Hi,
The below patch implements GCC plugin support on MinGW. Some notes on the
functionality and the patch:
1. Due to the way dynamic libraries work on Windows there are limitations.
Specifically, we have to produce import libraries for backend executables
(cc1, cc1plus) and plugins have to
Hi!
On Sun, Nov 12, 2017 at 12:33:03AM -0700, Tom Tromey wrote:
>if (argc < macro->paramc)
> {
> - /* As an extension, variadic arguments are allowed to not appear in
> + /* In C++2a (here the va_opt flag is used), and also as a GNU
> + extension, variadic arguments are all
On Mon, 13 Nov 2017, Jeff Law wrote:
> On 11/10/2017 01:00 AM, Richard Biener wrote:
> >
> > It's the usual issue with an optimizing compiler vs. a static analyzer.
> > We try to get rid of the little semantic details of the input languages
> > that in the end do not matter for code-generation bu
On Mon, 13 Nov 2017, Martin Sebor wrote:
> On 11/10/2017 01:00 AM, Richard Biener wrote:
> > On Thu, 9 Nov 2017, Jeff Law wrote:
> >
> > > On 11/02/2017 05:48 AM, Richard Biener wrote:
> > >
> > > >
> > > > There were elaborate transforms of ptr + CST to ptr->a.b.c[3] in the
> > > > past. We h
Hi,
this patch removes most of uses of frequencies in bb reorder. The last
remaining
case is the heap of traces which needs to be ordered by integer or turned into
the fibheap. I think integer frequencies works pretty well there.
I also noticed that while connecting traces bacwards we use proba
Hi,
there is no longer need to use counts because the path using frequencies
does precisely the same.
Bootstrapped/regtested x86_64-linux, comitted.
* ipa-inline.c (compute_uninlined_call_time, compute_inlined_call_time):
always use frequencies.
Index: ipa-inline.c
==
[ was: Re: [PATCH 3/4] enhance overflow and truncation detection in
strncpy and strncat (PR 81117) ]
On 08/06/2017 10:07 PM, Martin Sebor wrote:
* c-c++-common/Wstringop-truncation.c: Same.
+/* Verify warnings for VLAs. */
+
+void test_strncpy_vla (unsigned n, const char* s)
+{
+
Hi,
this patch moves predict.c away from frequencies and also updates name of
counts_to_freqs to update_max_bb_count because it no longer does any conversion
just determines max count in function.
Bootstrapped/regtested x86_64-linux, comitted.
2017-11-14 Jan Hubicka
* auto-profile.c (
Hi,
this patch allows openacc test-case asyncwait-1.c to run for non-nvidia
devices.
Committed as obvious.
Thanks,
- Tom
Allow asyncwait-1.c to run for non-nvidia devices
2017-11-14 Tom de Vries
* testsuite/libgomp.oacc-c-c++-common/asyncwait-1.c: Allow to run for
non-nvidia devices.
Hi,
this patch updates tracer, reg-stack and shrink wrap.
Bootstrapped/regtested x86_64-linux, comitted.
Honza
* tracer.c (better_p): Do not compare frequencies.
* reg-stack.c (better_edge): Likewise.
* shrink-wrap.c (try_shrink_wrapping): Do not convert to gcov counts
On 3 November 2017 at 15:38, Richard Biener wrote:
> On Fri, Nov 3, 2017 at 6:15 AM, Prathamesh Kulkarni
> wrote:
>> Hi Martin,
>> As mentioned in PR, the issue here for propagating value of 'm' from
>> f_c1 to foo() is that the jump function operation is FLOAT_EXPR, and
>> the type of input para
Martin Sebor writes:
> On 11/13/2017 04:36 PM, Richard Sandiford wrote:
>> Jeff Law writes:
>>> On 11/09/2017 04:06 AM, Richard Sandiford wrote:
>>>
> Let me say at the outset that I struggle to comprehend that a few
> instructions is even a consideration when not optimizing, especially
>
Hi,
This patch fixes PR82941 and PR82942 by adding vzeroupper generation on SKX.
Bootstrapped and tested.
14.11.2017 Sebastian Peryt
gcc/
* config/i386/i386.c (pass_insert_vzeroupper): Modify gate condition
to return true on Xeon and not on Xeon Phi.
(ix86_check_avx256
Didn't get in the list for some reason.
> -Original Message-
> From: Koval, Julia
> Sent: Tuesday, November 14, 2017 10:29 AM
> To: GCC Patches
> Cc: Kirill Yukhin
> Subject: [x86,avx][patch] Fix PR82983
>
> Hi, this patch fix GFNI check which didn't work properly in gfni+sse case.
>
>
Hi,
This patch adds missing intrinsics for _mm_mask[z]_scalef_round_[sd,ss].
14.11.2017 Olga Makhotina
gcc/
* config/i386/avx512fintrin.h (_mm_mask_scalef_round_sd,
_mm_maskz_scalef_round_sd, _mm_mask_scalef_round_ss,
_mm_maskz_scalef_round_ss): New intrinsics.
On Tue, Nov 14, 2017 at 09:45:12AM +, Peryt, Sebastian wrote:
> Hi,
>
> This patch fixes PR82941 and PR82942 by adding vzeroupper generation on SKX.
> Bootstrapped and tested.
>
> 14.11.2017 Sebastian Peryt
>
> gcc/
In that case the ChangeLog entry should list the PRs, i.e.
PR ta
Hi,
Attachment got lost by accident. Attaching it again.
Thanks,
Olga
-Original Message-
From: Makhotina, Olga
Sent: Tuesday, November 14, 2017 10:49 AM
To: gcc-patches@gcc.gnu.org
Cc: Uros Bizjak ; Kirill Yukhin ;
Makhotina, Olga ; Peryt, Sebastian
Subject: [patch][i386, AVX] Adding
Hi,
I thought I sent the following email last Friday but found it in my
drafts folder right now, so let me send it now so that anybody
interested can see what the patch does on Haswell.
I have only skimmed through new messages in the thread. I am now
looking into something else right now but wil
Hi Ian,
> With the change in the Solaris release model (no more major releases
> like Solaris 12 but only minor ones like 11.4), the Solaris 12
> references in GCC need to be adapted.
>
> The following patch does this, consisting mostly of comment changes.
>
> Only a few changes bear comment:
>
>
Currently, g++.dg/cpp0x/rv-trivial-bug.C FAILs on Solaris 10 only:
+FAIL: g++.dg/cpp0x/rv-trivial-bug.C -std=c++11 (test for excess errors)
+FAIL: g++.dg/cpp0x/rv-trivial-bug.C -std=c++14 (test for excess errors)
/vol/gcc/src/hg/trunk/local/gcc/testsuite/g++.dg/cpp0x/rv-trivial-bug.C:27:1:
war
Hi, this patch fix GFNI check which didn't work properly in gfni+sse case.
gcc/
* config/i386/gfniintrin.h: Add sse check.
* config/i386/i386.c (ix86_expand_builtin): Fix gfni check.
0001-fix-gfni.patch
Description: 0001-fix-gfni.patch
Mac OS X 10.7 bootstrap is currently broken in libsanitizer since
lsan/lsan_common_mac.cc uses VM_MEMORY_OS_ALLOC_ONCE which was only
introduced in Mac OS X 10.9. Upstream rejected the trivial patch to
make the use conditional since that's likely to be unreliable:
https://reviews.llvm.org
On Tue, Nov 14, 2017 at 11:44:34AM +0100, Rainer Orth wrote:
> 2017-11-13 Rainer Orth
>
> PR sanitizer/82824
> * configure.ac (LSAN_COMMON_SUPPORTED): New conditional. Disable
> before *-*-darwin13*.
> (WEAK_HOOKS_SUPPORT): New conditional.
> * configure: Regenera
I have updated tests and changelog according to Jakub's suggestions.
Please find attached v2 of my patch.
14.11.2017 Sebastian Peryt
gcc/
PR target/82941
PR target/82942
* config/i386/i386.c (pass_insert_vzeroupper): Modify gate condition
to return true on Xeo
> Le 13 nov. 2017 à 18:40, Mike Stump a écrit :
>
> On Nov 12, 2017, at 6:58 AM, H.J. Lu wrote:
>>
>> On Sun, Nov 12, 2017 at 6:22 AM, Dominique d'Humières
>> wrote:
>>> The following patch fixes PR68356, PR81210, and PR81693 on darwin.
>>> ...
>>
>> I wrote these tests. These tests don't
On Tue, Nov 14, 2017 at 3:18 AM, Peryt, Sebastian
wrote:
> I have updated tests and changelog according to Jakub's suggestions.
> Please find attached v2 of my patch.
>
>
> 14.11.2017 Sebastian Peryt
>
> gcc/
>
> PR target/82941
> PR target/82942
> * config/i386/i386.c (
On Mon, Nov 13, 2017 at 6:37 PM, Martin Sebor wrote:
> Richard, this thread may have been conflated with the one Re:
> [PATCH] enhance -Warray-bounds to detect out-of-bounds offsets
> (PR 82455) They are about different things.
>
> I'm still looking for approval of:
>
> https://gcc.gnu.org/ml/gc
This patch addresses c++/82836 & c++/82737. The root cause was a bad
assumption I made when moving the mangling alias machinery to its own
hash table.
I had thought that once we SET_DECL_ASSEMBLER_NAME, it never becomes
unset (or changed). That is false. There are paths in the compiler
tha
On Fri, Sep 15, 2017 at 11:45 PM, Jason Merrill wrote:
> The hash_map interface is a lot more convenient than that of
> hash_table for cases where it makes sense, but there hasn't been a way
> to get the ggc_cache_remove behavior with a hash_map. In other words,
> not marking elements during the
On Tue, Nov 14, 2017 at 10:31 AM, Prathamesh Kulkarni
wrote:
> On 3 November 2017 at 15:38, Richard Biener
> wrote:
>> On Fri, Nov 3, 2017 at 6:15 AM, Prathamesh Kulkarni
>> wrote:
>>> Hi Martin,
>>> As mentioned in PR, the issue here for propagating value of 'm' from
>>> f_c1 to foo() is that
On Fri, Nov 10, 2017 at 11:57 PM, Jeff Law wrote:
>
> And here's the monster that pulls the vr_values class out of tree-vrp.c.
> I've also pulled out the transitive closure of free routines that are
> used strictly by vr-values.c.
>
> Like the gimple-ssa-evrp.c change, this exposes various functi
Hi Jakub,
>> --- a/libsanitizer/configure.ac
>> +++ b/libsanitizer/configure.ac
>> @@ -140,6 +140,24 @@ case "$host" in
>> esac
>> AM_CONDITIONAL(USING_MAC_INTERPOSE, $MAC_INTERPOSE)
>>
>> +# lsan_common_mac.cc needs VM_MEMORY_OS_ALLOC_ONCE which was only
>> +# introduced in Mac OS X 10.9/Darw
This fixes a latent bug I ran into with RPO value-numbering work which
similar to all SSA propagators and existing FRE/PRE do not perform
elimination in unreachable code-regions. Those unreachable code-regions
are controlled by if (0 != 0) style conditions and CFG cleanup is supposed
to deal with
On Tue, Nov 14, 2017 at 01:57:21PM +0100, Rainer Orth wrote:
> >> --- a/libsanitizer/configure.ac
> >> +++ b/libsanitizer/configure.ac
> >> @@ -140,6 +140,24 @@ case "$host" in
> >> esac
> >> AM_CONDITIONAL(USING_MAC_INTERPOSE, $MAC_INTERPOSE)
> >>
> >> +# lsan_common_mac.cc needs VM_MEMORY_OS_
Hello.
Quite obvious fix for SH target. Joseph can you please continue with testing?
I don't have a machine to test the patch.
Ready for trunk?
Thanks,
Martin
gcc/ChangeLog:
2017-11-14 Martin Liska
PR target/82927
* config/sh/sh-mem.cc: Initialize probabilities.
(sh_
On Tue, 14 Nov 2017, Martin Liška wrote:
> Hello.
>
> Quite obvious fix for SH target. Joseph can you please continue with testing?
> I don't have a machine to test the patch.
I don't have SH hardware, but can confirm that, in conjunction with
Martin's glibc patch to work around strncpy warning
Please hold review on this until I investigate something that Carl brought up.
Thanks,
and sorry for the noise!
-- Bill
> On Nov 13, 2017, at 10:30 AM, Bill Schmidt
> wrote:
>
> Hi,
>
> Some previous patches to add support for vec_xl_be and fill in gaps in testing
> for vec_xl and vec_xst
On Wed, Oct 04, 2017 at 05:44:07PM +0100, James Greenhalgh wrote:
>
> On Thu, Jul 27, 2017 at 06:49:01PM +0100, James Greenhalgh wrote:
> >
> > On Mon, Jun 12, 2017 at 02:44:40PM +0100, James Greenhalgh wrote:
> > > [Sorry for the re-send. I spotted that the attributes were not right for
> > > th
On Tue, Nov 14, 2017 at 4:17 AM, H.J. Lu wrote:
> On Tue, Nov 14, 2017 at 3:18 AM, Peryt, Sebastian
> wrote:
>> I have updated tests and changelog according to Jakub's suggestions.
>> Please find attached v2 of my patch.
>>
>>
>> 14.11.2017 Sebastian Peryt
>>
>> gcc/
>>
>> PR target/82
On Tue, Nov 14, 2017 at 06:14:10AM -0800, H.J. Lu wrote:
> Just a thought, should we have a separate patch to add -mprefer-vzeroupper
> to cover all bases in the future, like
>
> /* opt_pass methods: */
> virtual bool gate (function *)
> {
> return TARGET_AVX && (!TARGET_AVX512ER ||
Attached is fixed patch.
Sebastian
> -Original Message-
> From: H.J. Lu [mailto:hjl.to...@gmail.com]
> Sent: Tuesday, November 14, 2017 1:18 PM
> To: Peryt, Sebastian
> Cc: Jakub Jelinek ; gcc-patches@gcc.gnu.org; Uros Bizjak
> ; Kirill Yukhin ; Lu, Hongjiu
>
> Subject: Re: [PATCH][i38
On Tue, Nov 14, 2017 at 2:09 AM, Rainer Orth
wrote:
>
>> With the change in the Solaris release model (no more major releases
>> like Solaris 12 but only minor ones like 11.4), the Solaris 12
>> references in GCC need to be adapted.
>>
>> The following patch does this, consisting mostly of comment
On Tue, Nov 14, 2017 at 6:20 AM, Jakub Jelinek wrote:
> On Tue, Nov 14, 2017 at 06:14:10AM -0800, H.J. Lu wrote:
>> Just a thought, should we have a separate patch to add -mprefer-vzeroupper
>> to cover all bases in the future, like
>>
>> /* opt_pass methods: */
>> virtual bool gate (function
Hi,
On 13/11/17 15:47, Christophe Lyon wrote:
On 30 October 2017 at 16:21, Maxim Ostapenko wrote:
On 30/10/17 17:08, Christophe Lyon wrote:
On 30/10/2017 11:12, Maxim Ostapenko wrote:
Hi,
sorry for the late response.
On 20/10/17 13:45, Christophe Lyon wrote:
Hi,
On 19 October 2017 at 13:
On Tue, Nov 14, 2017 at 06:40:32PM +0300, Maxim Ostapenko wrote:
> > But on the upstream sanitizer repo,
> > sanitizer_common/sanitizer_syscall_linux_arm.inc was added on Nov 8th,
> > and also checks for retval >= -4095, hence handling the clone() error
> > gracefully. So... can we merge again our
Hi All,
Dot Product is intended to only be available for Armv8.2-a and newer.
While this restriction is reflected in the intrinsics, the patterns
themselves were missing the Armv8.2-a bit.
While GCC would prevent invalid options e.g. `-march=armv8.1-a+dotprod`
we should prevent the pattern from b
Hi All,
This patch adds Armv8.3-a as an architecture to the compiler with
the feature set inherited from Armv8.2-a.
Bootstrapped regtested on arm-none-linux-gnueabihf and no issues.
gcc/
2017-11-14 Tamar Christina
* config/arm/arm-cpus.in (armv8_3, ARMv8_3a, armv8.3-a): New
*
Hi All,
Dot Product is intended to only be available for Armv8.2-a and newer.
While this restriction is reflected in the intrinsics, the patterns
themselves were missing the Armv8.2-a bit.
This means that using -march=armv8.1-a+dotprod incorrectly got the
auto-vectorizer to generate dot product i
Hi All,
This patch allows larger bitsizes to be used as copy size
when the target does not have SLOW_UNALIGNED_ACCESS.
fun3:
adrpx2, .LANCHOR0
add x2, x2, :lo12:.LANCHOR0
mov x0, 0
sub sp, sp, #16
ldrhw1, [x2, 16]
ldrbw2, [x2
Hi James,
I have split off the aarch64 bit off from the generic parts and processed your
feedback.
Attached is the reworked patch.
Ok for Tunk?
Thanks,
Tamar
Thanks,
Tamar
gcc/
2017-11-14 Tamar Christina
* config/aarch64/aarch64.c (aarch64_expand_movmem):
Add MEM to REG o
On 11/14/2017 10:20 AM, Jan Hubicka wrote:
@@ -4670,11 +4671,12 @@ expand_call_inline (basic_block bb, gimp
if (dump_file && (dump_flags & TDF_DETAILS))
{
- fprintf (dump_file, "Inlining ");
- print_generic_expr (dump_file, id->src_fn);
- fprintf (dump_file, " to ");
PING^1
On 11/06/2017 09:27 AM, Martin Liška wrote:
On 11/03/2017 04:21 PM, Jason Merrill wrote:
On Fri, Nov 3, 2017 at 10:25 AM, Martin Liška wrote:
On 10/27/2017 09:44 PM, Nathan Sidwell wrote:
On 10/27/2017 02:34 PM, Jakub Jelinek wrote:
But when singly inheriting a polymorphic base and
Tested on Linux-x64.
2017-11-14 Ville Voutilainen
Implement LWG 2733 and LWG 2759
* include/experimental/numeric (gcd): Reject cv-qualified bool.
(lcm): Likewise.
* include/std/numeric (gcd): Likewise.
(lcm): Likewise.
* testsuite/26_numerics/gcd/gcd_neg.cc: Add tests a
OK.
On Mon, Nov 6, 2017 at 3:27 AM, Martin Liška wrote:
> On 11/03/2017 04:21 PM, Jason Merrill wrote:
>> On Fri, Nov 3, 2017 at 10:25 AM, Martin Liška wrote:
>>> On 10/27/2017 09:44 PM, Nathan Sidwell wrote:
On 10/27/2017 02:34 PM, Jakub Jelinek wrote:
> But when singly inheriting
On Mon, Nov 06, 2017 at 05:22:36PM +0100, Jakub Jelinek wrote:
> I'd like to ping the:
>
> http://gcc.gnu.org/ml/gcc-patches/2017-10/msg01895.html
> PR debug/82718
> Fix DWARF5 .debug_loclist handling with hot/cold partitioning
>
> patch. Thanks
Ping^2.
Jakub
Hi!
On Fri, Nov 10, 2017 at 06:13:19PM -0500, Michael Meissner wrote:
> This patch updates the float128 built-in functions that get/set exponents, get
> mantissa, and do tests to work with _Float128 on the current system, and with
> long double when -mabi=ieeelongdouble is used.
>
> The issue is
On 14/11/17 18:14 +0200, Ville Voutilainen wrote:
Implement LWG 2733 and LWG 2759
* include/experimental/numeric (gcd): Reject cv-qualified bool.
(lcm): Likewise.
* include/std/numeric (gcd): Likewise.
(lcm): Likewise.
* testsuite/26_numerics/gcd/gcd_neg.cc: Add tests and adjust
On 11/11/17 19:14 +0100, Marc Glisse wrote:
Hello,
operator new can clobber memory, it is hard to teach the compiler
otherwise since it is replaceable. Here I cache a couple values before
the call to the allocator. I checked the result on this simple
example:
#include
void f(std::vector&v)
On Tue, Nov 14, 2017 at 3:26 PM, Peryt, Sebastian
wrote:
> Attached is fixed patch.
>
> Sebastian
>
>
>> -Original Message-
>> From: H.J. Lu [mailto:hjl.to...@gmail.com]
>> Sent: Tuesday, November 14, 2017 1:18 PM
>> To: Peryt, Sebastian
>> Cc: Jakub Jelinek ; gcc-patches@gcc.gnu.org; Uro
On 11/14/2017 05:28 AM, Richard Biener wrote:
On Mon, Nov 13, 2017 at 6:37 PM, Martin Sebor wrote:
Richard, this thread may have been conflated with the one Re:
[PATCH] enhance -Warray-bounds to detect out-of-bounds offsets
(PR 82455) They are about different things.
I'm still looking for appr
On Tue, Nov 14, 2017 at 04:05:12PM +, Tamar Christina wrote:
> Hi James,
>
> I have split off the aarch64 bit off from the generic parts and processed
> your feedback.
>
> Attached is the reworked patch.
>
> Ok for Tunk?
Thanks for the respin, I'm a bit confused by this comment.
> diff --
On 14 November 2017 at 19:06, Jonathan Wakely wrote:
> Both files can use remove_cv_t instead of remove_cv::type (and there's
> no need to std-qualify it in std::gcd).
>
> They could also use is_integral_v and is_same_v but that's
> pre-existing, and less important because there's no 'typename' th
On 14/11/17 15:04 +, Jonathan Wakely wrote:
On 14 November 2017 at 14:35, Sergey Nenakhov wrote:
Hello.
Excuse me if I'm posting to the wrong mailing list.
libstdc++ has its own mailing list (which Richard CC'd).
I've grabbed gcc-7.2.0
sources and noticed strange statement in the file
g
On 14/11/17 19:51 +0200, Ville Voutilainen wrote:
On 14 November 2017 at 19:06, Jonathan Wakely wrote:
Both files can use remove_cv_t instead of remove_cv::type (and there's
no need to std-qualify it in std::gcd).
They could also use is_integral_v and is_same_v but that's
pre-existing, and les
Ed,
gfortran uses libquadmath for support of its REAL(16)
intrinsic subprograms. I checked the list of intrinsics
and tgamma is current not in the list. I don't have
time to try Fortran's C interop feature to see if an
ordinary user can access __float128. While your patch
is probably useful for
Th 11/14/2017 17:48, James Greenhalgh wrote:
> On Tue, Nov 14, 2017 at 04:05:12PM +, Tamar Christina wrote:
> > Hi James,
> >
> > I have split off the aarch64 bit off from the generic parts and processed
> > your feedback.
> >
> > Attached is the reworked patch.
> >
> > Ok for Tunk?
>
> Th
While working on an unrelated issue I noticed that the compiler didn't
like some of these declarations after preprocessing, when they aren't
protected by system-header permissiveness.
I thought about limiting the permissiveness to only extern "C"
functions, but I believe that system headers are ad
I am building GCC (and binutils) with --with-sysroot and I build glibc in
that sysroot too. I want to test GCC using the glibc libraries I just
built, but by default GCC uses the dynamic linker that is in /lib and that
is loading the "normal" libraries from /lib or /lib64 at runtime and not
the on
On November 14, 2017 7:59:18 PM GMT+01:00, Steve Ellcey
wrote:
>I am building GCC (and binutils) with --with-sysroot and I build glibc
>in
>that sysroot too. I want to test GCC using the glibc libraries I just
>built, but by default GCC uses the dynamic linker that is in /lib and
>that
>is loadi
Howdy!
For some upcoming work I need some pass local data that I don't want to
be passing around as an argument. We have enough of those in the
threader as it is. So I moved the current pass local data into its own
class, and basically classified the entire pass, thus avoiding a lot of
argu
Hi,
this patch makes call time computations consistent again.
Bootstrapped/regtested x86_64-linux, comitted.
Honza
* ipa-inline.c (edge_badness): Dump sreal frequency.
(compute_inlined_call_time): Match natural implementaiton ...
* ipa-fnsummary.c (estimate_edge_size_and_ti
On Fri, Nov 3, 2017 at 10:48 AM, H.J. Lu wrote:
> On Wed, Nov 1, 2017 at 9:39 AM, H.J. Lu wrote:
>> On Wed, Nov 1, 2017 at 9:32 AM, Rich Felker wrote:
>>> On Sun, Oct 15, 2017 at 06:16:57AM -0700, H.J. Lu wrote:
crt1.o is used to create dynamic and non-PIE static executables. Static
P
On 11/13/2017 10:25 AM, Shalnov, Sergey wrote:
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index 4427328..bc9eb85 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -26087,15 +26087,24 @@ before a transfer of control flow out of the function
to minimize
the AVX to SSE tra
Hi,
this patch removes now unused variable. Bootstrapped/regtested x86_64-linux,
comitted.
Honza
* tree-ssa-threadupdate.c (compute_path_counts): Remove
unused path_in_freq_ptr parameter.
(ssa_fix_duplicate_block_edges): Do not pass around path_in_freq
Index: tree-ssa-thr
On Tue, 2017-11-14 at 14:08 -0500, Aldy Hernandez wrote:
> Howdy!
>
> For some upcoming work I need some pass local data that I don't want
> to
> be passing around as an argument. We have enough of those in the
> threader as it is. So I moved the current pass local data into its
> own
> class
This patch fixes 82878, which turned out to be a problem in genericizing
the call inside a __closure::_FUN () that we synthesize when we're
providing a non-member fn pointer for a non-capturing lambda. That call
is marked as CALL_FROM_THUNK_P, but I elided the check for that when
fixing 78495.
Hi,
I had a pasto in the function prototype for vec_xst_be. Fixed patch is below.
Thanks,
Bill
On 11/14/17 8:11 AM, Bill Schmidt wrote:
> Please hold review on this until I investigate something that Carl brought
> up. Thanks,
> and sorry for the noise!
>
> -- Bill
>> On Nov 13, 2017, at 10:
Hi!
strlen_to_stridx.get (rhs1) returns an address into the hash_map, and
strlen_to_stridx.put (lhs, *ps); (in order to be efficient) doesn't make a
copy of the argument just in case, first inserts the slot into it which
may cause reallocation, and only afterwards runs the copy ctor to assign
the
On 11/14/2017 02:04 PM, Jakub Jelinek wrote:
> Hi!
>
> strlen_to_stridx.get (rhs1) returns an address into the hash_map, and
> strlen_to_stridx.put (lhs, *ps); (in order to be efficient) doesn't make a
> copy of the argument just in case, first inserts the slot into it which
> may cause reallocati
Hi!
For targets that don't have {,u}mulv4 insn we try 3 different
expansions of the basic signed * signed -> signed or unsigned * unsigned ->
unsigned overflow computation. The first one is done if
if (GET_MODE_2XWIDER_MODE (mode).exists (&wmode)
&& targetm.scalar_mode_supporte
On 11/14/2017 02:04 PM, Jakub Jelinek wrote:
Hi!
strlen_to_stridx.get (rhs1) returns an address into the hash_map, and
strlen_to_stridx.put (lhs, *ps); (in order to be efficient) doesn't make a
copy of the argument just in case, first inserts the slot into it which
may cause reallocation, and on
Hi!
I'll be working on further store-merging improvements next two days
and hope to use some of the bswap pass APIs to handle stuff like:
void foo (char *__restrict p, char *__restrict q)
{
p[0] = q[3];
p[1] = q[2];
p[2] = q[1];
p[3] = q[0];
}
or
p[4] = data >> 8;
p[5] = data;
etc. As
On Tue, Nov 14, 2017 at 02:24:28PM -0700, Martin Sebor wrote:
> On 11/14/2017 02:04 PM, Jakub Jelinek wrote:
> > Hi!
> >
> > strlen_to_stridx.get (rhs1) returns an address into the hash_map, and
> > strlen_to_stridx.put (lhs, *ps); (in order to be efficient) doesn't make a
> > copy of the argument
On 11/14/2017 02:10 PM, Jeff Law wrote:
On 11/14/2017 02:04 PM, Jakub Jelinek wrote:
Hi!
strlen_to_stridx.get (rhs1) returns an address into the hash_map, and
strlen_to_stridx.put (lhs, *ps); (in order to be efficient) doesn't make a
copy of the argument just in case, first inserts the slot int
Hi guys,
I see this new test case failing on x86_64-linux-gnu:
FAIL: gfortran.dg/pr78240.f90 -O (test for excess errors)
$ gfortran-8 pr78240.f90
pr78240.f90:11:12:
integer x(n)/1/ ! { dg-error "Nonconstant array" }
1
Error: Variable ‘n’ cannot appear in the expression
On Nov 14, 2017, at 3:33 AM, Dominique d'Humières wrote:
>
>> Le 13 nov. 2017 à 18:40, Mike Stump a écrit :
>> On Nov 12, 2017, at 6:58 AM, H.J. Lu wrote:
>>>
>>> On Sun, Nov 12, 2017 at 6:22 AM, Dominique d'Humières
>>> wrote:
The following patch fixes PR68356, PR81210, and PR81693 on d
Hi,
Add support for gimple folding of vec_cmp_{eq,ge,gt,le,ne}
for the integer data types.
As part of this change, several define_insn stanzas have been added/updated
in vsx.md that specify the "ne: -> not: + eq: " combinations to allow for the
generation
of the desired vcmpne[bhw] ins
On Tue, Nov 14, 2017 at 4:58 PM, Janus Weil wrote:
> Hi guys,
>
> I see this new test case failing on x86_64-linux-gnu:
>
> FAIL: gfortran.dg/pr78240.f90 -O (test for excess errors)
>
>
> $ gfortran-8 pr78240.f90
> pr78240.f90:11:12:
>
>integer x(n)/1/ ! { dg-error "Nonconstant array"
2017-11-14 23:21 GMT+01:00 Fritz Reese :
> On Tue, Nov 14, 2017 at 4:58 PM, Janus Weil wrote:
>> Hi guys,
>>
>> I see this new test case failing on x86_64-linux-gnu:
>>
>> FAIL: gfortran.dg/pr78240.f90 -O (test for excess errors)
>>
>>
>> $ gfortran-8 pr78240.f90
>> pr78240.f90:11:12:
>>
>>
On Tue, Nov 14, 2017 at 05:21:41PM -0500, Fritz Reese wrote:
> On Tue, Nov 14, 2017 at 4:58 PM, Janus Weil wrote:
> > Error: The module or main program array ‘x’ at (1) must have constant shape
> > pr78240.f90:11:19:
> >
> >integer x(n)/1/ ! { dg-error "Nonconstant array" }
> >
Hi,
any feedback about this - apparently rather simple - issue?
On 03/11/2017 18:33, Paolo Carlini wrote:
Hi,
this ICE on valid (given GNU's designated initializers) is rather
simple to analyze: for the testcase, the gcc_assert in
process_init_constructor_array triggers because at that time
On Tue, 2017-11-14 at 20:04 +0100, Richard Biener wrote:
>
> I think the intent is to link even for just - lfoo while it makes
> sense to ignore -L/path - Wl,... Certainly counts as possible input.
>
> It seems you even break - lfoo.
>
> Richard.
Good point. I guess I need to take a differe
On 11/14/2017 02:30 PM, Jakub Jelinek wrote:
On Tue, Nov 14, 2017 at 02:24:28PM -0700, Martin Sebor wrote:
On 11/14/2017 02:04 PM, Jakub Jelinek wrote:
Hi!
strlen_to_stridx.get (rhs1) returns an address into the hash_map, and
strlen_to_stridx.put (lhs, *ps); (in order to be efficient) doesn't
Hi,
This patch fixes https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82235
For the following test case
struct Foo {
Foo() {}
explicit Foo(const Foo& aOther) {}
};
struct Bar {
Foo m[1];
};
void test() {
Bar a;
Bar b = a;
}
the compiler issues an error when the compiler generate
On 14/11/17 13:56 -0500, Jason Merrill wrote:
While working on an unrelated issue I noticed that the compiler didn't
like some of these declarations after preprocessing, when they aren't
protected by system-header permissiveness.
I thought about limiting the permissiveness to only extern "C"
fun
Ping?
Thanks,
Kugan
On 7 November 2017 at 15:10, Kugan Vivekanandarajah
wrote:
> Hi,
>
> Attached patch implements the vld1_*_x2 intrinsics as defined by the
> neon document.
>
> Bootstrap for the latest patch is ongoing on aarch64-linux-gnu. Is
> this OK for trunk if no regressions?
>
> Thanks
On 11/14/2017 02:30 PM, Martin Sebor wrote:
> On 11/14/2017 02:10 PM, Jeff Law wrote:
>> On 11/14/2017 02:04 PM, Jakub Jelinek wrote:
>>> Hi!
>>>
>>> strlen_to_stridx.get (rhs1) returns an address into the hash_map, and
>>> strlen_to_stridx.put (lhs, *ps); (in order to be efficient) doesn't
>>> mak
On 11/14/2017 01:42 AM, Richard Biener wrote:
>
> I suspect once you're dealing with C++ code you run into the issue
> that even early inlining exposes code with forwprop run on it
> before running forwprop again on the inlined-into body.
>
> So the IPA issues start very early. Of course if you
On 11/13/2017 06:21 PM, Martin Sebor wrote:
> I have a different concern with the general idea of moving these
> kinds of warnings into passes of their own. It would unavoidably
> result in duplicating some code from the optimization passes (at
> a minimum, the GIMPLE traversal, but likely more th
This patch to the Go frontend by Than McIntosh removes the LHS/RHS
context determination for variable references. It used to be needed
for a different backend but it's no longer necessary. Bootstrapped
and ran Go testsuite on x86_64-pc-linux-gnu. Committed to mainline.
Ian
2017-11-14 Than M
1 - 100 of 123 matches
Mail list logo
