On 11/8/18 9:56 AM, Martin Liška wrote:
> On 11/7/18 11:23 PM, Jeff Law wrote:
>> On 10/30/18 6:28 AM, Martin Liška wrote:
>>> On 10/30/18 11:03 AM, Jakub Jelinek wrote:
On Mon, Oct 29, 2018 at 04:14:21PM +0100, Martin Liška wrote:
> +hashtab_chk_error ()
> +{
> + fprintf
Robin Dapp writes:
>>> Bit tests on x86 also truncate [1], if the bit base operand specifies
>>> a register, and we don't use BT with a memory location as a bit base.
>>> I don't know what is referred with "(real or pretended) bit field
>>> operations" in the documentation for
在 2019/3/26 下午7:40, JunMa 写道:
Hi
According to gnu document of function attributes, neither weakref nor
alias
could be attached to a function defined in current translation unit.
Although GCC checks the attributes under some circumstances, it still
fails
on some cases and even causes ICE.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89100 see comment 4
Please can someone commit the attached patch as I do not have commit rights.
Change logs:
For gcc/fortran
Jeff Law
Mark Eggleston
PR fortran/89100
* gfortran.texi: Add Default widths for F, G and I format
On 10/05/19 23:12 -0400, Ed Smith-Rowland via libstdc++ wrote:
No complete patch because we could just accumulate a few of these.
Thanks. I've committed it to trunk and gcc-9-branch, because I'm
probably going to be regenerating the HTML again soon.
Index: doc/xml/manual/status_cxx2020.xml
With the GCC 9.1 release out of the door, this patch implements the
removal of Solaris 10 support that had been obsoleted in GCC 9.
The patch is mostly straightforward:
* removing Solaris 10-only code and documentation,
* simplifying configure triplets, using *-*-solaris2* everywhere since
Hi,
While working on other PR, I noticed that we can save lots of if-then-else in
accessing
cand_vec by placing an additional NULL element at front of it.
Bootstrap and test on x86_64. Is it OK?
Thanks,
bin
2019-05-13 Bin Cheng
* gimple-ssa-strength-reduction.c (lookup_cand):
> > No, this is not OK.
> >
> > I'd rather keep the simple current logic and either stick to cp -p, or
> > use a proper $(INSTALL_whatever) as done elsewhere rather than adding more
> > kludges.
>
> how do you mean, `proper $(INSTALL_whatever)'?
Using e.g. INSTALL_DATA from configure.
> I've
Another pick from the branch.
Bootstrapped / tested on x86_64-unknown-linux-gnu, applied.
Richard.
>From 7a2cf96fb0846d1fde52e11f9c1501f1ae70a94d Mon Sep 17 00:00:00 2001
From: Richard Guenther
Date: Mon, 21 Jan 2019 16:15:02 +0100
Subject: [PATCH] add-slp-vce-support
*
Hi Jonathan,
> On 13/05/19 13:39 +0200, Rainer Orth wrote:
>> libstdc++-v3:
>> * config/os/solaris/solaris2.10: Move to ...
>> * config/os/solaris: ... this.
>> * configure.host (os_include_dir): Adapt.
>> (abi_baseline_pair): Remove Solaris 10 handling.
>> *
Sorry for the slow reply.
Matthew Malcomson writes:
> @@ -326,16 +326,18 @@ int opt_ext_cmp (const void* a, const void* b)
> turns on as a dependency. As an example +dotprod turns on FL_DOTPROD
> and
> FL_SIMD. As such the set of bits represented by this option is
>
On 13/05/19 14:02 +0200, Christophe Lyon wrote:
Hi Jonathan,
On Fri, 10 May 2019 at 23:41, Jonathan Wakely wrote:
More Doxygenation.
Tested powerpc64le-linux. Committed to trunk.
You've probably noticed by now, but this introduced a regression (!):
FAIL: g++.dg/cpp0x/Wattributes1.C
On Mon, May 13, 2019 at 1:39 PM bin.cheng wrote:
>
> Hi,
> While working on other PR, I noticed that we can save lots of if-then-else in
> accessing
> cand_vec by placing an additional NULL element at front of it.
>
> Bootstrap and test on x86_64. Is it OK?
Nice.
OK.
Richard.
> Thanks,
>
Hi Arnaud,
>> Tested on x86_64-pc-linux-gnu installing both to a local filesystem and
>> an NFSv3 filesystem.
>>
>> Ok for mainline (and the gcc-9 and gcc-8 branches eventually)?
>
> No, this is not OK.
>
> I'd rather keep the simple current logic and either stick to cp -p, or
> use a proper
Hi Bence,
> I tried building a cross-compiler for `sparc-sun-solaris2`, but the
> resulting GCC executable wouldn't compile anything, always failing
> with: "as: unrecognized option '-m32'". After extensive research, I
> found that `as` takes `--32`/`--64`, instead of `-m32`/`-m64`. The
> patched
在 2019/5/13 下午5:02, Jakub Jelinek 写道:
On Mon, May 13, 2019 at 04:53:52PM +0800, JunMa wrote:
According to gnu document of function attributes, neither weakref nor
alias
could be attached to a function defined in current translation unit.
Although GCC checks the attributes under some
On 11/05/19 20:56 +0100, Jonathan Wakely wrote:
The Profile Mode extension is not used by anybody, nor maintained by
anybody. The containers do not support the full API specified in recent
standards, and so enabling Profile Mode is not source compatible with
much modern C++ code. The heuristics
Hi Richard,
On 5/9/19 9:06 AM, Richard Sandiford wrote:
Kyrill Tkachov writes:
+;; Helper expander for aarch64_abd_3 to save the callers
+;; the hassle of constructing the other arm of the MINUS.
+(define_expand "abd_3"
+ [(use (match_operand:VDQ_BHSI 0 "register_operand"))
+
Martin Liška writes:
> Hi.
>
> dump_printf_loc is a variadic function and the usafe in tree-vect-stmts.c
> is wrongly passing 2nd part of the string format as a first variadic argument.
> That's why I saw such a strange crashes.
>
> I'm going to install the patch as soon as it finishes tests.
Hi Jonathan,
On Fri, 10 May 2019 at 23:41, Jonathan Wakely wrote:
>
> More Doxygenation.
>
> Tested powerpc64le-linux. Committed to trunk.
>
>
You've probably noticed by now, but this introduced a regression (!):
FAIL: g++.dg/cpp0x/Wattributes1.C -std=c++14 (test for warnings, line 125)
This
On 2019/2/26 9:51 PM, Thomas Schwinge wrote:
On Tue, 26 Feb 2019 01:49:09 +0800, Chung-Lin Tang
wrote:
I have incorporated all your patches you've included in the last mail (with
some modifications, though pretty minor I think).
OK, thanks, that's good for next GCC development stage 1 as far
On Mon, May 13, 2019 at 04:53:52PM +0800, JunMa wrote:
> > According to gnu document of function attributes, neither weakref nor
> > alias
> > could be attached to a function defined in current translation unit.
> > Although GCC checks the attributes under some circumstances, it still
> > fails
>
On 5/10/19 11:21 AM, Jakub Jelinek wrote:
> On Fri, May 10, 2019 at 11:04:12AM +0200, Martin Liška wrote:
>> --- a/gcc/config/i386/i386.h
>> +++ b/gcc/config/i386/i386.h
>> @@ -1906,6 +1906,9 @@ typedef struct ix86_args {
>>
>> #define CLEAR_RATIO(speed) ((speed) ? MIN (6,
On Sun, May 12, 2019 at 10:07:28PM -0500, Jiufu Guo wrote:
> To elimiate the instructions which moving result of CMP to GPR, I'm
> wondering maybe we could do a combine(or a peephole) after bbro to let
> the condition jump directly using the result of CMP.
It will have to be a peephole. And
On 5/13/19 1:19 PM, Richard Sandiford wrote:
> Martin Liška writes:
>> Hi.
>>
>> dump_printf_loc is a variadic function and the usafe in tree-vect-stmts.c
>> is wrongly passing 2nd part of the string format as a first variadic
>> argument.
>> That's why I saw such a strange crashes.
>>
>> I'm
On Sun, May 12, 2019 at 2:51 PM Marc Glisse wrote:
>
> On Sun, 12 May 2019, Richard Sandiford wrote:
>
> > Marc Glisse writes:
> >> Hello,
> >>
> >> this patch lets gcc know that if a pointer existed before the call to
> >> malloc, the result of malloc cannot alias it. This is a bit ad hoc and
>
On Thu, May 9, 2019 at 11:55 AM Thomas Schwinge wrote:
> On Sun, 10 Feb 2019 20:51:39 +0100, Uros Bizjak wrote:
> > On Fri, Feb 8, 2019 at 1:24 PM Uros Bizjak wrote:
> > > Attached patch fixes --enable-frame-pointer handling [...]
>
> ACK.
>
> > Please note that this fix will re-enable frame
Hi all,
Vectorise __builtin_signbit (v2sf, v4sf) with unsigned shift right vector
instruction.
Bootstrapped and tested on aarch64-none-linux-gnu.
Assembly output for:
$ aarch64-elf-gcc -S -O3 signbitv2sf.c -dp
Before patch:
foo:
ldp w2, w1, [x1]// 37 [c=0 l=4]
The following switches PRE insertion from a dominator walk to a
RPO walk which should reduce the number of iterations because
we know AVAIL_OUT will be up-to-date for all predecessors
(apart from backedges).
I have a more elaborate patch but that somehow fails bootstrap
comparison so I'm doing
Hi Martin,
+++ b/gcc/testsuite/gcc.c-torture/pr90263.c
@@ -0,0 +1,9 @@
+/* PR middle-end/90263 */
+
+int *f (int *p, int *q, long n)
+{
+ return __builtin_mempcpy (p, q, n);
+}
+
+/* { dg-final { scan-assembler "mempcpy" { target { i?86-*-* x86_64-*-* } } }
} */
+/* { dg-final { scan-assembler
On Fri, May 10, 2019 at 2:40 PM Nathan Sidwell wrote:
>
> When poking at some dwarf bugs, which were ultimately fixed by Richard,
> I made a couple of cleanups. The first two are pretty obvious comment
> clarification, but the last fragment is a more invasive control flow
> change. In that case
On Mon, 13 May 2019, Segher Boessenkool wrote:
> On Sun, May 12, 2019 at 10:07:28PM -0500, Jiufu Guo wrote:
> > To elimiate the instructions which moving result of CMP to GPR, I'm
> > wondering maybe we could do a combine(or a peephole) after bbro to let
> > the condition jump directly using the
This makes /gcc-10/changes.html valid HTML5.
Committed to CVS.
Index: htdocs/gcc-10/changes.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-10/changes.html,v
retrieving revision 1.2
diff -u -r1.2 changes.html
---
Here's the libgo part of the Solaris 10 removal patch. I wonder if the
cleanup is worthwhile given the small size of the patch.
Bootstrapped without regressions on i386-pc-solaris2.11 and
sparc-sun-solaris2.11 together with the main patch.
Rainer
--
On 13/05/19 13:39 +0200, Rainer Orth wrote:
libstdc++-v3:
* config/os/solaris/solaris2.10: Move to ...
* config/os/solaris: ... this.
* configure.host (os_include_dir): Adapt.
(abi_baseline_pair): Remove Solaris 10 handling.
*
Some notes for /gcc-9/porting_to.html (I haven't seen this change
cause any issues in practice, but it doesn't hurt to document it).
Committed to CVS.
Index: htdocs/gcc-9/porting_to.html
===
RCS file:
On Mon, May 13, 2019 at 12:14:37PM +0200, Martin Liška wrote:
> On 5/10/19 11:21 AM, Jakub Jelinek wrote:
> > On Fri, May 10, 2019 at 11:04:12AM +0200, Martin Liška wrote:
> >> --- a/gcc/config/i386/i386.h
> >> +++ b/gcc/config/i386/i386.h
> >> @@ -1906,6 +1906,9 @@ typedef struct ix86_args {
> >>
Hi.
dump_printf_loc is a variadic function and the usafe in tree-vect-stmts.c
is wrongly passing 2nd part of the string format as a first variadic argument.
That's why I saw such a strange crashes.
I'm going to install the patch as soon as it finishes tests.
Thanks,
Martin
gcc/ChangeLog:
On Mon, 13 May 2019, Richard Biener wrote:
On Sun, May 12, 2019 at 2:51 PM Marc Glisse wrote:
On Sun, 12 May 2019, Richard Sandiford wrote:
Marc Glisse writes:
Hello,
this patch lets gcc know that if a pointer existed before the call to
malloc, the result of malloc cannot alias it. This
On 5/10/19 9:17 AM, Jakub Jelinek wrote:
> On Fri, May 10, 2019 at 01:46:07PM +, Wilco Dijkstra wrote:
>> And looking at the generated code, emitting a tailcall for this case is
>> actually a
>> de-optimization since the large eh_return epilog must be repeated for every
>> tailcall.
>
> On
> On 13 May 2019, at 20:10, Jakub Jelinek wrote:
>
> On Mon, May 13, 2019 at 12:52:54PM -0600, Jeff Law wrote:
>> On 5/10/19 9:17 AM, Jakub Jelinek wrote:
>>> On Fri, May 10, 2019 at 01:46:07PM +, Wilco Dijkstra wrote:
And looking at the generated code, emitting a tailcall for this
Hi Ian,
> Thanks. The patch to mksysinfo.sh wasn't quite right. This was the
> patch that I committed.
that was the only part I'd added at the very last minute without testing
it. I'd already noticed the mess myself earlier this evening when it
broke my own bootstraps. Sorry for the mess and
* doc/xml/manual/policy_data_structures.xml: Comment out stray
elements. Fix formatting of bibliography references.
Committed to trunk.
commit 97987cf2b2e7549e3c864e257f7e9f07c05db95a
Author: Jonathan Wakely
Date: Mon May 13 17:24:56 2019 +0100
Small markup changes to
Make the filesystem::path constructors SFINAE away for void* arguments,
instead of giving an error due to iterator_traits::reference.
PR libstdc++/90454.cc path construction from void*
* include/bits/fs_path.h (path::_Path): Use remove_pointer so that
pointers to void are
On 13/05/19 13:11 +0100, Jonathan Wakely wrote:
On 13/05/19 14:02 +0200, Christophe Lyon wrote:
Hi Jonathan,
On Fri, 10 May 2019 at 23:41, Jonathan Wakely wrote:
More Doxygenation.
Tested powerpc64le-linux. Committed to trunk.
You've probably noticed by now, but this introduced a
From: Rainer Orth
> Here's the libgo part of the Solaris 10 removal patch. I wonder if the
> cleanup is worthwhile given the small size of the patch.
>
> Bootstrapped without regressions on i386-pc-solaris2.11 and
> sparc-sun-solaris2.11 together with the main patch.
Thanks. The patch to
The releasing_vec class is useful anywhere we want to release_tree_vector on
exit, particularly in the presence of multiple returns.
Tested x86_64-pc-linux-gnu, applying to trunk.
* cp-tree.h (struct releasing_vec): Replace get_ref method with
operator&.
(vec_safe_push,
On 5/13/19 10:49 AM, Jakub Jelinek wrote:
On Mon, May 13, 2019 at 10:36:00AM -0600, Martin Sebor wrote:
On 5/11/19 5:33 PM, Marc Glisse wrote:
Hello,
this patch lets gcc know that if a pointer existed before the call to
malloc, the result of malloc cannot alias it. This is a bit ad hoc and
On 5/13/19 1:32 AM, Richard Sandiford wrote:
>
> IMO we should only consider deprecating the macro. I'm not sure how
> much practical effect that will have though. cc0 was deprecated ages
> ago but we're a still a long way from getting rid of it. :-)
Yea, but (in theory) removing
On 5/13/19 11:37 AM, Marc Glisse wrote:
On Mon, 13 May 2019, Martin Sebor wrote:
On 5/11/19 5:33 PM, Marc Glisse wrote:
Hello,
this patch lets gcc know that if a pointer existed before the call to
malloc, the result of malloc cannot alias it. This is a bit ad hoc
and fragile. A small
On Mon, May 13, 2019 at 12:52:54PM -0600, Jeff Law wrote:
> On 5/10/19 9:17 AM, Jakub Jelinek wrote:
> > On Fri, May 10, 2019 at 01:46:07PM +, Wilco Dijkstra wrote:
> >> And looking at the generated code, emitting a tailcall for this case is
> >> actually a
> >> de-optimization since the
On Mon, May 13, 2019 at 07:57:30PM +, Wilco Dijkstra wrote:
> Hi,
>
> > I think my worry would be all the targets that support tail calls, but
> > which we can't easily test. ie, we put in the hack, but when do we
> > know it can be removed?
>
> Well it would be easy to make a runnable
On 5/6/19 2:34 AM, Paul Hua wrote:
> The attached patch fix pr90357, bootstraped and regressed test on
> mips64el-linux-gnu target.
> Ok for commit ?
>
>
> 0001-PATCH-MIPS-Skip-forward-src-into-next-insn-when-the-.patch
>
> From a3db8763ee8460a5f63c567d58624a985f9924ce Mon Sep 17 00:00:00 2001
* g++.dg/cpp0x/Wattributes1.C: Adjust dg-error line number to fix
regression, by matching a note on any line.
* g++.dg/cpp0x/Wattributes2.C: Add another copy that checks the
correct line number is matched without depending on a library header.
Tested
On 5/13/19 1:55 PM, Jonathan Wakely wrote:
> * g++.dg/cpp0x/Wattributes1.C: Adjust dg-error line number to fix
> regression, by matching a note on any line.
> * g++.dg/cpp0x/Wattributes2.C: Add another copy that checks the
> correct line number is matched without depending on a
Hi,
> I think my worry would be all the targets that support tail calls, but
> which we can't easily test. ie, we put in the hack, but when do we
> know it can be removed?
Well it would be easy to make a runnable torture test. I'm not sure
how easy it would be to do it using a scan-assembler
On Mon, 13 May 2019, Martin Sebor wrote:
On 5/11/19 5:33 PM, Marc Glisse wrote:
Hello,
this patch lets gcc know that if a pointer existed before the call to
malloc, the result of malloc cannot alias it. This is a bit ad hoc and
fragile. A small improvement would be, when the 2 statements
sh3-linux-gnu and sh3eb-linux-gnu:
Tests that now fail, but worked before (3 tests):
gcc.target/sh/pr51244-11.c scan-assembler-not subc|and|bra
gcc.target/sh/pr51244-11.c scan-assembler-times bf\t0f 1
gcc.target/sh/pr51244-11.c scan-assembler-times bt\t0f 1
Previously we'd match this
On Mon, May 13, 2019 at 08:27:15AM -0600, Jeff Law wrote:
>
> sh3-linux-gnu and sh3eb-linux-gnu:
>
>
> Tests that now fail, but worked before (3 tests):
>
> gcc.target/sh/pr51244-11.c scan-assembler-not subc|and|bra
> gcc.target/sh/pr51244-11.c scan-assembler-times bf\t0f 1
>
I'm applying this obvious patch. Two hunks deal with inappropriate line
breaks, and the first changes the baroque:
var = (thing) ? thing : var;
into the much clearer
if (thing)
var = thing;
nathan
--
Nathan Sidwell
2019-05-13 Nathan Sidwell
gcc.c
* (execute): Simplify
On Mon, May 13, 2019 at 08:27:15AM -0600, Jeff Law wrote:
> sh3-linux-gnu and sh3eb-linux-gnu:
I test sh2 and sh4, but not sh3 :-)
> Tests that now fail, but worked before (3 tests):
>
> gcc.target/sh/pr51244-11.c scan-assembler-not subc|and|bra
> gcc.target/sh/pr51244-11.c scan-assembler-times
Tester started failing Wattributes1.C on various targets after this change:
> commit de3f1d9aabb765f78d127696ff9dd0a83b268aa2 (HEAD, refs/bisect/bad)
> Author: redi
> Date: Fri May 10 21:41:11 2019 +
>
> Improve API docs for and
>
> * include/bits/shared_ptr.h:
On 5/13/19 9:08 AM, Jonathan Wakely wrote:
> On 13/05/19 08:56 -0600, Jeff Law wrote:
>> Tester started failing Wattributes1.C on various targets after this
>> change:
>>
>>
>>> commit de3f1d9aabb765f78d127696ff9dd0a83b268aa2 (HEAD, refs/bisect/bad)
>>> Author: redi
>>> Date: Fri May 10
On Mon, May 13, 2019 at 11:41:58PM +0900, Oleg Endo wrote:
> On Mon, 2019-05-13 at 09:38 -0500, Segher Boessenkool wrote:
> > On Mon, May 13, 2019 at 08:27:15AM -0600, Jeff Law wrote:
> > > Tests that now fail, but worked before (3 tests):
> > >
> > > gcc.target/sh/pr51244-11.c scan-assembler-not
On 5/13/19 8:38 AM, Segher Boessenkool wrote:
> On Mon, May 13, 2019 at 08:27:15AM -0600, Jeff Law wrote:
>> sh3-linux-gnu and sh3eb-linux-gnu:
>
> I test sh2 and sh4, but not sh3 :-)
>
>> Tests that now fail, but worked before (3 tests):
>>
>> gcc.target/sh/pr51244-11.c scan-assembler-not
The Hollerith constants support section in the gfortran manual stated
that they can be used with the ASSIGN statement. This is not the case as
a compilation error occurs and in any event the value of any such
constant would be larger than the largest valid label (9).
I've added some extra
On 13/05/19 08:56 -0600, Jeff Law wrote:
Tester started failing Wattributes1.C on various targets after this change:
commit de3f1d9aabb765f78d127696ff9dd0a83b268aa2 (HEAD, refs/bisect/bad)
Author: redi
Date: Fri May 10 21:41:11 2019 +
Improve API docs for and
*
From: Rainer Orth
Date: Mon, May 13, 2019 at 4:39 AM
> With the GCC 9.1 release out of the door, this patch implements the
> removal of Solaris 10 support that had been obsoleted in GCC 9.
...
> The vast majority of the patch falls squarely under my Solaris
> maintainership. I'm uncertain
On 5/11/19 5:33 PM, Marc Glisse wrote:
Hello,
this patch lets gcc know that if a pointer existed before the call to
malloc, the result of malloc cannot alias it. This is a bit ad hoc and
fragile. A small improvement would be, when the 2 statements are in the
same bb but in the wrong order,
On Mon, May 13, 2019 at 10:36:00AM -0600, Martin Sebor wrote:
> On 5/11/19 5:33 PM, Marc Glisse wrote:
> > Hello,
> >
> > this patch lets gcc know that if a pointer existed before the call to
> > malloc, the result of malloc cannot alias it. This is a bit ad hoc and
> > fragile. A small
On Mon, May 13, 2019 at 09:38:00AM -0500, Segher Boessenkool wrote:
> I tested on 30-something targets (all *-linux), and only mips64 regressed
> a little, everything else improved.
I meant alpha, btw. 0.08% code size growth. And the biggest winner is
i386, with a similar shrink. Most targets
On Mon, 2019-05-13 at 09:38 -0500, Segher Boessenkool wrote:
> On Mon, May 13, 2019 at 08:27:15AM -0600, Jeff Law wrote:
> > sh3-linux-gnu and sh3eb-linux-gnu:
>
> I test sh2 and sh4, but not sh3 :-)
>
> > Tests that now fail, but worked before (3 tests):
> >
> > gcc.target/sh/pr51244-11.c
Following the same pattern as the AArch64 patch, this patch fixes the
PPC32 BE and AIX EH failures after the sibcall changes.
This does not address the Darwin failures because of SAVE_WORLD.
Bootstrapped on powerpc-ibm-aix7.2.0.0, powerpc64le-unknown-linux-gnu,
powerpc64-unknown-linux-gnu,
Here is a version of the patch with a cheaper test, and an extra testcase
for Martin.
(I kept the tree-ssa-loop-niter.c part although I am not using it anymore)
2019-05-15 Marc Glisse
gcc/
* tree-ssa-loop-niter.c (stmt_dominates_stmt_p): Handle NULL
basic block.
*
From: Kewen Lin
Previous version link for background:
https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00912.html
This hook is to predict whether one loop in gimple will
be transformed to low-overhead loop later in RTL, and
designed to be called in ivopts.
"Since the low-overhead loop optimize
From: Kewen Lin
Previous version link for background:
https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00912.html
Firstly, it's to call predict_doloop_p hook to check this
loop will be transformed to doloop or not, if yes, find
the expected comp iv use and its dependent original iv,
set the iv
From: Kewen Lin
Previous version link:
https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00912.html
This is a NFC (no functional change) patch. Ivopts has
some codes to expand gimple to RTL seq, but before call
expanding, we should call a preparation funciton
prepare_decl_rtl. This patch is to
This patch implements CWG 2096 which relaxes the constraints for literal unions.
[basic.types]/p10 now says: A type is a literal type if...
-- it is a union, at least one of its non-static data members is of
non-volatile literal type,
[...]
check_field_decls is called with CLASSTYPE_LITERAL_P
On 5/13/19 2:09 PM, Jakub Jelinek wrote:
> On Mon, May 13, 2019 at 07:57:30PM +, Wilco Dijkstra wrote:
>> Hi,
>>
>>> I think my worry would be all the targets that support tail calls, but
>>> which we can't easily test. ie, we put in the hack, but when do we
>>> know it can be removed?
>>
>>
On Mon, May 13, 2019 at 04:29:21PM -0600, Jeff Law wrote:
> > That is a serious misunderstanding of what __builtin_eh_return does.
> > For the most part, __builtin_eh_return works by forcing all call clobbered
> > registers to stack, having that described in the unwind info and having the
> >
Hi Jakub,
>> We could simpify __builtin_eh_return and implement it as a libgcc
>> function or automatically generated thunk. Basically all you need is
>> tailcall a 2-instruction helper which adjusts the stack and then jumps
>> to the given function pointer. This way you don't need the complex
On 5/10/19 3:38 PM, Jonathan Wakely wrote:
This seems generally OK, but ...
On Fri, 10 May 2019, 05:59 François Dumont wrote:
I remove several _M_map != nullptr checks cause in current
implementation it can't be null. I have several patches following this
one to support it but in this
On Tue, Mar 19, 2019 at 10:09 PM Alex Henrie wrote:
>
> On Tue, Mar 19, 2019 at 11:03 AM Martin Liška wrote:
> >
> > What's Joseph telling is that right now the patch can't be approved (and
> > installed).
> > It's due to fact that we are close to the release in stage4:
> >
Hi
This is the patch on vector to:
- Optimize sizeof in Versioned namespace mode. We could go one step
further by removing _M_p from _M_finish and just transform it into an
offset but it is a little bit more impacting for the code.
- Implement the swap optimization already done on main
84 matches
Mail list logo
