Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2017-08-02 Richard Biener
* tree-ssa-pre.c (bitmap_insert_into_set_1): Remove and inline
into ...
(bitmap_insert_into_set): ... this.
Index: gcc/tree-ssa-pre.c
On August 1, 2017 10:35:43 PM GMT+02:00, Jakub Jelinek wrote:
>Hi!
>
>In this function we insert 0-2 prologue sequences (which can sometimes
>contain jumps and other insns that need to end basic blocks) on edges,
>then commit edge insertions and then finally attempts to find out
>into which basic
On 05/19/2017 06:42 AM, Richard Biener wrote:
+ /* ??? In some cases the C++ FE (at least) fails to
+set DECL_CONTEXT properly. Simply globalize stuff
+in this case. For example
+__dso_handle created via iostream line 74 col 25. */
+ parent = comp_unit_die ();
On Tue, Aug 01, 2017 at 10:40:53AM -0400, David Malcolm wrote:
> On Mon, 2017-07-31 at 19:46 -0400, tbsaunde+...@tbsaunde.org wrote:
> > From: Trevor Saunders
> >
> > For most of the history of this see
> > https://sourceware.org/ml/gdb-patches/2016-10/msg00223.html
> > The changes are mostly s/
On Tue, Aug 01, 2017 at 04:21:41PM -0400, David Malcolm wrote:
> On Wed, 2017-07-12 at 09:13 -0400, Trevor Saunders wrote:
> > On Tue, Jul 11, 2017 at 11:24:45AM -0400, David Malcolm wrote:
> > > + public:
> > > + /* token_pair's ctor. */
> > > + token_pair () : m_open_loc (UNKNOWN_LOCATION) {}
I'm wondering if the messages could use a slight rewording, to give a
clue to the user about the reason *why* the expression has changed
signedness. The old message "signed and unsigned type in conditional
expression" gave the clue (but failed to underline the subexpression
changing sign, and te
This is a small readability tweak to the tree printer to have
it consistently use dashes and colons in tree attribute names
in favor of spaces. It's been tested on x86_64-linux.
The tree printer tends to avoid using spaces to separate tree
attribute names (not the __attribute__ kind but things l
On Tue, Aug 01, 2017 at 06:21:34PM +0200, Jakub Jelinek wrote:
> Apparently I broke power bootstrap with this, because two new spots were
> introduced after I wrote the patch and my cross-compiler which didn't have
> HAVE_AS_POWER9 defined didn't reveal that. Fixed thusly, committed as
> obvious t
It was broken by the recent removal of TYPE_METHODS.
Bootstrapped/regtested on x86_64-suse-linux, applied on the mainline.
2017-08-01 Eric Botcazou
c-family/
* c-ada-spec.c (has_static_fields): Look only into fields.
(dump_generic_ada_node): Small tweak.
(dump_nested_
On Tue, Aug 1, 2017 at 11:35 PM, H.J. Lu wrote:
> On Tue, Aug 1, 2017 at 2:25 PM, Uros Bizjak wrote:
>> On Tue, Aug 1, 2017 at 11:23 PM, H.J. Lu wrote:
>>> On Tue, Aug 1, 2017 at 2:11 PM, Uros Bizjak wrote:
On Tue, Aug 1, 2017 at 11:05 PM, H.J. Lu wrote:
> On Tue, Aug 1, 2017 at 1:49
On Tue, Aug 1, 2017 at 2:25 PM, Uros Bizjak wrote:
> On Tue, Aug 1, 2017 at 11:23 PM, H.J. Lu wrote:
>> On Tue, Aug 1, 2017 at 2:11 PM, Uros Bizjak wrote:
>>> On Tue, Aug 1, 2017 at 11:05 PM, H.J. Lu wrote:
On Tue, Aug 1, 2017 at 1:49 PM, Uros Bizjak wrote:
> On Tue, Aug 1, 2017 at 9:
On Tue, Aug 1, 2017 at 11:23 PM, H.J. Lu wrote:
> On Tue, Aug 1, 2017 at 2:11 PM, Uros Bizjak wrote:
>> On Tue, Aug 1, 2017 at 11:05 PM, H.J. Lu wrote:
>>> On Tue, Aug 1, 2017 at 1:49 PM, Uros Bizjak wrote:
On Tue, Aug 1, 2017 at 9:46 PM, H.J. Lu wrote:
> Add some tests for implementi
On Tue, Aug 1, 2017 at 2:11 PM, Uros Bizjak wrote:
> On Tue, Aug 1, 2017 at 11:05 PM, H.J. Lu wrote:
>> On Tue, Aug 1, 2017 at 1:49 PM, Uros Bizjak wrote:
>>> On Tue, Aug 1, 2017 at 9:46 PM, H.J. Lu wrote:
Add some tests for implementing interrupt handlers with naked attribute.
O
On Tue, Aug 1, 2017 at 11:05 PM, H.J. Lu wrote:
> On Tue, Aug 1, 2017 at 1:49 PM, Uros Bizjak wrote:
>> On Tue, Aug 1, 2017 at 9:46 PM, H.J. Lu wrote:
>>> Add some tests for implementing interrupt handlers with naked attribute.
>>>
>>> OK for trunk?
>>>
>>> H.J.
>>> ---
>>> * gcc.dg/gual
On Tue, Aug 1, 2017 at 1:49 PM, Uros Bizjak wrote:
> On Tue, Aug 1, 2017 at 9:46 PM, H.J. Lu wrote:
>> Add some tests for implementing interrupt handlers with naked attribute.
>>
>> OK for trunk?
>>
>> H.J.
>> ---
>> * gcc.dg/guality/pr25967-1.c: New test.
>> * gcc.dg/guality/pr25
On Tue, 2017-08-01 at 16:15 +0200, Marek Polacek wrote:
> On Mon, Jul 31, 2017 at 11:31:44AM -0400, David Malcolm wrote:
> > On Mon, 2017-07-31 at 16:14 +0200, Marek Polacek wrote:
> > > This patch improves the diagnostic of -Wsign-compare for ?: by
> > > also
> > > printing
> > > the types, simila
Plus a couple of minor tweaks left and right.
Bootstrapped/regtested on x86_64-suse-linux, applied on the mainline.
2017-08-01 Eric Botcazou
c-family/
* c-ada-spec.c (print_generic_ada_decl):Pass correctly-typed constant
(dump_ada_function_declaration): Likewise.
(dum
On Tue, Aug 1, 2017 at 9:46 PM, H.J. Lu wrote:
> Add some tests for implementing interrupt handlers with naked attribute.
>
> OK for trunk?
>
> H.J.
> ---
> * gcc.dg/guality/pr25967-1.c: New test.
> * gcc.dg/guality/pr25967-2.c: Likewise.
> * gcc.dg/torture/pr25967-1.c: Lik
On Tue, Aug 1, 2017 at 12:50 PM, Jakub Jelinek wrote:
> On Thu, Jul 20, 2017 at 08:59:29AM +0200, Martin Liška wrote:
>> Hello.
>>
>> Following patch does sharing of expansion for mem{p,}cpy and also strpcy
>> (with a known constant as source)
>> so that we use same type of expansion (direct insn
Hi!
In this function we insert 0-2 prologue sequences (which can sometimes
contain jumps and other insns that need to end basic blocks) on edges,
then commit edge insertions and then finally attempts to find out
into which basic blocks the sequences were inserted and calls
find_many_sub_basic_bloc
On Tue, Aug 1, 2017 at 10:20 PM, H.J. Lu wrote:
> OK for trunk?
>
> H.J.
> ---
> gcc/
>
> PR target/81654
> * config/i386/i386.c (ix86_set_func_type): Disallow naked
> attribute with interrupt attribute.
>
> gcc/testsuite/
>
> PR target/81654
> * gcc.target/
OK for trunk?
H.J.
---
gcc/
PR target/81654
* config/i386/i386.c (ix86_set_func_type): Disallow naked
attribute with interrupt attribute.
gcc/testsuite/
PR target/81654
* gcc.target/i386/pr81654.c: New test.
---
gcc/config/i386/i386.c |
On Mon, 2017-07-31 at 18:05 +0200, Marek Polacek wrote:
> On Mon, Jul 31, 2017 at 09:54:03AM -0600, Martin Sebor wrote:
> > On 07/31/2017 08:14 AM, Marek Polacek wrote:
> > > This patch improves the diagnostic of -Wsign-compare for ?: by
> > > also printing
> > > the types, similarly to my recent p
On Thu, Jul 20, 2017 at 08:59:29AM +0200, Martin Liška wrote:
> Hello.
>
> Following patch does sharing of expansion for mem{p,}cpy and also strpcy
> (with a known constant as source)
> so that we use same type of expansion (direct insns emission, direct emission
> with a loop instruction and
>
Changed in v2:
* Renamed template argument to traits_t; eliminated subclasses, just
using traits struct.
* Moved constants into struct bodies.
* Fixed typo.
OK for trunk?
gcc/cp/ChangeLog:
* parser.c (cp_parser_error): Update for new param to
c_parse_error.
(class token
Changed in v2:
* Renamed template argument to traits_t; eliminated subclasses, just
using traits struct.
* Moved enum constants into struct bodies (string constants can't be
without constexpr, which isn't available in C++98).
* Fixed typo.
OK for trunk?
gcc/c/ChangeLog:
* c-parser.c
On Wed, 2017-07-12 at 09:13 -0400, Trevor Saunders wrote:
> On Tue, Jul 11, 2017 at 11:24:45AM -0400, David Malcolm wrote:
> > +/* Some tokens naturally come in pairs e.g.'(' and ')'.
> > + This class is for tracking such a matching pair of symbols.
> > + In particular, it tracks the location o
(Unchanged since v1; already approved by Marek, assuming rest is approved)
gcc/c-family/ChangeLog:
* c-common.c (c_parse_error): Add rich_location * param, using it
rather implicitly using input_location.
* c-common.h (c_parse_error): Add rich_location * param.
gcc/testsui
Add some tests for implementing interrupt handlers with naked attribute.
OK for trunk?
H.J.
---
* gcc.dg/guality/pr25967-1.c: New test.
* gcc.dg/guality/pr25967-2.c: Likewise.
* gcc.dg/torture/pr25967-1.c: Likewise.
* gcc.dg/torture/pr25967-2.c: Likewise.
---
gcc/
On Jun 24, 2017, at 10:52 AM, Iain Buclaw wrote:
> Added a few extra comments for procedures, altering dmd2dg to write
> out flags converted to dejagnu in-place, instead on newlines.
>
> In the other testsuite patch, added new tests to accompany fixes that
> have been made since the last patch.
Hi all,
This is a rewrite of contrib/mklog in Python. I started adding features
suggested by Trevor some time ago but this quickly turned into a full
rewrite of existing Perl mklog and then I decided to just fully rewrite
it in Python (given that this has been requested several times).
The
HI Paul,
This reminds me of project that I once started to translate fortran
into C using a similar option. I gave up in the end because I found it
more convenient to use a tree dump and modify the declarations by
hand. In respect of your query about suggestions, how about outputting
non_interop
Hi Thomas,
This is 'obvious, I think. Yes, OK for trunk.
Thanks
Paul
On 1 August 2017 at 16:09, Thomas Koenig wrote:
> Am 24.07.2017 um 23:27 schrieb Thomas Koenig:
>>
>> Hello world,
>>
>> the attached patch fixes the PR; patch and test case are rather
>> self-explanatory.
>>
>> Regression-t
Hi Thomas,
This reminds me of project that I once started to translate fortran
into C using a similar option. I gave up in the end because I found it
more convenient to use a tree dump and modify the declarations by
hand. In respect of your query about suggestions, how about outputting
non_interop
richi's LTO debug patch mentioned in a comment that __dso_handle
didn't have a proper DECL_CONTEXT; this fixes that.
Tested x86_64-pc-linux-gnu, applying to trunk.
commit b96b9abccb1afd301fb907dcfb327ffac05998b1
Author: Jason Merrill
Date: Mon Jul 31 16:46:46 2017 -0400
* decl.c (d
On 01/08/2017 00:08, Joseph Myers wrote:
On Wed, 26 Jul 2017, Jeff Law wrote:
TYPE_SIZE, according to my understanding, should be a tree for the size
of the expression in bits.
The problem is for msp430 that size varies depending on where it's used.
ie, in a register an object might have a b
On Wed, Jul 26, 2017 at 06:41:23AM -0500, Segher Boessenkool wrote:
> > That is to follow aarch64 iterator naming convention, where they have
>
> Ugh, for some reason I thought this was in rs6000/ as well. I have
> fresh coffee now. Sorry.
Apparently I broke power bootstrap with this, because t
On Tue, Aug 01, 2017 at 08:40:28AM +0200, Jakub Jelinek wrote:
> Here is the variant patch. In addition to fixing the ICE for vec_ld, for
> vec_st it just moves the premature computation of aligned to the point where
> it is used and that is after we've also verified that the types of the call
> a
On Aug 1, 2017, at 8:50 AM, Bill Schmidt wrote:
>
> On Aug 1, 2017, at 7:44 AM, Bill Schmidt wrote:
>>
>>>
>>> On Aug 1, 2017, at 3:46 AM, Richard Biener
>>> wrote:
>>>
>>> On Mon, Jul 31, 2017 at 4:03 PM, Bill Schmidt
>>> wrote:
> On Jul 31, 2017, at 8:19 AM, Bill Schmidt
I pushed this patch to openacc-gcc-7-branch that fixes an ICE in
libgomp.oacc-c/asyncwait-2.c caused by the recent async backport from
gomp-4_0-branch. Before, expand_omp_target was expecting the wait clause
argument to be a constant value. This patch teaches that function to be
more flexible and a
On Wed, Jul 19, 2017 at 11:59 PM, Martin Liška wrote:
> Hello.
>
> Following patch does sharing of expansion for mem{p,}cpy and also strpcy
> (with a known constant as source)
> so that we use same type of expansion (direct insns emission, direct emission
> with a loop instruction and
> library
Description:
* This patch provides an initial support for DWARF debug sections in XCOFF.
Tests:
* AIX: Build: SUCCESS
- build made by means of gmake.
ChangeLog:
* xcoff.c: Initial support for DWARF debug sections in XCOFF.
Cordialement,
Tony Reix
Bull - ATOS
IBM Coop Architect & Technic
Hi Fritz,
Regtests on x86_64-redhat-linux. OK for trunk?
Patch looks good in principle; I really find all these DEC extensions
strange, but if they are needed for old code, why not?
Just one point:
+ gfc_error ("%s not allowed outside STRUCTURE at %C", "%FILL");
This should ha
Hi Fritz,
This is a simple patch. The original intent was for -fdec to set
-fd-lines-as-comments by default if flag_d_lines was unspecified by
the user. However, currently flag_d_lines is interrogated in
set_dec_flags(), usually before its final value is determined. The
attached patch fixes this
Am 24.07.2017 um 23:27 schrieb Thomas Koenig:
Hello world,
the attached patch fixes the PR; patch and test case are rather
self-explanatory.
Regression-testing as I write this. OK for trunk if it passes?
Regards
Thomas
OK?
Regards
Thomas
Hello world,
here is a slight update on the patch, with the following changes:
Fixed one ICE (yes, there was one)
Added a bit to the documentation to recommend to edit
function pointers
Translates c_size_t into ssize_t now - we only have a signed
type, unsigned makes little sense.
OK for trun
On Mon, Jul 31, 2017 at 03:53:42PM +0100, Jonathan Wakely wrote:
> This __has_bultin check only exists for Clang, so should be replaced
> by the correct __is_identifier check, not left there in addition to
> it.
I see. Actually I've guessed so, and thank you for clarifying it.
I'm attaching a repl
Hi All,
real_to_target seems to return the order of the elements in the array
differently depending on the endiannes. This undoes the endianness when
combining the values back to a HOST_WIDE_INT.
Regtested on aach64-none-linux-gnu and aarch64_be-none-linux-gnu and no issues.
Thanks,
Tamar
gcc/
On Mon, 2017-07-31 at 19:46 -0400, tbsaunde+...@tbsaunde.org wrote:
> From: Trevor Saunders
>
> For most of the history of this see
> https://sourceware.org/ml/gdb-patches/2016-10/msg00223.html
> The changes are mostly s/gdb/gtl/g
>
> include/ChangeLog:
>
> 2017-07-29 Trevor Saunders
>
>
Hi,
I would like to kindly remind you of the following patches,
which are already waiting for over 6 months:
[PATCH, ARM] correctly encode the CC reg data flow
https://gcc.gnu.org/ml/gcc-patches/2017-01/msg01351.html
[PATCH, ARM] Further improve stack usage in sha512 (PR 77308)
https://gcc.gnu.o
> On Aug 1, 2017, at 16:31 , Olivier Hainque wrote:
>
> This patch adds a variant implementation of _clear_cache
> for arm-vxworks*, needed for proper functioning of trampolines
> on targets with separate instruction/data caches.
Forgot to mention:
Tested by verifying success of an in-house bu
Hello,
On top of previous changes reworking the arm-vxworks support
https://gcc.gnu.org/ml/gcc-patches/2017-08/msg00085.html
https://gcc.gnu.org/ml/gcc-patches/2017-08/msg00075.html
https://gcc.gnu.org/ml/gcc-patches/2017-08/msg00078.html
This patch adds a variant implementation of _clear_
On Mon, Jul 31, 2017 at 11:31:44AM -0400, David Malcolm wrote:
> On Mon, 2017-07-31 at 16:14 +0200, Marek Polacek wrote:
> > This patch improves the diagnostic of -Wsign-compare for ?: by also
> > printing
> > the types, similarly to my recent patch. But we can do even better
> > here if we
> > ac
On 08/01/2017 03:46 PM, Richard Biener wrote:
On Tue, Aug 1, 2017 at 2:02 PM, Martin Liška wrote:
On 07/27/2017 01:48 PM, Richard Biener wrote:
On Thu, Jul 27, 2017 at 12:12 PM, Martin Liška wrote:
Hello.
As reported in mentioned PR, we segfault in gcov tool when one uses -a.
It's caused
On Tue, Aug 1, 2017 at 3:50 PM, Bill Schmidt
wrote:
> On Aug 1, 2017, at 7:44 AM, Bill Schmidt wrote:
>>
>>>
>>> On Aug 1, 2017, at 3:46 AM, Richard Biener
>>> wrote:
>>>
>>> On Mon, Jul 31, 2017 at 4:03 PM, Bill Schmidt
>>> wrote:
> On Jul 31, 2017, at 8:19 AM, Bill Schmidt
> w
Hi,
this patch simplifies the nvtpx/slp* test-cases by using signed loop
iteration variables, in order to work around PR81635.
Committed.
Thanks,
- Tom
Simplify nvptx/slp* test-cases
Use signed loop iteration variable in nvtpx/slp* test-cases to work around
PR tree-optimizaion/81635.
2017-0
Hello,
As mentioned in the thread rooted at
https://gcc.gnu.org/ml/gcc-patches/2017-07/msg00720.html,
the arm-vxworks port needs refreshing. As mentioned earlier in other vxworks
related threads, it was on my list of things to do and this patch implements
a first batch of changes to this effect.
On Aug 1, 2017, at 7:44 AM, Bill Schmidt wrote:
>
>>
>> On Aug 1, 2017, at 3:46 AM, Richard Biener
>> wrote:
>>
>> On Mon, Jul 31, 2017 at 4:03 PM, Bill Schmidt
>> wrote:
>>>
On Jul 31, 2017, at 8:19 AM, Bill Schmidt
wrote:
That would certainly be much simpler! I'll
On Tue, Aug 01, 2017 at 03:19:28PM +0200, Richard Biener wrote:
> + if (lvectype != vectype)
> + {
> + tree tem = make_ssa_name (lvectype);
> + gimple *pun= gimple_build_assign (tem, build1
> + (VIEW_CONVERT_EXPR, lvectyp
On Fri, Jul 28, 2017 at 3:15 PM, Richard Sandiford
wrote:
> "Bin.Cheng" writes:
>> On Fri, Jul 28, 2017 at 12:55 PM, Richard Sandiford
>> wrote:
>>> Bin Cheng writes:
Hi,
This simple patch fixes the ICE by adding LTGT in
vec_cmp pattern.
I also modified the original test cas
On Tue, Aug 1, 2017 at 2:02 PM, Martin Liška wrote:
> On 07/27/2017 01:48 PM, Richard Biener wrote:
>>
>> On Thu, Jul 27, 2017 at 12:12 PM, Martin Liška wrote:
>>>
>>> Hello.
>>>
>>> As reported in mentioned PR, we segfault in gcov tool when one uses -a.
>>> It's caused by fact
>>> that vectors b
On Aug 01 2017, Maxim Kuvyrkov wrote:
> Do you know a reliable way of checking whether target can issue nops in
> simple code?
Try inspecting one of the rtl dumps.
Andreas.
--
Andreas Schwab, SUSE Labs, sch...@suse.de
GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7
"
The following fixes another case of endless compute_antic iteration
in PRE. After a lengthy four-eyes discussion here we concluded that
the way clean () operates on expressions rather than values can
cause oscillation in the dataflow problem.
Thus we have to delay it (at the cost of some interme
Hello,
libgcc/config/t-vxworks twists LIBGCC2_INCLUDE to workaround a problem of
header file name conflicts between the system headers and gcc headers during
libgcc builds:
# This ensures that the correct target headers are used; some
# VxWorks system headers have names that collide with GCC's
#
The following adds optab checks to see whether the target prefers
vector from vector extracts or integer from punned integer vector
extracts. But instead of falling back to elementwise operation
we fall back to the vector from vector extract path as spilling
the vector to extract from and then do
> On Aug 1, 2017, at 1:52 PM, Andreas Schwab wrote:
>
> On Aug 01 2017, Gerald Pfeifer wrote:
>
>> On Mon, 31 Jul 2017, Maxim Kuvyrkov wrote:
>>> I don't see an easy way to correctly differentiate between "attribute"
>>> nops and "bundle" nops, so XFAILing these tests on ia64 seems like a
>>>
Hello,
libgcc/config/vxlib*.c implement parts-of/helpers-for the gthreads API
to support EH services for VxWorks.
This patch adjusts config/t-vxworks to add them to LIB2ADDEH instead
of modifying LIB2ADD, so the object files get bundled together with the
other EH related modules.
Tested by verif
> On Aug 1, 2017, at 3:46 AM, Richard Biener wrote:
>
> On Mon, Jul 31, 2017 at 4:03 PM, Bill Schmidt
> wrote:
>>
>>> On Jul 31, 2017, at 8:19 AM, Bill Schmidt
>>> wrote:
>>>
>>> That would certainly be much simpler! I'll regstrap it and test it on the
>>> other
>>> occurrence I've found
> On Aug 1, 2017, at 1:40 AM, Jakub Jelinek wrote:
>
> Hi!
>
> On Mon, Jul 31, 2017 at 02:42:21PM -0500, Bill Schmidt wrote:
>>> On Jul 31, 2017, at 11:27 AM, Jakub Jelinek wrote:
>>> On Mon, Jul 31, 2017 at 11:19:26AM -0500, Bill Schmidt wrote:
https://gcc.gnu.org/bugzilla/show_bug.cgi?i
Hi Richard,
> On Jul 31, 2017, at 11:58 , Richard Earnshaw (lists)
> wrote:
>
>> Regarding removal of old ABI support, which release were you
>> targeting ?
>>
>> On the VxWorks front, where we adapt to what the system toolchains
>> do, it will mean dropping support for VxWorks versions prior
Hi,
Sorry this fix is incorrect, I have reverted the patch to address the rest of
the
Big endian failures.
Sorry,
Tamar
> -Original Message-
> From: gcc-patches-ow...@gcc.gnu.org [mailto:gcc-patches-
> ow...@gcc.gnu.org] On Behalf Of Tamar Christina
> Sent: 01 August 2017 12:43
> To: gc
On Tue, Aug 1, 2017 at 12:51 PM, Tamar Christina
wrote:
>>
>> Given review comment already pointed out big-endian issue and patch was
>> updated to address it, I would expect reg-test on a big-endian target before
>> applying patch, right?
>
> The patch spent 6 months in external review.
> Given t
On 07/27/2017 01:48 PM, Richard Biener wrote:
On Thu, Jul 27, 2017 at 12:12 PM, Martin Liška wrote:
Hello.
As reported in mentioned PR, we segfault in gcov tool when one uses -a. It's
caused by fact
that vectors blocks and block_lists have indices kept in sync and as one
removes an element
f
>
> Given review comment already pointed out big-endian issue and patch was
> updated to address it, I would expect reg-test on a big-endian target before
> applying patch, right?
The patch spent 6 months in external review.
Given that, I simply forgot to rerun big endian before the commit as I d
On Sat, 15 Apr 2017, Gerald Pfeifer wrote:
> On Sat, 8 Apr 2017, Ionut Vatavu wrote:
>> I would like to announce a new mirror in Germany Gunzenhausen:
>>
>> http://www.bothelp.net/mirrors/gcc - updated daily by rsync
> This is now part of our mirrors list per the patch below.
And here is an updat
On Mon, Jun 26, 2017 at 11:50 AM, Tamar Christina
wrote:
> Hi all,
>
> Here's the re-spun patch.
> Aside from the grouping of the split patterns it now also uses h register for
> the fmov for HF when available,
> otherwise it forces a literal load.
>
> Regression tested on aarch64-none-linux-gnu
Hi All,
The big-endian tests were failing because it failed to take into account that
in order to generate mov/movk pairs for doubles the bit order are different from
le.
I have updated the tests with conditional results for both endianness.
Committed as r250770.
Regtested on aach64-none-linux-
I am testing the following pair of patches (first for trunk, 2nd for GCC 7
branch) to fix PR81633. On trunk recent refactoring made the PR71752
change obsolete, on the branch the patch installs the simpler originally
suggested patch which works within the constraints vect_get_slp_defs
is used on
Using -O -masm=intel following testcase (gcc.target/i386/addr-space-2.c):
--cut here--
int test(void)
{
int __seg_fs *f = (int __seg_fs *)16;
int __seg_gs *g = (int __seg_gs *)16;
return *f + *g;
}
--cut here--
compiles to:
mov eax, DWORD PTR gs:ds:16
add eax, DWORD
Hi Bin,
> Hi,
> I saw below failure after svn+ssh://gcc.gnu.org/svn/gcc/trunk@250672
>
> FAIL: gcc.target/aarch64/advsimd-intrinsics/vcvt_high_1.c -O1
> (internal compiler error)
This should be fixed by r 250766
Cheers,
Tamar
>
> Regression in patch updates?
>
> Thanks,
> bin
> >
> > OK fo
On Wed, Jun 7, 2017 at 12:38 PM, Tamar Christina
wrote:
> Hi All,
>
>
> This patch lays the ground work to fix the immediate moves for floats
> to use a combination of mov, movi, fmov instead of ldr and adrp to load
> float constants that fit within the 16-bit limit of movz.
>
> The idea behind it
Hello!
Naked functions should not sibcall, since in naked functions epilogue
point (placed just above sibcal insn) is unreachable and marked with a
trap insn.
2017-08-01 Uros Bizjak
PR target/81639
* config/i386/i386.c (ix86_funciton_naked): New prototype.
(ix86_function_ok_for_si
On Thu, Jul 27, 2017 at 11:36 PM, Richard Sandiford
wrote:
> This is a minimal-ish backport of the fix for PR80769. The trunk version
> also replaced open-coded instances of get_next_strinfo with calls to the
> new function. It also added asserts in various other places to try to
> ensure that r
On Tue, Aug 1, 2017 at 12:08 PM, James Greenhalgh
wrote:
>
> On Thu, Jul 27, 2017 at 03:21:01PM +0100, James Greenhalgh wrote:
>> On Thu, Jul 27, 2017 at 02:26:03PM +0200, Richard Biener wrote:
>> > On Thu, Jul 27, 2017 at 2:08 PM, Jakub Jelinek wrote:
>> > > On Thu, Jul 27, 2017 at 01:54:21PM +0
On Aug 01 2017, Gerald Pfeifer wrote:
> On Mon, 31 Jul 2017, Maxim Kuvyrkov wrote:
>> I don't see an easy way to correctly differentiate between "attribute"
>> nops and "bundle" nops, so XFAILing these tests on ia64 seems like a
>> valid approach.
>
> Make sense, given that the use of Itanium h
On Tue, 1 Aug 2017, Richard Biener wrote:
>
> When working on PR81181 I ran into some things I wanted to clean up
> several times. First a few PRE cleanups done for the fix. Second,
> the fake exit edges we add for infinite loops happen to start from
> loop headers rather than latches which is
On Mon, 31 Jul 2017, Maxim Kuvyrkov wrote:
> I don't see an easy way to correctly differentiate between "attribute"
> nops and "bundle" nops, so XFAILing these tests on ia64 seems like a
> valid approach.
Make sense, given that the use of Itanium has gone done drastically.
Gerald
On 07/31/2017 09:21 AM, Martin Liška wrote:
On 07/26/2017 07:45 PM, Jeff Law wrote:
On 07/12/2017 07:38 AM, Martin Liška wrote:
Hi.
Following patch adds -lspp when one uses -mstack-protector-guard=global.
Patch can bootstrap on ppc64le-redhat-linux and survives regression tests.
Ready to be
ping
From: Wilco Dijkstra
Sent: 20 July 2017 13:49
To: GCC Patches; James Greenhalgh
Cc: nd
Subject: [PATCH][AArch64] Improve addressing of TI/TFmode
In https://gcc.gnu.org/ml/gcc-patches/2017-06/msg01125.html Jiong
pointed out some addressing inefficiencies due to a recent change in
regcpr
ping
This patch further improves aarch64_legitimate_constant_p. Allow all
integer, floating point and vector constants. Allow label references
and non-anchor symbols with an immediate offset. This allows such
constants to be rematerialized, resulting in smaller code and fewer stack
sp
ping
From: Wilco Dijkstra
Sent: 17 January 2017 15:14
To: Richard Earnshaw; GCC Patches; James Greenhalgh
Cc: nd
Subject: Re: [PATCH v3][AArch64] Fix symbol offset limit
Here is v3 of the patch - tree_fits_uhwi_p was necessary to ensure the size of a
declaration is an integer. So the
ping
Wilco Dijkstra wrote:
> James Greenhalgh wrote:
>
> > I note this is still marked as an RFC, are you now proposing it as a
> > patch to be merged to trunk?
>
> Absolutely. It was marked as an RFC to get some comments - I thought it
> may be controversial to separate the frame po
ping
From: Wilco Dijkstra
Sent: 25 July 2017 14:58
To: GCC Patches; James Greenhalgh; Jeff Law
Cc: nd
Subject: [PATCH][AArch64] Simplify frame layout for stack probing
This patch makes some changes to the frame layout in order to simplify
stack probing. We want to use the save of LR as a p
ping
From: Wilco Dijkstra
Sent: 26 July 2017 14:46
To: GCC Patches; James Greenhalgh
Cc: nd
Subject: [PATCH][AArch64] Remove '*' from movsi/di/ti patterns
Remove the remaining uses of '*' from the movsi/di/ti patterns.
Using '*' in alternatives is typically incorrect at it tells the registe
When working on PR81181 I ran into some things I wanted to clean up
several times. First a few PRE cleanups done for the fix. Second,
the fake exit edges we add for infinite loops happen to start from
loop headers rather than latches which is somewhat confusing and
making PRE dataflow order more
On Thu, Jul 27, 2017 at 03:21:01PM +0100, James Greenhalgh wrote:
> On Thu, Jul 27, 2017 at 02:26:03PM +0200, Richard Biener wrote:
> > On Thu, Jul 27, 2017 at 2:08 PM, Jakub Jelinek wrote:
> > > On Thu, Jul 27, 2017 at 01:54:21PM +0200, Richard Biener wrote:
> > >> --- gcc/common.opt (revis
On Mon, Jun 26, 2017 at 11:49 AM, Tamar Christina
wrote:
> Hi,
>
> With the changes in the patches the testsuite had a minor update in the
> assembler scan.
> I've posted the patch but will assume it's OK based on the previous OK for
> trunk and
> the fact that this can fall in the obvious rule.
On Tue, Aug 01, 2017 at 08:35:06AM +0100, Tamar Christina wrote:
> Hi All,
>
> Previously I allowed 0s unconditionally through aarch64_can_const_movi_rtx_p
> because we should always be able to use movi with 0 regardless of the mode.
>
> However this was causing issues when a vector contained a 0
On Tue, Jul 25, 2017 at 8:26 AM, Richard Biener
wrote:
> On Mon, Jul 24, 2017 at 10:43 AM, Bin Cheng wrote:
>> Hi,
>> This is a followup patch to PR81388's fix. According to Richi,
>> POINTER_TYPE_OVERFLOW_UNDEFINED was added in -fstrict-overflow
>> warning work. Given:
>> A) strict-overflow
On Tue, Aug 1, 2017 at 11:23 AM, Richard Biener
wrote:
> On Tue, Aug 1, 2017 at 4:27 AM, Martin Sebor wrote:
>> Richard,
>>
>> in discussing this work Jeff mentioned that your comments on
>> the tree-ssa-alias.c parts would be helpful. When you have
>> a chance could you please give it a once ov
1 - 100 of 125 matches
Mail list logo
