Re: Early jump threading

Hi, On 18 September 2016 at 20:48, Jan Hubicka wrote: > Hi, > this is the patch compensating testsuite I commited after re-testing > on x86_64-linux. > > Other placements of early_thread_jumps does not work veyr well (at least in > current implementation). Putting it before forwprop disables abo

[PATCH] More trivial bits from early LTO debug merge

This merges moving filename and related CU annotation from late finish to early finish. With this all changes to cgraphunit.c have been merged. [LTO] Bootstrap and test running on x86_64-unknown-linux-gnu. Richard. 2016-09-19 Richard Biener * debug.h (gcc_debug_hooks): Add filename

[PATCH] manual: _Bool has trap representations

Defined behavior for extraordinary values of _Bool would be preferable, but we can at least document what we have now. Thanks, Florian 2016-09-19 Florian Weimer * doc/implement-c.texi (Integers implementation): _Bool has trap representations. Index: gcc/doc/implement-c.te

Re: [PR lto/77458] Avoid ICE in offloading with differing _FloatN, _FloatNx types (was: Advice sought for debugging a lto1 ICE (was: Implement C _FloatN, _FloatNx types [version 6]))

On Fri, Sep 16, 2016 at 7:07 PM, Joseph Myers wrote: > On Fri, 16 Sep 2016, Thomas Schwinge wrote: > >> That's what I was afraid of: for example, I can't tell if it holds for >> all GCC configurations (back ends), that complex types' component types >> will always match one of the already existing

Re: [PR lto/77458] Avoid ICE in offloading with differing _FloatN, _FloatNx types (was: Advice sought for debugging a lto1 ICE (was: Implement C _FloatN, _FloatNx types [version 6]))

On Fri, Sep 16, 2016 at 3:32 PM, Thomas Schwinge wrote: > Hi! > > On Fri, 16 Sep 2016 10:59:16 +0200, Richard Biener > wrote: >> On Fri, Sep 16, 2016 at 9:05 AM, Thomas Schwinge >> wrote: >> > On Thu, 08 Sep 2016 13:43:30 +0200, I wrote: >> >> On Wed, 7 Sep 2016 14:23:18 +0200, Richard Biener

Re: [PATCH 2/2][AArch64] Add missing support for poly64x1_t

On 17/09/16 07:20, James Greenhalgh wrote: Hi James, The convention when fixing a bugzilla is to place a line in your ChangeLog which looks like: PR component/x Oh, I wasn't aware of this, thanks! I'll add it. FAIL: gcc.target/aarch64/advsimd-intrinsics/vdup-vmov.c -O0

Re: [accaf, fortran, 4/4] Allocatable components support in derived typed coarrays

This time in plain text! On 19 September 2016 at 10:48, Paul Richard Thomas wrote: > Dear Andre, > > I have taken a look at the patches that you posted at: >> >> >> 0 of 4: https://gcc.gnu.org/ml/fortran/2016-08/msg00189.html >> 1 of 4: https://gcc.gnu.org/ml/fortran/2016-08/msg00190.html >> 2 of

Re: [PATCHv2, PING 4][ARM] -mpure-code option for ARM

On 30/08/16 09:01, Andre Vieira (lists) wrote: > On 11/08/16 15:13, Andre Vieira (lists) wrote: >> On 25/07/16 11:52, Andre Vieira (lists) wrote: >>> On 11/07/16 17:56, Andre Vieira (lists) wrote: On 07/07/16 13:30, mickael guene wrote: > Hi Andre, > > Another feedback on your pur

Re: Early jump threading

> On Mon, Sep 19, 2016 at 2:48 AM, Jan Hubicka wrote: > > Hi, > > this is the patch compensating testsuite I commited after re-testing > > on x86_64-linux. > > > > Other placements of early_thread_jumps does not work veyr well (at least in > > current implementation). Putting it before forwprop di

Re: [PATCH] Fix PR tree-optimization/77550

On Sun, 18 Sep 2016, Bernd Edlinger wrote: > Hi, > > this PR shows that in vectorizable_store and vectorizable_load > as well, the vector access always uses the first dr as the alias > type for the whole access. But that is not right, if they are > different types, like in this example. > > So

Re: Do not drom MEM_EXPR when accessing structure fields

On Sun, 18 Sep 2016, Jan Hubicka wrote: > > Hi, > > > > > when expanding code for > > > struct a {short a,b,c,d;} *a; > > > a->c; > > > > > > we first produce correct BLKmode MEM rtx representing the whole > > > structure *a, > > > then we use adjust_address to turn it into HImode MEM and

Re: Fortran, committed: segfault with allocate and stat for derived types with default initialization (pr 68078)

Hi, On 18 September 2016 at 21:19, Louis Krupp wrote: > Two possibilities: > > 1. malloc() doesn't silently return NULL on Darwin when it runs out of > memory; it always generates an error message. > > 2. setrlimit() doesn't work the same on Darwin as it does on Linux, and the > test program

Re: [PATCH] Allow FP to be used as a call-saved registe

On 15/09/16 17:36, Jeff Law wrote: > On 09/13/2016 05:10 AM, Tamar Christina wrote: >> Hi Jeff, >> >> >> On 12/09/16 18:16, Jeff Law wrote: >>> On 09/05/2016 08:59 AM, Tamar Christina wrote: Hi All, This patch allows the FP register to be used as a call-saved register when -fomi

Re: [PATCH][Libiberty] Support empty arguments in pex-win32

On 16/09/16 19:12, DJ Delorie wrote: This is OK. Thanks! Thanks, committed. Andrew

Re: [PR lto/77458] Avoid ICE in offloading with differing _FloatN, _FloatNx types (was: Advice sought for debugging a lto1 ICE (was: Implement C _FloatN, _FloatNx types [version 6]))

Hi! On Mon, 19 Sep 2016 10:12:48 +0200, Richard Biener wrote: > On Fri, Sep 16, 2016 at 7:07 PM, Joseph Myers wrote: > > On Fri, 16 Sep 2016, Thomas Schwinge wrote: > > > >> That's what I was afraid of: for example, I can't tell if it holds for > >> all GCC configurations (back ends), that comp

Re: [PATCH, PR71602, 4/4] Make canonical_va_list_type more strict

On 29/08/16 17:12, Tom de Vries wrote: > On 29/08/16 17:51, Joseph Myers wrote: >> On Wed, 24 Aug 2016, Tom de Vries wrote: >> >>> This patch fixes PR71602 by making canonical_va_list_type more strict. > 2016-08-22 Tom de Vries > > PR C/71602 > * builtins.c (std_canonical_va_list_t

Re: [PATCH, PR71602, 4/4] Make canonical_va_list_type more strict

On 19/09/16 13:02, Szabolcs Nagy wrote: On 29/08/16 17:12, Tom de Vries wrote: > On 29/08/16 17:51, Joseph Myers wrote: >> On Wed, 24 Aug 2016, Tom de Vries wrote: >> >>> This patch fixes PR71602 by making canonical_va_list_type more strict. > 2016-08-22 Tom de Vries > >PR C/71602 >

Re: [PR lto/77458] Avoid ICE in offloading with differing _FloatN, _FloatNx types (was: Advice sought for debugging a lto1 ICE (was: Implement C _FloatN, _FloatNx types [version 6]))

Hi! On Mon, 19 Sep 2016 10:18:35 +0200, Richard Biener wrote: > On Fri, Sep 16, 2016 at 3:32 PM, Thomas Schwinge > wrote: > > --- gcc/tree-core.h > > +++ gcc/tree-core.h > > @@ -553,20 +553,6 @@ enum tree_index { > >TI_BOOLEAN_FALSE, > >TI_BOOLEAN_TRUE, > > > > - TI_COMPLEX_INTEGER_TYP

Re: [PR lto/77458] Avoid ICE in offloading with differing _FloatN, _FloatNx types (was: Advice sought for debugging a lto1 ICE (was: Implement C _FloatN, _FloatNx types [version 6]))

On Mon, Sep 19, 2016 at 1:19 PM, Thomas Schwinge wrote: > Hi! > > On Mon, 19 Sep 2016 10:18:35 +0200, Richard Biener > wrote: >> On Fri, Sep 16, 2016 at 3:32 PM, Thomas Schwinge >> wrote: >> > --- gcc/tree-core.h >> > +++ gcc/tree-core.h >> > @@ -553,20 +553,6 @@ enum tree_index { >> >TI_BO

Re: [PATCH 0/9] RFC: selftests based on RTL dumps

On 09/16/2016 11:12 PM, David Malcolm wrote: There are some interrelated questions here: (a) where do the dumps live? (string fragments embedded in the source file vs external files) (b) -fself-test vs DejaGnu tests that use a real frontend. In the latter case, is the frontend "rtl1", or an exte

[PATCH] Fix PR c++/77639 (ICE during error recovery)

Some syntactically invalid code can confuse the compiler into trying to define a specialization while processing_template_parmlist != 0, during which there is a dummy parameter level (a TREE_VEC of length 0) on current_template_parms which the TEMPLATE_DECL of the specialization then inherits. Thi

Re: Do not drom MEM_EXPR when accessing structure fields

On 09/18/2016 02:51 PM, Jan Hubicka wrote: to_rtx = expand_expr (tem, NULL_RTX, VOIDmode, EXPAND_WRITE); - - /* If the field has a mode, we want to access it in the -field's mode, not the computed mode. -If a MEM has VOIDmode (external with incomplete type), -

[PATCH] libstdc++/77641 fix std::variant for literal types

As noted in bugzilla PR 77641 I don't think is_literal_type is the right condition for _Uninitialized, because a type can have a non-trivial default constructor but still be literal, but that means it can't be used in the union in _Variant_storage. I'm not sure if the right condition is is_litera

Re: [PR lto/77458] Avoid ICE in offloading with differing _FloatN, _FloatNx types (was: Advice sought for debugging a lto1 ICE (was: Implement C _FloatN, _FloatNx types [version 6]))

On Mon, 19 Sep 2016, Thomas Schwinge wrote: > > The question is whether such a complex type could be a global tree which I > > don't think it could. > > Specifically, my question was whether for every complex type that is part > of the global trees, it holds that the complex type's component type

Re: [RFC][IPA-VRP] Early VRP Implementation

On Sun, Sep 18, 2016 at 10:50 PM, kugan wrote: > Hi Richard, > > > > On 16/09/16 20:21, Richard Biener wrote: >> >> On Fri, Sep 16, 2016 at 7:59 AM, kugan >> wrote: >>> >>> Hi Richard, >>> >>> Thanks for the review. >>> >>> On 14/09/16 22:04, Richard Biener wrote: On Tue, Aug 23, 2

Re: [PATCH 2/2][AArch64] Add missing support for poly64x1_t

Hi, On 19 September 2016 at 10:39, Tamar Christina wrote: > > > On 17/09/16 07:20, James Greenhalgh wrote: > > > Hi James, > >> The convention when fixing a bugzilla is to place a line in your ChangeLog >> which looks like: >> >> PR component/x > > Oh, I wasn't aware of this, thanks!

Re: [PATCH 2/2][AArch64] Add missing support for poly64x1_t

On Mon, Sep 19, 2016 at 03:22:39PM +0200, Christophe Lyon wrote: > Hi, > > > On 19 September 2016 at 10:39, Tamar Christina > wrote: > > > > > > On 17/09/16 07:20, James Greenhalgh wrote: > > > > > > Hi James, > > > >> The convention when fixing a bugzilla is to place a line in your ChangeLog >

Re: [PR72835] Incorrect arithmetic optimization involving bitfield arguments

On Sun, Sep 18, 2016 at 10:21 PM, kugan wrote: > Hi Richard, > > > On 14/09/16 21:31, Richard Biener wrote: >> >> On Fri, Sep 2, 2016 at 10:09 AM, Kugan Vivekanandarajah >> wrote: >>> >>> Hi Richard, >>> >>> On 25 August 2016 at 22:24, Richard Biener >>> wrote: On Thu, Aug 11, 2016 at

Re: Do not drom MEM_EXPR when accessing structure fields

On 09/19/16 11:28, Richard Biener wrote: > On Sun, 18 Sep 2016, Jan Hubicka wrote: > >>> Hi, >>> >>> > when expanding code for >>> > struct a {short a,b,c,d;} *a; >>> > a->c; >>> > >>> > we first produce correct BLKmode MEM rtx representing the whole >>> > structure *a, >>> > then we

Re: Do not drom MEM_EXPR when accessing structure fields

On 19/09/16 15:49, Bernd Edlinger wrote: > On 09/19/16 11:28, Richard Biener wrote: >> On Sun, 18 Sep 2016, Jan Hubicka wrote: >> Hi, > when expanding code for > struct a {short a,b,c,d;} *a; > a->c; > > we first produce correct BLKmode MEM rtx represent

Re: [PATCH] Fix PR c++/77639 (ICE during error recovery)

>if (tree t = maybe_new_partial_specialization (type)) > { > + if (processing_template_parmlist) > + { > +/* Some syntactically invalid code can confuse the compiler into > + defining a specialization from inside a template parameter > +

Re: [PATCH][simplify-rtx] (GTU (PLUS a C) (C - 1)) --> (LTU a -C)

On 16/09/16 12:12, Kyrill Tkachov wrote: On 16/09/16 11:45, Bernd Schmidt wrote: On 09/16/2016 10:40 AM, Kyrill Tkachov wrote: 2016-09-16 Kyrylo Tkachov * simplify-rtx.c (simplify_relational_operation_1): Add transformation (GTU (PLUS a C) (C - 1)) --> (LTU a -C). 2016-09-16 Ky

RE: [PATCH 4/4] [MIPS] Add tests for MSA

Hi Robert, Sorry for the long delay. Just a couple of cleanup bits in here but otherwise OK. Unless you want me to read through again please go ahead and commit once you've addressed the comments. We could do more in mips.exp for inferring -mfp64 and -mhard-float but I'm not sure it is worth it c

Re: Do not drom MEM_EXPR when accessing structure fields

On 09/19/16 16:23, Richard Biener wrote: > On 19/09/16 15:49, Bernd Edlinger wrote: >> On 09/19/16 11:28, Richard Biener wrote: >>> On Sun, 18 Sep 2016, Jan Hubicka wrote: >>> > Hi, > >> when expanding code for >> struct a {short a,b,c,d;} *a; >> a->c; >> >>>

[PATCH test]Revise gcc.dg/vect/pr57558-1.c by using int instead of long

Hi, Case gcc.dg/vect/pr57558-1.c uses unsigned long at the moment which can't be vectorized on Sparc64. This patch revises it by using unsigned int. I think it's an obvious change. Test result checked, is it OK? Thanks, bin gcc/testsuite/ChangeLog 2016-09-15 Bin Cheng * gcc.dg/vec

Re: [PATCH, GCC/LRA] Teach LRA to not use same register value for multiple output operands of an insn

On 07/08/2016 11:07 AM, Thomas Preudhomme wrote: Hi, While investigating the root cause a testsuite regression for the ARM/embedded-5-branch GCC in gcc.dg/vect/slp-perm-5.c, we found that the bug seems to also affect trunk. The bug manifests itself as an ICE in cselib due to a parallel insn wi

Re: [PATCH] Fix PR c++/77639 (ICE during error recovery)

On Mon, 19 Sep 2016, Jason Merrill wrote: > >if (tree t = maybe_new_partial_specialization (type)) > > { > > + if (processing_template_parmlist) > > + { > > +/* Some syntactically invalid code can confuse the compiler > > into > > + defi

Re: [PATCH 1/3] Put a TARGET_LRA_P into every target

On Fri, Sep 16, 2016 at 12:04:23PM -0700, Mike Stump wrote: > On Sep 16, 2016, at 6:40 AM, Segher Boessenkool > wrote: > > > > Does just "The default version of this target hook returns true." sound > > better? I.e. delete "always". > > That is fine. Okay, I'll commit the following (yes I'm s

[PATCH] libstdc++/77645 Fix xmethods for std::list

The std::list xmethods have been broken since I added support for allocators with fancy pointers, and the switch to std::gnu++14 by default. This makes them work for any -std option. PR libstdc++/77645 * python/libstdcxx/v6/xmethods.py (DequeWorkerBase.index): Rename argum

Re: [Fortran, Patch] First patch for coarray FAILED IMAGES (TS 18508)

Hi Alessandro, there are still some violations of the style guide: contrib/check_GNU_style.sh first_complete_patch_REV2.diff emits: Lines should not exceed 80 characters. 154:+ add_sym_2 ("failed_images", GFC_ISYM_FAILED_IMAGES, CLASS_TRANSFORMATIONAL, ACTUAL_NO, BT_INTEGER, 155:+

Ping Re: Make max_align_t respect _Float128 [version 3]

Ping. This patch is pending review. -- Joseph S. Myers jos...@codesourcery.com

Re: [PATCH] Define pretty printers for C++17 library components

On 17/09/16 16:20 +0100, Jonathan Wakely wrote: This tweaks the existing printers for Fundamentals TS components to work for the C++17 versions, and adds a printer for std::variant. * python/libstdcxx/v6/printers.py (StdVariantPrinter): Define. (StdExpAnyPrinter, StdExpOptionalPr

Re: [PATCH] Fix PR tree-optimization/77550

On 09/19/16 11:25, Richard Biener wrote: > On Sun, 18 Sep 2016, Bernd Edlinger wrote: > >> Hi, >> >> this PR shows that in vectorizable_store and vectorizable_load >> as well, the vector access always uses the first dr as the alias >> type for the whole access. But that is not right, if they are >

Re: [PATCH][simplify-rtx] (GTU (PLUS a C) (C - 1)) --> (LTU a -C)

On 09/19/2016 04:43 PM, Kyrill Tkachov wrote: Ok? Sure. Bernd

Re: [PATCH test]Revise gcc.dg/vect/pr57558-1.c by using int instead of long

On 09/19/2016 08:54 AM, Bin Cheng wrote: Hi, Case gcc.dg/vect/pr57558-1.c uses unsigned long at the moment which can't be vectorized on Sparc64. This patch revises it by using unsigned int. I think it's an obvious change. Test result checked, is it OK? Thanks, bin gcc/testsuite/ChangeLog 20

[PATCH, wwwdocs, committed] Update gcc-7/changes.html for LRA

Hi! I'm committing the following for changes-7.html (and actually ran the w3c validator thing this time ;-) ) Segher Index: htdocs/gcc-7/changes.html === RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-7/changes.html,v retrieving revision 1

Re: [PATCH] Allow FP to be used as a call-saved registe

On 09/19/2016 03:45 AM, Richard Earnshaw (lists) wrote: So those don't seem to me to imply that the frame pointer needs to be a fixed register. So the first thing I'd do is fix the aarch64 port to not do that and see what fallout there is and how to fix it. Most ports simply don't mark the fra

Re: Ping Re: Make max_align_t respect _Float128 [version 3]

On 09/19/2016 08:58 AM, Joseph Myers wrote: Ping. This patch is pending review. Thanks, the patch looks good to me. It should be safe for the uses of max_align_t that I know of (e.g., Emacs).

Re: [PATCH 5/9] regrename: Don't run if function was separately shrink-wrapped

Hi Jeff, On Wed, Sep 14, 2016 at 01:33:04PM -0600, Jeff Law wrote: > On 09/14/2016 01:03 PM, Segher Boessenkool wrote: > >>If you think about it, conceptually we want the return insn to make the > >>callee saved registers "used" so that DCE, regrename and friends don't > >>muck with them. The fac

[PATCH] libstdc++/77645 fix deque and vector xmethods for Python 3

The problem for these xmethods is that in Python 3 division of two integers produces a float, and GDB then doesn't allow that value to be used for pointer arithmetic (https://sourceware.org/PR20622). PR libstdc++/77645 * python/libstdcxx/v6/xmethods.py (DequeWorkerBase.__init__)

Re: [patch] Fix ICE on ACATS test for Aarch64 at -O

On 09/16/2016 04:05 PM, Eric Botcazou wrote: Hi, for the attached reduced testcase, the ICE is: eric@polaris:~/gnat/bugs/P901-028> ~/build/gcc/aarch64-linux/gcc/gnat1 -quiet p.adb -O -I ~/build/gcc/aarch64-linux/gcc/ada/rts +===GNAT BUG DETECTED==

Re: [build] Fix race condition during libgcc build

On 09/16/2016 03:54 PM, Eric Botcazou wrote: Hi, we have some new machines which seem to be very good at stumbling upon race conditions during bootstrap. Another example with libgcc: /azun.a/gnatmail/sandbox/gnatcross-cont/x86-linux/gcc/build/./gcc/xgcc - B/azun.a/gnatmail/sandbox/gnatcross-co

Re: [f...@deneb.enyo.de: [PATCH] ada/77535: GNAT.Perfect_Hash_Generators for non-1-based strings]

* Thomas Quinot, 2016-09-18 : > Sorry for the delay in reviewing your patch. It looks good on the > surface; we're QAing it internally to make sure it does not break > PolyORB (which is the primary consumer of g-pehage). I'll let you know > as soon as I have the results from that run. QA went fin

Re: [PATCH] Fix PR c++/77639 (ICE during error recovery)

That looks good, OK. On Mon, Sep 19, 2016 at 11:17 AM, Patrick Palka wrote: > On Mon, 19 Sep 2016, Jason Merrill wrote: > >> >if (tree t = maybe_new_partial_specialization (type)) >> > { >> > + if (processing_template_parmlist) >> > + { >> > +/* Some

Re: Make regcprop to eliminate noop moves better

On 09/18/2016 06:42 AM, Jan Hubicka wrote: Hi, while working on the GCN port I ended up with many redundant register copies of the form mov reg, exec do something mov reg, exec do something ... these copies are generated by LRA because exec is small register class and needs a lot of reloadin

RE: [PATCH] libstdc++/77645 fix deque and vector xmethods for Python 3

Python has a distinct integer division operator, "//". 7 // 3 returns the integer 2. -Original Message- From: libstdc++-ow...@gcc.gnu.org [mailto:libstdc++-ow...@gcc.gnu.org] On Behalf Of Jonathan Wakely Sent: Monday, September 19, 2016 10:11 AM To: libstd...@gcc.gnu.org; gcc-patches@gc

Re: [PATCH 9/9] cse.c selftests

On 09/16/2016 03:19 PM, David Malcolm wrote: When possible I don't think we want the tests to be target specific. Hmm, I'm probably about to argue for Bernd's work... The 71779 testcase is a great example -- it uses high/lo_sum. Not all targets support that -- as long as we don't try to recog

Re: [PATCH] More trivial bits from early LTO debug merge

OK. On Mon, Sep 19, 2016 at 3:53 AM, Richard Biener wrote: > > This merges moving filename and related CU annotation from late finish > to early finish. With this all changes to cgraphunit.c have been merged. > > [LTO] Bootstrap and test running on x86_64-unknown-linux-gnu. > > Richard. > > 2016

Re: [PATCH 0/9] RFC: selftests based on RTL dumps

On 09/16/2016 03:27 PM, David Malcolm wrote: We should also twiddle how we represent registers in the dumps. Identifying hard regs by name (so we can map back to a hard reg if the hard regs change), identifying pseudos by number that isn't affected if the hard register set changes ie, p0, p1, p2,

Re: [PATCH] libstdc++/77645 fix deque and vector xmethods for Python 3

On 19/09/16 17:24 +, Joe Buck wrote: Python has a distinct integer division operator, "//". 7 // 3 returns the integer 2. Python 3 does, but Python 2 doesn't have it unless you import it from __future, and I don't know how far back that works. I don't want to introduce a fix for Python 3

Re: [PATCH, wwwdocs, committed] Update gcc-7/changes.html for LRA

On 9/19/16, Segher Boessenkool wrote: > Hi! > > I'm committing the following for changes-7.html (and actually ran the > w3c validator thing this time ;-) ) > > > Segher > > > Index: htdocs/gcc-7/changes.html > === > RCS file: /cvs/gcc

Re: PR35503 - warn for restrict pointer

On Sun, Sep 18, 2016 at 1:16 PM, Prathamesh Kulkarni wrote: > Sorry for late response. In the attached patch, I removed obstack > building on fmt, and used pp_format for formatting arg_positions by adding > specifier %I (name > chosen arbitrarily). Does that look OK ? > > However it results in fo

Re: [PATCH, wwwdocs, committed] Update gcc-7/changes.html for LRA

On Mon, Sep 19, 2016 at 01:49:57PM -0400, Eric Gallager wrote: > Perhaps add a link to the wiki page you created about the transition > to LRA, for easy access? Like this. Thanks for the suggestion, committed. Segher Index: htdocs/gcc-7/changes.html ===

Re: [C++ PATCH] Fix constexpr switch handling (PR c++/77467)

On Fri, Sep 16, 2016 at 4:44 PM, Jakub Jelinek wrote: > On Fri, Sep 16, 2016 at 03:51:11PM -0400, Jason Merrill wrote: >> On Mon, Sep 5, 2016 at 1:11 PM, Jakub Jelinek wrote: >> > + /* If body is a statement other than STATEMENT_LIST or BIND_EXPR, >> > + it should be skipped. E.g. switch (a

Re: C/C++ PATCH to implement -Wbool-operation (PR c/77490)

On 09/16/2016 05:01 AM, Marek Polacek wrote: @@ -5853,7 +5853,16 @@ cp_build_unary_op (enum tree_code code, tree xarg, bool noconvert, arg, true))) errstring = _("wrong type argument to bit-complement"); else if (!noconvert && CP_

Re: C/C++ PATCH to implement -Wpointer-compare warning (PR c++/64767)

On Thu, Sep 15, 2016 at 8:16 AM, Marek Polacek wrote: > On Wed, Sep 14, 2016 at 12:07:49AM -0400, Jason Merrill wrote: >> On Sat, Sep 10, 2016 at 10:58 AM, Marek Polacek wrote: >> > Spurred by the recent findings, I >> > decided to >> > implement a warning that

Re: [PATCHv3, resent] Add a warning for suspicious use of conditional expressions in boolean context

On 09/15/2016 05:19 AM, Bernd Edlinger wrote: + if (warn_int_in_bool_context) + { + tree val1 = fold_for_warn (TREE_OPERAND (expr, 1)); + tree val2 = fold_for_warn (TREE_OPERAND (expr, 2)); + if (TREE_CODE (val1) == INTEGER_CST + && TREE_CODE (val2)

Re: [PATCHv3, resent] Add a warning for suspicious use of conditional expressions in boolean context

On 09/15/2016 03:19 AM, Bernd Edlinger wrote: On 09/14/16 20:11, Jason Merrill wrote: >> >> Yes. The reasoning I initially had was that it is completely >> pointless to have something of the form "if (x ? 1 : 2)" or >> "if (x ? 0 : 0)" because the result does not even depend on x >> in this cas

Re: debug container mutex association

Hi Following our conversation here is a much simpler patch. I just consider that all debug containers will have the same alignment. Even if I submit this patch as a whole I will commit into pieces, at least one for the pure cleanup parts and one for the debug.cc change. Among th

Re: [PATCH] libstdc++/77641 fix std::variant for literal types

On Mon, Sep 19, 2016 at 5:34 AM, Jonathan Wakely wrote: > As noted in bugzilla PR 77641 I don't think is_literal_type is the > right condition for _Uninitialized, because a type can have a > non-trivial default constructor but still be literal, but that means > it can't be used in the union in _Va

Re: [PATCH] Fix PR64078

On 09/15/2016 04:29 AM, Tom de Vries wrote: On 31/08/16 07:42, Tom de Vries wrote: On 30/08/16 11:38, Bernd Edlinger wrote: On 08/30/16 10:21, Tom de Vries wrote: On 29/08/16 18:43, Bernd Edlinger wrote: Thanks! Actually my patch missed to fix one combination: -m32 with -fpic make check-gcc

Re: Ping Re: Define TS 18661-1 CR_DECIMAL_DIG in

On 09/15/2016 06:56 AM, Joseph Myers wrote: Ping. This patch is pending review. I suspect you know more about this than anyone else here, so ISTM you're the authority on correctness. Rubber stamp OK. :-) jeff

Re: [PATCH 0/8] make next_*_insn take rtx_insn * arguments

On 09/16/2016 04:10 AM, Trevor Saunders wrote: On Thu, Sep 15, 2016 at 10:27:56AM -0600, Jeff Law wrote: On 09/15/2016 10:10 AM, Trevor Saunders wrote: On Thu, Sep 15, 2016 at 12:52:59PM +0200, Bernd Schmidt wrote: On 09/14/2016 09:21 PM, tbsaunde+...@tbsaunde.org wrote: Basically $subject.

Re: [PATCHv3, resent] Add a warning for suspicious use of conditional expressions in boolean context

On 09/19/16 21:17, Jason Merrill wrote: > On 09/15/2016 05:19 AM, Bernd Edlinger wrote: >> + if (warn_int_in_bool_context) >> +{ >> + tree val1 = fold_for_warn (TREE_OPERAND (expr, 1)); >> + tree val2 = fold_for_warn (TREE_OPERAND (expr, 2)); >> + if (TREE_CODE (val1) == INT

Re: [PATCH] Handle VECTOR_CST in integer_nonzerop()

On Wed, Sep 14, 2016 at 1:58 AM, Marc Glisse wrote: > On Fri, 19 Aug 2016, Patrick Palka wrote: > >> On Fri, Aug 19, 2016 at 7:30 PM, Patrick Palka >> wrote: >>> >>> integer_nonzerop() currently unconditionally returns false for a >>> VECTOR_CST argument. This is confusing because one would expe

Re: [PATCH 5/8] make prev_real_insn take rtx_insn *

On 09/14/2016 01:21 PM, tbsaunde+...@tbsaunde.org wrote: From: Trevor Saunders gcc/ChangeLog: 2016-09-13 Trevor Saunders * emit-rtl.c (prev_real_insn): Change argument type to rtx_insn *. * rtl.h: Adjust prototype. * config/sh/sh.md: Adjust. * dwarf2out.c (a

Re: [PATCH 6/8] make next/prev nonnote_nondebug_insn take rtx_insn *

On 09/14/2016 01:21 PM, tbsaunde+...@tbsaunde.org wrote: From: Trevor Saunders gcc/ChangeLog: 2016-09-14 Trevor Saunders * config/cris/cris.c (cris_asm_output_case_end): Change argument type to rtx_insn *. * emit-rtl.c (next_nonnote_nondebug_insn): Likewise. (prev_n

Re: [PATCH] Fix PR64078

On 09/19/16 22:19, Jeff Law wrote: > On 09/15/2016 04:29 AM, Tom de Vries wrote: >> On 31/08/16 07:42, Tom de Vries wrote: >>> On 30/08/16 11:38, Bernd Edlinger wrote: On 08/30/16 10:21, Tom de Vries wrote: > On 29/08/16 18:43, Bernd Edlinger wrote: >> Thanks! >> >> Actually my

Re: PR35503 - warn for restrict pointer

On Mon, 2016-09-19 at 14:21 -0400, Jason Merrill wrote: > On Sun, Sep 18, 2016 at 1:16 PM, Prathamesh Kulkarni > wrote: > > Sorry for late response. In the attached patch, I removed obstack > > building on fmt, and used pp_format for formatting arg_positions by > > adding specifier %I (name > > ch

Re: PR35503 - warn for restrict pointer

On Sun, 2016-09-18 at 22:46 +0530, Prathamesh Kulkarni wrote: > On 2 September 2016 at 23:14, David Malcolm > wrote: > > On Thu, 2016-09-01 at 14:55 +0530, Prathamesh Kulkarni wrote: > > > > [...] > > > > > The attached version passes bootstrap+test on ppc64le-linux-gnu. > > > Given that it only

Re: [PATCH 7/8] make next/prev active_insn and active_insn_p take rtx_insn *

On 09/14/2016 01:21 PM, tbsaunde+...@tbsaunde.org wrote: From: Trevor Saunders gcc/ChangeLog: 2016-09-14 Trevor Saunders * emit-rtl.c (next_active_insn): Change argument type to rtx_insn *. (prev_active_insn): Likewise. (active_insn_p): Likewise. * r

Re: [PATCH 3/8] make next/prev _nonnote_insn take rtx_insn *

On 09/14/2016 01:21 PM, tbsaunde+...@tbsaunde.org wrote: From: Trevor Saunders gcc/ChangeLog: 2016-09-13 Trevor Saunders * emit-rtl.c (next_nonnote_insn): Change argument type to rtx_insn *. (prev_nonnote_insn): Likewise. * jump.c (reversed_comparison_code_p

Re: [PATCH] manual: _Bool has trap representations

On powerpc-darwin _Bool is 4-byte not 1-byte, so saying values are represented as bytes isn't accurate for all systems supported by GCC. -- Joseph S. Myers jos...@codesourcery.com

Re: [PATCH] Fix PR64078

On 09/19/2016 03:08 PM, Bernd Edlinger wrote: Would it work to break this up into distinct tests, exit()-ing from each function rather than returning back to main? Yes. I think how this test is designed, each function must be inlined, or it will fail anyway. It was for instance impossible t

Re: [PATCHv3, resent] Add a warning for suspicious use of conditional expressions in boolean context

On 09/19/16 21:51, Jeff Law wrote: > On 09/15/2016 03:19 AM, Bernd Edlinger wrote: >> On 09/14/16 20:11, Jason Merrill wrote: >> >> Yes. The reasoning I initially had was that it is completely >> pointless to have something of the form "if (x ? 1 : 2)" or >> "if (x ? 0 : 0)" be

Re: [PATCH 8/9] final.c selftests

On 09/16/2016 03:34 PM, David Malcolm wrote: On Fri, 2016-09-16 at 14:45 -0600, Jeff Law wrote: On 09/08/2016 06:30 PM, David Malcolm wrote: gcc/ChangeLog: * final.c: Include selftest.h and selftest-rtl.h. (class selftest::temp_asm_out): New subclass of selftest::named_t

Re: [PATCHv3, resent] Add a warning for suspicious use of conditional expressions in boolean context

On 09/19/2016 03:32 PM, Bernd Edlinger wrote: On 09/19/16 21:51, Jeff Law wrote: On 09/15/2016 03:19 AM, Bernd Edlinger wrote: On 09/14/16 20:11, Jason Merrill wrote: Yes. The reasoning I initially had was that it is completely pointless to have something of the form "if (x ? 1 : 2)" or "if

Re: [PATCH 5/9] Introduce class function_reader

On 09/16/2016 03:42 PM, David Malcolm wrote: On Fri, 2016-09-16 at 15:28 -0600, Jeff Law wrote: On 09/08/2016 06:30 PM, David Malcolm wrote: This patch generalizes the RTL-reading capabilities so that they can be run on the host as well as the build machine. The available rtx in rtl.def changes

Re: [PATCH] libstdc++/77645 fix deque and vector xmethods for Python 3

On Mon, 19 Sep 2016, Jonathan Wakely wrote: > On 19/09/16 17:24 +, Joe Buck wrote: > > Python has a distinct integer division operator, "//". 7 // 3 returns the > > integer 2. > > Python 3 does, but Python 2 doesn't have it unless you import it from > __future, and I don't know how far back

patch to fix PR77416

The following patch fixes https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77416 The patch was tested on ppc64 and bootstrapped on x86-64. Committed as rev. 240247. Index: ChangeLog === --- ChangeLog (revision 240246) +++ ChangeLog

Re: PR35503 - warn for restrict pointer

On Mon, 19 Sep 2016, David Malcolm wrote: > But does the new specifier need to be added to c-family/c-format.c, so > that stage 2 and 3 compilers do know about it? Yes. -- Joseph S. Myers jos...@codesourcery.com

[C++ PATCH] Fix parser ICE with [[...]] and [[unused,...]] (PR c++/77637)

Hi! As the testcase shows, while the grammar has attribute-list: attribute[opt] attribute-list , attribute[opt] attribute ... attribute-list , attribute ... we are actually parsing it as if the last 2 lines were attribute[opt] ... attribute-list , attribute[opt] ... and ICEing when the

Re: [BUILDROBOT] vax-netbsdelf / vax-linux: ‘ELIMINABLE_REGS’ was not declared in this scope

On 09/17/2016 05:29 PM, Bernd Edlinger wrote: On 09/17/16 22:29, Jan-Benedict Glaw wrote: On Fri, 2016-09-09 21:40:38 +, Bernd Edlinger wrote: Hi, I think it is time to remove support for INITIAL_FRAME_POINTER_OFFSET, which is no longer used by any target today. This removes a bunch of

[C++ PATCH] Fix ICE on operator"" template (PR c++/77638)

Hi! As the testcase shows for C++14 and up, for 1 argument template we don't ICE even if the template argument is invalid, because it checks if (TREE_TYPE (parm) != char_type_node || !TEMPLATE_PARM_PARAMETER_PACK (DECL_INITIAL (parm))) and if parm is error_mark_node, then it doesn't have c

[C++ PATCH] Fix ICE with class fields with invalid types (PR c++/77626)

Hi! layout_class_type for FIELD_DECLs with error_mark_node skips further processing, so such fields don't have DECL_FIELD_OFFSET and DECL_FIELD_BIT_OFFSET, thus byte_position ICEs on it. So, when we walk in constexpr handling all FIELD_DECLs, this patch fixes it by skipping over fields with error

[PATCH], Use VMRGEW on PowerPC power8/power9 to construct V4SFmode

The original VSX instruction set did not have a simple way to merge the two vectors with the upper words set in each double word that hold the float values in V4SFmode format after the XVCVDPSP instructions. So, the code used a VPERM instruction to reorder the parts. vector float combine (float a

[PATCH] Fix ICE with __atomic_{always,is}_lock_free (PR middle-end/77624)

Hi! fold_builtin_atomic_always_lock_free has an assertion that if the 2nd argument isn't INTEGER_CST, it has POINTER_TYPE_P, but additionally for casts to void * it looks through the cast. That means though for invalid code if an integral expression etc. is converted to pointer, we violate the as

[C++ PATCH] Fix C++ violation in g++.jason/init3.C (PR testsuite/63299)

Hi! As valgrind reports, this testcase allocates the string with new [], but deallocates with delete str. Fixed thusly, regtested on x86_64-linux and i686-linux, ok for trunk? 2016-09-19 Maxim Ostapenko Jakub Jelinek PR testsuite/63299 * g++.old-deja/g++.jason/i

Re: [PATCH] Optimise the fpclassify builtin to perform integer operations when possible

On Mon, Sep 12, 2016 at 04:19:32PM +, Tamar Christina wrote: > Hi All, > > This patch adds an optimized route to the fpclassify builtin > for floating point numbers which are similar to IEEE-754 in format. > > The goal is to make it faster by: > 1. Trying to determine the most common case fir

Re: [BUILDROBOT] vax-netbsdelf / vax-linux: ‘ELIMINABLE_REGS’ was not declared in this scope

On 09/19/16 23:51, Jeff Law wrote: > On 09/17/2016 05:29 PM, Bernd Edlinger wrote: >> On 09/17/16 22:29, Jan-Benedict Glaw wrote: >>> On Fri, 2016-09-09 21:40:38 +, Bernd Edlinger >>> wrote: Hi, I think it is time to remove support for INITIAL_FRAME_POINTER_OFFSET, which is

  1   2   >