Hello Richard,
On Apr 25, 2012, at 00:06 , Richard Sandiford wrote:
> STRICT_LOW_PART is OK too.
Ah, right.
> Would be nice to use a single function that knows about the extra
> contraints here. Maybe something like the attached?
>
> I'm deliberately requiring the SET to the first rtx in the
On Wed, Apr 25, 2012 at 11:29 PM, Olivier Hainque wrote:
> Thanks for your feedback Richard,
>
> On Apr 25, 2012, at 16:16 , Richard Guenther wrote:
>> I think much better would be to simply disallow any toplevel
>> VIEW_CONVERT_EXPR of BLKmode,
>
>> Does that fix your problems, too? If so I pref
Hello,
This is a followup submission for a change first proposed at
http://gcc.gnu.org/ml/gcc-patches/2007-06/msg00767.html
The exact same misbehavior is still observable on mainline, exposed by a
slightly adjusted version of the original testcase (part of the patch):
bld/gcc $ gcc -B./ -c -DS
When front-end layout is enabled (such as for non-GCC targets like AAMP),
the RM_Size value of variant records with a static-sized single variant is
computed incorrectly as storage units rather than bits (which can cause
incorrect code for reading and writing objects). This case is now tested for
a
This patch corrects the mechanism which determines whether a construct appears
at the library level. This in turn allows for proper detection of cases where
a Finalize_Storage_Only object appears in a nested scope and requires
finalization.
-- Source --
-- main.adb
wit
If the subtype indication for a formal parameter carries a non null indicator,
the compiler creates an internal subtype for it. This subtype must be frozen
after its designated type, so its freezing must be delayed, but only if the
designated type is itself not frozen yet.
The following must compi
A new package Clean with attributes Object_Artifact_Extensions and
Source_Artifact_Extensions is created in project files.
Tested on x86_64-pc-linux-gnu, committed on trunk
2012-04-26 Vincent Celier
* prj-attr.adb: New package Clean with attributes
Object_Artifact_Extensions a
On Wed, 25 Apr 2012, Aldy Hernandez wrote:
> On 04/25/12 06:45, Richard Guenther wrote:
> > On Tue, Apr 24, 2012 at 7:43 PM, Aldy Hernandez wrote:
> > > On 04/13/12 03:46, Richard Guenther wrote:
> > > >
> > > > On Fri, Apr 13, 2012 at 12:11 AM, Aldy Hernandez
> > > > wrote:
>
> > + /* ?? Perha
This patch adds code to retrieve the index type of a string literal. Since
string literals do not use attribute First_Index, the proper index type is
obtained from their low bound. No changes in compiler behavior.
Tested on x86_64-pc-linux-gnu, committed on trunk
2012-04-26 Hristian Kirtchev
This patch causes pragmas to be ignored when checking for unreachable
code, which catches additional cases not caught before, as shown by
the following example, compiled with -gnatwe.w -gnatld7
1. with Ada.Text_IO; use Ada.Text_IO;
2. procedure Main4 is
3. begin
4.loop
This patch modifies System.Finalization_Root to avoid a dependency on streams.
Type Root_Controlled is now abstract tagged null record.
Tested on x86_64-pc-linux-gnu, committed on trunk
2012-04-26 Hristian Kirtchev
* s-finroo.ads: Remove with clause for
Ada.Streams. Type Root_
Function to_host_path_spec is removed (was unused).
No functional change.
Tested on x86_64-pc-linux-gnu, committed on trunk
2012-04-26 Tristan Gingold
* adaint.c (to_host_path_spec): Removed (unused).
Minor reformatting.
Index: adaint.c
===
Tested on x86_64-pc-linux-gnu, committed on trunk
2012-04-26 Vasiliy Fofanov
* a-stzunb-shared.adb, gnat_ugn.texi, a-strunb-shared.adb,
a-strunb-shared.ads, uintp.adb, a-stwiun-shared.adb,
a-stwiun-shared.ads, a-cbhama.ads, vms_data.ads, a-cobove.adb,
a-convec.a
Jason Merrill a écrit:
> It seems like we'll do this for every line in the header, which could
> lead to a lot of leaked memory. Instead, we should canonicalize when
> setting ORDINARY_MAP_FILE_NAME.
[...]
Manuel López-Ibáñez a écrit:
> On 21 April 2012 14:56, Jason Merrill wrote:
>> It see
On 26 April 2012 11:12, Dodji Seketeli wrote:
>
> So maybe it'd be better to canonicalize the _cpp_file::path when it's
> first build? One drawback of that approach would be that
> _cpp_file::path will then permanently loose the information about the
> current directory, that is indirectly encoded
On Wed, Apr 25, 2012 at 11:56 PM, Tom de Vries wrote:
> On 25/04/12 11:57, Richard Guenther wrote:
>
>> Hmm. I'm not sure we can conclude that they have the same value!
>>
>> +int bar (int);
>> +void baz (int);
>> +void bla (int);
>> +
>> +void
>> +foo (in
On 26 April 2012 12:12, Dodji Seketeli wrote:
>
> So maybe it'd be better to canonicalize the _cpp_file::path when it's
> first build? One drawback of that approach would be that
> _cpp_file::path will then permanently loose the information about the
> current directory, that is indirectly encode
Hello all, and sorry for getting into this a bit late. I have a
question concerning this patch:
> +++ cp/call.c (working copy)
[...]
> +static const char *
> +op_error_string (const char *err_msg, int ntypes, bool match)
> +{
> + const char *msg;
> +
> + const char *msgt = concat (match ? G_(
Diego Novillo a écrit:
> On Mon, Apr 16, 2012 at 10:23, Richard Guenther wrote:
[...]
>> I'm not sure PPH will replace PCH - will it?
[...]
> In theory, it may. But it's not in the immediate plans.
>
> The big thing that PCH has going for it is load speed. PCH is almost
> instantaneous, PP
Hi,
I would like to remove old code handling aliases. The catch is that in new
representation I am not accepting aliases from functions to variables or
vice versa. It is not impossible to add, but it is another weird side
case (for exmaple you can then call variable).
I know of only one case whe
On Wed, Apr 25, 2012 at 8:06 AM, Maxim Kuvyrkov wrote:
> On 18/04/2012, at 9:17 PM, Richard Guenther wrote:
>
>> On Wed, Apr 18, 2012 at 4:15 AM, Maxim Kuvyrkov
>> wrote:
>>> Steven,
>>> J"orn,
>>>
>>> I am looking into fixing performance regression on EEMBC's bitmnp01, and a
>>> version of you
The current code in reg_fits_class_p appears to be incorrect; since
offset may be negative, it's necessary to check both ends of the range
otherwise an array overrun or underrun may occur when calling
in_hard_reg_set_p. in_hard_reg_set_p should also be checked for each
register in the range of
We're creating new pseudos, and while we're resizing some data
structures, we aren't doing it for everything.
Bootstrapped and tested on i686-linux, committed as obvious.
Bernd
Index: ChangeLog
===
--- ChangeLog (revision 186874)
On Thu, Apr 26, 2012 at 5:34 AM, Dodji Seketeli wrote:
> Hello all, and sorry for getting into this a bit late. I have a
> question concerning this patch:
>
>> +++ cp/call.c (working copy)
>
> [...]
>
>> +static const char *
>> +op_error_string (const char *err_msg, int ntypes, bool match)
>> +{
On Wed, Apr 25, 2012 at 3:58 PM, Marc Glisse wrote:
> On Wed, 25 Apr 2012, Richard Guenther wrote:
>
>> On Wed, Apr 25, 2012 at 10:12 AM, Marc Glisse
>> wrote:
>>>
>>> Hello,
>>>
>>> a conversion like int->double->int is just the identity, as long as
>>> double
>>> is big enough to represent all
On Wed, Apr 25, 2012 at 11:09 PM, Marc Glisse wrote:
> On Wed, 25 Apr 2012, H.J. Lu wrote:
>
>> You can download an AVX2 emulator (SDE) from
>>
>> http://software.intel.com/en-us/avx/
>>
>> to test AVX2 binaries.
>
>
> Hello,
>
> not related to this specific patch, but do you have a dejagnu board
> "Ollie" == Ollie Wild writes:
Ollie> 2012-04-22 Ollie Wild
Ollie> * gcc/c-family/c-common.c:
Ollie> * gcc/c-family/c-opts.c (c_common_handle_option):
Ollie> * gcc/c-family/c.opt:
Ollie> * gcc/doc/invoke.texi (struct A):
Ollie> * gcc/testsuite/g++.dg/cpp0x/Wreserved-user-defined-lit
On Thu, Apr 26, 2012 at 06:43, Dodji Seketeli wrote:
> I guess it's also worth noting one limitation of PPHs that is, if I
> believe the wiki:
>
> In essence, the only headers that can be pre-parsed are those that
> produce the same result when they are compiled in isolation or as
> part
On 4/26/12 9:35 AM, Diego Novillo wrote:
On Thu, Apr 26, 2012 at 06:43, Dodji Seketeli wrote:
I guess it's also worth noting one limitation of PPHs that is, if I
believe the wiki:
In essence, the only headers that can be pre-parsed are those that
produce the same result when they are
On Thu, 26 Apr 2012, Diego Novillo wrote:
> On 4/26/12 9:35 AM, Diego Novillo wrote:
> > On Thu, Apr 26, 2012 at 06:43, Dodji Seketeli wrote:
> >
> > > I guess it's also worth noting one limitation of PPHs that is, if I
> > > believe the wiki:
> > >
> > > In essence, the only headers that can
On 04/25/2012 10:25 AM, Paolo Carlini wrote:
Thus, the below elementary patch appears to work fine (I also double
checked that in such cases the type remains trivial). It's all there is
to it?
Unfortunately, I don't think so; there's a lot of code in the compiler
that assumes that trivial cons
On Mon, Apr 23, 2012 at 08:45:18AM -0600, Tom Tromey wrote:
> Tom> Here is a new patch for gcc.
> Tom> I still haven't updated the src side, but there's little to do there
> Tom> that isn't already done in this patch.
>
> Tom> Ok?
>
> Tom> Ping.
>
> Tom> Ping.
>
> This is the third ping.
>
> P
Hello!
Attached patch implements post-reload compare elimination pass for x86
target. The patch converts arithmetic and logic patterns (that is:
plus, minus, and, or, xor, neg, one complement) to the patterns,
recognized by post-reload cmp elimination pass, together with all
relevant splitters and
On Wed, Apr 25, 2012 at 9:14 PM, Jakub Jelinek wrote:
> We have a splitter for reg1 = reg2 & 0x, but only if regnums
> are different. But movl %edi, %edi is a cheaper variant of
> andq $0x, %rdi even with the same register and doesn't clobber
> flags, so this patch attempts to ex
The following patch has some parts that still need approval, i.e.
* Makefile.in (TEXI_GCC_FILES): Add avr-mmcu.texi.
* doc/avr-mmcu.texi: New auto-generated file.
* doc/invoke.texi (AVR Options): Include avr-mmcu.texi in order
to document all valid -mmcu= arguments.
http://gcc.gnu.org/ml/gcc-pa
On Thu, Apr 26, 2012 at 05:25:31PM +0200, Uros Bizjak wrote:
> On Wed, Apr 25, 2012 at 9:14 PM, Jakub Jelinek wrote:
>
> > We have a splitter for reg1 = reg2 & 0x, but only if regnums
> > are different. But movl %edi, %edi is a cheaper variant of
> > andq $0x, %rdi even with the
On Thu, Apr 26, 2012 at 5:28 PM, Jakub Jelinek wrote:
>> > We have a splitter for reg1 = reg2 & 0x, but only if regnums
>> > are different. But movl %edi, %edi is a cheaper variant of
>> > andq $0x, %rdi even with the same register and doesn't clobber
>> > flags, so this patch at
"Ulrich Weigand" writes:
> Richard Sandiford wrote:
>> Vladimir Makarov writes:
>> > Taking your results for S390 and ARM with Neon into account, I guess it
>> > should be included and probably made by default for these 2 targets (for
>> > sure for s390).
>>
>> OK, thanks to both of you.
>>
>
Il 23/04/2012 14:53, Gabriel Dos Reis ha scritto:
>> > I chose -Wreserved-user-defined-literal because that's the name used
>> > by Clang. It seemed better to maintain consistency between front ends
>> > than to invent a new, concise option name.
>> >
>> > Do you still want me to shorten in?
> yes
Hello,
The attached patch re-organizes some code in tree-switch-conversion.c.
All information about a GIMPLE_SWITCH is now collected by one
function, so that my switch lowering code can use the same
switch_conv_info. Bootstrapped&tested on x86_64-unknown-linux-gnu. OK
for trunk?
Ciao!
Steven
Diego Novillo a écrit:
> On Thu, Apr 26, 2012 at 06:43, Dodji Seketeli wrote:
>
>> I guess it's also worth noting one limitation of PPHs that is, if I
>> believe the wiki:
>>
>> In essence, the only headers that can be pre-parsed are those that
>> produce the same result when they are comp
Jonathan Wakely a écrit:
> On 26 April 2012 11:12, Dodji Seketeli wrote:
>>
>> So maybe it'd be better to canonicalize the _cpp_file::path when it's
>> first build? One drawback of that approach would be that
>> _cpp_file::path will then permanently loose the information about the
>> current dir
Manuel López-Ibáñez a écrit:
> Why not remove this comment and free file here with XDELETEVEC (file) ?
>
>> + canonical_path = maybe_shorter_path (path);
>> + if (canonical_path != NULL && canonical_path != path)
>> + {
>> + /* The canonical path was newly allocated. Let's free the
>>
I will not oppose adding more unrelated stuff to libiberty, but
neither will I approve it. I will let one of the other maintainers or
a global maintainer approve it.
On 26 April 2012 20:11, Dodji Seketeli wrote:
> Manuel López-Ibáñez a écrit:
>
>> Why not remove this comment and free file here with XDELETEVEC (file) ?
>>
>>> + canonical_path = maybe_shorter_path (path);
>>> + if (canonical_path != NULL && canonical_path != path)
>>> + {
>>> + /* The
On Thu, 26 Apr 2012, Richard Guenther wrote:
On Wed, Apr 25, 2012 at 3:58 PM, Marc Glisse wrote:
Here is take 2 on this patch, which seems cleaner. Bootstrapped and
regression tested.
gcc/ChangeLog
2012-04-25 Marc Glisse
PR middle-end/27139
* tree-ssa-forwprop.c (combine_co
On 04/05/2012 05:14 AM, Richard Guenther wrote:
Ok. Remember that you should use non-tree things if you can in GIMPLE
land. This probably means that both the size and the memmodel "operands"
should be
+ struct GTY(()) gimple_statement_atomic
+ {
+ /* [ WORD 1-8 ] */
+ struct gimple_state
On 4/26/12 2:15 PM, Dodji Seketeli wrote:
One more question, for my education. Does the PPH machinery detect that
a given header doesn't respect that requirement and thus refuses to
serialize it?
Yup.
Some requirements can be enforced when the image is being generated
(e.g., double inclusio
Manuel López-Ibáñez a écrit:
> On 26 April 2012 20:11, Dodji Seketeli wrote:
>> Manuel López-Ibáñez a écrit:
>>
>>> Why not remove this comment and free file here with XDELETEVEC (file) ?
>>>
+ canonical_path = maybe_shorter_path (path);
+ if (canonical_path != NULL && canonical_pat
On 26 April 2012 20:56, Dodji Seketeli wrote:
> Manuel López-Ibáñez a écrit:
>
>> On 26 April 2012 20:11, Dodji Seketeli wrote:
>>> Manuel López-Ibáñez a écrit:
>>>
Why not remove this comment and free file here with XDELETEVEC (file) ?
> + canonical_path = maybe_shorter_path (pa
On 04/05/2012 05:14 AM, Richard Guenther wrote:
+ static inline bool
+ gimple_atomic_has_fail_order (const_gimple gs)
+ {
+ return gimple_atomic_kind (gs) == GIMPLE_ATOMIC_COMPARE_EXCHANGE;
+ }
btw, these kind of predicates look superfluous to me - if they are true
exactly for one atomic kind
On Apr 26, 2012, at 3:43 AM, Dodji Seketeli wrote:
> How hard would it be to drop that limitation?
It would be a mistake to try... My take away from the work on a compile server
I worked on is that a header whose semantics don't change is critical. Just
checking this property for a header, is
On Mon, Apr 23, 2012 at 5:36 PM, Michael Hope wrote:
> 2012-04-24 Michael Hope
> Richard Earnshaw
>
> * config/arm/linux-eabi.h (GLIBC_DYNAMIC_LINKER_SOFT_FLOAT): Define.
> (GLIBC_DYNAMIC_LINKER_HARD_FLOAT): Define.
> (GLIBC_DYNAMIC_LINKER_DEFAULT): Define.
>
On Thu, Apr 26, 2012 at 01:52:31PM -0400, DJ Delorie wrote:
>
> I will not oppose adding more unrelated stuff to libiberty, but
> neither will I approve it. I will let one of the other maintainers or
> a global maintainer approve it.
The original libiberty patch is ok for trunk then.
Ja
Hi,
this is the spurious error on asm statements of the form:
error: 'asm' operand requires impossible reload
present for IA-64 on mainline and 4.7 branch. As diagnosed by Ulrich, the code
responsible for the error implicitly assumes that constraints accepting memory
operands also accept ps
On 04/26/2012 04:15 PM, Jason Merrill wrote:
On 04/25/2012 10:25 AM, Paolo Carlini wrote:
Thus, the below elementary patch appears to work fine (I also double
checked that in such cases the type remains trivial). It's all there is
to it?
Unfortunately, I don't think so; there's a lot of code in
On 27 April 2012 08:20, Carlos O'Donell wrote:
> On Mon, Apr 23, 2012 at 5:36 PM, Michael Hope wrote:
>> 2012-04-24 Michael Hope
>> Richard Earnshaw
>>
>> * config/arm/linux-eabi.h (GLIBC_DYNAMIC_LINKER_SOFT_FLOAT): Define.
>> (GLIBC_DYNAMIC_LINKER_HARD_FLOAT): Defin
Hi,
On 04/26/2012 03:27 PM, Gabriel Dos Reis wrote:
yes, it does. On the other hand, the program is going to exit soon...
-- Gaby
In any case, a bit of sloppiness on my part. Sorry about that. Now, all
in all I don't have a strong opinion, but we may want to apply something
like the below (bo
Quoting Richard Guenther :
the function implementation is odd. cost is always 1 when used, and
both memory loads and calls are always cheap, but for example casts are not?
Isn't
return EDGE_COUNT (block->preds) * cost >= inserts_needed;
always true? Or is inserts_needed not what it suggests?
Hi!
On Wed, 25 Apr 2012 13:51:16 +0200, Richard Guenther
wrote:
> On Wed, Apr 25, 2012 at 1:27 PM, Thomas Schwinge
> wrote:
> > On Thu, 19 Apr 2012 19:46:17 +0200, I wrote:
> >> Here is my current test case, reduced from gcc.dg/tree-ssa/20030922-1.c:
> >>
> >> extern void abort (void);
> >>
On 26/04/12 12:20, Richard Guenther wrote:
> On Wed, Apr 25, 2012 at 11:56 PM, Tom de Vries wrote:
>> On 25/04/12 11:57, Richard Guenther wrote:
>>
>>> Hmm. I'm not sure we can conclude that they have the same value!
>>>
>>> +int bar (int);
>>> +void baz (int);
>>> +void bla
2012/4/26 Janis Johnson :
> A couple of months ago Kai Tietz fixed several tests for MS layout of
> bitfields, including:
>
> 2012-02-23 Kai Tietz
>
> * gcc.dg/bf-ms-layout-2.c: Adjust offsets to fit ms-bitfield
> structure-layout.
>
> He missed changes to gcc.dg/bf-ms-layout.c
Remove extra whitespace from doc.
--
Ricardo (http://r.untroubled.be/)
Index: gcc/doc/extend.texi
===
--- gcc/doc/extend.texi (revision 186860)
+++ gcc/doc/extend.texi (working copy)
@@ -7115,7 +7115,7 @@
release stores in all threa
I noticed these problems while writing the Niagara4 scheduling
description.
The ultra3 scheduling bug is particularly heinous, it means all
integer ALU op reservations were essentially wrong since the
ialu and ialuX reservations matched the same exact instructions.
Commited to trunk.
gcc/
Niagara4 is a dual issue processor, but with a twist. The dual issue
pipeline is shared amongst a set of cpu threads.
So in the case that there is only one thread active, then up to two
instructions from that cpu thread can execute at a time. If multiple
threads are active, we can instead end u
65 matches
Mail list logo
