing.
>>>
>>> Good. Do you get rid of CONST_DOUBLE (for integers) at the same time?
>>
>> Yeah. I initial
Hi,
this is the udpated patch I comitted after testing. I suppose we will need to
find way
to make SOC smaller for simple loops - it is way too overestimated currently.
* tree-vectorizer.h (vect_estimate_min_profitable_iters): Remove.
* tree-vect-loop.c (vect_estimate_min_profita
On Fri, Oct 5, 2012 at 4:21 AM, Rainer Orth
wrote:
>
> I think this should work. The only complication might be that users of
> DEFAULT_PROCESS_FILENAME need to include on Solaris to get
> the declaration of getexecname(). Given that this header is already
> used unconditionally, its inclusion
On Fri, Oct 5, 2012 at 2:39 PM, Richard Guenther
wrote:
>
> Ok, I see where you are going. Let me look at the patch again.
* The introduction and use of CONST_SCALAR_INT_P could be split out
(obvious and good)
* DEF_RTL_EXPR(CONST_WIDE_INT, "const_wide_int", "", RTX_CONST_OBJ)
defining that
On 10/02/2012 08:56 AM, Jakub Jelinek wrote:
Perhaps maybe_constant_value already does all that integral_constant_value
does, but I'm not sure about it.
It does, you can remove that call.
OK with that change.
Jason
On Fri, Oct 5, 2012 at 3:21 AM, Rainer Orth
wrote:
> With the strnlen-related Go failures now gone on Solaris, another issue
> crept up: several tests were failing like this:
>
> throw: /proc/self/exe
>
> goroutine 4 [running]:
> panic during panic
> FAIL: log
>
> This is due to the hardcoded use
On Fri, Oct 05, 2012 at 02:49:07PM +0200, Jakub Jelinek wrote:
> I believe that gsi_insert_before in another function
> isn't going to work well.
> E.g. update_modified_stmt starts with
> if (!ssa_operands_active (cfun))
> return;
>
> Or is it ok to use gsi_insert_before_without_update and e
Richard Sandiford writes:
How is CONST_WIDE_INT variable size?
>>>
>>> It's just the usual trailing variable-length array thing.
>>
>> Good. Do you get rid of CONST_DOUBLE (for integers) at the same time?
>
> Yeah. I initially thought it might be OK to keep them and have
> CONST_INT, intege
richi,
let me address a bunch of issues that are randomly spread thru the thread.
1) unlike the double int's current relationship to int cst, we do not
currently wrap a wide-int into an CONST_WIDE_INT nor (in the patch that
you have not seen) do we wrap a wide-int into the int cst.wide-int
Richard Guenther writes:
> On Fri, Oct 5, 2012 at 2:26 PM, Richard Sandiford
> wrote:
>> Richard Guenther writes:
>>> On Fri, Oct 5, 2012 at 1:24 PM, Richard Sandiford
>>> wrote:
Richard Guenther writes:
> The issue is that unlike RTL where we "construct" double-ints from
> CONST_
Oleg Endo wrote:
> Do you mean something like the attached patch as a preparation step?
> (checked with 'make all')
Yes. The patch is OK with removing the first line of the ChangeLog
entry for PR number.
Regards,
kaz
1 - 100 of 136 matches
Mail list logo
