On 30/08/18 21:56 +0200, François Dumont wrote:
--- a/libstdc++-v3/include/debug/safe_iterator.h
+++ b/libstdc++-v3/include/debug/safe_iterator.h
@@ -36,6 +36,28 @@
#include
#include
+#define _GLIBCXX_DEBUG_VERIFY_OPERANDS(_Lhs, _Rhs, _BadMsgId, _DiffMsgId) \
+ _GLIBCXX_DEBUG_VERIFY(!_Lhs._M_
On Fri, Aug 31, 2018 at 1:32 PM, Jason Merrill wrote:
> On Fri, Aug 31, 2018 at 3:33 PM, H.J. Lu wrote:
>> On Thu, Aug 30, 2018 at 10:21 AM, Jason Merrill wrote:
>>>
r138335 allowed arg_pointer_rtx to be eliminated by either FP or SP,
but only when dynamic stack alignment is supported.
On Fri, Aug 24, 2018 at 11:38 AM, Tom de Vries wrote:
> [ was: Re: [PATCH][debug] Add -gdescriptive-dies ]
> On Fri, Aug 24, 2018 at 12:44:38PM +0200, Richard Biener wrote:
>> On Wed, 22 Aug 2018, Tom de Vries wrote:
>>
>> > [ was: Re: [PATCH][debug] Add -gforce-named-dies ]
>> >
>> > On 08/22/201
On 07/23/2018 05:24 PM, H.J. Lu wrote:
On Mon, Jun 18, 2018 at 12:26 PM, Joseph Myers wrote:
On Mon, 18 Jun 2018, Jason Merrill wrote:
On Mon, Jun 18, 2018 at 11:59 AM, Joseph Myers wrote:
On Mon, 18 Jun 2018, Jason Merrill wrote:
+ if (TREE_CODE (rhs) == COND_EXPR)
+{
+ /* Chec
OK. And I'd say that this sort of error recovery qualifies as obvious.
Jason
On Fri, Aug 31, 2018 at 5:32 AM, Paolo Carlini wrote:
> Hi,
>
> another straightforward ICE on invalid. I spent however quite a bit of time
> on it, because I tried to catch the error_mark_node as early as possible,
On Fri, Aug 31, 2018 at 3:33 PM, H.J. Lu wrote:
> On Thu, Aug 30, 2018 at 10:21 AM, Jason Merrill wrote:
>>
>>> r138335 allowed arg_pointer_rtx to be eliminated by either FP or SP,
>>> but only when dynamic stack alignment is supported. In this case,
>>> arg_pointer_rtx is eliminated by FP even
On Thu, Aug 30, 2018 at 10:21 AM, Jason Merrill wrote:
>
>> r138335 allowed arg_pointer_rtx to be eliminated by either FP or SP,
>> but only when dynamic stack alignment is supported. In this case,
>> arg_pointer_rtx is eliminated by FP even when frame_pointer_needed
>> is false and there is no d
On 2018-08-13 07:58 PM, Michael Ploujnikov wrote:
> Ping and I've updated the patch since last time as follows:
>
> - unittest scans assembly rather than the constprop dump because its
> forward changed
> - unittests should handle different hosts where any of
> NO_DOT_IN_LABEL, NO_DOLL
Ping. Any feedback from the Aarch64 maintainers?
Steve Ellcey
sell...@cavium.com
On Mon, 2018-08-20 at 10:37 -0700, Steve Ellcey wrote:
> On Tue, 2018-08-07 at 12:15 -0500, Segher Boessenkool wrote:
>
> >
> > >
> > > +/* { dg-final { scan-assembler-not "\[ \t\]stp\tq\[01234567\]" }
> > >
On 08/31/18 16:42, Jeff Law wrote:
> On 08/31/2018 01:08 AM, Bernd Edlinger wrote:
>> Hi,
>>
>>
>> when re-testing the new STRING_CST semantic patch here:
>> https://gcc.gnu.org/ml/gcc-patches/2018-08/msg01569.html
>>
>> I noticed that the (my) fix for PR 87053 does still use
>> semantic properties
On 08/31/18 18:45, Jeff Law wrote:
> On 08/26/2018 01:45 AM, Bernd Edlinger wrote:
>
cp:
2018-08-24 Bernd Edlinger
* decl.c (eval_check_narrowing): Remove.
(check_initializer): Move call to braced_list_to_string from here ...
* typeck2.c (store_init_value)
On 08/26/2018 01:45 AM, Bernd Edlinger wrote:
>>> cp:
>>> 2018-08-24 Bernd Edlinger
>>>
>>> * decl.c (eval_check_narrowing): Remove.
>>> (check_initializer): Move call to braced_list_to_string from here ...
>>> * typeck2.c (store_init_value): ... to here.
>>> (digest_init_r): Re
On 08/28/2018 11:53 PM, James Greenhalgh wrote:
On Wed, Aug 01, 2018 at 10:07:23AM -0500, Sam Tebbs wrote:
+ ones from the MSB. */
+bool
+aarch64_is_left_consecutive (HOST_WIDE_INT i)
+{
+ return (i | (i - 1)) == HOST_WIDE_INT_M1;
exact_log2(-i) != HOST_WIDE_INT_M1?
I could change this
On Fri, 31 Aug 2018, Jason Merrill wrote:
> > Anonymous structures and unions are in C11 (and before that a widely
> > accepted extension).
>
> This isn't an anonymous union, it's named "x1". I don't think that
> line declares a field in C, either.
Indeed, the case where the field has a tag is
On Fri, 31 Aug 2018, Martin Jambor wrote:
> diff --git a/gcc/common.opt b/gcc/common.opt
> index ebc3ef43ce2..2950760fb2a 100644
> --- a/gcc/common.opt
> +++ b/gcc/common.opt
> @@ -815,6 +815,10 @@ Wvector-operation-performance
> Common Var(warn_vector_operation_performance) Warning
> Warn when
On Fri, Aug 31, 2018 at 10:35 AM, Michael Matz wrote:
> Hi,
>
> On Thu, 30 Aug 2018, Nathan Sidwell wrote:
>
>> > struct foo
>> >{
>> > unsigned a : 21;
>> > union x1 { char x; };
>> > unsigned b : 11;
>> > union x1 u;
>> >};
>>
>> (for C++) this happens to be a case we
On Fri, 31 Aug 2018, Kyrill Tkachov wrote:
> > That sounds good! Although I'm confused about what the 'p' means.
>
> It stands for "predicate" meaning a boolean function with no side-effects.
> It's the preferred way to name these kinds of functions in GCC (though I can't
> seem to find the docume
On 31/08/18 16:27, Sam Tebbs wrote:
On 08/31/2018 11:59 AM, Kyrill Tkachov wrote:
>
> On 30/08/18 16:53, Sam Tebbs wrote:
>>
>>
>> On 08/28/2018 11:53 PM, James Greenhalgh wrote:
>> > Hm, I'm not very sure about the naming here; "left consecutive"
>> isn't a
>> > common phrase to denote the ma
On 08/31/2018 11:59 AM, Kyrill Tkachov wrote:
On 30/08/18 16:53, Sam Tebbs wrote:
On 08/28/2018 11:53 PM, James Greenhalgh wrote:
> Hm, I'm not very sure about the naming here; "left consecutive"
isn't a
> common phrase to denote the mask you're looking for (exact_log2
(-i) != -1
> if I'm
On 28/08/18 22:58, James Greenhalgh wrote:
On Tue, Aug 28, 2018 at 03:59:25AM -0500, Vlad Lazar wrote:
Gentle ping.
On 08/08/18 17:38, Vlad Lazar wrote:
On 01/08/18 18:35, James Greenhalgh wrote:
On Wed, Aug 01, 2018 at 07:13:53AM -0500, Vlad Lazar wrote:
On 31/07/18 22:48, James Greenhalgh
On 08/31/2018 01:08 AM, Bernd Edlinger wrote:
> Hi,
>
>
> when re-testing the new STRING_CST semantic patch here:
> https://gcc.gnu.org/ml/gcc-patches/2018-08/msg01569.html
>
> I noticed that the (my) fix for PR 87053 does still use
> semantic properties of the STRING_CST that is not compatible
On Fri, 31 Aug 2018, Vinay Kumar wrote:
> diff --git a/gcc/ChangeLog b/gcc/ChangeLog
> index 1fbcbd5..8dc9fb4 100644
> --- a/gcc/ChangeLog
> +++ b/gcc/ChangeLog
> @@ -1,3 +1,8 @@
> +2018-08-31 Vinay Kumar
> +
> + * doc/invoke.texi (-Wreturn-type): Document new warning
> + -Wprio-ctor-dt
Hi,
On Thu, 30 Aug 2018, Nathan Sidwell wrote:
> > struct foo
> > {
> > unsigned a : 21;
> > union x1 { char x; };
> > unsigned b : 11;
> > union x1 u;
> > };
>
> (for C++) this happens to be a case we already get right. I think
> that'd be a C vendor extension, I don
On Fri, Aug 31, 2018 at 04:31:09PM +0200, Richard Biener wrote:
> On August 31, 2018 4:11:26 PM GMT+02:00, Michael Matz wrote:
> >Hi,
> >
> >as Jakub correctly identified, uses of values produced by the inner
> >loop
> >that occur inside the outer loop prevent fusion as well. We are
> >in LCSSA
On August 31, 2018 4:11:26 PM GMT+02:00, Michael Matz wrote:
>Hi,
>
>as Jakub correctly identified, uses of values produced by the inner
>loop
>that occur inside the outer loop prevent fusion as well. We are
>in LCSSA so the check is easy. (Uses outside the outer loop are okay,
>see
>the comm
On Fri, Aug 31, 2018 at 10:04:07AM -0400, Nathan Sidwell wrote:
> On 08/30/2018 04:27 PM, Jason Merrill wrote:
>
> > On Thu, Aug 30, 2018 at 3:31 PM, Julian Brown
> > wrote:
> >
> > > "Apart from parsing, it's necessary to prevent the "cannot take the
> > > address of 'this', which is an rvalue
On Wed, 29 Aug 2018 at 13:01, Kyrill Tkachov
wrote:
>
> Hi Christophe,
>
> On 13/07/18 17:11, christophe.l...@st.com wrote:
> > From: Christophe Lyon
> >
> > In FDPIC mode, the trampoline generated to support pointers to nested
> > functions looks like:
> >
> >.word trampoline address
Attached is an nvptx patch that adds support for a new, albeit rarely
used, compiler option -misa. At present, there are only two valid ISA
arguments, SM_30 and SM_35. Without that flag, GCC will default to
SM_30. The major advantage of using the SM_35 ISA is to enable the use
PTX atom instructions
On Wed, 29 Aug 2018 at 12:46, Kyrill Tkachov
wrote:
>
>
> On 13/07/18 17:10, christophe.l...@st.com wrote:
> > From: Christophe Lyon
> >
> > The new arm-uclinuxfdpiceabi target behaves pretty much like
> > arm-linux-gnueabi. In order the enable the same set of features, we
> > have to update seve
Hi,
as Jakub correctly identified, uses of values produced by the inner loop
that occur inside the outer loop prevent fusion as well. We are
in LCSSA so the check is easy. (Uses outside the outer loop are okay, see
the comment)
Regstrapping on x86-64-linux in progress. Okay for trunk?
Cia
On Wed, 29 Aug 2018 at 12:46, Kyrill Tkachov
wrote:
>
> Hi Christophe,
>
> On 13/07/18 17:10, christophe.l...@st.com wrote:
> > From: Christophe Lyon
> >
> > 2018-XX-XX Christophe Lyon
> > Mickaël Guêné
> >
> > gcc/
> > * config/arm/arm.opt: Add -mfdpic option.
> >
>
Hi Joseph,
>> Isn't it enabled by default, not by -Wall (so the main option is
>> -Wno-prio-ctor-dtor to disable the warning)?
Thanks for reviewing the patch and your suggestions.
Please find attached the modified patch as per your review comments.
Please review the patch and let me know if it
On 08/30/2018 04:27 PM, Jason Merrill wrote:
On Thu, Aug 30, 2018 at 3:31 PM, Julian Brown wrote:
"Apart from parsing, it's necessary to prevent the "cannot take the
address of 'this', which is an rvalue expression" error from appearing
Breaking a rather fundamental language attribute does
These functions require the shift amount to be an integer constant
expression in the range [1, N]. The patch adds a new hook so that
targets can check this.
Also, the ACLE asm tests try many combinations per file, so I'd added
code to parallelise them at single-file granularity. However,
the cod
> On Aug 30, 2018, at 9:02 PM, Jeff Law wrote:
>
> On 08/30/2018 10:58 AM, Richard Henderson wrote:
>> On 08/28/2018 07:13 AM, Jeff Law wrote:
>>> Please consider using function descriptors rather than trampolines.
>>> This allows you to make the stack non-executable at all times which is
>>>
I am testing the following patch to fix a missed optimization and
wrong-code issue in region VN.
{,O1-}Bootstrap and regtest running on x86_64-unknown-linux-gnu.
2018-08-31 Richard Biener
PR tree-optimization/87168
* tree-ssa-sccvn.c (SSA_VAL): Add visited output parameter.
>
> 2018-08-23 Martin Liska
>
> PR middle-end/59521
> * predict.c (set_even_probabilities): Add likely_edges
> argument and handle cases where we have precisely one
> likely edge.
> (combine_predictions_for_bb): Catch also likely_edges.
> (tree_predict_by_op
Hi,
thank you very much for your comments.
On Fri, Aug 24 2018, Joseph Myers wrote:
> On Fri, 24 Aug 2018, Martin Jambor wrote:
>
>> +/* Assuming we have encountered a call to a probably wrong kind of abs,
>> issue a
>> + warning. LOC is the location of the call, FNKIND is a string
>> charac
This fixes 87155 where an inline namespace caused us to find an inner
anonymous namespace.
The literal wording of the standard is wrong. Nobody implements that.
They implement essentially 1 of 2 different interpretations, which were
equivalent until inline namespaces and DR2061. Then one of
On Fri, Aug 31, 2018 at 1:31 PM Martin Jambor wrote:
>
> Hi,
>
> I have discovered the following thinko in IPA-CP's
> estimate_local_effects added during conversion to use nonspecialized
> time. The intent clearly was to add an upper bound to the time
> difference, not a lower one.
>
> The patch
On Fri, Aug 31, 2018 at 12:49 PM Martin Liška wrote:
>
> Hi.
>
> As noticed here:
> https://gcc.gnu.org/ml/gcc/2018-08/msg4.html
>
> There's ugly usage of a static variable in sem_item_optimizer
> that's called from sem_item.
>
> I believe logically the hash should live in sem_item.
>
> Patch
Hi,
I have discovered the following thinko in IPA-CP's
estimate_local_effects added during conversion to use nonspecialized
time. The intent clearly was to add an upper bound to the time
difference, not a lower one.
The patch introduces a guality failure:
gcc.dg/guality/pr41616-1.c -O3 -g
On 08/30/2018 06:24 PM, David Malcolm wrote:
Adding "const" to these indicates to both humans and the compiler in which
direction information is intended to flow. For example, compare:
extract_range_from_binary_expr_1, which takes two vr and modifies a third
with:
ranges_from_anti_range
Hi.
There are sometimes situations where a line is executed:
1G: 307:read_size (unsigned char *p, int size)
in such case a more fine value is appreciated:
1.4G: 307:read_size (unsigned char *p, int size)
Thus I'm suggesting to add one digit after dot.
Patch survives gcov.exp, will
On 30/08/18 19:19 -0600, Sandra Loosemore wrote:
One of the simulators we use for testing nios2-elf builds doesn't
include full semihosted fileio support. (IIRC it can do I/O to the
console and open and close files, but doesn't support fseek,
stat/fstat, unlink, etc). There are a bunch of lib
On Thu, 30 Aug 2018, Kyrill Tkachov wrote:
> Ping.
>
> https://gcc.gnu.org/ml/gcc-patches/2018-08/msg01496.html
>
> Thanks,
> Kyrill
>
> On 23/08/18 18:09, Kyrill Tkachov wrote:
> > Hi Richard,
> >
> > On 23/08/18 11:13, Richard Sandiford wrote:
> > > Kyrill Tkachov writes:
> > > > Hi all,
>
On 30/08/18 16:53, Sam Tebbs wrote:
On 08/28/2018 11:53 PM, James Greenhalgh wrote:
> Hm, I'm not very sure about the naming here; "left consecutive" isn't a
> common phrase to denote the mask you're looking for (exact_log2 (-i) != -1
> if I'm reading right), and is misleading 0x is '
Hi.
As noticed here:
https://gcc.gnu.org/ml/gcc/2018-08/msg4.html
There's ugly usage of a static variable in sem_item_optimizer
that's called from sem_item.
I believe logically the hash should live in sem_item.
Patch can bootstrap on x86_64-linux-gnu and survives regression tests.
Ready to
Hi all,
I'm committing this as obvious.
Thanks,
Kyrill
2018-08-31 Kyrylo Tkachov
* ipa-inline.c (can_inline_edge_by_limits_p): Fix typos in comment.
diff --git a/gcc/ipa-inline.c b/gcc/ipa-inline.c
index a84d1d9ad3e45935eff51f00b8f6b5058a691c9e..025788522fbe602e1d190a30ccea006b9ab053c5
On Thu, Aug 30, 2018 at 7:39 PM Martin Sebor wrote:
>
> On 08/30/2018 11:22 AM, Richard Biener wrote:
> > On August 30, 2018 6:54:21 PM GMT+02:00, Martin Sebor
> > wrote:
> >> On 08/30/2018 02:35 AM, Richard Biener wrote:
> >>> On Thu, Aug 30, 2018 at 2:12 AM Martin Sebor
> >> wrote:
>
> >
On Thu, Aug 30, 2018 at 5:09 PM JonY <10wa...@gmail.com> wrote:
>
> On 08/30/2018 11:59 AM, Nathan Sidwell wrote:
> > On 08/29/2018 11:06 PM, Liu Hao wrote:
> >
> >> It is strictly an ABI break but I doubt whether code in real world
> >> that got broken by this bug ever exists. Usually when people
Hi David,
On 28/08/2018 02:20, David Malcolm wrote:
This patch adds a fix-it hint to missing "typename" errors in the C++
frontend, suggesting the insertion of "typename ".
... it occurs to me, that probably we could have a fix-it hint for
another kind of missing typename (class) in testcases l
Hi,
another straightforward ICE on invalid. I spent however quite a bit of
time on it, because I tried to catch the error_mark_node as early as
possible, but that doesn't seem to work well error-recovery-wise: if,
say, we catch it in deduce_constrained_parameter we end-up emitting a
second r
On 08/30/2018 12:16 PM, Richard Biener wrote:
> On Wed, Aug 29, 2018 at 2:47 PM Martin Liška wrote:
>>
>> On 08/29/2018 01:06 PM, Richard Biener wrote:
>>> On Mon, Aug 27, 2018 at 12:00 PM Martin Liška wrote:
On 08/13/2018 03:00 PM, Martin Liška wrote:
> On 08/13/2018 02:54 PM, Rama
TARGET_FPRND should be on for everything ISA 2.04 and later, and
TARGET_VSX implies ISA 2.06 or later; but it is possible to disable
TARGET_FPRND (separately via -mno-fprnd, but also implicitly)
currently, and then things fall down. This patch makes things not
fall down.
Committing to trunk, and
On Thu, 30 Aug 2018, Jan Hubicka wrote:
> > On Fri, 10 Aug 2018, Jan Hubicka wrote:
> >
> > > Hi,
> > > this patch should fix merging of PIC and PIE options so we always resort
> > > to the least common denominator of the object files compiled (i.e.
> > > linking together -fpic and -fPIE will re
Hi,
when re-testing the new STRING_CST semantic patch here:
https://gcc.gnu.org/ml/gcc-patches/2018-08/msg01569.html
I noticed that the (my) fix for PR 87053 does still use
semantic properties of the STRING_CST that is not compatible
with the new proposed STRING_CST semantics.
That means that c
57 matches
Mail list logo
