Fix libjava build failure with glibc-2.26, replacing 'struct ucontext' with
ucontext_t. Ok for the gcc-5 and gcc-6 branches? Doesn't apply to 7 and trunk,
because libjava is removed.
Matthias
* include/x86_64-signal.h (HANDLE_DIVIDE_OVERFLOW): Replace
'struct ucontext' with ucon
Hi
The trace-div and trace-gep options seems be used to evaluate corpus
to trigger specific kind of bugs. And they don't have strong effect to coverage.
The trace-pc-guard is useful, but it may be much more complex than trace-pc.
I think the best benefit of trace-pc-guard is avoiding dealing ASLR
Fix asan build failure with glibc-2.26, not finding the SIGSEGV define. Solved
by explicitly including the signal.h header. Ok for the gcc-5 branch?
Matthias
PS: libsanitizer still needs the backport for PR 81066
* asan/asan_linux.cc: Include
--- a/src/libsanitizer/asan/asan_linux.c
On 14.07.2017 11:14, Jakub Jelinek wrote:
> Hi!
>
> glibc recently changed a couple of headers. One change was
> that typedef struct sigaltstack { ... } stack_t; is now
> typedef struct { ... } stack_t;
> and the other change is that resolv.h now ignores the former (private glibc
> macro) __need_
On Thu, Sep 7, 2017 at 9:01 AM, Matthias Klose wrote:
> Fix libjava build failure with glibc-2.26, replacing 'struct ucontext' with
> ucontext_t. Ok for the gcc-5 and gcc-6 branches? Doesn't apply to 7 and
> trunk,
> because libjava is removed.
Ok.
Richard.
> Matthias
>
> * include/x
On Thu, Sep 7, 2017 at 9:05 AM, Matthias Klose wrote:
> Fix asan build failure with glibc-2.26, not finding the SIGSEGV define.
> Solved
> by explicitly including the signal.h header. Ok for the gcc-5 branch?
Ok.
Richard.
> Matthias
>
> PS: libsanitizer still needs the backport for PR 81066
On Thu, Sep 7, 2017 at 9:08 AM, Matthias Klose wrote:
> On 14.07.2017 11:14, Jakub Jelinek wrote:
>> Hi!
>>
>> glibc recently changed a couple of headers. One change was
>> that typedef struct sigaltstack { ... } stack_t; is now
>> typedef struct { ... } stack_t;
>> and the other change is that r
This fixed default-def handling in VN. We can't really make those
VN_TOP as VN_TOP means unvisited rather than uninitialized. When
keeping things VN_TOP if unintialized we run into issues like
value-numbering a PHI without undefs equal to a PHI with undefs
and replacing the well-defined one with
On Wed, Sep 6, 2017 at 10:18 PM, Richard Sandiford
wrote:
> The next main step in the SVE submission is to add support for
> offsets and sizes that are a runtime invariant rather than a compile
> time constant. This is an RFC about our approach for doing that.
> It's an update of https://gcc.gnu.
Hi!
The C and C++ FE handle resolve_overloaded_builtin differently, the C FE
calls it when e.g. array-to-pointer and function-to-pointer conversions
are already done on the arguments, while C++ FE does that only much later.
The c-common code e.g. for __sync/__atomic builtins deals with
that e.g. b
Hi!
The powerpc patch I've just posted led me to try __atomic_* builtins
on arrays as in the testcase below. While it works fine if the
array is just on the first argument or in C, in C++ for arrays in 2nd
or 3rd argument the atomics are rejected (complaining that the argument
is not a pointer),
This enhances VN to do the same PHI handling as CCP, meeting
undefined and constant to constant. I've gone a little bit
further (and maybe will revisit this again) in also meeting
all-undefined to undefined taking one of the undefined args
as the value number. I feel like this might run into
the
On Thu, 7 Sep 2017, Richard Biener wrote:
>
> This enhances VN to do the same PHI handling as CCP, meeting
> undefined and constant to constant. I've gone a little bit
> further (and maybe will revisit this again) in also meeting
> all-undefined to undefined taking one of the undefined args
> as
This patch improve the error reporting on a null procedure with misplaced
aspect specification, which the parser first attempts to interpret as a
malformed expression function, in spite of the initial keyword in the
declaration.
Compiling
---
package Null_Proc_With_Contract is
procedu
This patch fixes a bug where if a limited object is initialized with a
build-in-place function call, and the call does not return on the
secondary stack, and the function raises an exception, so that the
object is not (successfully) created, the uninitialized object is
incorrectly finalized.
The f
This fixes a small regression introduced by the recent improvement to the
algorithm used by Hide_Public_Entities to compute the final set of external
visible entities of a package. It now needs to recurse on the Actions list
of freeze nodes to find references to subprograms.
The following procedu
This patch fixes a performance regression. The compiler was using the
secondary stack for a nondispatching build-in-place call, which is
unnecessary, and indeed older compilers did not do so. The compiler no
longer uses the secondary stack for such calls. Note that the secondary
stack is necessary
This patch fixes spurious visibility errors on the expression for a dynamic
predicate in a subtype declaration, when the enclosing package includes private
declarations.
The following packages much compile quietly:
---
package foo is
type Kind_Type is (None, Known);
type Auction_State_Ty
Ghost types are allowed to mention ghost entities in their predicate.
Because Ghost is propagated from type to the generated predicate function,
GNAT correctly identified valid from invalid cases, but this modification
simplifies the reason for correction.
Tested on x86_64-pc-linux-gnu, committed
This fixes a small regression introduced by the recent improvement to the
algorithm used by Hide_Public_Entities to compute the final set of external
visible entities of a package. Subprogram renamings not only need to be
dealt with as declaring a subprogram but also as referencing another one.
T
On Wed, Sep 06, 2017 at 11:48:00AM -0500, Segher Boessenkool wrote:
> On Wed, Sep 06, 2017 at 06:26:10PM +0200, Jakub Jelinek wrote:
> > > Maybe this "switch to the other section" thing should be abstracted out?
> > > Messing with in_cold_section_p is a bit dirty.
> >
> > But it reflects the reali
This patch fixes a spurious error on a classwide precondition for a subprogram
S that is a primitive of some type T, when the precondition includes a
dispatching call on a classwide formal of S whose type is urelated to T.
The following must compile quietly:
gnatmake -q main
---
with Derived_
This patches fixes a bug in which a function returns an anonymous access
type whose designated type is a class-wide interface type, and the
return statement returns a raise expression, the compiler crashes.
Tested on x86_64-pc-linux-gnu, committed on trunk
2017-09-07 Bob Duff
* sem_ch
This patch adds switch -gnatd.y to disable the generation of
implicit Elaborate_All on a package X when a task body calls
a procedure in the same package, and that procedure calls a
procedure in another package X.
As documented in the GNAT User Guide, when sources cannot be
modified, the recommend
This patch implements the proper handling of dimension information on type
conversions. Given a conversion T (Expr), where the expression has type TE,
the following cases arise:
a) If TE has dimension information, the dimensions of the conversion are those
of TE.
b) If TE has no dimension informa
Thanks for the quick feedback.
Richard Biener writes:
> On Wed, Sep 6, 2017 at 10:18 PM, Richard Sandiford
> wrote:
>> The next main step in the SVE submission is to add support for
>> offsets and sizes that are a runtime invariant rather than a compile
>> time constant. This is an RFC about ou
On 08/27/2017 09:36 PM, Jon Beniston wrote:
> I have an out-of-tree GCC port and it is struggling supporting
> auto-vectorization on some dot product instructions. For example, I have an
> instruction that takes three operands which are all 32-bit general
> registers. The second and third operands
On Thu, 7 Sep 2017, Jakub Jelinek wrote:
> On Wed, Sep 06, 2017 at 11:48:00AM -0500, Segher Boessenkool wrote:
> > On Wed, Sep 06, 2017 at 06:26:10PM +0200, Jakub Jelinek wrote:
> > > > Maybe this "switch to the other section" thing should be abstracted out?
> > > > Messing with in_cold_section_p
Hi Bernd,
>This seems to ring a bell. I think I submitted a patch for this here
>- is this the same problem?
>https://gcc.gnu.org/ml/gcc-patches/2010-12/msg01724.html
Looks like it.
Cheers,
Jon
This is a ping for a small MIPS specific patch adds `short_call`
synonym for existing `near` attribute. The patch has not been approved
or commented on.
Add 'short_call' attribute for MIPS targets.
https://gcc.gnu.org/ml/gcc-patches/2017-08/msg01280.html
--
Simon Atanasyan
On Thu, Sep 07, 2017 at 11:46:09AM +0200, Jakub Jelinek wrote:
> On Wed, Sep 06, 2017 at 11:48:00AM -0500, Segher Boessenkool wrote:
> > On Wed, Sep 06, 2017 at 06:26:10PM +0200, Jakub Jelinek wrote:
> > > > Maybe this "switch to the other section" thing should be abstracted out?
> > > > Messing wi
As discussed in PR 79433, the recommended way to test for new features
such as std::optional has problems. The current version of SD-6 at
https://isocpp.org/std/standing-documents/sd-6-sg10-feature-test-recommendations
says to simply check __has_include(optional). This test will be true
even when
On September 7, 2017 1:20:13 PM GMT+02:00, Jon Beniston
wrote:
>Hi Bernd,
>
>>This seems to ring a bell. I think I submitted a patch for this here
>>- is this the same problem?
>>https://gcc.gnu.org/ml/gcc-patches/2010-12/msg01724.html
>
>Looks like it.
Looks my suggestion was exactly what we h
On 7 September 2017 at 17:18, Jonathan Wakely wrote:
> Does anybody object to this change? Is getting a #error still useful,
> given the existence of __has_include and __cpp_lib macros?
I wholeheartedly agree with this change. And I should point out that
vast amounts of optional's implementation
On Mon, Aug 28, 2017 at 10:13 AM, H.J. Lu wrote:
> On Mon, Aug 28, 2017 at 9:10 AM, Joseph Myers wrote:
>> On Tue, 8 Aug 2017, H.J. Lu wrote:
>>
>>> This patch adds -static-pie to GCC driver to create static PIE. A static
>>> position independent executable (PIE) is similar to static executable,
A small oversight in the implementation of component reordering.
Tested on x86_64-suse-linux, applied on mainline.
2017-09-07 Eric Botcazou
PR ada/82127
* gcc-interface/decl.c (copy_and_substitute_in_layout): Put the fields
in order of increasing position in more case
On Thu, 7 Sep 2017, Jakub Jelinek wrote:
> Hi!
>
> The powerpc patch I've just posted led me to try __atomic_* builtins
> on arrays as in the testcase below. While it works fine if the
> array is just on the first argument or in C, in C++ for arrays in 2nd
> or 3rd argument the atomics are rejec
This is the Ada bootstrap failure on SPARC64/Linux that I worked around at
some point; it turns out that we can do something plausible in the back-end.
Tested on SPARC64/Linux, applied on mainline.
2017-09-07 Eric Botcazou
PR target/80897
* config/sparc/sparc.c (sparc_emit_s
Jakub Jelinek writes:
> On Wed, Sep 06, 2017 at 10:08:01PM +0200, David Edelsohn wrote:
>> This change broke bootstrap on AIX because sancov.c now references a
>> macro that is defined as a function on AIX. sancov.c needs to include
>> tm_p.h to pull in the target-dependent prototypes. The foll
On 09/05/2017 03:48 PM, Segher Boessenkool wrote:
> Hi!
>
> On Sat, Sep 02, 2017 at 12:31:16AM -0600, Jeff Law wrote:
>> On 08/29/2017 05:14 PM, Segher Boessenkool wrote:
>>> Actually, everywhere it is used it has a Pmode == SImode wart before
>>> it, to emit the right update_stack insn... So fol
On Thu, Sep 7, 2017 at 6:57 PM, Rainer Orth
wrote:
> Jakub Jelinek writes:
>
>> On Wed, Sep 06, 2017 at 10:08:01PM +0200, David Edelsohn wrote:
>>> This change broke bootstrap on AIX because sancov.c now references a
>>> macro that is defined as a function on AIX. sancov.c needs to include
>>>
Hi!
I've bootstrapped/regtested and committed to gcc-7-branch backports
of 10 trunk commits.
Jakub
2017-09-07 Jakub Jelinek
Backported from mainline
2017-07-27 Jakub Jelinek
PR c/45784
* c-omp.c (c_finish_omp_for): If the condition is wrapped in
Hi
Following this report I had a closer look to this debug-dependent
behavior which I agree about.
I have added some checks to 2.cc as shown in the patch. Initially the
test was failing because:
/home/fdt/dev/gcc/git/libstdc++-v3/testsuite/24_iterators/istreambuf_iterator/2.cc:45:
void test0
On Thu, Sep 07, 2017 at 03:35:00PM +, Joseph Myers wrote:
> > The powerpc patch I've just posted led me to try __atomic_* builtins
> > on arrays as in the testcase below. While it works fine if the
> > array is just on the first argument or in C, in C++ for arrays in 2nd
> > or 3rd argument th
On Thu, 7 Sep 2017, Jakub Jelinek wrote:
> Ah, ok, so like this instead?
>
> 2017-09-07 Jakub Jelinek
>
> PR target/82112
> * c-common.c (sync_resolve_size): Instead of c_dialect_cxx ()
> assertion check that in the condition.
> (get_atomic_generic_size): Likewise. Be
On Fri, 25 Aug 2017, Alexandre Oliva wrote:
> for gcc/ChangeLog
>
> * common.opt (Wa, Wl, Wp, g, gz=): Add
> RejectNegative.
> (gno-column-info): Remove.
> (gcolumn-info): Drop RejectNegative.
> (gno-): New prefix.
> (gno-record-gcc-switches): Remove.
>
-gcc-patches
On Thu, 7 Sep 2017 23:02:15 +0200
François Dumont wrote:
> + _M_c = _M_sbuf->sgetc();
> + if (_S_at_eof(_M_c))
> + _M_sbuf = 0;
_M_sbuf = 0; <--- Is not what I axpect here.
Suggestions will be later, after we finish copyright assignment for
changes procedure (in pro
47 matches
Mail list logo
