On 09/05/2018 09:28 PM, Indu Bhagat wrote:
Hi.
Thanks for working on that. I believe it's useful enhancement. Note that I'm
not profile feedback maintainer,
but I'll provide some feedback:
> Patch for PR 86957 " gcc should warn about missing profiles for a compilation
> unit or a new function
On 09/05/2018 09:28 PM, Indu Bhagat wrote:
> Patch for PR 86957 " gcc should warn about missing profiles for a compilation
> unit or a new function with -fprofile-use".
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86957
>
> The patch adds -Wmissing-profile warning flag to alert user about cases
We can calculate wide_int_range_shift_undefined_p() without having to
pass down the sign bit of the operand. Also, vrp_shift_undefined_p is a
brain dead wrapper so I'm removing it.
OK for trunk?
commit 9aeb62d4c33b50bc007b07ec5097e8f3edd4b31b
Author: Aldy Hernandez
Date: Mon Sep 10 17:46:10
Doing this was approved by Richi at some point when I rewrote the
division code. I finally got around to testing it.
Committed.
gcc/
* tree-vrp (extract_range_from_binary_expr_1): Treat all divisions
by zero as VR_UNDEFINED.
diff --git a/gcc/tree-vrp.c b/gcc/tree-vrp.c
index 8f16713300c..a8
On 05/09/18 14:41, David Malcolm wrote:
Please add the:
auto_diagnostic_group d;
line to the top of the function.
OK with that change.
Here's what I committed.
Andrew
Add sorry_at diagnostic function.
The plain "sorry" diagnostic only gives the "current" location, which is
typically th
For addition and subtraction, we're currently punting on things like
[0,0] - VR_VARYING. However, if we normalize VR_VARYING to the entire
domain, we can see that the above is actually [-MIN+1, +MAX].
We would've normally caught this sort of things with my rewrite of the
binary operators, but
On 2018/9/11 1:22 AM, Cesar Philippidis wrote:
On 09/10/2018 08:04 AM, Chung-Lin Tang wrote:
GOACC_2.0 {
Index: libgomp/oacc-mem.c
===
--- libgomp/oacc-mem.c (revision 264192)
+++ libgomp/oacc-mem.c (working copy)
@@ -153,8 +15
LWG DR 2905 says that is_constructible_v, P, D const &>
should be false when D is not copy constructible. This commit implements
the changes from the DR and simplifies the signatures as per
https://github.com/cplusplus/draft/issues/1530
* include/bits/unique_ptr.h (__uniq_ptr_impl): Add a
ping!
Am Mo., 3. Sep. 2018 um 21:19 Uhr schrieb Janus Weil :
>
> Hi all,
>
> attached is a simple patch that fixes a regression which was
> introduced by one of my recent commits (spotted by Dominique).
>
> My first impulse to fix the spurious error was to check for the
> 'intrinsic' attribute, how
PR libstdc++/87278
* include/bits/shared_ptr.h (make_shared): Use remove_cv instead of
remove_const.
* testsuite/20_util/shared_ptr/creation/87278.cc: New test.
Tested x86_64-linux, committed to trunk and gcc-8-branch.
commit 29eed23367a6a01a5fb60d7e88a49f3470cb39
Hi.
This patch makes the scheduler prefer instructions with higher cost if two
given instructions are equally good.
Issuing more restricted instructions first is particularly useful on in-order
cores because it increases the
number of dual issue opportunities.
For example, on AArch64, instead
Hi everyone,
this patch optimizes a few boolean expressions,
and fixes some unneeded whitespace.
Will do a full make check soon.
Thanks for your time.
2018-09-11 MCC CS
gcc/
PR tree-optimization/87261
* match.pd: Add boolean optimizations,
fix whitespace.
2018
On 08/29/2018 06:08 PM, Alexander Monakov wrote:
> On Wed, 29 Aug 2018, Martin Liška wrote:
>>> Can you shortly tell why the testcase in the PR segfaults? Does the issue
>>> only affect indirect call profiling?
>>
>> What happens is that there will exist 2 instances of:
>> void * __gcov_indirect_c
On 10.09.2018 19:53, Jakub Jelinek wrote:
> On Mon, Sep 10, 2018 at 04:05:26PM +0200, Andreas Krebbel wrote:
>> On 20.08.2018 16:30, Jeff Law wrote:
>>> On 08/18/2018 03:20 AM, Eric Botcazou wrote:
> Eric, didn't your patches explicitely handle this case of a non-constant
> inbetween?
On 05/09/18 13:43, Richard Biener wrote:
No. You might want to look into the x86 backend if there's maybe more tweaks
needed when using non-vector mask modes.
I tracked it down to the vector alignment configuration.
Apparently the vectorizer likes to build a "truth" vector, but is
perfectly
Tamar Christina writes:
> Hi Richard,
>
> The 08/28/2018 21:58, Richard Sandiford wrote:
>> Tamar Christina writes:
>> > + HOST_WIDE_INT guard_used_by_caller = STACK_CLASH_CALLER_GUARD;
>> > + /* When doing the final adjustment for the outgoing argument size
>> > we can't
>> > + assume that LR w
On Tue, 11 Sep 2018, 14:38 Vlad Lazar, wrote:
> Hi.
>
> This patch makes the scheduler prefer instructions with higher cost if two
> given instructions are equally good.
> Issuing more restricted instructions first is particularly useful on
> in-order cores because it increases the
> number of du
Hi Ramana,
On 11/09/18 15:55, Ramana Radhakrishnan wrote:
On Tue, 11 Sep 2018, 14:38 Vlad Lazar, wrote:
> Hi.
>
> This patch makes the scheduler prefer instructions with higher cost if two
> given instructions are equally good.
> Issuing more restricted instructions first is particularly usefu
The driver gcc.c has 2 internal error reporting routines
perror_with_name & pfatal_with_name, which take a single string argument.
Firstly the latter forwards to the former, then calls exit after doing
some cleanup. However, that cleanup is already registered via atexit,
so it'll get done any
>
> > This to me feels like the wrong approach as it feels like you are assuming
> > INSN_COST is latency in some way ? Surely, we shouldn't be introducing
> > INSN_COST based stuff into the scheduler.
> >
> > Have you investigated using TARGET_SCHED_ADJUST_COST (IIRC, look for the
> > right name
On Fri, 2018-09-07 at 16:54 -0500, Segher Boessenkool wrote:
> Hi!
>
> On Fri, Sep 07, 2018 at 04:27:34PM -0500, Will Schmidt wrote:
> > The expected instructions for this test (p8-vec-xl-*) were incorrect for
> > some
> > of the power targets. Add codegen variations as appropriate for
> > the
On Tue, 11 Sep 2018, Martin Liška wrote:
> I've discussed the topic with Alexander on the Cauldron and we hoped
> that the issue with unique __gcov_root can be handled with DECL_COMMON in
> each DSO.
> Apparently this approach does not work as all DSOs in an executable have
> eventually
> just a
Hi,
This is a simple patch to add a new first-class option
-finline-visibility={all|extern|static}
to finer control inlining based on function’s visibility.
'-finline-visibility=[all|extern|static]'
By default, GCC inlines functions without considering their
visibility. This flag
On Tue, Sep 04, 2018 at 10:13:43AM -0500, Sam Tebbs wrote:
> Hi James,
>
> Thanks for the feedback. Here is an update with the changes you proposed
> and an updated changelog.
>
> gcc/
> 2018-09-04 Sam Tebbs
>
> PR target/85628
> * config/aarch64/aarch64.md (*aarch64_bfxil)
Tamar Christina writes:
>> > +
>> > + /* No probe leave. */
>> > + ASM_OUTPUT_INTERNAL_LABEL (asm_out_file, loop_end_lab);
>> > + return "";
>>
>> With the CFA stuff and constant load, I think this works out as:
>>
>> -
>> # 12 insns
>> mov
* Vineet Gupta [2018-09-10 11:26:35 -0700]:
>* config/arc/arc.c: object attributes for core4 not reflected correctly
>* config/arc/arc.h: Don't restrict DBNZ to core3 (core4 includes core3)
>
> Signed-off-by: Vineet Gupta
> ---
> gcc/ChangeLog| 7 +++
> gcc/config/arc/arc.c
On 09/01/2018 08:10 PM, Tom de Vries wrote:
>> Please add more of this description to the one-line documentation
>> patch you have now;
> Done.
>
>> there are many DIEs that have no name because they
>> don't need one, and this patch doesn't add names to all of them.
>>
> I've added the qualificat
Hi Richard,
> -Original Message-
> From: Richard Sandiford
> Sent: Tuesday, September 11, 2018 15:49
> To: Tamar Christina
> Cc: Jeff Law ; gcc-patches@gcc.gnu.org; nd
> ; James Greenhalgh ;
> Richard Earnshaw ; Marcus Shawcroft
>
> Subject: Re: [PATCH][GCC][AArch64] Updated stack-clash
On Fri, Sep 07, 2018 at 11:03:28AM -0500, Tamar Christina wrote:
> Hi Richard,
>
> The 08/28/2018 21:58, Richard Sandiford wrote:
> > Tamar Christina writes:
> > > + HOST_WIDE_INT guard_used_by_caller = STACK_CLASH_CALLER_GUARD;
> > > + /* When doing the final adjustment for the outgoing argume
I'd expect all the cases where you pass a string containing '%s' to the
generic diagnostic functions to use %qs instead of '%s' for quoting, to
get appropriate locale-specific quotes.
--
Joseph S. Myers
jos...@codesourcery.com
On Tue, Sep 4, 2018 at 9:01 AM, H.J. Lu wrote:
> On Tue, Aug 28, 2018 at 11:04 AM, H.J. Lu wrote:
>> With -mavx, for
>>
>> [hjl@gnu-cfl-1 skx-2]$ cat foo.i
>> extern float f;
>> extern double d;
>> extern int i;
>>
>> void
>> foo (void)
>> {
>> d = f;
>> f = i;
>> }
>>
>> we need to generate
On Mon, Sep 10, 2018 at 08:31:49PM -0400, Julian Brown wrote:
> #pragma omp target enter data map(to:arr[0:32])
>
> //#pragma omp target // works, mapped as zero-length array section (*)
> //#pragma omp target map(tofrom:myptr) // crashes (**)
In this case OpenMP doesn't allow the implementation
On Sat, Sep 1, 2018 at 7:10 PM, Tom de Vries wrote:
> On 08/31/2018 11:19 PM, Jason Merrill wrote:
>> 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 Au
On 05/09/18 15:44, Joseph Myers wrote:
On Wed, 5 Sep 2018, Andrew Stubbs wrote:
I suppose someone might choose to compile things on an alternative system for
running on a compatible system, in which case we'd want to simply skip this
binary.
How does one normally do this?
I'd expect a config
On Tue, 11 Sep 2018, Andrew Stubbs wrote:
> On 05/09/18 15:44, Joseph Myers wrote:
> > On Wed, 5 Sep 2018, Andrew Stubbs wrote:
> >
> > > I suppose someone might choose to compile things on an alternative system
> > > for
> > > running on a compatible system, in which case we'd want to simply ski
Hello!
Attached patch removes unnecessary mixed-mode log patterns. These are
not needed, because all x87 float_extend RTXes degenerate to a plain
move (or a no-op move). The patch also includes a couple of cleanups
with no functional changes.
2018-09-11 Uros Bizjak
* config/i386/i386.md (
On Mon, Aug 06, 2018 at 11:14:17AM -0500, Vlad Lazar wrote:
> Hi,
>
> The patch adds support for the TARGET_COMPUTE_FRAME_LAYOUT hook on AArch64
> and removes unneeded frame layout recalculation.
>
> The removed aarch64_layout_frame calls are unnecessary because the functions
> in which
> they a
2018-09-11 Uros Bizjak
* config/i386/i386.md (define_attr "type"): Remove mpxmov, mpxmk,
mpxchk, mpxld and mpxst types.
(define_attr length_immediate): Remove all processing of mpx types.
(define_attr prefix_0f): Ditto.
(define_attr memory): Ditto.
Patch was bootstrapped an
On Wed, Sep 5, 2018 at 4:38 AM, Kyrill Tkachov
wrote:
> On 04/09/18 17:52, Kyrill Tkachov wrote:
>>
>> On 04/09/18 15:31, Richard Biener wrote:
>>> On Tue, 4 Sep 2018, Kyrill Tkachov wrote:
Hi Richard,
On 31/08/18 12:07, Richard Biener wrote:
> On Thu, 30 Aug 2018, Kyrill Tkach
Fixed as 'obvious' on trunk and 8-branch in revisions 264210 and
264213 respectively.
Paul
2018-09-11 Paul Thomas
PR fortran/87277
* expr.c (is_subref_array): Add the check of dimensionality for
class, dummy, pointer arrays.
2018-09-11 Paul Thomas
PR fortran/87277
*
On Tue, 11 Sep 2018, Qing Zhao wrote:
> '-finline-visibility=[all|extern|static]'
> By default, GCC inlines functions without considering their
> visibility. This flag allows finer control of inlining based on
> their visibility.
>
> The value 'extern' tells the compiler to on
Hi again,
On Tue, Sep 11, 2018 at 10:04:45AM -0500, Will Schmidt wrote:
> On Fri, 2018-09-07 at 16:54 -0500, Segher Boessenkool wrote:
> > On Fri, Sep 07, 2018 at 04:27:34PM -0500, Will Schmidt wrote:
> > > -/* { dg-final { scan-assembler-times "lvx" 4 } } */
> > > -/* { dg-final { scan-assembler
Applied, after visual inspection before/after.
Gerald
Index: c99status.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/c99status.html,v
retrieving revision 1.66
diff -u -r1.66 c99status.html
--- c99status.html 1 Sep 2018 23:41:59 -00
Applied.
Gerald
Index: gcc-3.3/gcj-status.html
===
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-3.3/gcj-status.html,v
retrieving revision 1.11
diff -u -r1.11 gcj-status.html
--- gcc-3.3/gcj-status.html 2 Sep 2018 19:10:52 - 1.
Hello!
The mode of clobbers was changed from QImode to reg_raw_mode from a
long time ago...
2018-09-11 Uros Bizjak
* reg-stack.c (subst_asm_stack_regs): Call replace_reg also
for clobbers. Remove obsolete comment.
Bootstrapped and regression tested on x86_64-linux-gnu {,-m32}.
Comm
On Sun, 2 Sep 2018, Gerald Pfeifer wrote:
> Committed.
Actually there still was one warning, though technically not a
validation error.
Fixed thusly, using CSS and using instead of on the
way, so improving this even beyond that.
Gerald
Index: projects/tree-ssa/vectorization.html
On 9/11/18 9:00 AM, Kyrill Tkachov wrote:
> Hi Ramana,
>
> On 11/09/18 15:55, Ramana Radhakrishnan wrote:
>> On Tue, 11 Sep 2018, 14:38 Vlad Lazar, wrote:
>>
>> > Hi.
>> >
>> > This patch makes the scheduler prefer instructions with higher cost
>> if two
>> > given instructions are equally good.
On 9/11/18 7:38 AM, Vlad Lazar wrote:
> Hi.
>
> This patch makes the scheduler prefer instructions with higher cost if
> two given instructions are equally good.
> Issuing more restricted instructions first is particularly useful on
> in-order cores because it increases the
> number of dual issue
Hi, Alexander,
thanks for your comments.
> On Sep 11, 2018, at 12:48 PM, Alexander Monakov wrote:
>
> On Tue, 11 Sep 2018, Qing Zhao wrote:
>> '-finline-visibility=[all|extern|static]'
>> By default, GCC inlines functions without considering their
>> visibility. This flag allows finer
(just some very quick comments, they may be off, and I haven't looked
closely)
On Tue, 11 Sep 2018, MCC CS wrote:
+/* (~x & y) | ~(x | y) -> ~x */
+(simplify
+ (bit_ior:c (bit_and:c (bit_not @0) @1) (bit_not (bit_ior:s @0 @1)))
+ (bit_not @0))
Did you mean :c instead of :s?
You should write
On 9/11/18 3:58 PM, Joseph Myers wrote:
I'd expect all the cases where you pass a string containing '%s' to the
generic diagnostic functions to use %qs instead of '%s' for quoting, to
get appropriate locale-specific quotes.
yes of course, slipped my mind.
Committed the attached after manual te
On Tue, 2018-09-11 at 13:20 -0500, Segher Boessenkool wrote:
> Hi again,
>
> On Tue, Sep 11, 2018 at 10:04:45AM -0500, Will Schmidt wrote:
> > On Fri, 2018-09-07 at 16:54 -0500, Segher Boessenkool wrote:
> > > On Fri, Sep 07, 2018 at 04:27:34PM -0500, Will Schmidt wrote:
> > > > -/* { dg-final {
On 9/5/18 5:48 AM, a...@codesourcery.com wrote:
>
> GCN has some registers which are special purpose, but not "fixed" because we
> want the register allocator to track their usage and select alternatives that
> use different special registers (e.g. scalar cc vs. vector cc).
>
> Sometimes this lea
On 9/5/18 5:49 AM, a...@codesourcery.com wrote:
>
> The vec_select operator is documented to require a const_int for the lane
> selector operand, but GCN has an instruction that can select the lane at
> runtime, so it seems reasonable to remove this restriction.
>
> This patch simply replaces ass
On 9/5/18 5:49 AM, a...@codesourcery.com wrote:
>
> GCN's 64-lane vectors tend to make RTL dumps very long. This patch makes them
> far more bearable by eliding long sequences of the same element into
> "repeated"
> messages.
>
> 2018-09-05 Andrew Stubbs
> Jan Hubicka
>
On 9/5/18 5:51 AM, a...@codesourcery.com wrote:
>
> This patch contains the GCN port of libgfortran. We use the minimal
> configuration created for NVPTX. That's all that's required, besides the
> target-independent bug fixes posted already.
>
> 2018-09-05 Andrew Stubbs
> Kwok Cheu
On 9/7/18 3:12 AM, Aldy Hernandez wrote:
> Semi obvious patch here...
>
> As the comment for alloca_type_and_limit says:
>
> // For ALLOCA_BOUND_MAYBE_LARGE and ALLOCA_BOUND_DEFINITELY_LARGE
> // types, this field indicates the assumed limit if known or
> // integer_zero_node if unknown. F
On 9/5/18 5:48 AM, a...@codesourcery.com wrote:
>
> The HSA GPU drivers can't cope with binaries that have the same symbol defined
> multiple times, even though the names are not exported. This happens whenever
> there are file-scope static variables with matching names. I believe it's
> also
>
On 9/11/18 4:09 AM, Aldy Hernandez wrote:
> We can calculate wide_int_range_shift_undefined_p() without having to
> pass down the sign bit of the operand. Also, vrp_shift_undefined_p is a
> brain dead wrapper so I'm removing it.
>
> OK for trunk?
>
> curr.patch
>
> commit 9aeb62d4c33b50bc007b07
On Mon, 10 Sep 2018 20:31:49 -0400
Julian Brown wrote:
> [...] I think the handling of references can and should match between
> the two APIs (though implementation details of the patch to make that
> so need a little work still).
Here's a new version of the patch, somewhat simplified and slight
60 matches
Mail list logo
