On Tue, Sep 10, 2013 at 3:30 AM, Steven Bosscher wrote:
> On Mon, Sep 9, 2013 at 10:01 AM, Richard Biener wrote:
>>> >> First, the loop passes that at the moment preceede IVOPTs leave
>>> >> around IL that is in desparate need of basic re-optimization
>>> >> like CSE, constant propagation and DCE.
On 2013.09.05 at 14:30 -0600, Jeff Law wrote:
>
> Bootstrapped and regression tested on x86_64-unknown-linux-gnu.
> Installed onto the trunk.
This patch caused http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58373
--
Markus
On Mon, Sep 9, 2013 at 11:35 PM, Bill Schmidt
wrote:
>
>> > I rely on size_binop to convert T2 into sizetype, because T2' may be in
>> > other kind of type. Otherwise there will be ssa_verify error later.
>>
>> OK, I see now. I had thought this was handled by fold_build2, but
>> apparently not
Hello,
On 09 Sep 13:50, Uros Bizjak wrote:
> OK.
Checked into main trunk: http://gcc.gnu.org/ml/gcc-cvs/2013-09/msg00286.html
--
Thanks, K
On Mon, 9 Sep 2013, Steven Bosscher wrote:
> On Mon, Sep 9, 2013 at 10:01 AM, Richard Biener wrote:
> >> >> First, the loop passes that at the moment preceede IVOPTs leave
> >> >> around IL that is in desparate need of basic re-optimization
> >> >> like CSE, constant propagation and DCE. That put
On 09/09/13 17:32, Kyrylo Tkachov wrote:
> Hi all,
>
> Shift operations with an immediate can generate a 16-bit encoding when the
> immediate is 5-bit wide, i.e. in the range [0-31]. Therefore we can use them
> in IT blocks even with the -mrestrict-it rules.
>
> I decided to reuse the "N" constra
Hi,
Looking at the way we handle the generic scheduler in AArch64, the order
of includes suggests that we will always match to the generic unit, even
when tuning for a non-generic pipeline.
This patch explicitly prevents that by defining a "generic_sched" attribute,
which is only true when the g
On Mon, Sep 9, 2013 at 9:32 PM, David Malcolm wrote:
> This patch is the handwritten part of the conversion of these types
> to C++; it requires the followup patch, which is autogenerated.
>
> It converts:
> struct GTY(()) symtab_node_base
> to:
> class GTY((user)) symtab_node_base
>
> and con
Updated to http://gcc.gnu.org/ml/gcc-patches/2012-12/msg01097.html
Build passes on arm-none-eabi and bootstrap passes on x86.
OK to trunk?
ChangeLog
* libgcc/Makefile.in: Include JAVA_IS_ENABLED in CFLAGS.
* libgcc/configure.ac (java_is_enabled): New variable.
* libgcc/configur
On Mon, 9 Sep 2013, Chung-Ju Wu wrote:
> Is it OK to apply on the cvs wwwdoc trunk? :)
I would perhaps say:
"A new nds32 port supports the 32-bit architecture from Andes
Technology Corporation"
or similar.
If you prefer the current wording, that needs an extra "the",
as in "to support the
This cleans up the RDG API and avoids computing data references
twice for each loop processed in loop distribution.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
Richard.
2013-09-10 Richard Biener
* tree-data-ref.h (build_rdg): Drop all parameters but loop.
On Mon, Sep 09, 2013 at 01:56:29PM -0700, Caroline Tice wrote:
> The patch looks good to me, but somebody else needs to approve it.
Why? You are listed as libvtv maintainer, the patch is fully contained in
libvtv, thus you can review it and approve.
On an unrelated note, what kind of testing was
Hi!
If two instructions are rtl_equal_p, and cross-jumping decides to cross-jump
them together, merge_memattrs takes care of clearing various MEM_ATTRS
or setting alignment to minimum, size to maximum etc.
But, it didn't handle flags on the MEMs.
The following patch fixes that. MEM_KEEP_ALIAS_SE
On Tue, Sep 10, 2013 at 12:38 PM, Jakub Jelinek wrote:
> Hi!
>
> If two instructions are rtl_equal_p, and cross-jumping decides to cross-jump
> them together, merge_memattrs takes care of clearing various MEM_ATTRS
> or setting alignment to minimum, size to maximum etc.
> But, it didn't handle fla
On Tue, Sep 10, 2013 at 01:18:34PM +0200, Richard Biener wrote:
> Ok.
Thanks.
> > BTW, I wonder about REG_ATTRS and REG_POINTER, REG_USERVAR_P,
> > REG_FUNCTION_VALUE_P, is it fine to keep them as is
>
> Definitely REG_EXPR is used to check for signedness of a register, but I
> suppose
> that i
Ping^4
Could please someone look at this patch? It is mostly i386 target
specific and is basic for further MPX based features.
Thanks,
Ilya
2013/9/2 Ilya Enkovich :
> Ping^3
>
> Attached is the same patch but against the current trunk.
>
> 2013/8/26 Ilya Enkovich :
>> Ping
>>
>> 2013/8/19 Ilya E
On 09/10/2013 01:27 AM, Jakub Jelinek wrote:
On Mon, Sep 09, 2013 at 08:39:16PM -0600, Jeff Law wrote:
On 09/09/2013 06:59 PM, Andrew MacLeod wrote:
Whilst poking around, I discovered these hast table functions and the
macro are no longer used anywhere in the compiler.
bootstraps on x86_64-unkn
On Fri, 6 Sep 2013, Wei Mi wrote:
> SCHED_GROUP works after I add chain_to_prev_insn after
> add_branch_dependences, in order to chain control dependences to prev
> insn for sched group.
chain_to_prev_insn is done in the end of deps_analyze_insn, why is that not
sufficient?
Alexander
Hi,
until recently we did not have thunks with incoming edges in the call
graph. However, speculative devirtualization can introduce them which
may lead to creation of new SCCs in the call graph. What is more,
ipa_reduced_postorder called with reduce parameter set to true may
decide to represent
On Tue, Sep 10, 2013 at 07:42:42AM -0400, Andrew MacLeod wrote:
> On 09/10/2013 01:27 AM, Jakub Jelinek wrote:
> >On Mon, Sep 09, 2013 at 08:39:16PM -0600, Jeff Law wrote:
> >>On 09/09/2013 06:59 PM, Andrew MacLeod wrote:
> >>>Whilst poking around, I discovered these hast table functions and the
>
> Hi,
>
> until recently we did not have thunks with incoming edges in the call
> graph. However, speculative devirtualization can introduce them which
We did, just not very often. Normal devirtualization code can produce them,
too.
> may lead to creation of new SCCs in the call graph. What i
As mentioned in the PR, we have a jump threading request with a loop
header in the middle of the jump threading path. In the past this would
have been rejected earlier but due to recent changes it gets registered.
The code to update the CFG didn't know how to detect this particular
case and
On 09/10/2013 02:00 AM, Richard Biener wrote:
On Mon, 9 Sep 2013, Steven Bosscher wrote:
On Mon, Sep 9, 2013 at 10:01 AM, Richard Biener wrote:
First, the loop passes that at the moment preceede IVOPTs leave
around IL that is in desparate need of basic re-optimization
like CSE, constant propag
On 10 September 2013 09:23, James Greenhalgh wrote:
> 2013-09-10 James Greenhalgh
>
> * config/aarch64/aarch64.md (generic_sched): New.
> * config/aarch64/aarch64-generic.md (load): Make conditional
> on generic_sched attribute.
> (nonload): Likewise.
OK
/Marcu
> Here's version 2 of this patch series.
>
> This is now 6 patches.
>
> The first two patches correspond to the two patches from the original
> patch series (and must both be applied together to make sense). They
> convert cgraph_node and varpool_node to inherit from symtab_node_base,
> eliminat
Hello Richard,
Thanks for inputs.
On 09 Sep 10:39, Richard Henderson wrote:
> gen_andhi_1 is not used, nor is it likely to be in the future, therefore this
> should still have "*".
We're using it in patch 6/8 when introducing plugins:
+ { OPTION_MASK_ISA_AVX512F, CODE_FOR_andhi_1, "__builtin_ia3
On 09/10/2013 08:30 AM, Jakub Jelinek wrote:
On Tue, Sep 10, 2013 at 07:42:42AM -0400, Andrew MacLeod wrote:
On 09/10/2013 01:27 AM, Jakub Jelinek wrote:
On Mon, Sep 09, 2013 at 08:39:16PM -0600, Jeff Law wrote:
On 09/09/2013 06:59 PM, Andrew MacLeod wrote:
Whilst poking around, I discovered
> On 09/09/13 17:32, Kyrylo Tkachov wrote:
> > Hi all,
> >
> > Shift operations with an immediate can generate a 16-bit encoding when
> > the immediate is 5-bit wide, i.e. in the range [0-31]. Therefore we
> > can use them in IT blocks even with the -mrestrict-it rules.
> >
> > I decided to reuse t
On Tue, Sep 10, 2013 at 09:04:41AM -0400, Andrew MacLeod wrote:
> Really? If it were useful I would have thought they'd be using it now.
>
> In any case, it doesn't belong in tree-flow.h.. . shouldn't these
> functions and macros be in hashtab.h then?
It doesn't belong to hashtab.h, because that
On Tue, 10 Sep 2013, Kugan wrote:
> On 09/09/13 19:01, Richard Biener wrote:
> > On Mon, Sep 9, 2013 at 1:09 AM, Kugan
> > wrote:
> > >
> > > On 06/09/13 16:16, Richard Biener wrote:
> > > >
> > > > On 9/3/13 2:15 PM, Kugan wrote:
> > > > >
> > > > > Thanks Richard for reviewing.
> > > > >
>
On Tue, 10 Sep 2013, Jeff Law wrote:
> On 09/10/2013 02:00 AM, Richard Biener wrote:
> > On Mon, 9 Sep 2013, Steven Bosscher wrote:
> >
> > > On Mon, Sep 9, 2013 at 10:01 AM, Richard Biener wrote:
> > > > > > > First, the loop passes that at the moment preceede IVOPTs leave
> > > > > > > around I
On 09/10/2013 07:23 AM, Richard Biener wrote:
update-ssa for example just queues to release SSA names. But yes,
re-using SSA names from a freelist has issues and advantages ;)
Yup. I've wondered from time to time if we could do away with the name
manager -- IIRCC it was written when we took e
On Tue, 2013-09-10 at 15:41 +0800, bin.cheng wrote:
> On Mon, Sep 9, 2013 at 11:35 PM, Bill Schmidt
> wrote:
> >
> >> > I rely on size_binop to convert T2 into sizetype, because T2' may be in
> >> > other kind of type. Otherwise there will be ssa_verify error later.
> >>
> >> OK, I see now.
On 10/09/13 14:11, Kyrylo Tkachov wrote:
>> On 09/09/13 17:32, Kyrylo Tkachov wrote:
>>> Hi all,
>>>
>>> Shift operations with an immediate can generate a 16-bit encoding when
>>> the immediate is 5-bit wide, i.e. in the range [0-31]. Therefore we
>>> can use them in IT blocks even with the -mrestr
On Mon, Sep 9, 2013 at 9:55 PM, Xinliang David Li wrote:
> looks fine to me.
>
> In the long run, I wonder if the machinery in diagnostic messages can
> be reused for opt-info dumping -- i.e., support different streams. It
> has many nice features including %qD specifier for printing tree
> decls.
On Tue, 10 Sep 2013, Jeff Law wrote:
> On 09/10/2013 07:23 AM, Richard Biener wrote:
> >
> > update-ssa for example just queues to release SSA names. But yes,
> > re-using SSA names from a freelist has issues and advantages ;)
> Yup. I've wondered from time to time if we could do away with the
> This patch is the handwritten part of the conversion of these types
> to C++; it requires the followup patch, which is autogenerated.
>
> It converts:
> struct GTY(()) symtab_node_base
> to:
> class GTY((user)) symtab_node_base
>
> and converts:
> struct GTY(()) cgraph_node
> to:
> stru
> Amongst other things, the rename_symtab.py script converts
> "symtab_node" to "symtab_node *".
>
> This will lead to broken code on declarations that declare
> more than one variable (only the first would get a "*"), so split
> up such declarations.
>
> gcc/
> * cgraphunit.c (analyze_func
> This patch is 439KB in size, so to avoid mailing-list limitations I've
> uploaded it to:
> http://dmalcolm.fedorapeople.org/gcc/large-patches/fd607e77883b8d3f8fb7c8091d0626dbbd31b9a5-0002-Automated-conversion-of-symtab-to-class-hierarchy.patch
OK,
thanks!
Honza
> gcc/
>
> * ipa-ref.h (symtab_node): Remove typedef to pointer type, as it
> clashes with the preferred name for the base class.
> (const_symtab_node): Remove redundant typedef.
This is OK, too.
Actually it is not only clash - it was really inconsistent to write
symtab_node nod
> gcc/
>
> Patch autogenerated by rename_symtab.py from
> https://github.com/davidmalcolm/gcc-refactoring-scripts
> revision 714e57b66b08fb29f47826fbb642707bdfa19520
OK,
thanks!
Honza
> * cgraph.c (gt_ggc_mx): Update for renaming of symtab_node_base
> to symtab_node
> (gt_pch_nx): Likewise.
> ---
> gcc/cgraph.c | 24
> 1 file changed, 12 insertions(+), 12 deletions(-)
>
> diff --git a/gcc/cgraph.c b/gcc/cgraph.c
> index bb2626d..abbff
On Tue, Sep 10, 2013 at 03:17:50PM +0200, Richard Biener wrote:
> > unsigned short s;
> > s.1_3 = (short unsigned int) l.0_2;
> > l.0_2: VARYING
> > s.1_3: [0, +INF]
>
> Note that [0, +INF] is the same as VARYING and [-INF, +INF] and VARYING for
> l.0_2 is the same as [-INF, +INF].
Yeah, I don't
On 09/10/2013 09:15 AM, Jakub Jelinek wrote:
On Tue, Sep 10, 2013 at 09:04:41AM -0400, Andrew MacLeod wrote:
Really? If it were useful I would have thought they'd be using it now.
In any case, it doesn't belong in tree-flow.h.. . shouldn't these
functions and macros be in hashtab.h then?
It do
On 09/09/2013 05:00 AM, Richard Biener wrote:
More discussion on rationale at the thread:
http://gcc.gnu.org/ml/gcc-patches/2013-09/msg00300.html
Tested on x86_64.
The split looks ok to me, given the plan. But as there is still discussion
around it please hold off until that died down with a
Hi,
fold_ctor_reference expects VAR or FUNCTION_DECL as FROM_DECL parameter.
gimple_get_virt_method_for_binfo instead passes expression
(pointer_plus_expr (addr_expr vtable_decl) (integer_cst XXX))
This bug makes can_refere_from_current_unit_p to mistakely think that
the references are safe because
Hi,
this patch adds polymorphic call targets into the boundary of an unit. This
enables
more late devirtualization. I briefly measured size on effect of WPA->ltrans
files
on firefox and they still seem to be around 2GB.
Bootstrapped/regtested x86_64-linux, comitted.
* lto-cgraph.c: Inc
Hi,
I've added references to fxsr, xsave and xsaveopt options and builtins
to doc/[invoke,extend].texi.
Is it OK?
doc_fxsr_xsave_fxsave.patch
Description: Binary data
On Fri, Sep 06, 2013 at 11:04:45AM +0100, Iain Sandoe wrote:
> Hi Kirill,
>
> Thanks for Ilya's input on the PR thread.
>
> We've done some testing/checking across the Darwin versions last night and
> I've bootstrapped all langs including Ada, and tested the patch below
> (together with the fra
This patch allows the compiler to parse and analyze aspect specifications that
apply to package bodies, protected bodies, task bodies, [generic] renaming
declarations, and body stubs.
-- Source --
-- bodies.adb
package body Bodies with Warnings => On is
protected bo
This patch implements properly the handling of categorization aspects that
apply to a package instantiation.
Compiling inst.ads must yield:
inst.ads:1:06: cannot depend on "Gen" (wrong categorization)
inst.ads:1:06: pure unit cannot depend on non-pure unit
generic
type T is private;
package G
This patch forbids the use of the pragma Warnings REASON parameter in
compiler units (since this would cause bootstrap issues). This is for
internal development purposes only. No functionality change.
Tested on x86_64-pc-linux-gnu, committed on trunk
2013-09-10 Robert Dewar
* sem_prag
This implements a warning that is output if the switch -gnateE (Extra
Exception Information) is specified on the command line, and the
configuration in use does not support it (true of configurable
run-times with the restriction No_Exception_Handlers or
No_Exception_Propagation. If a trivial progra
This patch corrects the check of a Global item of mode In_Out or Out that
appear as an input in the Global aspect of an enclosing subprogram. Prior to
this patch, the check caused an infinite loop in certain scenarios.
-- Source --
-- stack_overflow.adb
procedure Stack
This patch adds the procedures For_Each and Sort_Table to GNAT.Table
making its interface more similar to that of GNAT.Dynamic_Tables.
The following test:
1. with GNAT.Table;
2. with Text_IO; use Text_IO;
3. procedure GTableTestFS is
4.package T is new GNAT.Table
5.
The use of -gnatR in check semantics only (-gnatc) mode does not work
since code must be generated for -gnatR to work. The attempt to combine
the two switches is now properly diagnosed. The following trivial package
1. package gnatrgnatc is
2. end;
If compiled with switches -gnatc -gnat
Previously if an Ada 2012 feature was used in an earlier Ada mode, the
error message always advised the use of the -gnat2012 switch. But that
is not helpful if the mode was set by an explicit use of pragma Ada_xx.
This patch improves the error message in this case by pointing to the
relevant pragma
This patch makes sure that the routine Check_For_BOM cannot
read past the end of file. In practice it is probably the
case that this cannot cause a real error, but valgrind can
see that this is happening. So this change will avoid the
annoying false positive from valgrind. It's not worth setting
up
The pragma Warnings (Off, msg) is intended only to delete warnings,
but it was deleting real errors as well. This patch corrects that
mistake.
The following compiles with the indicated message (previously
this message was suppressed).
1. procedure ErrMsgDelete is
2.X : Integer;
Hi All,
Here is a simple fix which improves SLM performance by avoiding memory
operand in conditional move instructions. Our measurements have shown
significant speedup (~+10%) for routelookup bench from eembc2.0 suite.
Bootstrapping and regression testing were successful for x86-64.
Is it OK fo
This patch modifies the processing of aspect/pragma SPARK_Mode to properly
handle the cases where the aspect/pragma apply to a [library-level] package or
subprogram [body].
-- Source --
-- func.ads
function Func return Integer with SPARK_Mode => On;
-- pack.adb
pack
This patch handles properly additional constructs of the form F.all (I), where
F is an access_to_function that can be called without parameters, and that
returns an array type.
Compiling err.adb must yield:
err.adb:23:09: too many arguments in call to "A"
err.adb:24:09: too many arguments in cal
A class-wide type has anonymous discriminants, because type extensions can
add discriminants at will. A constraint on a class-wide type is thus a partial
constraint that applies only to the known discriminants of the root type. Such
a partial constraint is a language pathology that the ARG has dec
On 09/10/2013 05:57 AM, Kirill Yukhin wrote:
> Do you still think we need "*"?
No, I suppose that's fine.
r~
On Tue, Sep 10, 2013 at 9:10 AM, Jakub Jelinek wrote:
> On Tue, Sep 10, 2013 at 08:55:35AM -0700, Caroline Tice wrote:
>> Based on the errors you are reporting, my guess is that you did not
>> configure with --enable-vtable-verifiy. The testsuite in libvtv will
>
> I'm not, but it doesn't make se
Hi,
Following Bin's patch in
http://gcc.gnu.org/ml/gcc-patches/2013-09/msg00695.html, this patch
tweaks backtrace_base_for_ref () to strip of any widening conversion
after the first TREE_CODE check fails. Without this patch, the test
(gcc.dg/tree-ssa/slsr-39.c) in Bin's patch will fail on AA
On Tue, Sep 10, 2013 at 3:17 AM, Jakub Jelinek wrote:
> On Mon, Sep 09, 2013 at 01:56:29PM -0700, Caroline Tice wrote:
>> The patch looks good to me, but somebody else needs to approve it.
>
> Why? You are listed as libvtv maintainer, the patch is fully contained in
> libvtv, thus you can review
On Wed, Sep 04, 2013 at 03:29:43PM +0100, Gary Benson wrote:
> I've added the result to the demangler test suite, but I know of
> no way to check the validity of the demangled symbol other than by
> inspection (and I am no expert here!) If anybody knows a way to
> check this then please let me kno
> "Jakub" == Jakub Jelinek writes:
Jakub> On Tue, Sep 10, 2013 at 09:04:41AM -0400, Andrew MacLeod wrote:
>> Really? If it were useful I would have thought they'd be using it now.
>>
>> In any case, it doesn't belong in tree-flow.h.. . shouldn't these
>> functions and macros be in hashtab.h
This patch corrects the analysis of pragma SPARK_Mode to handle the case where
the pragma appears without an argument.
It also corrects the analysis of aspect/pragma Global to propery process
an item that appears as a selected component and is later converted into an
expanded name.
Tested on x86_
Oops, now attach the correct patch and change log.
Thanks,
Yufeng
gcc/
* config/aarch64/aarch64-builtins.c (aarch64_simd_expand_args):
Call aarch64_simd_expand_args to update op[argc].
On 09/10/13 18:08, Yufeng Zhang wrote:
This patch fixes a number of test failures in gcc.ta
Hi,
This seems to have caused PR58373. The bug occurs where GEN_INT would
previously have then been used to build a constant of vector mode.
These pop in a few places when building for AArch64, though I did
the debugging using gcc.target/aarch64/vect-fcm-eq-d.c
Here we could get in to the situa
The following patch switches on optional reloads. The optional
reload patch resolves PR55342. The original code was switched off as it
resulted in new GCC failures. The current code contains a new
version of optional reload implementation. I've tried many different
strategies to use optional
PR target/58361 is a wrong code bug where we treat an instruction as
conditional but fail to put the conditional marker on the assembly.
Since the instruction can be correctly executed as a conditional
instruction, the fix is to enable that and to put out the condition code
accordingly.
The partic
> Yeah, but that's because strip_address_mutations doesn't consider
> SIGN_EXTRACT to be a "mutation" as things stand. My point was that
> I think it should, at least for the special extract-from-lsb case.
> It then shouldn't be necessary to handle SIGN_EXTRACT in the other
> address-analysis rout
Hello,
On 10 Sep 09:17, Richard Henderson wrote:
> On 09/10/2013 05:57 AM, Kirill Yukhin wrote:
> > + { OPTION_MASK_ISA_AVX512F, CODE_FOR_andhi_1, "__builtin_ia32_kandhi",
> > IX86_BUILTIN_KAND16, UNKNOWN, (int) HI_FTYPE_HI_HI },
>
> Alternately, why not use the standard CODE_FOR_andhi3 expander
On Tue, Sep 10, 2013 at 4:56 PM, Yuri Rumyantsev wrote:
> Here is a simple fix which improves SLM performance by avoiding memory
> operand in conditional move instructions. Our measurements have shown
> significant speedup (~+10%) for routelookup bench from eembc2.0 suite.
>
> Bootstrapping and r
I believe I know how to fix this problem; I am in the middle of
testing my patch. I should be able to submit the patch either late
today or early tomorrow. In the meantime, if this is impacting your
ability to make progress, you can configure gcc with --disable-libvtv
and that should remove the p
On 09/10/2013 11:25 AM, Kirill Yukhin wrote:
> Hello,
> On 10 Sep 09:17, Richard Henderson wrote:
>> On 09/10/2013 05:57 AM, Kirill Yukhin wrote:
>>> + { OPTION_MASK_ISA_AVX512F, CODE_FOR_andhi_1, "__builtin_ia32_kandhi",
>>> IX86_BUILTIN_KAND16, UNKNOWN, (int) HI_FTYPE_HI_HI },
>>
>> Alternately
Yvan Roux writes:
>> Yeah, but that's because strip_address_mutations doesn't consider
>> SIGN_EXTRACT to be a "mutation" as things stand. My point was that
>> I think it should, at least for the special extract-from-lsb case.
>> It then shouldn't be necessary to handle SIGN_EXTRACT in the other
On 09/10/2013 05:57 AM, Kirill Yukhin wrote:
> + { OPTION_MASK_ISA_AVX512F, CODE_FOR_andhi_1, "__builtin_ia32_kandhi",
> IX86_BUILTIN_KAND16, UNKNOWN, (int) HI_FTYPE_HI_HI },
Alternately, why not use the standard CODE_FOR_andhi3 expander?
r~
James Greenhalgh writes:
> This seems to have caused PR58373. The bug occurs where GEN_INT would
> previously have then been used to build a constant of vector mode.
>
> These pop in a few places when building for AArch64, though I did
> the debugging using gcc.target/aarch64/vect-fcm-eq-d.c
>
> H
Hi,
On Fri, Sep 06, 2013 at 11:19:18AM +0200, Richard Biener wrote:
> On Fri, Sep 6, 2013 at 10:35 AM, Bernd Edlinger
> wrote:
> > Richard,
> >
> >> But the movmisalign path skips all this code and with the
> >> current code thinks the actual access is in the mode of the
> >> whole structure. (an
I added a new macro to replace the use of TREE_TYPE to get
constraints. It's called TEMPLATE_PARMS_CONSTRAINTS. Patch attached:
2013-09-10 Andrew Sutton
* gcc/cp/cp-tree.h (TEMPLATE_PARMS_CONSTRAINTS): New.
* gcc/cp/parser.c (cp_parser_template_declaration_after_export),
(cp_par
On Tue, Sep 10, 2013 at 08:36:02PM +0200, Paolo Carlini wrote:
> On 09/10/2013 05:34 PM, Jakub Jelinek wrote:
> >On Wed, Sep 04, 2013 at 03:29:43PM +0100, Gary Benson wrote:
> >>I've added the result to the demangler test suite, but I know of
> >>no way to check the validity of the demangled symbol
On 09/10/2013 08:38 PM, Jakub Jelinek wrote
Agreed, can you please revert it?
Sure, I'll do that momentarily.
Paolo.
Hi,
On 09/10/2013 05:34 PM, Jakub Jelinek wrote:
On Wed, Sep 04, 2013 at 03:29:43PM +0100, Gary Benson wrote:
I've added the result to the demangler test suite, but I know of
no way to check the validity of the demangled symbol other than by
inspection (and I am no expert here!) If anybody kno
Hi,
this seems to be a trivial one.
ChangeLog
2013-09-10 Kai Tietz
PR/54314
* config/abi/pre/gnu-versioned-namespace.ver: Add thunk _ZTv0_n12_NS* like
in gnu,ver.
Tested for i686-w64-mingw32, x86_64-w64-mingw32. Ok for apply?
Regards,
Kai
Index: config/abi/pre/gnu-versioned-n
This splits out tree-ssaname related things to tree-ssanames.h. This is
then included from tree-ssa.h
similar treatment can be given to tree-phinodes.c
I notice a number of the other ssa passes only export a couple of
functions, and thats it.. no structs or anything like that. (like
tree-ss
Hi,
All the latent issues wthat ipa-devirt has triggered has been hopefully
fixed. So I am enabling ipa-devirt with LTO.
To answer original David's comment on non-LTO performance, I did not really
noticed
any significant improvements/dergradations at -O2 with our benchmarks (SPEC2k,
2k6
and C++
On Sep 9, 2013, at 4:49 PM, Mike Stump wrote:
>> I don't see how it can be appropriate here; my impression
>> is that this code should never fail for any compiler input,
Oh, I might have missed what you meant by this… If so, sorry… The nicest of
options would be to generate a libcall for the
Hi!
If exp has side effects, for [-,-] ranges we would optimize them away.
Fixed thusly, bootstrapped/regtested on x86_64-linux and i686-linux, ok for
trunk/4.8?
2013-09-10 Jakub Jelinek
PR tree-optimization/58385
* fold-const.c (build_range_check): If both low and high are NU
This patch fixes a number of test failures in gcc.target/aarch64/v*.c in
ILP32.
The corresponding RTL patterns for some load/store builtins have Pmode
(i.e. DImode) specified for their address operands. However, coming
from a pointer-typed function argument, op[argc] will have SImode in
ILP3
The following patch enables GCC to record system include paths in
module infos. This allows more precise parsing context to be
established when compiling auxiliary modules -- FE behaves slightly
differently when parsing system headers. There are also a couple of
other cleanups in the patch.
David
On Tue, Sep 10, 2013 at 08:55:35AM -0700, Caroline Tice wrote:
> Based on the errors you are reporting, my guess is that you did not
> configure with --enable-vtable-verifiy. The testsuite in libvtv will
I'm not, but it doesn't make sense if you don't configure gcc in certain
non-default way, you
Because deps_analyze_insn only analyzes data deps but no control deps.
Control deps are included by add_branch_dependences. Without the
chain_to_prev_insn in the end of add_branch_dependences, jmp will be
control dependent on every previous insn in the same bb, and the cmp
and jmp group could still
On Tue, Sep 10, 2013 at 8:55 AM, Caroline Tice wrote:
> On Tue, Sep 10, 2013 at 3:17 AM, Jakub Jelinek wrote:
>> On Mon, Sep 09, 2013 at 01:56:29PM -0700, Caroline Tice wrote:
>>> The patch looks good to me, but somebody else needs to approve it.
>>
>> Why? You are listed as libvtv maintainer, t
On Tue, Sep 10, 2013 at 08:09:42PM +0100, Richard Sandiford wrote:
> Sorry for the breakage. gen_int_mode and GEN_INT really are only for
> scalar integers though. (So is plus_constant.) Vector constants should
> be CONST_VECTORs rather than CONST_INTs.
>
> I think the gcc.target/aarch64/vect-f
Hello!
> I've added references to fxsr, xsave and xsaveopt options and builtins
> to doc/[invoke,extend].texi.
>
> Is it OK?
>
> 2013-09-10 Alexander Ivchenko
>
> * doc/invoke.texi: Document fxsr, xsave and xsaveopt options.
> * doc/extend.texi: Document fxsr, xsave and xsaveopt builtins.
Pl
Andrew Sutton writes:
| I added a new macro to replace the use of TREE_TYPE to get
| constraints. It's called TEMPLATE_PARMS_CONSTRAINTS. Patch attached:
|
| 2013-09-10 Andrew Sutton
| * gcc/cp/cp-tree.h (TEMPLATE_PARMS_CONSTRAINTS): New.
| * gcc/cp/parser.c (cp_parser_template_declar
1 - 100 of 109 matches
Mail list logo
