When not verifying we can actually short-circuit verifying the individual
chains.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied to trunk.
Richard.
2017-05-22 Richard Biener
* df-scan.c (df_insn_refs_verify): Speedup when not verifying.
Index: gcc/df-scan.c
==
In preparation for an upcoming improvement.
Tested on x86_64-suse-linux, applied on the mainline.
2017-05-22 Eric Botcazou
libada/
* configure.ac: Add check for sys/capability.h header.
(have_capability): New substitution.
* configure: Regenerate.
* Makefile.i
On Fri, May 19, 2017 at 7:41 PM, Dominique d'Humières
wrote:
>
>> Le 19 mai 2017 à 19:26, Bernd Edlinger a écrit :
>>
>> No, unfortunately it looks like incremental does not work in this case.
>>
>> You can try to remove gcc/s-gtyp-input, maybe it un-breaks your
>> build. However, I gave up at t
On Sat, May 20, 2017 at 8:27 PM, Marc Glisse wrote:
> Hello,
>
> as asked, I am adding some replacement in match.pd instead of just deleting
> this (redundant with CCP). It is not clear how general the match.pd version
> needs to be. (for instance it could handle some casts in addition to what I
>
Ping?
Best regards,
Thomas
On 11/05/17 16:13, Thomas Preudhomme wrote:
Hi,
When ARMv8-M Baseline support was introduced no testcase was added for
support of the MOVT instruction because GCC would always prefer a load
from literal pool. With the recent work to fix -mslow-flash-data by
Andre Si
GNAT has historically enforced strong restrictions on volatile components, but
this is no basis for that in the RM.
Tested on x86_64-suse-linux, applied on the mainline.
2017-05-22 Eric Botcazou
* gcc-interface/decl.c (gnat_to_gnu_field): Do not enforce strict
alignment for
On 22/05/17 08:48, Thomas Preudhomme wrote:
Ping?
Best regards,
Thomas
On 11/05/17 16:13, Thomas Preudhomme wrote:
Hi,
When ARMv8-M Baseline support was introduced no testcase was added for
support of the MOVT instruction because GCC would always prefer a load
from literal pool. With the re
This change makes it possible for the compiler to inline into a child package
a subprogram declared in a parent package, even if the parent package has a
with clause for the child package in the body. It also fixes segfaults that
can occur when inlining into a package a subprogram coming from an u
Hi,
this is rather boring patch to turn some of inline summary manipulatotions
into member functions. More work is needed on this.
Bootstrapped/regtested x86_64-linux, comitted.
Honza
* ipa-inline-analysis.c (account_size_time): Rename to ...
(inline_summary::account_size_time):
It's a regression present on the mainline and 7 branch: the compiler aborts on
a package containing a declaration of a subprogram taking an access parameter
with null exclusion and a reference to the Access attribute applied to this
subprogram.
Tested on x86_64-suse-linux, applied on the mainli
On 20/05/17 15:10 +0800, Xi Ruoyao wrote:
On 2017-05-19 15:38 +0100, Jonathan Wakely wrote:
On 18/05/17 19:10 +0800, Xi Ruoyao wrote:
> This UB has been hiding so long...
Indeed! Thanks for the patch.
> 2017-03-11 Xi Ruoyao
>
>PR libstdc++/67214
>* include/bits/locale_facets.tcc (_M
Hi,
self_time is currently write only and there is quite good reason for it (one
should not care about running time of unoptimized body without context at all).
Bootstrapped/regtested x86_64-linux, comitted.
Honza
* ipa-inline-analysis.c (inline_summary::reset): Do not reset
self
name-lookup.c is somewhat messy in which functions are external
interfaces and which are internal workers. One impact is the
conditional timekeeping code.
In an attempt to clean that up, I'm trying to move the external
interfaces to the end of the file and consistently name worker functions
This patch neatens up the recent change in except.c to register GTY
roots. I've made them all global trees.
Whilst there I noticed except's declare_library_fn could be doing more
work, rather than its callers repeat themselves. Also renamed a couple
of global trees to be more mnemonic and co
This patch simplifies genmultilib.awk so that it generates
MULTILIB_REQUIRED instead of the complement, MULTILIB_EXCEPTIONS.
The current awk script dates back to the days when we mapped
all the > 200 devices to the mmcu core and also parts of
--help=target to list all devices were generated from
2017-05-22 15:37 GMT+04:00 Georg-Johann Lay :
> This patch simplifies genmultilib.awk so that it generates
> MULTILIB_REQUIRED instead of the complement, MULTILIB_EXCEPTIONS.
>
> The current awk script dates back to the days when we mapped
> all the > 200 devices to the mmcu core and also parts of
On 05/21/2017 03:35 AM, Tom de Vries wrote:
On 12/02/2015 04:09 PM, Nathan Sidwell wrote:
+/* Output a pattern for a move instruction. */
+
+const char *
+nvptx_output_mov_insn (rtx dst, rtx src)
+{
src_inner uses dst_mode rather than GET_MODE (src). I'm trying to
understand if that is inten
gtfiles in config-lang.in is a massive line of confusion. While very
rarely modified, it's completely impossible^W^Wreally hard to tell
what's changed.
Committed this patch to break the line and sort it. Although sh can
cope with newlines in the middle of a string, the value gets fed to
mak
This patch moves the -fdump-translation-unit into the new langhook for
registering dumps. On the way the option gets renamed to -fdump-lang-raw.
Committed as obvious.
nathan
--
Nathan Sidwell
2017-05-22 Nathan Sidwell
gcc/
* doc/invoke.texi (fdump-translation-unit): Delete documentation.
Hi!
OK for trunk?
commit 7b2f06b7d2fd23b20d81fda8be6ec7453e8b3fe3
Author: Thomas Schwinge
Date: Thu Dec 22 08:30:04 2016 +0100
libgomp nvptx plugin: Debugging output for cuInit failure
libgomp/
* plugin/plugin-nvptx.c (nvptx_get_num_devices): Debugging output
On 18/05/17 19:27 +0800, Xi Ruoyao wrote:
This patch use ioctl to get entropy of std::random_device using
/dev/random and /dev/urandom.
This is a nice addition, thanks.
N.B. I couldn't apply your patch, several lines had U+00A0 (i.e.
NO-BREAK SPACE) characters where normal spaces should be. I
On 19/05/17 22:40 -0700, Tim Shen via libstdc++ wrote:
diff --git a/libstdc++-v3/include/std/variant b/libstdc++-v3/include/std/variant
index 0e04a820d69..b9824a5182c 100644
--- a/libstdc++-v3/include/std/variant
+++ b/libstdc++-v3/include/std/variant
@@ -936,9 +936,9 @@ _GLIBCXX_BEGIN_NAMESPACE_
On 2017-05-22 14:00 +0100, Jonathan Wakely wrote:
> On 18/05/17 19:27 +0800, Xi Ruoyao wrote:
> > This patch use ioctl to get entropy of std::random_device using
> > /dev/random and /dev/urandom.
>
> This is a nice addition, thanks.
>
> N.B. I couldn't apply your patch, several lines had U+00A0 (
On 22/05/17 21:41 +0800, Xi Ruoyao wrote:
On 2017-05-22 14:00 +0100, Jonathan Wakely wrote:
On 18/05/17 19:27 +0800, Xi Ruoyao wrote:
> This patch use ioctl to get entropy of std::random_device using
> /dev/random and /dev/urandom.
This is a nice addition, thanks.
N.B. I couldn't apply your pa
On 2017-05-22 14:50 +0100, Jonathan Wakely wrote:
> On 22/05/17 21:41 +0800, Xi Ruoyao wrote:
> > On 2017-05-22 14:00 +0100, Jonathan Wakely wrote:
> > > On 18/05/17 19:27 +0800, Xi Ruoyao wrote:
> > > > This patch use ioctl to get entropy of std::random_device using
> > > > /dev/random and /dev/ur
From: Andi Kleen
When a verification check fails it is useful to dump the current
function to the dump file, so it's easier to figure out what
actually went wrong.
v2: Updated version now using a hook in internal_error, and
also prints the pass name and the dump file name.
gcc/:
2017-05-21 An
Hi!
In , we currently describe acc_pcopyin, acc_pcreate as "old
names", but they're not "old" but really "alternative names", with the
intention to provide them at symbol level, not via "#define"s. OK for
trunk?
commit 89c6599cc343a2f3b087caf9cb47c2c42bb02074
Author: Thomas Schwinge
Date: Wed
On 2017-05-22Mon的 14:50 +0100, Jonathan Wakely wrote:
> On 22/05/17 21:41 +0800, Xi Ruoyao wrote:
> > On 2017-05-22 14:00 +0100, Jonathan Wakely wrote:
> > > On 18/05/17 19:27 +0800, Xi Ruoyao wrote:
> > > > This patch use ioctl to get entropy of std::random_device using
> > > > /dev/random and /de
Hi!
On Mon, 22 May 2017 16:26:48 +0200, I wrote:
> C/C++ OpenACC: acc_pcopyin, acc_pcreate
> libgomp/
> [...]
> * testsuite/libgomp.oacc-c-c++-common/lib-38.c: Remove, merging
> its content into...
> * testsuite/libgomp.oacc-c-c++-co
Hi!
On Thu, 13 Apr 2017 19:45:05 +0200, I wrote:
> On Fri, 10 Oct 2014 12:42:06 +0200, I wrote:
> > In the OpenACC Runtime Library, acc_pcopyin and acc_pcreate are to be
> > aliases for acc_present_or_copyin and acc_present_or_create,
> > respectively.
> >
> > This is implemented in openacc.f90 a
On 19/05/2017 04:13, Martin Sebor wrote:
On 05/18/2017 10:14 AM, Jozef Lawrynowicz wrote:
A patch for this bug was originally posted here:
https://gcc.gnu.org/ml/gcc-patches/2017-02/msg01054.html
There were some issues with that patch which I've now fixed.
The MSP430 target supports the automat
On 05/11/2017 04:44 PM, Rainer Orth wrote:
Hi Tom,
2017-05-01 Tom de Vries
PR testsuite/65941
* lib/target-supports.exp (check_effective_target_rdrand): New proc.
the new effective-target keyword needs documenting in sourcebuild.texi.
Attached patch adds the missing docu
This is the first half of using directive cleanup. I split the current
interface in two, one for namespace-level directives and one for local
directives. That avoids the need for a load if 'if (namespace_level_p
())' tests in the bodies of the worker functions. I also merged and
renamed do_u
Ping, original post:
https://gcc.gnu.org/ml/gcc-patches/2017-03/msg01560.html
---
For the msp430-elf target, the gcc-6-branch fails to build when
configured to enable C++ support.
The gcc-5-branch successfully builds with C++ support.
Backporting r244727 and r243310 to the gcc-6-branch fixes
On 05/20/2017 01:38 AM, DJ Delorie wrote:
>
> Pedro Alves writes:
>> Ah, yeah. AFAICS, all the declaration checks in libiberty.h are
>> HAVE_DECL checks. This suggests to me that this declaration guard
>> should be HAVE_DECL too [1].
>
> Except the ones in the $funcs list, which includes str
On 05/19/2017 11:26 AM, Bernd Edlinger wrote:
On 05/19/17 19:05, Dominique d'Humières wrote:
Hi Bernd,
Your patches are causing troubles when I try to do "incremental updates »:
After revision r248290 it fails with
../../work/gcc/cp/init.c:4916:10: fatal error: gt-cp-init.h: No such file or
On 05/16/2017 03:12 PM, Rainer Orth wrote:
Hi Tom,
sorry for chiming in so very late: I've been on vacation and sick in
between...
thanks for review anyway.
On 04/24/2017 05:20 PM, David Malcolm wrote:
On Sat, 2017-04-22 at 19:49 +0200, Tom de Vries wrote:
Hi,
there are currently two typ
2017-05-21 23:56 GMT+02:00 Jerry DeLisle :
> On 05/21/2017 09:14 AM, Janus Weil wrote:
>>
>> Hi all,
>>
>> the attached patch fixes an ICE-on-valid regression by making sure
>> that the relevant vtype symbol is resolved properly (for further
>> discussion see the PR).
>>
>> The patch regtests clean
On Fri, May 19, 2017 at 1:51 PM, Richard Biener
wrote:
> On Mon, May 15, 2017 at 5:58 PM, Bin.Cheng wrote:
>> On Thu, May 11, 2017 at 12:02 PM, Richard Biener
>> wrote:
>>> On Tue, Apr 18, 2017 at 12:54 PM, Bin Cheng wrote
Hi,
Based on vect_peeling algorithm, we know for sure that vec
Hi!
For OpenACC 1.0 compatibility, we need to provide the aliases
acc_async_wait, and acc_async_wait_all for acc_wait, and acc_wait_all,
respectively. OK for trunk?
commit 0f8302913db6d2c23804a3463c51a47e623e76b2
Author: Thomas Schwinge
Date: Mon May 22 19:22:24 2017 +0200
OpenACC 1.0 co
On 05/20/2017 02:27 AM, DJ Delorie wrote:
>
> Pedro Alves writes:
>> That sounds to me like the root issue that should be fixed,
>> so that these fallback definitions don't come into into play at all.
>> I.e., why isn't HAVE_ENVIRON_DECL defined on mingw when
>> setenv.o is built? Sounds like a
On Mon, May 22, 2017 at 6:21 AM, Jonathan Wakely wrote:
> On 19/05/17 22:40 -0700, Tim Shen via libstdc++ wrote:
>>
>> diff --git a/libstdc++-v3/include/std/variant
>> b/libstdc++-v3/include/std/variant
>> index 0e04a820d69..b9824a5182c 100644
>> --- a/libstdc++-v3/include/std/variant
>> +++ b/lib
On Mon, May 22, 2017 at 11:05 AM, Tim Shen wrote:
> On Mon, May 22, 2017 at 6:21 AM, Jonathan Wakely wrote:
> I suggest to cc a front-end person (Jason?) to take a look, as I
> suggested in the bug, and the example: https://godbolt.org/g/AxUv16.
See more discussion in pr80737. Basically in the g
On Mon, May 22, 2017 at 12:28 PM, Martin Sebor wrote:
> On 05/19/2017 11:26 AM, Bernd Edlinger wrote:
>>
>> On 05/19/17 19:05, Dominique d'Humières wrote:
>>>
>>> Hi Bernd,
>>>
>>> Your patches are causing troubles when I try to do "incremental updates
>>> »:
>>>
>>> After revision r248290 it fail
On 05/19/17 05:17, Daniel Santos wrote:
> On 05/18/2017 08:37 AM, Bernd Edlinger wrote:
>> On 05/17/17 04:01, Daniel Santos wrote:
- if (ignore_outlined && cfun->machine->call_ms2sysv
- && in_hard_reg_set_p (stub_managed_regs, DImode, regno))
-return false;
+ if (igno
When I was comparing spec 2006 numbers between GCC 6.3 and 7.1, there was one
benchmark that was noticeably slower (milc). In looking at the code generated,
the #1 hot function (mult_adj_su3_mat_vec) had some cases where automatic
vectorization generated splat of double from memory.
The register
Hi!
For pointer based array sections, we are not requiring the pointer to be
shared in outer context and it might be not shared (e.g. function argument
or just a private pointer initialized to point to something shared etc.),
which is right, but we shouldn't require it during lowering either.
Thi
Hi!
If a VLA is shared e.g. on parallel or task construct, we firstprivatize
a pointer pointing to it, and add a private VLA variable that will have a
DECL_VALUE_EXPR for debug info purposes. But, standardwise, the VLA is
still shared on the parallel/task, it is just implementation detail what we
Hi!
This PR shows another bug, the computation of the size argument
for GOMP_task call could include uninitialized variable uses.
The problem is that the task arg_size needs to be finalized only
after mwe return from omp_low_1 of the task construct, thus it
is hard to gimplify it somewhere, but i
Hi!
On Fri, May 12, 2017 at 09:48:28PM +0200, Jakub Jelinek wrote:
> On Fri, May 12, 2017 at 09:37:27PM +0200, Marek Polacek wrote:
> > @@ -565,6 +564,25 @@ c_fully_fold_internal (tree expr, bool in_init, bool
> > *maybe_const_operands,
> > appropriate in any particular case. */
> >
Eli Zaretskii writes:
> Hmm... no, this doesn't solve the problem. The expansion of AC_LIBOBJ
> for waitpid is gone from the configure script, but the value of
> LIBOBJS in libiberty/Makefile still includes waitpid.o. What else is
> related to this?
After re-reading the sources a bit, I come t
Hi,
The subject test requires little endian, but the target selector doesn't
specify this. This patch fixes that. Tested on BE/LE P8 systems,
committed as obvious.
Thanks,
Bill
2017-05-22 Bill Schmidt
* gcc.target/powerpc/p8-vec-xl-xst.c: Fix target string to
LE-only.
In
On Mon, 22 May 2017, Bill Schmidt wrote:
> Hi,
>
> The subject test requires little endian, but the target selector doesn't
> specify this. This patch fixes that. Tested on BE/LE P8 systems,
> committed as obvious.
It's not obvious; it's wrong. You can configure for a BE default while
suppor
On Mon, May 22, 2017 at 9:21 AM, Jonathan Wakely wrote:
> On 19/05/17 22:40 -0700, Tim Shen via libstdc++ wrote:
>>
>> diff --git a/libstdc++-v3/include/std/variant
>> b/libstdc++-v3/include/std/variant
>> index 0e04a820d69..b9824a5182c 100644
>> --- a/libstdc++-v3/include/std/variant
>> +++ b/lib
On Mon, May 22, 2017 at 4:14 PM, Tim Song wrote:
> assuming core issue 1227's resolution
Actually, 1227 doesn't touch default template arguments :( OTOH, the
paragraph dealing with default template arguments seems to be full of
issues - it says "invalid type" rather than "invalid type or
expressi
On Mon, May 22, 2017 at 2:22 PM, Jason Merrill wrote:
> On Mon, May 22, 2017 at 12:28 PM, Martin Sebor wrote:
>> On 05/19/2017 11:26 AM, Bernd Edlinger wrote:
>>>
>>> On 05/19/17 19:05, Dominique d'Humières wrote:
Hi Bernd,
Your patches are causing troubles when I try to do "i
On Mon, May 22, 2017 at 08:14:13PM +, Joseph Myers wrote:
> > The subject test requires little endian, but the target selector doesn't
> > specify this. This patch fixes that. Tested on BE/LE P8 systems,
> > committed as obvious.
>
> It's not obvious; it's wrong. You can configure for a BE
On Mon, May 22, 2017 at 02:32:44PM -0400, Michael Meissner wrote:
> The register allocator did not use the load with splat instruction (LXVDSX)
> because all of the loads were register+offset. For the scalar values that it
> could load into the FPR registers, it used the normal register+offset loa
Hi Sterling,
this is an RFC series that makes xtensa gcc dynamically configurable,
i.e. on a platform with shared library support single toolchain binary
becomes capable of building code for arbitrary xtensa configuration.
At the same time it fully preserves the traditional way of configuring
usin
XCHAL_* macros from the xtensa-config.h are used in a number of places
that require them to be preprocessor constants. Rewrite these places so
that non-constant XCHAL_* definitions could be used there.
2017-05-22 Max Filippov
gcc/
* config/xtensa/xtensa.c (xtensa_option_override): Appen
2017-05-22 Max Filippov
gcc/
* config/xtensa/xtensa-config.c (dlfcn.h, windows.h): New
#include'd headers.
(RTLD_LAZY): New macro definition.
(dlopen, dlsym, dlclose, dlerror): New function definitions.
---
gcc/config/xtensa/xtensa-config.c | 38 +
Now that XCHAL_* macros don't have to be preprocessor constants add
include/xtensa-dynconfig.h that defines them as fields of a structure
returned from the xtensa_get_config function.
Define that structure and fill it with default parameter values
specified in the include/xtensa-config.h.
Define re
Provide essential XCHAL_* configuration parameters as __XCHAL_* built-in
macros. This way it will be possible to use them in libgcc and libc
without need to patch libgcc or libc source for the specific xtensa core
configuration.
2017-05-22 Max Filippov
gcc/
* config/xtensa/xtensa-config
Now that gcc provides __XCHAL_* definitions use them instead of XCHAL_*
definitions from the include/xtensa-config.h. That makes libgcc
dynamically configurable for the target xtensa core.
2017-05-22 Max Filippov
include/
* xtensa-config-builtin.h: New File.
libgcc/
* config/xt
On Mon, May 22, 2017 at 2:09 PM, Max Filippov wrote:
> XCHAL_* macros from the xtensa-config.h are used in a number of places
> that require them to be preprocessor constants. Rewrite these places so
> that non-constant XCHAL_* definitions could be used there.
This whole series of patches is goin
On Mon, May 22, 2017 at 2:30 PM, augustine.sterl...@gmail.com
wrote:
> On Mon, May 22, 2017 at 2:09 PM, Max Filippov wrote:
>> XCHAL_* macros from the xtensa-config.h are used in a number of places
>> that require them to be preprocessor constants. Rewrite these places so
>> that non-constant XCH
Ping:
https://gcc.gnu.org/ml/gcc-patches/2017-04/msg00547.html
On 05/15/2017 09:01 PM, Martin Sebor wrote:
Ping: Jeff, is this patch approved?
https://gcc.gnu.org/ml/gcc-patches/2017-04/msg00547.html
On 05/08/2017 08:38 PM, Martin Sebor wrote:
On 04/28/2017 12:35 PM, Jeff Law wrote:
On 0
Ping:
https://gcc.gnu.org/ml/gcc-patches/2017-05/msg01428.html
On 05/17/2017 06:04 PM, Martin Sebor wrote:
While working on a new warning for unsafe conversion I noticed
that the existing warnings that diagnose these kinds of problems
are missing some useful detail. For example, given declarat
68 matches
Mail list logo
