C++ PATCH for c++/51009 (ICE regression with aggregate initializer)

The problem here was that split_nonconstant_init depends on the value of stmts_are_full_exprs_p(), but it wasn't being set properly at namespace scope. Fixed thus. build_aggr_init_full_exprs seems to be an ancient workaround for this same issue. Tested x86_64-pc-linux-gnu, applied to trunk.

Re: [C++ Patch] PR 51327

On 11/30/2011 06:41 PM, Paolo Carlini wrote: Ok. The point is, locate_ctor turns an error_mark_node returned by locate_fn_flags - meaning indeed not callable - into NULL_TREE. In fact, uses elsewhere of locate_ctor / get_default_ctor always check for it. Ah, I see. I guess what we want here is

[pph] Prepare for not streaming identifier bindings. (issue5437095)

The identifier bindings are causing trouble in merging. This patch is phase one of getting rid if binding information for identifiers within pph files. The bindings will be reconstructed after pph files have been read in. The core code enabling this switch has been disabled. We determine when a

[patch] original function and TM clone has to be marked needed

In the current version, the original function and its clone are marked as taken with cgraph_mark_address_taken_node. But it seems not enough and it has to be marked as needed. It comes with a testcase (testsuite/g++.dg/tm/ctor-used.C). Passed all TM tests. PS: Note that there is still a proble

Re: [Patch Darwin] adjust crts for Darwin >= 10.

On Nov 30, 2011, at 7:27 AM, Iain Sandoe > wrote: The vendor's tools no longer use crts for dylibs or bundles for Darwin 10. They also use "crt1.10.6.o" for exes. OK for trunk? Ok.

Re: [Patch] Remove flag_objc-sjlj_exceptions as a User control, move ObjC exception mechanism choices to gcc/objc/

On Nov 30, 2011, at 3:32 PM, Iain Sandoe > wrote: This removes the flag_objc_sjlj_exceptions (or, more correctly, localizes its effect to the ObjC FEs). OK for trunk? Ok.

Re: [C++ Patch] PR 51367

OK. Jason

[committed] Add new stubs to config/pa/stublib.c

The addition of libitm broke bootstrap on hppa64-hpux. The attached patch adds two new stubs to avoiding link errors and fixes a number of warnings caused by misplaced __attribute__'s. Tested on hppa64-hp-hpux11.11. Committed to trunk. Dave -- J. David Anglin d

Re: [RFC] Port libitm to powerpc

On Wed, Nov 30, 2011 at 8:05 PM, Richard Henderson wrote: > This is a tad rough, but not too bad. Cool. Maybe I don't understand what they are suppose to represent, but why the choice of values for cacheline size? Is that suppose to be a value chosen by ITM or suppose to be the hardware cacheli

Re: [PATCH] Re: [v3] updated atomic configury

Thanks Andrew. Here's the updated patch, as checked in. This removes one level of cruft WRT atomic configure-ness (leaving us with the most primordial...). tested x86/linux tested x86/linux x arm-eabi tested x86/linux x cris-elf -benjamin 2011-11-30 Benjamin Kosnik * acinclude.m4 (

[committed] Disallow frame related insns in branch delay solots on PA

A change to the EH stack adjust code made it necessary to disallow frame related instructions in branch delay slots on the parisc architecture. Tested on hppa2.0w-hp-hpux11.11, hppa64-hp-hpux11.11 and hppa-unknown-linux-gnu. Committed to trunk. Dave -- J. David Anglin

[google] dump inline decisions to stderr under -fopt-info

This patch is for google-{main|gcc_4.6} only. Tested with bootstrap and regression tests. Dump inline decisions, also output the inline chain. Dehao 2011-12-01 Dehao Chen * ipa-inline.c (dump_inline_decision): New function. (inline_small_functions): Use it to dump the inline

[RFC] Port libitm to powerpc

This is a tad rough, but not too bad. No hwcap stuff. Probably something similar to what I did for ARM will work. I'd prefer that someone else figure out what to do with all that, honestly. No support for non-ELF, aka AIX and Darwin. I'm not 100% sure how to handle the assembly markup for th

[PATCH 1/5] arm: Convert to atomic optabs.

At the same time, perform post-reload splitting. --- gcc/config/arm/arm-protos.h |7 +- gcc/config/arm/arm.c | 816 +++-- gcc/config/arm/arm.h | 18 - gcc/config/arm/arm.md | 26 +- gcc/config/arm/constraints.md |5 + gcc

[PATCH 5/5] arm-linux: Add libitm support.

--- libitm/Makefile.am |3 + libitm/Makefile.in | 20 +++-- libitm/config/arm/hwcap.cc | 67 + libitm/config/arm/hwcap.h| 41 ++ libitm/config/arm/sjlj.S | 135 +

[PATCH 2/5] arm: Emit swp for pre-armv6.

--- gcc/config/arm/arm.h |6 gcc/config/arm/sync.md | 63 +++- 2 files changed, 68 insertions(+), 1 deletions(-) diff --git a/gcc/config/arm/arm.h b/gcc/config/arm/arm.h index 31f4856..33e5b8e 100644 --- a/gcc/config/arm/arm.h +++ b/gcc/con

[PATCH 4/5] arm: Set predicable on more instructions.

Make sure its set for all CMP, CMN, TST instructions, which do work inside IT blocks. Split the TEQ pattern so that it can be predicated in ARM mode for better compare-and-swap generation. --- gcc/config/arm/arm.md | 40 ++-- 1 files changed, 30 insertions(+)

[PATCH 3/5] arm: Use CC_REGNUM as success output from compare-and-swap.

--- gcc/config/arm/arm.c | 52 +--- gcc/config/arm/arm.md |4 +- gcc/config/arm/sync.md | 62 +-- 3 files changed, 62 insertions(+), 56 deletions(-) diff --git a/gcc/config/arm/arm.c b/gcc/config/arm/arm.c

[RFC][PATCH v2 0/5] Convert ARM to atomics, libitm

Changes since v1: * Emit SWP/SWPB for pre-armv6, non-linux (i.e. no external __sync routines). The assumption here is that we must be on some sort of embedded OS, and further that the system is not SMP. I'm more or less assuming that all of the embedded ARM multi-core are at least

Re: [C++ Patch] PR 51327

Hi, On 11/30/2011 04:42 PM, Paolo Carlini wrote: add_implicitly_declared_members sets TYPE_HAS_DEFAULT_CONSTRUCTOR (t) = 1 for B, but then, in explain_non_literal_class, the true value doesn't imply that locate_ctor actually finds the constructor, due to the nature of A. It should find a delete

[C++ Patch] PR 51367

Hi, for c++/51230 I tried to change unify_inconsistency to unconditionally use %qE but apparently that doesn't really work: quickly we end up duplicating the whole dump_type inside dump_expr, first POINTER_TYPE (as per this specific PR), and then REFERENCE_TYPE, and then OFFSET_TYPE, and then

Re: [Patch,AVR] Was/Fix: error linking lto1 for target avr

Steven Bosscher writes: > On Tue, Nov 29, 2011 at 6:04 PM, Georg-Johann Lay wrote: >> I attached a patch but I fail to find the right configure options for >> gcc/binutils as the testsuite complains >> >> ./avr/bin/ld: bad -plugin option >> >> Configured gcc with --enable-lto and binutils 2.21 w

Re: [Patch, fortran] PR 25708 Reduce seeks when loading module files

Dominique pointed out an extra unused variable in peek_atom(). Updated patch attached On Thu, Dec 1, 2011 at 00:04, Janne Blomqvist wrote: > On Wed, Nov 30, 2011 at 22:36, Mikael Morin wrote: >> Hi, >> >>> >>> modseek1.diff >>>   diff --git a/gcc/fortran/module.c b/gcc/fortran/module.c >>> index

[PATCH] Swap the action and jmpbuf arguments to GTM_longjmp

On several ABIs, the first register parameter is the same as the return value register. Moving the action parameter first avoids the need for register shuffling on these targets. * libitm_i.h (GTM_longjmp): Swap first and second arguments. * beginend.c (_ITM_abortTransaction): Upd

Predict compare-and-swap to succeed.

This came up while amodra and I were discussing libgomp vs atomic builtins, where he was annotating all of the compare-and-swap instances with __builtin_expect. Which is a bit silly, since any users almost certainly want the same treatment. This isn't 100% ideal, since the resulting prediction

Re: [Patch, fortran] PR 25708 Reduce seeks when loading module files

On Wed, Nov 30, 2011 at 22:36, Mikael Morin wrote: > Hi, > >> >> modseek1.diff >>   diff --git a/gcc/fortran/module.c b/gcc/fortran/module.c >> index 70f8565..982425d 100644 >> --- a/gcc/fortran/module.c >> +++ b/gcc/fortran/module.c >> @@ -1069,51 +1069,49 @@ module_unget_char (void) >>  static v

Re: [Patch,AVR] Was/Fix: error linking lto1 for target avr

On Tue, Nov 29, 2011 at 6:04 PM, Georg-Johann Lay wrote: > I attached a patch but I fail to find the right configure options for > gcc/binutils as the testsuite complains > > ./avr/bin/ld: bad -plugin option > > Configured gcc with --enable-lto and binutils 2.21 with --enable-plugin. Huh, wait. I

Re: [PATCH] Fix emit_inc_dec_insn_before in DSE (PR rtl-optimization/51044)

On 11/30/2011 12:09 AM, Jakub Jelinek wrote: > PR rtl-optimization/51044 > * dse.c (emit_inc_dec_insn_before): Ensure new_insn is an instruction > rather than just insn body. Ok. r~

Re: [C++ Patch] PR 51327

On 11/30/2011 04:42 PM, Paolo Carlini wrote: add_implicitly_declared_members sets TYPE_HAS_DEFAULT_CONSTRUCTOR (t) = 1 for B, but then, in explain_non_literal_class, the true value doesn't imply that locate_ctor actually finds the constructor, due to the nature of A. It should find a deleted co

[C++ Patch] PR 51327

Hi, the below fixes an ICE on invalid which didn't occur in 4.6 (because we didn't produce the notes). The issue is that for this kind of test: struct A { A(int); }; struct B : A {}; constexpr int foo(B) { return 0; } add_implicitly_declared_members sets TYPE_HAS_DEFAULT_CONSTRUCTOR (t) =

Re: [IA64] Hookize PRINT_OPERAND, PRINT_OPERAND_ADDRESS and PRINT_OPERAND_PUNCT_VALID_P

On 11/30/2011 09:30 AM, Anatoly Sokolov wrote: > * config/ia64/ia64.h (PRINT_OPERAND, PRINT_OPERAND_ADDRESS, > PRINT_OPERAND_PUNCT_VALID_P): Remove macros. > * config/ia64/ia64-protos.h (ia64_print_operand, > ia64_print_operand_address): Remove. > * config/ia

Re: [PATCH] PR c++/51289 - ICE with alias template for bound template

Jason Merrill writes: > I guess let's check DECL_ORIGINAL_TYPE instead of TREE_TYPE for alias > templates. Like the below that I am currently bootstrapping? From: Dodji Seketeli Date: Sat, 26 Nov 2011 11:50:43 +0100 Subject: [PATCH] PR c++/51289 - ICE with alias template for bound template te

Re: [PATCH v2 0/5] Convert Sparc to atomic optabs

From: Richard Henderson Date: Wed, 30 Nov 2011 13:12:35 -0800 > On 11/30/2011 12:33 PM, David Miller wrote: >> From: Richard Henderson >> Date: Wed, 30 Nov 2011 10:40:36 -0800 >> >>> On 11/30/2011 09:44 AM, David Miller wrote: So one thing to consider doing is, if there is some doubt, to s

Re: [PATCH v2 0/5] Convert Sparc to atomic optabs

On 11/30/2011 12:33 PM, David Miller wrote: > From: Richard Henderson > Date: Wed, 30 Nov 2011 10:40:36 -0800 > >> On 11/30/2011 09:44 AM, David Miller wrote: >>> So one thing to consider doing is, if there is some doubt, to set >>> it explicitly to TSO in some crt code. >> >> ... easier to keep

Re: [PATCH] Avoid adding second edge between the same src and dest bb in ehcleanup (PR middle-end/51089)

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/30/11 10:58, Jakub Jelinek wrote: > Hi! > > This testcase ICEs starting with the {CLOBBER} additions, because > cleanup_empty_eh_merge_phis attempts to redirect an edge to a bb > from a bb that already has another edge to that same destination

Re: [PATCH] Fix shrink-wrapping detection with crtl->drap_reg (PR target/50725)

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/30/11 10:38, Jakub Jelinek wrote: > Hi! > > As the following testcase shows, when the prologue computes > crtl->drap_reg (when stack_realign_drap) and something e.g. in the > first bb uses that register, we can incorrectly shrink-wrap it and > u

Re: [Patch, fortran] PR 25708 Reduce seeks when loading module files

On Wednesday 30 November 2011 18:32:11 Janne Blomqvist wrote: > Hi, > > this patch expands a bit on the recent work done by Thomas Koenig. > Using aermod.f90 from the polyhedron benchmark suite as a test case, > the lseek() calls as reported by strace -c -f go roughly as > > - trunk before Thomas

Re: [PATCH v2 0/5] Convert Sparc to atomic optabs

From: Richard Henderson Date: Wed, 30 Nov 2011 10:40:36 -0800 > On 11/30/2011 09:44 AM, David Miller wrote: >> So one thing to consider doing is, if there is some doubt, to set >> it explicitly to TSO in some crt code. > > ... easier to keep the RMO/PSO default in that case. I'd rather you use

[Patch] Remove flag_objc-sjlj_exceptions as a User control, move ObjC exception mechanism choices to gcc/objc/

There is really no point in having a flag to control the Objective C (or Objective C++) exceptions [@throw, as opposed to throw] scheme, since we actually only support one scheme for each permutation of runtime and ABI. This removes the flag_objc_sjlj_exceptions (or, more correctly, lo

[Patch, fortran] Fix PR 51338 - ICE with front-end optimization and assumed character lengths

Hello world, the attached patch fixes PR 51338. I chose a different approach from what Bud did in the PR, because I rather wanted to fix the specific code path when two NULL expressions were handled to gfc_dep_compare_expr where determining the right answer is straightforward rather than possibl

Re: [PATCH] Fix C/51321, ICE With incomplete types with __builtin_types_compatible_p

On Wed, 30 Nov 2011, Andrew Pinski wrote: > > ChangeLog: > > * c-parser.c (c_parser_postfix_expression): Check groktypename results > > before looking at the main variant. > > > > testsuite/ChangeLog: > > * testsuite/gcc.dg/pr51321.c: New testcase OK. -- Joseph S. Myers jos...@codesourcery.com

Re: [Patch] Fix Bug 51162

On 11/24/2011 05:42 AM, Sameera Deshpande wrote: - if (TREE_CODE (fn) == ADDR_EXPR) + if (fn != NULL && TREE_CODE (fn) == ADDR_EXPR) fn = TREE_OPERAND (fn, 0); - if (TREE_CODE (fn) == FUNCTION_DECL && decl_is_tm_clone (fn)) + if (fn != NULL && TREE_CODE (fn) == FUNCTION_DECL && decl_is_t

Re: [PATCH] Fix C/51321, ICE With incomplete types with __builtin_types_compatible_p

Woops, I forgot to attach the patch. Thanks, Andrew Pinski On Wed, Nov 30, 2011 at 11:23 AM, Andrew Pinski wrote: > Hi, >  The problem here is that right after groktypename we take the > TYPE_MAIN_VARIANT of the result which is incorrect.  There needs a > check for error_mark_node.  This adds th

[PATCH] Fix C/51321, ICE With incomplete types with __builtin_types_compatible_p

Hi, The problem here is that right after groktypename we take the TYPE_MAIN_VARIANT of the result which is incorrect. There needs a check for error_mark_node. This adds the obvious check and fixes the ICE. OK? Bootstrapped and tested on x86_64-linux-gnu with no regressions. Thanks, Andrew Pi

[STORMY16] Hookize FUNCTION_VALUE_REGNO_P

Hello. This patch removes obsolete FUNCTION_VALUE_REGNO_P macro from ARM back end in the GCC and introduces equivalent TARGET_FUNCTION_VALUE_REGNO_P target hook. Since the LIBCALL_VALUE macro should be poisoned in the future, this patch replaces it with the arm_libcall_value_1 function.

[PATCH] Fix PR 51198, DECL_INITIAL still contains stuff for FIELD_DECLs

Hi, With C++11's decl initialization for non static members, the DECL_INITIAL for FIELD_DECLs contains stuff which we don't need to keep around after the front-end is done. This patch clears them in the free_lang_data pass. OK? Bootstrapped and tested on x86_64-linux-gnu with no regressions. T

Re: [google] Backport r171347 and r181549 from trunk (strict volatile bitfield) (issue5434084)

On 11/29/11 20:59, Doug Kwan wrote: Hi Diego, This is a backport for two upstream patches into our 4.6 branch. I submitted the first patch by Julian a while ago for backport but Richard Earnshaw pointed out a problem with the first patch. The second patch from Joey fixes that problem. Thi

Re: [PATCH v2 0/5] Convert Sparc to atomic optabs

On 11/30/2011 09:44 AM, David Miller wrote: > Linux originally did RMO but I killed that several years ago, no > new chips are implementing RMO and this means all the necessary > membar instructions are nothing but pure overhead. Well, you've now got the option for -mmemory-model=sc to omit every

Re: [C++ Patch] PR 51299

On 11/30/2011 06:27 PM, Jason Merrill wrote: On 11/24/2011 11:34 PM, Paolo Carlini wrote: + /* Avoid -Wzero-as-null-pointer-constant warnings. */ + ++c_inhibit_evaluation_warnings; neq = c_common_truthvalue_conversion (input_location, result); + --c_inhi

[Patch, Fortran] PR50923 - [4.4-4.7] Print warning function does not return a value

The bug has been introduced when changing the warning system to do more in the front end. The problem is that for: module m contains function f() end end the sym->attr.referenced gets set - and no warning is printed. I now ignore the sym->attr.referenced attribute as the RESULT ==

Re: [PATCH] Treat tail calls as scheduling barriers (PR rtl-optimization/48721)

On 11/30/2011 12:54 PM, Jakub Jelinek wrote: Hi! As discussed in the PR, I think tail calls are best handled as full scheduling barriers, scheduling tail calls before other insns effectively means we remove those other insns, but isn't dead insn removal the job of other passes instead? In any ca

Re: [PATCH] Fix up begin_move_insn (PR rtl-optimization/48721)

On 11/30/2011 12:43 PM, Jakub Jelinek wrote: Hi! When INSN is followed by BARRIER, we should add the new dummy basic block after the barrier, not before it. Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk? 2011-11-30 Jakub Jelinek PR rtl-optimization/48721

[PATCH] Avoid adding second edge between the same src and dest bb in ehcleanup (PR middle-end/51089)

Hi! This testcase ICEs starting with the {CLOBBER} additions, because cleanup_empty_eh_merge_phis attempts to redirect an edge to a bb from a bb that already has another edge to that same destination bb. Richard fixed a similar issue last year in one of the cleanup_empty_eh_merge_phis caller's, bu

[PATCH] Treat tail calls as scheduling barriers (PR rtl-optimization/48721)

Hi! As discussed in the PR, I think tail calls are best handled as full scheduling barriers, scheduling tail calls before other insns effectively means we remove those other insns, but isn't dead insn removal the job of other passes instead? In any case, tail calls need to come up after all restor

Re: [google] ThreadSanitizer instrumentation pass (issue 5303083)

ok for google branches. David On Wed, Nov 30, 2011 at 6:09 AM, wrote: > On 2011/11/30 13:17:04, Diego Novillo wrote: >> >> On 11-11-30 03:53 , mailto:dvyu...@google.com wrote: >> > On 2011/11/14 16:48:45, davidxl wrote: >> >> Ok for google/main after compiler bootstrap and regression test >> >>

Re: [PATCH v2 0/5] Convert Sparc to atomic optabs

From: Richard Henderson Date: Wed, 30 Nov 2011 08:39:44 -0800 > I've committed this sequence to mainline after re-testing on > sparc64-linux. I've got a regstrap running on Niagara-3, sorry for taking so long :-/ > As a follow-up, I have a patch to change the default memory model to > TSO for a

[PATCH] Fix up begin_move_insn (PR rtl-optimization/48721)

Hi! When INSN is followed by BARRIER, we should add the new dummy basic block after the barrier, not before it. Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk? 2011-11-30 Jakub Jelinek PR rtl-optimization/48721 * sched-ebb.c (begin_move_insn): Insert empt

[PATCH] Avoid messages about non-existent gnatls command (PR bootstrap/51201)

Hi! := assigned variable gets evaluated right away, including the case when host doesn't have any Ada compiler installed. In that case we remove ada from enabled languages, but still RTS_DIR is sometimes computed. With this patch we could call gnatls 4 times (or even 6 times) instead of just onc

[PATCH] Fix shrink-wrapping detection with crtl->drap_reg (PR target/50725)

Hi! As the following testcase shows, when the prologue computes crtl->drap_reg (when stack_realign_drap) and something e.g. in the first bb uses that register, we can incorrectly shrink-wrap it and use uninitialized %ecx/%r10 etc. register. Fixed by making the drap regno also added to SET_UP_BY_P

[Patch, fortran] PR 25708 Reduce seeks when loading module files

Hi, this patch expands a bit on the recent work done by Thomas Koenig. Using aermod.f90 from the polyhedron benchmark suite as a test case, the lseek() calls as reported by strace -c -f go roughly as - trunk before Thomas' patch: 21 million - current trunk: 5.7 million - with attached patch: 2.

[IA64] Hookize PRINT_OPERAND, PRINT_OPERAND_ADDRESS and PRINT_OPERAND_PUNCT_VALID_P

Hello. This patch removes obsolete PRINT_OPERAND, PRINT_OPERAND_ADDRESS and PRINT_OPERAND_PUNCT_VALID_P macros from IA64 back end in the GCC and introduces equivalent TARGET_PRINT_OPERAND, TARGET_PRINT_OPERAND_ADDRESS and TARGET_PRINT_OPERAND_PUNCT_VALID_P target hooks.  Bootstrapped and re

Re: [C++ Patch] PR 51299

On 11/24/2011 11:34 PM, Paolo Carlini wrote: + /* Avoid -Wzero-as-null-pointer-constant warnings. */ + ++c_inhibit_evaluation_warnings; neq = c_common_truthvalue_conversion (input_location, result); + --c_inhibit_evaluation_warnings; How about

Re: [PATCH] PR c++/51289 - ICE with alias template for bound template

On 11/30/2011 06:32 AM, Dodji Seketeli wrote: Here is an example of what I had in mind for keeping the walking of the underlying type there: template struct S {}; template using A = S;//#1<-- I think we should error here. We certainly should; we do need to walk the pattern

Re: [PATCH, lto]: Handle *tm regparm attribute

> Hello! > > Attached patch handles "*tm regparm" attribute, to avoid "*tm regparm" > attribute ignored warnings in lto compile on non-x86 targets. > > 2011-11-30 Uros Bizjak > > * lto-lang.c (lto_attribute_table): Handle *tm regparm. > (ignore_attribute): New. > > Tested on alph

Re: [Patch,AVR] Was/Fix: error linking lto1 for target avr

Georg-Johann Lay writes: > I just can't find the right configure options to get the LTO stuff > work, i.e. how to enable plugins in binutils and LTO in gcc. For binutils: configure with --enable-plugins (note: not --enable-plugin). For gcc: should be on by default, or configure with --enable-lt

Re: [PATCH v2 0/5] Convert Sparc to atomic optabs

On 11/28/2011 10:12 AM, Richard Henderson wrote: > Richard Henderson (5): > sparc: Convert to mem_thread_fence. > sparc: Convert to atomic_load/store. > sparc: Convert to atomic_compare_and_swap. > sparc: Convert to atomic_exchange. > sparc: Add -mmemory-model command-line option. > > g

Re: [PATCH] Implement stap probe on ARM's unwinder

Sergio Durigan Junior writes: > Hello, > > This is the implementation of > for the ARM > unwinder. Since ARM has a different unwinder, I basically replicated > the existing code (on unwind-dw2.c) into it, with a few modifications in > ord

Re: [lra] patch mostly implementing pseudo live range split

On 11/30/2011 12:09 AM, Jeff Law wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/29/11 15:03, Vladimir Makarov wrote: This patch contains a part of my several last weeks work on LRA. 1. The major change is to add pseudo live range split transformation to the existing inheritance tra

Re: Run ThreadSanitizer test only on x86/linux (issue 5437087)

On 2011/11/30 14:08:34, dvyukov wrote: The patch is for google-main branch. Add directives to run ThreadSanitizer tests only on i386/x86_64-*-linux targets. Index: gcc/ChangeLog.google-main === --- gcc/ChangeLog.google-main (rev

[Patch PPC/Darwin] some tidy-ups for save_world (and a prelude to splitting it out of the rs6000 code).

While trying to track down the vector unwind problems on ppc-darwin, I made some tidy-ups for "save_world()". In the end, that was not where the main problem, lay - but I did find a few things wrong there on the way - they should be fixed, even if there's no specific bug filed at present. I

Run ThreadSanitizer test only on x86/linux (issue5437087)

The patch is for google-main branch. Add directives to run ThreadSanitizer tests only on i386/x86_64-*-linux targets. Index: gcc/ChangeLog.google-main === --- gcc/ChangeLog.google-main (revision 181841) +++ gcc/ChangeLog.google-main

[RFA/testsuite] Use rand instead of random (again)

All, The attached patch changes another use of random in the testsuite to rand. Tested on an arm-none-eabi target using QEmu. Please can someone review. Thanks, Matt gcc/testsuite/ChangeLog: 2011-11-30 Matthew Gretton-Dann * gcc.dg/torture/vec-cvt-1.c (FLTTEST): Call rand instea

Re: [RFA/ARM] Add an integer pipeline description for Cortex-A15

On 29/11/11 11:04, Ramana Radhakrishnan wrote: gcc/ChangeLog: 2011-11-28 Matthew Gretton-Dann * config/arm/arm.c (arm_issue_rate): Cortex-A15 can triple issue. * config/arm/arm.md (mul64): Add new attribute. (generic_sched): Cortex-A15 is not scheduled genericall

Re: [google] ThreadSanitizer instrumentation pass (issue 5303083)

On 11-11-30 03:53 , dvyu...@google.com wrote: On 2011/11/14 16:48:45, davidxl wrote: Ok for google/main after compiler bootstrap and regression test (without ftsan), and some large tests with tsan turned on (as many as you can but at your discretion). Hi David, Perhaps a bit late question...

Re: [Patch.AVR,4.6] Fix PR51002

As Georg-Johann Lay wrote: > Then avr-mcus.def adopted this bug from the manual for ATtiny4313 at least: > > AVR_MCU ("attiny4313", ARCH_AVR25, "__AVR_ATtiny4313__", 1 /* short_sp, should > be 0 ? */, 0, 0x0060, "tn4313") Not unlikely. I just ordered one. Hopefully, it will be here by tomorrow

[Patch Darwin] adjust crts for Darwin >= 10.

The vendor's tools no longer use crts for dylibs or bundles for Darwin 10. They also use "crt1.10.6.o" for exes. our crts amended to match (checked on i686-darwin9 and x86-64-darwin10). OK for trunk? Iain gcc: config/darwin.h (STARTFILE_SPEC): Do not use -lbundle1.o when Darwin >= 10.

Re: [Patch.AVR,4.6] Fix PR51002

Joerg Wunsch wrote: > As Georg-Johann Lay wrote: > >> There is ATtiny4313 (and maybe others) that have 8-bit SP and 0x100 RAM. >> As RAM starts at 0x60, I wonder what the meaning of SP is? > > I think this is simply a bug in the datasheet. The device description > XML file of the ATtiny4313 ment

Re: [PATCH] PR c++/51289 - ICE with alias template for bound template

Jason Merrill writes: > On 11/29/2011 10:41 AM, Dodji Seketeli wrote: >> + cp_walk_tree (&DECL_ORIGINAL_TYPE (TYPE_NAME (t)), >> +&find_parameter_packs_r, >> +ppd, ppd->visited); > > I still don't think we want to walk the underlying type, since it > isn't part of

[PATCH, lto]: Handle *tm regparm attribute

Hello! Attached patch handles "*tm regparm" attribute, to avoid "*tm regparm" attribute ignored warnings in lto compile on non-x86 targets. 2011-11-30 Uros Bizjak * lto-lang.c (lto_attribute_table): Handle *tm regparm. (ignore_attribute): New. Tested on alphaev68-pc-linux-gnu

Re: [Patch.AVR,4.6] Fix PR51002

As Georg-Johann Lay wrote: > There is ATtiny4313 (and maybe others) that have 8-bit SP and 0x100 RAM. > As RAM starts at 0x60, I wonder what the meaning of SP is? I think this is simply a bug in the datasheet. The device description XML file of the ATtiny4313 mentions an SPH register, and it wou

Re: [Patch,AVR] Light-weight DImode implementation.

On 11/22/2011 01:15 AM, Georg-Johann Lay wrote: ldi r30,lo8(1) ; 25*movqi/2[length = 1] cp r10,r18 ; 26*cmpqi/2[length = 1] brlo .L2 ; 27branch[length = 1] ldi r30,lo8(0) ; 28*movqi/1[length = 1] .L2: add r11,r19 ;

Re: [Patch.AVR,4.6] Fix PR51002

Georg-Johann Lay wrote: > For devices with 8-bit SP reading the high byte SP_H of SP will get garbage. > > The patch uses CLR instead of IN SP_H to "read" the high part of SP. > > There are two issues with this patch: > > == 1 == > > I cannot really test it because for devices that small runni

[PATCH/Committed] Add self to write-after-approval section of MAINTAINERS file

All, I have just committed the attached patch to add myself to the write-after-approval section of the MAINTAINERS file. Thanks, Matt ChangeLog: 2011-11-30 Matthew Gretton-Dann * MAINTAINERS (write-after-approval): Add self. -- Matthew Gretton-Dann Principal Engineer, PD Softwa

Re: RFA: AVR: Provide local version of c_addr_space_name

Nick Clifton wrote: > Hi Denis, Hi Anatoly, Hi Eric, > > The patch below fixes the current problem building gcc for the avr-elf > toolchain by providing a local copy of the c_addr_space_name > function. I am not sure if the space names are correct, but at least > this allows me to build t

Re: Find more shrink-wrapping opportunities

Bernd Schmidt writes: > On 11/29/11 20:02, Richard Sandiford wrote: >> Bernd Schmidt writes: The reason I'm suddenly "reviewing" the code now is that it doesn't prevent shrink-wrapping, because nothing adds register 2 to the liveness info of the affected blocks. The temporary prol

RFA: AVR: Provide local version of c_addr_space_name

Hi Denis, Hi Anatoly, Hi Eric, The patch below fixes the current problem building gcc for the avr-elf toolchain by providing a local copy of the c_addr_space_name function. I am not sure if the space names are correct, but at least this allows me to build the port. OK to apply ? Cheer

Re: [Patch,AVR] Was/Fix: error linking lto1 for target avr

Denis Chertykov wrote: 2011/11/29 Georg-Johann Lay: I attached a patch but I fail to find the right configure options for gcc/binutils as the testsuite complains ./avr/bin/ld: bad -plugin option Configured gcc with --enable-lto and binutils 2.21 with --enable-plugin. Maybe the patch can be

Re: [PATCH] PR target/50038 fix: redundant zero extensions removal

Ping 2011/11/22 Ilya Enkovich : > 2011/11/11 Eric Botcazou : >>> I have already signed copyright agreement with the FSF. Will I need>> the >>> separate one for this particular commit?>> No, if your contributions are >>> already covered by a copyright agreement with the> FSF, nothing more needs

[PATCH] Fix emit_inc_dec_insn_before in DSE (PR rtl-optimization/51044)

Hi! While gen_move_insn returns an INSN (or more), gen_add3_insn can return just the pattern of an insn. The rest of the function and the helper called through note_stores assumes it is always instruction, so this patch ensures it. Bootstrapped/regtested on x86_64-linux and i686-linux and tested