Hi,
int_bit_position is used by ipa-devirt's type walking code. It is currently a
bottleneck
since I introduced speculation into contextes (I plan to solve this by changing
the
way i cache results). But this patch seems to make sense anyway: we do not need
to go
through folding:
tree
bit_from_p
Hi,
libreoffice fails to build with TLS because of "Cannot load any more object
with static TLS". Iant pointed out to me the difference that the initial exec
TLS model is also used by static TLS variables.
This patch prevents turning TLS variables into static that lets me to finish
libreoffice bui
Hi,
this patch fixes ordering issue that prevents ipa-reference folding with LTO
Bootstrapped/regtested x86_64-linux, comitted.
Honza
PR tree-optimization/63255
* ipa.c (symbol_table::remove_unreachable_nodes): Fix ordering
issue in setting body_removed flag.
Index: ipa.c
mark_exp_read doesn't know how to deal with a CONSTRUCTOR. In this case
it doesn't need to, since we're about to pull out the single element
anyway, so let's do that before calling mark_exp_read.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit cd87c48098daa3f1f31a7877d6fb0bf799b10d7b
Aut
This code was first allowed by DR 850: as of C++11, in unevaluated
context it's OK to refer to a non-static data member without an
associated object. The ABI says that this is mangled like an unresolved
name, i.e. as the plain identifier.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit
Hi,
this patch makes symbol==NULL comparsion to be resolved during parsing again.
I made this foldable only after symbol table is built because during parsing
the visibility of symbols may change affecting the outcome of folding.
Sadly it seems that C++ FE needs this to be folded early. This patch
On Fri, Sep 19, 2014 at 6:41 AM, Tom de Vries wrote:
> So it's a question of predictability (always do the same or do nothing) vs.
> robustness (do as much as you can given the circumstances). I'm not sure
> which one is better in this case.
I think it's fine the way it is now. Thanks for the pa
Hi,
As with the earlier patch for sh
( https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01627.html ), this fixes the
fallout caused by https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01615.html.
These are build failures, and the fixes are "obvious", but I don't know
my way around the failing ports, so
Hi Richard,
I also ran the gcc testsuite with
RUNTESTFLAGS="--tool_opts=-mcopyrelocs" to check for issues. The only
test that failed was g++.dg/tsan/default_options.C. It uses -fpie
-pie and BFD ld to link. Since BFD ld does not support copy
relocations with -pie, it does not link. I linked with
On 09/15/14 08:33, Jiong Wang wrote:
Jeff,
thanks, I partially understand your meaning here.
take the function "ira_implicitly_set_insn_hard_regs" in ira-lives.c
for example,
when generating address rtl, gcc will automatically generate "const"
operator to prefix
the address express
On Thu, 2014-09-18 16:55:35 -0500, Joel Sherrill
wrote:
> I only see one RTEMS target that has been built in the top page.
> Are more than the powerpc-rtems being built? How can I check?
The builders work down their list, and when done will start a new
round (with a list that would now contain t
On Thu, 2014-09-18 10:27:46 -0500, Joel Sherrill
wrote:
> On 9/18/2014 6:51 AM, Jan-Benedict Glaw wrote:
> > And to tell the whole story, Sebastian approached me with
> > extending the target lists in use by those targets he sent a patch
> > for; I just asked him to go this route, because I gue
On 09/19/14 13:36, Ilya Enkovich wrote:
Hi,
During my work on enabling pseudo PIC register I've found that cfg cleaunp may
remove lables with LABEL_PRESERVE_P set to 1. In my case I generated SET_RIP
during expand pass and cfg cleanup removed label it used as an operand. Below
is a patch th
On 09/16/14 03:09, Ilya Enkovich wrote:
I must be misunderstanding something then. I fundamentally don't see how
the return bounds are any different here than the return value. If we have
exposed the bounds in the IL, then aren't they going to be handled just like
any other object in the IL?
On 09/19/2014 09:29 AM, Marek Polacek wrote:
But we also started to warn on
CPP_KEYWORD and two others: "case value not in enumerated type".
Fixed by moving CPP_KEYWORD, CPP_TEMPLATE_ID, and CPP_NESTED_NAME_SPECIFIER
into enum cpp_ttype. Not sure if this is going to hurt something else.
Wait,
Hi,
During my work on enabling pseudo PIC register I've found that cfg cleaunp may
remove lables with LABEL_PRESERVE_P set to 1. In my case I generated SET_RIP
during expand pass and cfg cleanup removed label it used as an operand. Below
is a patch that fixes it. It is not actually required
Still no feedback regarding this proposal ?
On 19/08/2014 22:14, François Dumont wrote:
Any news regarding this proposal ?
Thanks
François
On 30/07/2014 23:39, François Dumont wrote:
Hi
Now that patch on testsuite allocator is in I would like to
reactivate this one. Here it is agai
Hi,
this patch fixes ICE by avoiding mangling of types with variadic size (those are
not really supported). Bootstrapped/regtested x86_64-linux, tested with
libreoffice,
comitted.
PR lto/63286
* tree.c (need_assembler_name_p): Do not mangle variadic types.
Index: tree.c
=
When the Go frontend builds a thunk, for a go or defer statement, it
sets up a struct to hold the arguments to the function, passes a pointer
to that struct to the thunk, and the thunk then calls the function
passing the fields of the struct. Something along these lines is
necessary when calling a
Hi,
On 09/19/2014 07:32 PM, Jason Merrill wrote:
On 09/19/2014 11:40 AM, Paolo Carlini wrote:
if (TREE_CODE (CP_DECL_CONTEXT (decl)) == FUNCTION_DECL
- || (TREE_CODE (decl) == FUNCTION_DECL && DECL_LOCAL_FUNCTION_P
(decl)))
+ || (TREE_CODE (decl) == FUNCTION_DECL
+ && (DECL_
On 09/09/14 11:53, Jakub Jelinek wrote:
On Fri, Jun 27, 2014 at 09:13:07AM +0200, Jakub Jelinek wrote:
The patch adds two new (trivial handlers) to libubsan, as it is maintained
in llvm's compiler-rt, will talk to them if they are interested in those
and what exact wording and form (AFAIK clang
On 09/09/14 11:52, Jakub Jelinek wrote:
Hi!
On Fri, Jun 27, 2014 at 09:13:07AM +0200, Jakub Jelinek wrote:
The patch adds two new (trivial handlers) to libubsan, as it is maintained
in llvm's compiler-rt, will talk to them if they are interested in those
and what exact wording and form (AFAIK c
On 09/18/14 04:26, Richard Sandiford wrote:
Patch 4 should make it possible to relax i386'a CANNOT_CHANGE_MODE_CLASS,
solving the missed optimisation that triggered the original thread.
gcc/
* config/i386/i386.c (ix86_cannot_change_mode_class): Remove
GET_MODE_SIZE (to) < GET_MO
On 09/18/14 04:25, Richard Sandiford wrote:
This is the main patch for the bug. We should treat a register as invalid
for a mode change if simplify_subreg_regno cannot provide a new register
number for the result. We should treat a class as invalid for a mode change
if all registers in the clas
On Fri, Sep 19, 2014 at 11:19:14AM -0400, Jason Merrill wrote:
> On 09/19/2014 09:29 AM, Marek Polacek wrote:
> >- orig_type = TREE_TYPE (cond);
> >+ /* Handle enum bit-fields. */
> >+ tree field;
> >+ if (TREE_CODE (cond) == COMPONENT_REF
> >+ && (field = TREE_OPERAND (co
On 09/19/2014 11:40 AM, Paolo Carlini wrote:
if (TREE_CODE (CP_DECL_CONTEXT (decl)) == FUNCTION_DECL
- || (TREE_CODE (decl) == FUNCTION_DECL && DECL_LOCAL_FUNCTION_P (decl)))
+ || (TREE_CODE (decl) == FUNCTION_DECL
+ && (DECL_LOCAL_FUNCTION_P (decl) || LAMBDA_FUNCTION_P (decl
This patch makes -fsched-pressure the default on AArch64, like on ARM. This
improves performance and
reduces codesize due to fewer unnecessary spills.
OK for commit?
ChangeLog:
2014-09-19 Wilco Dijkstra
* gcc/common/config/aarch64/aarch64-common.c:
(default_options aarch_opti
On 09/19/14 01:23, Richard Sandiford wrote:
Jeff Law writes:
On 09/18/14 04:07, Richard Sandiford wrote:
This series is a cleaned-up version of:
https://gcc.gnu.org/ml/gcc/2014-03/msg00163.html
The underlying problem is that the semantics of subregs depend on the
word size. You can't
This is the patch I committed. Should fix everyone's issues.
Fix mcount test cases to only run on supported targets
* gcc.dg/pg-override.c: Only run on x86 Linux.
* gcc.dg/pg.c: Dito.
* gcc.target/i386/fentry-override.c: Exclude for PIC.
* gcc.
On 09/11/2014 01:29 AM, James Greenhalgh wrote:
> +;; Predicates used by the various SIMD shift operations. These
> +;; fall in to 3 categories.
> +;; Shifts with a range 0-(bit_size - 1) (aarch64_simd_shift_imm)
> +;; Shifts with a range 1-bit_size (aarch64_simd_shift_imm_offset)
> +;; Shif
On 09/18/14 13:55, Jason Merrill wrote:
On 09/18/2014 03:25 PM, Aldy Hernandez wrote:
This patch disables removing of
DW_AT_{declaration,object_pointer,formal_parameter} tags while
generating a DIE for subprograms. Now that we generate dwarf info
early, we will always have an old_die the second
On Thu, 18 Sep 2014, Joseph S. Myers wrote:
> On Thu, 18 Sep 2014, Uros Bizjak wrote:
>
> > OK for mainline and release branches.
>
> I've omitted ia64 from the targets in the testcase in the release branch
> version, given the lack of any definition of FP_TRAPPING_EXCEPTIONS at all
> there.
>
On 09/19/14 06:21, Segher Boessenkool wrote:
On Thu, Sep 18, 2014 at 11:54:46PM -0600, Jeff Law wrote:
Shouldn't you be testing if the register is fixed rather than its class?
Or maybe both?
register_operand (via general_operand) uses operand_reg_set for this; it is
initialised via the regcl
On Sep 19, 2014, at 2:37 AM, Segher Boessenkool
wrote:
> On Thu, Sep 18, 2014 at 01:44:55PM -0500, Segher Boessenkool wrote:
>> I am testing a patch that is just
>>
>>
>> diff --git a/contrib/dg-extract-results.py b/contrib/dg-extract-results.py
>> index cccbfd3..3781423 100644
>> --- a/contrib
On 09/19/14 07:43, Felix Yang wrote:
+2014-09-19 Felix Yang
+
+* cfgrtl.c ira.c ira-color.c ira-conflicts ira-lives.c:
+Clean up code: update comments and fix spacing to conform to coding style.
THanks. Committed.
jeff
On Fri, Sep 19, 2014 at 2:53 PM, Ilya Enkovich wrote:
>> > This patch adds i386 target hooks for Pointer Bounds Checker.
> New version with fixes and better documentation for ix86_load_bounds and
> ix86_store_bounds is below.
> +/* Expand pass uses this hook to load bounds for function paramet
On 09/19/14 09:48, Martin Liška wrote:
Hello.
The following patch adds a test for PR/63270. The test has been tested before
and with r215328, where the bug was fixed.
Thank you,
Martin
pr63270.changelog
gcc/testsuite/ChangeLog:
2014-09-19 Martin Liska
PR LTO/63270
* g++.
On 09/19/14 10:16, James Greenhalgh wrote:
On Fri, Sep 19, 2014 at 04:10:34PM +0100, Andreas Krebbel wrote:
On 09/19/2014 02:59 PM, James Greenhalgh wrote:
Hi,
After https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01615.html we error
on the use of constraints in define_splits, define_expands a
On 09/19/14 10:02, Jiong Wang wrote:
On 19/09/14 16:49, Jeff Law wrote:
On 09/19/14 05:27, Jiong Wang wrote:
On 19/09/14 06:51, Jeff Law wrote:
On 09/16/14 00:55, Zhenqiang Chen wrote:
-Original Message-
From: gcc-patches-ow...@gcc.gnu.org [mailto:gcc-patches-
ow...@gcc.gnu.org] On B
On Fri, Sep 19, 2014 at 04:10:34PM +0100, Andreas Krebbel wrote:
> On 09/19/2014 02:59 PM, James Greenhalgh wrote:
> >
> > Hi,
> >
> > After https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01615.html we error
> > on the use of constraints in define_splits, define_expands and
> > define_peephole2s. T
On 19/09/14 16:49, Jeff Law wrote:
On 09/19/14 05:27, Jiong Wang wrote:
On 19/09/14 06:51, Jeff Law wrote:
On 09/16/14 00:55, Zhenqiang Chen wrote:
-Original Message-
From: gcc-patches-ow...@gcc.gnu.org [mailto:gcc-patches-
ow...@gcc.gnu.org] On Behalf Of Jiong Wang
Sent: Monday, Sept
On 09/19/14 01:23, Richard Sandiford wrote:
Jeff Law writes:
On 09/18/14 04:07, Richard Sandiford wrote:
This series is a cleaned-up version of:
https://gcc.gnu.org/ml/gcc/2014-03/msg00163.html
The underlying problem is that the semantics of subregs depend on the
word size. You can't
On 09/19/14 05:27, Jiong Wang wrote:
On 19/09/14 06:51, Jeff Law wrote:
On 09/16/14 00:55, Zhenqiang Chen wrote:
-Original Message-
From: gcc-patches-ow...@gcc.gnu.org [mailto:gcc-patches-
ow...@gcc.gnu.org] On Behalf Of Jiong Wang
Sent: Monday, September 15, 2014 11:28 PM
To: Zhenqi
Hello.
The following patch adds a test for PR/63270. The test has been tested before
and with r215328, where the bug was fixed.
Thank you,
Martin
gcc/testsuite/ChangeLog:
2014-09-19 Martin Liska
PR LTO/63270
* g++.dg/lto/pr63270_0.C: New test.
* g++.dg/lto/pr63270_1.
On Wed, Sep 17, 2014 at 12:47 PM, Michael Meissner
wrote:
> This patch is an intermediate step of what I want to do to improve power8
> fusion.
>
> In the current trunk, the fusion support for gpr loads is done by a peephole2
> to find the addis followed by the load instruction where the only cons
Hi,
I had a quick look at this issue and it immediately reminded me
c++/60605, which Jason fixed by handling local functions at the
beginning of check_default_tmpl_args. In the present case of:
template< class = void >
struct S
{
friend void foo( S )
{
[](){};
}
};
it occurs to me
Michael, I really don't understand the need for this change in your
original patch. I don't know if this was a temporary testing change or
what.
For instance, you had the following commented out in your original patch:
if (/*DECL_ABSTRACT (decl) &&*/ DECL_ABSTRACT_ORIGIN (decl) == decl)
On Thu, Sep 18, 2014 at 12:50 PM, Ulrich Weigand wrote:
> Hello,
>
> the ABI warning messages I introduced in recent patches refer to a GCC
> version "4.10". As GCC has since adopted a new version naming scheme,
> this patch updates those messages to refer to "GCC 5" instead.
>
> Tested on powerp
Jan Hubicka wrote:
> I am bit in hurry, so just quickly (will give it second look at evening). The
> pass is a simple dataflow propagating preferred sections. NULL means that the
> symbol in question was not visited yet. Assuming that all symbols are
> reachable,
> this should not happen by the
On 09/19/14 08:11, Sebastian Pop wrote:
Jeff Law wrote:
On 09/17/14 13:59, Sebastian Pop wrote:
Hi,
I got an ICE while building libstdc++ of a cross compiler x86 to aarch64. I
have a testcase that ICEs on current GCC trunk. I was trying to painfully
reduce it with creduce, and it is still se
This patch to libiberty fixes writing out ELF files that have a large
number of sections (more than 0xFF00). We already handled reading those
files correctly; this fixes writing them, for absurdly large programs
when using LTO. Bootstrapped and ran testsuite on
x86_64-unknown-linux-gnu. This pat
On 09/19/2014 09:29 AM, Marek Polacek wrote:
- orig_type = TREE_TYPE (cond);
+ /* Handle enum bit-fields. */
+ tree field;
+ if (TREE_CODE (cond) == COMPONENT_REF
+ && (field = TREE_OPERAND (cond, 1))
+ && DECL_BIT_FIELD_TYPE (field)
+ && TREE_CODE (DE
On 09/19/2014 02:59 PM, James Greenhalgh wrote:
>
> Hi,
>
> After https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01615.html we error
> on the use of constraints in define_splits, define_expands and
> define_peephole2s. These are never looked at by the compiler, and so
> have no reason to be set.
>
This patch fixes ipa/61998, where simple testcases would crash when
using -Wsuggest-final-types, by bailing out early when odr_types_ptr is
NULL in ipa_devirt().
Approved by Honza on IRC.
Boostrapped and tested on x86_64-unknown-linux-gnu, applied.
diff --git a/gcc/ChangeLog b/gcc/ChangeLog
i
On 09/19/14 06:59, James Greenhalgh wrote:
Hi,
After https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01615.html we error
on the use of constraints in define_splits, define_expands and
define_peephole2s. These are never looked at by the compiler, and so
have no reason to be set.
Right.
I expec
> Jiong Wang wrote:
>
> when generating instructions to access local variable, for example a local
> array,
>
> if the array size very big, then we need a temp reg to keep the intermediate
> index,
> then use that temp reg as base reg, so that ldr is capable of indexing the
> element.
>
> whi
> Jeff Law wrote:
> > On 09/17/14 13:59, Sebastian Pop wrote:
> > >Hi,
> > >
> > >I got an ICE while building libstdc++ of a cross compiler x86 to aarch64.
> > >I
> > >have a testcase that ICEs on current GCC trunk. I was trying to painfully
> > >reduce it with creduce, and it is still several t
Jeff Law wrote:
> On 09/17/14 13:59, Sebastian Pop wrote:
> >Hi,
> >
> >I got an ICE while building libstdc++ of a cross compiler x86 to aarch64. I
> >have a testcase that ICEs on current GCC trunk. I was trying to painfully
> >reduce it with creduce, and it is still several thousand lines of c++
This fix typos and spacing in comments for several source files.
Please commit this patch if OK for trunk.
Index: gcc/ira-conflicts.c
===
--- gcc/ira-conflicts.c(revision 215391)
+++ gcc/ira-conflicts.c(working copy)
@@ -60,
On Fri, Sep 19, 2014 at 03:29:44PM +0200, Marek Polacek wrote:
> Jakub, does the asan.c change look ok?
> --- gcc/gcc/asan.c
> +++ gcc/gcc/asan.c
> @@ -2024,6 +2024,8 @@ maybe_instrument_call (gimple_stmt_iterator *iter)
> case BUILT_IN_TRAP:
> /* Don't instrument these. */
This patch makes -Wswitch{,-enum} work even with enum bit-fields (we got
several bugreports about that).
The problem is that enum e:2; has an integral type with precision 2,
and c_do_switch_warnings doesn't like that:
6195 /* From here on, we only care about about enumerated types. */
6196 if
when generating instructions to access local variable, for example a local
array,
if the array size very big, then we need a temp reg to keep the intermediate
index,
then use that temp reg as base reg, so that ldr is capable of indexing the
element.
while this will cause trouble, because the
Hi,
After https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01615.html we error
on the use of constraints in define_splits, define_expands and
define_peephole2s. These are never looked at by the compiler, and so
have no reason to be set.
I expect there will be more fallout as Jan's auto-builder make
On 18 Sep 21:34, Uros Bizjak wrote:
> 2014-06-11 18:00 GMT+04:00 Ilya Enkovich :
> > Hi,
> >
> > This patch adds i386 target hooks for Pointer Bounds Checker.
> >
> > Bootstrapped and tested on linux-x86_64.
> >
> > Thanks,
> > Ilya
> > --
> > gcc/
> >
> > 2014-06-11 Ilya Enkovich
> >
> >
On Fri, Sep 19, 2014 at 12:41:53PM +0200, Tom de Vries wrote:
> I've followed up on the explanation by Segher about 2.15 File module
> version and fixed the comment.
>
> I've not added the 2.15 file module version check on copy Segher also
> mentioned, since I'm not sure about that one. AFAIU th
On Thu, Sep 18, 2014 at 11:54:46PM -0600, Jeff Law wrote:
> Shouldn't you be testing if the register is fixed rather than its class?
> Or maybe both?
register_operand (via general_operand) uses operand_reg_set for this; it is
initialised via the regclass NO_REGS too (and other things).
This wou
On 18 Sep 21:09, Jeff Law wrote:
> On 09/18/14 13:34, Uros Bizjak wrote:
> >2014-06-11 18:00 GMT+04:00 Ilya Enkovich :
> >>Hi,
> >>
> >>This patch adds i386 target hooks for Pointer Bounds Checker.
> >>
> >>Bootstrapped and tested on linux-x86_64.
> >>
> >>Thanks,
> >>Ilya
> >>--
> >>gcc/
> >>
> >>
I've corrected my patch accordingly to what you said. To diffirentiate
second case in destructor I've added pthread_setspecific
(gomp_tls_key, NULL) at the end of gomp_thread_start. So, destructor
can simply skip the case when pthread_getspecific (gomp_tls_key)
returns 0. I also think that it's bet
On 19/09/14 06:51, Jeff Law wrote:
On 09/16/14 00:55, Zhenqiang Chen wrote:
-Original Message-
From: gcc-patches-ow...@gcc.gnu.org [mailto:gcc-patches-
ow...@gcc.gnu.org] On Behalf Of Jiong Wang
Sent: Monday, September 15, 2014 11:28 PM
To: Zhenqiang Chen
Cc: gcc-patches@gcc.gnu.org;
On 18/09/14 20:38, Charles Baylis wrote:
This patch replaces the inline assembler implementations of the
vld[234](q?)_lane_* intrinsics with new versions which exploit the new builtin
functions added in patch 1.
Tested (with the rest of the patch series) with make check on aarch64-oe-linux
with
On Fri, Sep 19, 2014 at 08:17:28AM +0400, Yury Gribov wrote:
> > I was talking about mbr's plugin here :-)
>
> Ah, ok. Then I'll mention thinca's plugin as a secondary option with a
> disclaimer then.
Why? There are more plugins that also do the same thing, all more popular
(on vim.org at least
*Ping*
Thanks,
James
On Thu, Sep 11, 2014 at 09:29:52AM +0100, James Greenhalgh wrote:
> gcc/
>
> 2014-09-11 James Greenhalgh
>
> * config/aarch64/aarch64-protos.h (aarch64_simd_const_bounds): Change
> return type to bool.
> * config/aarch64/aarch64-simd.md (aarch64_qshl):
+(define_expand "aarch64_ld2_lane"
+ [(match_operand:OI 0 "register_operand" "=w")
+ (match_operand:DI 1 "register_operand" "w")
+ (match_operand:OI 2 "register_operand" "0")
+ (match_operand:SI 3 "immediate_operand" "i")
+ (unspec:VQ [(const_int 0)] UNSPEC_VSTRUCTDUMMY)]
On 18-09-14 19:46, Diego Novillo wrote:
On Thu, Sep 18, 2014 at 10:56 AM, Yury Gribov wrote:
On 08/04/2014 12:14 PM, Tom de Vries wrote:
On 04-08-14 08:45, Yury Gribov wrote:
Thanks! My 2 (actually 4) cents below.
Hi Yuri,
thanks for the review.
> +if ($#ARGV == 1 && ("$ARGV[0]" eq
On Thu, Sep 18, 2014 at 09:45:59PM +0100, Jeff Law wrote:
> On 09/18/14 04:19, James Greenhalgh wrote:
> >
> > Hi,
> >
> > As discussed in https://gcc.gnu.org/ml/gcc-patches/2014-09/msg01334.html
> > The construct
> >
> >(clobber (match_scratch 0 "r"))
> >
> > is invalid - operand 0 must be mar
tsaund...@mozilla.com writes:
> diff --git a/gcc/ChangeLog b/gcc/ChangeLog
> index c048672..5b27aa8 100644
> --- a/gcc/ChangeLog
> +++ b/gcc/ChangeLog
> @@ -1,3 +1,8 @@
> +2014-09-12 Trevor Saunders
> +
> + * hash-table.h (gt_pch_nx): don't call gt_pch_note_object within an
> + assert.
>
On Thu, Sep 18, 2014 at 01:44:55PM -0500, Segher Boessenkool wrote:
> I am testing a patch that is just
>
>
> diff --git a/contrib/dg-extract-results.py b/contrib/dg-extract-results.py
> index cccbfd3..3781423 100644
> --- a/contrib/dg-extract-results.py
> +++ b/contrib/dg-extract-results.py
> @@
On 18 September 2014 11:15, James Greenhalgh wrote:
> gcc/
>
> 2014-09-18 James Greenhalgh
>
> * config/aarch64/aarch64.md (stack_protect_test_): Mark
> scratch register as an output to placate register renaming.
>
> gcc/testsuite/
>
> 2014-09-18 James Greenhalgh
>
>
On Sep 19, 2014, at 01:22 , Jeff Law wrote:
>> * varasm.c (default_section_type_flags): Flag .persistent.bss
>> sections as SECTION_BSS.
> Ok.
>
> Jeff
Thanks for this review and the other ones Jeff :-)
Hi Charles,
Good to see these intrinsics being brought into the modern world :)
Some style comments inline.
On 18/09/14 20:38, Charles Baylis wrote:
This patch adds new patterns and builtins to represent single lane structure
loads instructions, which will be used to implement the vld[234](q?)_
Hi,
On 09/19/2014 08:03 AM, Gerald Pfeifer wrote:
Hi Paolo,
On Wed, 17 Sep 2014, Paolo Carlini wrote:
clang recently, in 3.5.0 if I remember correctly, stopped -Wnon-virtual-dtor
warning for final classes. I think it makes sense to do the same.
mind adding a note to wwwdocs/htdocs/gcc-5/chang
> thanks for your good question. I think it is equivalent, as it seems that
> GFC_DESCRIPTOR_TYPE_P (type) implies either sym->attr.allocatable or
> sym->attr.pointer. To check, I rank a check-fortran with the explicit patch
> below, and this made no difference. Code gen for a number of addition
On 18 Sep 21:54, Uros Bizjak wrote:
> Hello!
>
> > 2014-06-11 Ilya Enkovich
> >
> > * config/i386/i386.c (ix86_option_override_internal): Do not
> > support x32 with MPX.
> > is not available.
> > (init_cumulative_args): Init stdarg, bnd_regno, bnds_in_bt
> >
On 4 August 2014 16:52, Ian Lance Taylor wrote:
> On Sun, Aug 3, 2014 at 11:12 AM, Iain Buclaw wrote:
>>
>> - I haven't signed any copyright assignments to GCC. But I have
>> papers from Donald ready to send across.
>
> Definitely necessary before we can consider this. Please get this
> squared
Hi Guys,
I am applying the patch below to fix a problem with the RL78's popping
of pushed registers in G10 mode. The problem was that the pop uses a
two instruction sequence and dead code elimination was deleting the
second instruction (the move of the popped value from register A to
th
> I found this a bit difficult to parse, so I'm going to try and summarize,
> please tell me if I've got it right or wrong.
>
> The code in question is not explicitly marked as being Darwin specific;
> however, to date we've only managed to exercise it on Darwin. Therefore, any
> fix is likely
Hello Jeff, all,
On 19 Sep 2014, at 05:14, Jeff Law wrote:
> On 09/18/14 16:20, Iain Sandoe wrote:
>>
>> 1. There has been a change made "to make the upper path like the lower path"
>> (as you said on IRC).
>> - apparently (from our conversation) you don't expect this to be a general
>> opti
On 18 Sep 19:33, Uros Bizjak wrote:
> On Thu, Sep 18, 2014 at 3:47 PM, Ilya Enkovich wrote:
>
> > Thanks for your comments. Below is a fixed verison.
> >
> > Ilya
> > --
>
> OK with a few nits below.
>
> Thanks,
> Uros.
>
> > +
> > +case IX86_BUILTIN_BNDRET:
> > + arg0 = CALL_EXPR_AR
Jeff Law writes:
> On 09/18/14 04:07, Richard Sandiford wrote:
>> This series is a cleaned-up version of:
>>
>> https://gcc.gnu.org/ml/gcc/2014-03/msg00163.html
>>
>> The underlying problem is that the semantics of subregs depend on the
>> word size. You can't have a subreg for byte 2 of a 4
>> The following fixes PR63152 zeroing the data field only for allocatables,
>> not pointers. The benefit of the patch is a >small speedup, and it avoids
>> that code starts to rely on behavior that is undefined in the standard. With
>> this patch, >something like
>>
>> INTEGER, DIMENSION(:), PO
90 matches
Mail list logo
