> Hi
> this patch implements the trik for punting if we get too many nested
> pointers.
> This fixes the ada tstcases. Curiously enough I would like to replace
> safe_push
> by quick_push but doing so I get weird error about freeing non-heap object
> in the auto_vec desructor...
>
> Bootstraping/r
This fixes the observed memory leak in the correct way, asserting that
overwriting of a vinfo doesn't happen.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2015-12-09 Richard Biener
* tree-vect-stmts.c (vectorizable_load): Set new vinfo only
if it wa
On 12/08/2015 07:27 AM, Richard Biener wrote:
I wonder if it makes more sense to integrate this with the
domwalker itself. Add a constructor flag to it and do everything
in itself. By letting the before_dom_children return a taken edge
(or NULL if unknown) it can drive the outgoing edge markin
Well, not really in the end - the actual testcase would need code
hoisting to make the refs equal enough for ifcvt. Making the testcase
simpler doesn't seem to capture the issue so for now without one
(I'll keep trying).
Bootstrapped and tested on x86_64-unknown-linux-gnu.
Richard.
2015-12-09
On Wed, 9 Dec 2015, Jan Hubicka wrote:
> Hi,
> this patch fixes base+offset alias oracles to consider variable aliases.
> With this patch and one extra hack I can LTO bootstrap with variale symbol
> merging disabled in lto-symtab.c
>
> The basic idea is simple - there is new comparer compare_base
On Wed, 9 Dec 2015, Jan Hubicka wrote:
> Hi, while debugging an renaming issue I run into ADDR_EXPR of MEM_REF of
> ADDR_EXPR which seemed somewhat odd + we don't really get the CONSTANT
> and other flags right because recompute_tree_invariant_for_addr_expr
> punts on ADDR_EXPR inside ADDR_EXPR
On Wed, 9 Dec 2015, Jan Hubicka wrote:
> > Hi,
> > while debugging an renaming issue I run into ADDR_EXPR of MEM_REF of
> > ADDR_EXPR
> > which seemed somewhat odd + we don't really get the CONSTANT and other flags
> > right because recompute_tree_invariant_for_addr_expr punts on ADDR_EXPR
> > i
On Wed, 9 Dec 2015, Jan Hubicka wrote:
> Hi
> this patch implements the trik for punting if we get too many nested pointers.
> This fixes the ada tstcases. Curiously enough I would like to replace
> safe_push
> by quick_push but doing so I get weird error about freeing non-heap object
> in the au
On Wed, 9 Dec 2015, Arnaud Charlet wrote:
> > Hi
> > this patch implements the trik for punting if we get too many nested
> > pointers.
> > This fixes the ada tstcases. Curiously enough I would like to replace
> > safe_push
> > by quick_push but doing so I get weird error about freeing non-heap ob
On 8 December 2015 at 19:48, Eric Botcazou wrote:
>> I think that is ok if the testing passes.
>
> Thanks. It did on the 3 architectures so I have applied the patch.
>
Thanks, I confirm it also fixes the regressions I noticed on ARM.
Christophe.
> --
> Eric Botcazou
On 12/09/2015 12:47 AM, Gerald Pfeifer wrote:
> On Tue, 8 Dec 2015, Martin Liška wrote:
>> I would like to add a missing configure option.
>
> I saw that Jeff approved while I was mulling over the patch,
> but still have two questons:
>
> 1. Why did you sort this in where you did? (It's not alph
c-c++-common/attr-simd-3.c fails to compile on arm-none-eabi targets due to
-fcilkplus needing -pthread which is not available for those targets. This
patch solves this issue by adding a condition to the cilkplus effective target
that compiling with -fcilkplus succeeds and requires cilkplus as a
gcc.target/arm/lto/pr65837 fails on M profile ARM targets because of lack of
neon instructions. This patch adds the necessary arm_neon_ok effective target
requirement to avoid running this test for such targets.
ChangeLog entry is as follows:
* gcc/testsuite/ChangeLog ***
2015-12-08 Thomas P
[ was: Re: [PATCH, PR46032] Handle BUILT_IN_GOMP_PARALLEL in ipa-pta ]
On 30/11/15 13:11, Tom de Vries wrote:
On 30/11/15 10:16, Richard Biener wrote:
On Mon, 30 Nov 2015, Tom de Vries wrote:
Hi,
this patch fixes PR46032.
It handles a call:
...
__builtin_GOMP_parallel (fn, data, num_thre
On 2015.12.09 at 10:33 +0100, Martin Liška wrote:
> On 12/09/2015 12:47 AM, Gerald Pfeifer wrote:
> > On Tue, 8 Dec 2015, Martin Liška wrote:
> >> I would like to add a missing configure option.
> >
> > +Specify that the compiler should interact with valgrind runtime, where
> > +selected invalid m
On Wed, Dec 09, 2015 at 11:01:31AM +0100, Tom de Vries wrote:
> PR tree-optimization/68716
> * tree-ssa-structalias.c (find_func_clobbers): Fix handling of
> BUILT_IN_GOMP_PARALLEL and BUILT_IN_GOMP_PARALLEL.
Pasto in ChangeLog entry?
Jakub
On Wed, 9 Dec 2015, Jakub Jelinek wrote:
> On Wed, Dec 09, 2015 at 11:01:31AM +0100, Tom de Vries wrote:
> > PR tree-optimization/68716
> > * tree-ssa-structalias.c (find_func_clobbers): Fix handling of
> > BUILT_IN_GOMP_PARALLEL and BUILT_IN_GOMP_PARALLEL.
>
> Pasto in ChangeLog entr
On 09/12/15 11:03, Jakub Jelinek wrote:
On Wed, Dec 09, 2015 at 11:01:31AM +0100, Tom de Vries wrote:
PR tree-optimization/68716
* tree-ssa-structalias.c (find_func_clobbers): Fix handling of
BUILT_IN_GOMP_PARALLEL and BUILT_IN_GOMP_PARALLEL.
Pasto in ChangeLog entry?
On Tue, Dec 8, 2015 at 7:56 PM, Steve Ellcey wrote:
> I have an ivopts optimization question/proposal. When compiling the
> attached program the ivopts pass prefers the original ivs over new ivs
> and that causes us to generate less efficient code on MIPS. It may
> affect other platforms too.
>
On Tue, 8 Dec 2015, Jakub Jelinek wrote:
> Hi!
>
> As written in the PR, with MASK_{LOAD,STORE} ifns we lose info on the
> alignment of the point, unless the referenced SSA_NAME has pointer info
> with alignment mask, but that is just an optimization issue rather than
> requirement.
>
> As the s
On Tue, Dec 8, 2015 at 11:24 PM, Ajit Kumar Agarwal
wrote:
> Based on the comments on RFC patch this patch incorporates all the comments
> from Jeff. Thanks Jeff for the valuable feedback.
>
> This patch enables the better register pressure estimate for Loop Invariant
> code motion. This patch C
On Tue, Dec 8, 2015 at 5:21 PM, Marek Polacek wrote:
> The following is a conservative fix for this PR. This is an ICE transpiring
> in the new "Factor conversion in COND_EXPR" optimization added in r225722.
>
> Before this optimization kicks in, we have
> :
> ...
> p1_32 = (short unsigned
On Wed, Dec 9, 2015 at 9:31 AM, Jeff Law wrote:
> On 12/08/2015 07:27 AM, Richard Biener wrote:
>>>
>>>
>>> I wonder if it makes more sense to integrate this with the
>>> domwalker itself. Add a constructor flag to it and do everything
>>> in itself. By letting the before_dom_children return a t
On 12/09/2015 03:18 AM, Bernd Edlinger wrote:
Furthermore there is a documented use for asm(""): The empty assembler string
is used to make a function
volatile, thus calls can not be optimized away. But I think it is not
necessary to make this clobber anything,
nor should it be an instruction
This is about the 3 -Wuninitialized warnings for the build of the Ada library:
g-expect.adb: In function 'GNAT.EXPECT.SET_UP_CHILD_COMMUNICATIONS':
g-expect.adb:1356:7: warning: 'INPUT' is used uninitialized in this function
[-Wuninitialized]
g-expect.adb:1357:7: warning: 'OUTPUT' is used uniniti
Hi.
This is simple follow-up of the previous patch that fixes last remaining
leak in vectorizer.
Patch can regbootstrap on x64_64-linux-gnu.
Ready for trunk?
Martin
>From 0d61420eb49dec0f5d14108373a546a8f1b52571 Mon Sep 17 00:00:00 2001
From: marxin
Date: Wed, 9 Dec 2015 10:14:00 +0100
Subject:
-Original Message-
From: Richard Biener [mailto:richard.guent...@gmail.com]
Sent: Wednesday, December 09, 2015 4:06 PM
To: Ajit Kumar Agarwal
Cc: Jeff Law; GCC Patches; Vinod Kathail; Shail Aditya Gupta; Vidhumouli
Hunsigida; Nagaraju Mekala
Subject: Re: [Patch,rtl Optimization]: Better
On Fri, Nov 27, 2015 at 01:01:01PM +, James Greenhalgh wrote:
> This patch follow Richard Henderson's advice to tighten up
> CANNOT_CHANGE_MODE_CLASS for AArch64 to avoid a simplification bug in
> the middle-end.
>
> There is nothing AArch64-specific about the testcase which triggers this,
> s
Hi.
I would like to backport forgotten patch to GCC-5 branch.
Bootstrap and regression tests have been running, ready after it finishes?
Thanks,
Martin
>From 626df2a92db87f7c0668c7f2902f314ed6a9bc4c Mon Sep 17 00:00:00 2001
From: marxin
Date: Wed, 9 Dec 2015 12:49:22 +0100
Subject: [PATCH] Backp
On Mon, Dec 07, 2015 at 12:19:57PM +0100, Martin Jambor wrote:
> +/* Flag set when the subsequent element in the device-specific argument
> + values. */
> +#define GOMP_TARGET_ARG_SUBSEQUENT_PARAM (1 << 7)
> +
> +/* Bitmask to apply to a target argument to find out the value identifier.
>
On Mon, Dec 07, 2015 at 12:20:49PM +0100, Martin Jambor wrote:
> +#include
> +#include
> +#include
> +#include
> +#include "libgomp-plugin.h"
> +#include "gomp-constants.h"
> +#include "hsa.h"
> +#include "hsa_ext_finalize.h"
If these 2 headers are coming from outside of gcc, better use <> for
On Mon, Dec 07, 2015 at 12:21:22PM +0100, Martin Jambor wrote:
> Subject: Make copy_gimple_seq_and_replace_locals copy seqs in omp clauses
>
> Hi,
>
> this is https://gcc.gnu.org/ml/gcc-patches/2015-12/msg00477.html with
> the early return requested by Jakub. Please refer to that previous
> emai
This time with a simplified testcase - the remaining issue with
the original one is if-conversion looking at DR_REF and nothing
hoisting the address compute of a[i+1] in both arms making it
see they are the same (yeah, SCEV to the rescue, but not in this
patch).
This patch re-organizes things so
On Wed, 9 Dec 2015, Richard Biener wrote:
>
> Well, not really in the end - the actual testcase would need code
> hoisting to make the refs equal enough for ifcvt. Making the testcase
> simpler doesn't seem to capture the issue so for now without one
> (I'll keep trying).
>
> Bootstrapped and t
On 12/09/2015 12:50 PM, Martin Liška wrote:
I would like to backport forgotten patch to GCC-5 branch.
Bootstrap and regression tests have been running, ready after it finishes?
Ok.
Bernd
On 27 November 2015 at 13:01, James Greenhalgh wrote:
> 2015-11-27 James Greenhalgh
>
> * config/aarch64/aarch64-protos.h
> (aarch64_cannot_change_mode_class): Bring back.
> * config/aarch64/aarch64.c
> (aarch64_cannot_change_mode_class): Likewise.
> * c
On Wed, Dec 9, 2015 at 12:14 PM, Martin Liška wrote:
> Hi.
>
> This is simple follow-up of the previous patch that fixes last remaining
> leak in vectorizer.
>
> Patch can regbootstrap on x64_64-linux-gnu.
>
> Ready for trunk?
Ok.
Richard.
> Martin
On Mon, Dec 07, 2015 at 12:22:43PM +0100, Martin Jambor wrote:
> it creates a copy of the entire target body and expands it slightly
> differently for concurrent execution on a GPU. Note that both teams
> and distribute constructs are mandatory. Moreover, currently the
> distribute has to be in a
On Mon, Dec 07, 2015 at 12:24:05PM +0100, Martin Jambor wrote:
> the pass manager changes required for HSA have already been committed
> to trunk so all that remains are these additions to the pass pipeline.
>
> Thanks,
>
> Martin
>
>
> 2015-12-04 Martin Jambor
> Martin Liska
>
This fixes vectorization re-start without SLP when SLP analysis
removed an SLP instance - I was relying on the reduction chains
still being in the SLP instances. The following properly
detects this case and also handles SLP reductions properly
(they can be vectorized non-SLP as well).
Bootstrap
On 04/12/2015 23:48, Jeff Law wrote:
>>
>> Why would pointer types be shifted at all (at the ubsan level,
>> which is basically the AST)?
> BTW, if you argument is that we can never get into this code with a
> shift of a pointer object, I'd like to see some kind of analysis to back
> up that asse
On Wed, Dec 9, 2015 at 2:20 PM, Jakub Jelinek wrote:
> On Mon, Dec 07, 2015 at 12:24:05PM +0100, Martin Jambor wrote:
>> the pass manager changes required for HSA have already been committed
>> to trunk so all that remains are these additions to the pass pipeline.
>>
>> Thanks,
>>
>> Martin
>>
>>
This removes the dubious dependence computation in if-conversion
just for the sake of testing if that might fail (given the vectorizer
itself doesn't fail that easily). It's a quadratic operation after all
which we should avoid at all cost.
Bootstrap and regtest running on x86_64-unknown-linux-g
On Tue, Dec 8, 2015 at 5:22 PM, H.J. Lu wrote:
> On Mon, Nov 23, 2015 at 12:53 PM, H.J. Lu wrote:
>> On Mon, Nov 23, 2015 at 1:57 AM, Richard Biener
>> wrote:
>>> On Sat, Nov 21, 2015 at 12:46 AM, H.J. Lu wrote:
On Fri, Nov 20, 2015 at 2:17 PM, Jason Merrill wrote:
> On 11/20/2015 01:
On 8 December 2015 at 21:35, Evandro Menezes wrote:
>Emit square root using the Newton series
>
>2015-12-03 Evandro Menezes
>
>gcc/
> * config/aarch64/aarch64-protos.h (aarch64_emit_swsqrt):
>Declare new
> function.
> * config/aarch64/aarch64-
On 12/09/2015 12:22 PM, Ajit Kumar Agarwal wrote:
This is because the available_regs = 6 and the regs_needed = 1 and
new_regs = 0 and the regs_used = 10. As the reg_used that are based
on the Liveness given above is greater than the available_regs, then
> it's candidate of spill and the estima
This patch adds the test for the fixed PR.
2015-12-09 Uros Bizjak
PR c++/68348
PR c++/68464
* g++.dg/pr68348.C: New test.
Tested on x86_64-linux-gnu.
OK for mainline?
Uros.
Index: g++.dg/pr68348.C
===
--- g++.dg/pr6
On Wed, Dec 09, 2015 at 11:41:44AM +0100, Richard Biener wrote:
> But I don't see why we have the asserts at all - they seem to be originated
> from
> development. IMHO factor_out_conditonal_conversion should simply return
> the new PHI it generates instead of relying on
> signle_non_signelton_ph
OK
I will add this text before "*memory_barrier" pattern:
;; For ARCHS, we use a hardware data memory barrier that waits for
;; completion of current data memory operations before initiating
;; similar data memory operations.
Once done, I will commit it.
Thanks,
Claudiu
>
> Tested with dg.exp (wh
On Wed, 2015-12-09 at 11:35 +0100, Richard Biener wrote:
> On Tue, Dec 8, 2015 at 11:24 PM, Ajit Kumar Agarwal
> wrote:
> > Based on the comments on RFC patch this patch incorporates all the comments
> > from Jeff. Thanks Jeff for the valuable feedback.
> >
> > This patch enables the better regis
On Wed, Dec 9, 2015 at 3:22 PM, Marek Polacek wrote:
> On Wed, Dec 09, 2015 at 11:41:44AM +0100, Richard Biener wrote:
>> But I don't see why we have the asserts at all - they seem to be originated
>> from
>> development. IMHO factor_out_conditonal_conversion should simply return
>> the new PHI
On 08/12/15 09:25, Tobias Burnus wrote:
On Mon, Dec 07, 2015 at 02:09:22PM +, Matthew Wahab wrote:
I wonder whether using
__asm__ __volatile__ ("":::"memory");
would be sufficient as it has a way lower overhead than
__sync_synchronize().
I don't know anything about Fortran or coarrays and
-Original Message-
From: Bernd Schmidt [mailto:bschm...@redhat.com]
Sent: Wednesday, December 09, 2015 7:34 PM
To: Ajit Kumar Agarwal; Richard Biener
Cc: Jeff Law; GCC Patches; Vinod Kathail; Shail Aditya Gupta; Vidhumouli
Hunsigida; Nagaraju Mekala
Subject: Re: [Patch,rtl Optimization]
On 7 December 2015 at 17:10, Matthew Wahab wrote:
> On 27/11/15 17:11, Matthew Wahab wrote:
>>
>> On 27/11/15 13:44, Christophe Lyon wrote:
On 26/11/15 16:02, Matthew Wahab wrote:
>>
>>
> This patch adds ARMv8.1 support to GCC Dejagnu, to allow ARM
> tests to specify targest and
Hi,
On 09.12.2015 12:06 Bernd Schmidt wrote:
> On 12/09/2015 03:18 AM, Bernd Edlinger wrote:
>> Furthermore there is a documented use for asm(""): The empty
>> assembler string is used to make a function
>> volatile, thus calls can not be optimized away. But I think it is
>> not necessary to ma
On Mon, 2015-11-02 at 16:41 -0700, Jeff Law wrote:
> On 11/02/2015 12:35 PM, David Malcolm wrote:
>
> >
> >> diff --git a/gdb/ada-lang.c b/gdb/ada-lang.c
> >> index fff4862..2559a36 100644
> >> --- a/gdb/ada-lang.c
> >> +++ b/gdb/ada-lang.c
> >> @@ -11359,9 +11359,11 @@ ada_evaluate_subexp (struct
> The main point of having rtl epilogues is to allow such optimizations.
> Traditionally, we have said that at -O1, it is OK to curb optimizations for
> the
> sake of having programs that are saner to debug, while -O2 and above should
> just optimize, and the debug info generation is just thrown i
This adds a testcase for the already fixed PR68348.
Tested on x86_64-linux, ok for trunk?
2015-12-09 Marek Polacek
PR c++/68348
* g++.dg/cpp0x/pr68348.C: New test.
diff --git gcc/testsuite/g++.dg/cpp0x/pr68348.C
gcc/testsuite/g++.dg/cpp0x/pr68348.C
index e69de29..9033bba 100
On 12/09/2015 04:38 PM, David Malcolm wrote:
+/* The following function contains examples of bad indentation that's
+ arguably not misleading, due to a blank line between the guarded and the
+ non-guarded code. Some of the blank lines deliberately contain
+ redundant whitespace, to verify
On 09/12/15 15:38, Marek Polacek wrote:
This adds a testcase for the already fixed PR68348.
Tested on x86_64-linux, ok for trunk?
2015-12-09 Marek Polacek
PR c++/68348
* g++.dg/cpp0x/pr68348.C: New test.
diff --git gcc/testsuite/g++.dg/cpp0x/pr68348.C
gcc/testsuite/g++.dg/
On Wed, Dec 09, 2015 at 03:39:50PM +, Kyrill Tkachov wrote:
> On 09/12/15 15:38, Marek Polacek wrote:
> >This adds a testcase for the already fixed PR68348.
> >
> >Tested on x86_64-linux, ok for trunk?
> >
> >2015-12-09 Marek Polacek
> >
> > PR c++/68348
> > * g++.dg/cpp0x/pr68348.C:
On 12/09/2015 04:09 PM, Bernd Edlinger wrote:
So would you agree on the general direction of the patch,
if I drop the hunk in sched-deps.c ?
I'm not sure there was any consensus in that other thread, but I think
assuming that basic asms clobber memory and CC, can be justified. That
certainly
I have committed the following testsuite patch as obvious.
The test calls a variadic function that extracts pointers with va_arg, but the
terminating NULL is passed as an int, not a pointer. This wouldn't trip on
32-bit architectures, and even on 64-bit the test simply iterates until it
gets a NU
This patch replaces some of the '%s' quotes of diagnostic strings
by the nicer quotes.
First, it replaces some leftovers of '%s' -> %qs in directly used
error strings.
It then also converts some (well: resolve.c only) '%s' to %%<%s%%>,
which are using with sprintf(), but which are still passed
Done.
I have permission for contributing but I don't have write permission
on the repository.
2015-12-09 8:23 GMT+01:00 Tobias Burnus :
> Alessandro Fanfarillo wrote:
>>
>> in attachment the new patch. I also checked the behavior with
>> move_alloc: it synchronizes right after the deregistration
On Wed, Dec 9, 2015 at 8:18 AM, Uros Bizjak wrote:
> Saying that, I see we don't need to define ASM_SUFFIX anymore. I'll
> prepare the patch that removes these #defines.
2015-12-09 Uros Bizjak
* gcc.target/i386/sse4_1-roundps-1.c: Remove ASM_SUFFIX define.
* gcc.target/i386/sse4_1-ro
Hi!
I've been meaning to suggest this for some time already:
On Wed, 9 Dec 2015 14:19:30 +0100, Jakub Jelinek wrote:
> As for omp-low.c changes, the file is already large enough that it would be
> nice if it is easy to find out what routines are for gridification purposes
> only, use some specia
Hi, Marcus.
I've run Geekbench, SPEC CPU2000 and synthetic benchmarks.
I can share these results iterating an array with values between 1 and 100
and taking their square root:
Million Operations/sJuno
A53 @850MHz A57 @1100MHz
X^½ DP
Hi,
On 09.12.2015 16:48 Bernd Schmidt wrote:
> On 12/09/2015 04:09 PM, Bernd Edlinger wrote:
>
>> So would you agree on the general direction of the patch,
>> if I drop the hunk in sched-deps.c ?
>
> I'm not sure there was any consensus in that other thread, but I think
> assuming that basic asms
Cesar,
On 12/08/2015 11:10 AM, Cesar Philippidis wrote:
On 12/08/2015 08:22 AM, James Norris wrote:
2. It appears that deviceptr code in GOACC_parallel_keyed is mostly
identical to GOACC_data_start. Can you put that duplicate code into
a function? That would be easier to maintai
On Wed, 2015-11-04 at 14:56 +0100, Bernd Schmidt wrote:
> On 10/30/2015 04:16 PM, David Malcolm wrote:
> > The idea is to more gracefully handle merger conflict markers
> > in the source code being compiled. Specifically, in the C and
> > C++ frontends, if we're about to emit an error, see if the
On Wed, 2015-12-09 at 16:40 +0100, Bernd Schmidt wrote:
> On 12/09/2015 04:38 PM, David Malcolm wrote:
> > +/* The following function contains examples of bad indentation that's
> > + arguably not misleading, due to a blank line between the guarded and the
> > + non-guarded code. Some of the b
Hi Evandro,
On 08/12/15 21:35, Evandro Menezes wrote:
Emit square root using the Newton series
2015-12-03 Evandro Menezes
gcc/
* config/aarch64/aarch64-protos.h (aarch64_emit_swsqrt):
Declare new
function.
* config/aarch64/aarch64-simd.md (sqrt2)
> > + bool in_symtab1 = decl_in_symtab_p (base1);
> > + bool in_symtab2 = decl_in_symtab_p (base2);
> > +
> > + /* Declarations of non-automatic variables may have aliases. All other
> > + decls are unique. */
> > + if (in_symtab1 != in_symtab2 || !in_symtab1)
> > +return 0;
> > + re
> On Wed, 9 Dec 2015, Arnaud Charlet wrote:
>
> > > Hi
> > > this patch implements the trik for punting if we get too many nested
> > > pointers.
> > > This fixes the ada tstcases. Curiously enough I would like to replace
> > > safe_push
> > > by quick_push but doing so I get weird error about fre
On 12/09/2015 10:52 AM, Kyrill Tkachov wrote:
Hi Evandro,
On 08/12/15 21:35, Evandro Menezes wrote:
Emit square root using the Newton series
2015-12-03 Evandro Menezes
gcc/
* config/aarch64/aarch64-protos.h (aarch64_emit_swsqrt):
Declare new
function.
On 09/12/15 16:59, Evandro Menezes wrote:
On 12/09/2015 10:52 AM, Kyrill Tkachov wrote:
Hi Evandro,
On 08/12/15 21:35, Evandro Menezes wrote:
Emit square root using the Newton series
2015-12-03 Evandro Menezes
gcc/
* config/aarch64/aarch64-protos.h (aarch64_emit_swsqrt)
Left shifts into the sign bit is a kind of overflow, and the
standard chooses to treat left shifts of negative values the
same way.
However, the -fwrapv option modifies the language to one where
integers are defined as two's complement---which also defines
entirely the behavior of shifts. Disable
On 09/12/15 17:02, Kyrill Tkachov wrote:
On 09/12/15 16:59, Evandro Menezes wrote:
On 12/09/2015 10:52 AM, Kyrill Tkachov wrote:
Hi Evandro,
On 08/12/15 21:35, Evandro Menezes wrote:
Emit square root using the Newton series
2015-12-03 Evandro Menezes
gcc/
* config/aarc
On 12/09/2015 04:49 PM, David Malcolm wrote:
> This is about managing the signal:noise ratio for something in -Wall.
>
> The distinction I want to make here is between badly indented code vs
> misleadingly indented code. Yes, the code is badly indented, but to my
> eyes the code is sufficiently
On 12/09/2015 05:24 PM, Thomas Schwinge wrote:
In addition to that, how about we split up gcc/omp-low.c into several
files? Would it make sense (I have not yet looked in detail) to do so
along the borders of the several passes defined therein? Or, can you
tell already that there would be too m
Hi,
this patch ensures that the &b[0][0][0] initializer of the acc loop in
the testcase is properly compiled.
Committed to gomp-4_0-branch.
Thanks,
- Tom
Handle ptr iterator in acc loop
2015-12-09 Tom de Vries
* omp-low.c (expand_omp_for_generic): Gimplify seq_loop loop start and
limit
Hi Christian,
On 08/12/15 12:53, Christian Bruel wrote:
Hi,
The order of the NEON builtins construction has led to complications since the attribute target support. This was not a problem when driven from the command line, but was causing various issues when the builtins was mixed between fpu
On 12/09/2015 05:49 PM, David Malcolm wrote:
+void
+fn_40_implicit_level_1 (int arg)
+{
+if (flagA)
+ foo (0);
+
+ foo (1);
+
The distinction I want to make here is between badly indented code vs
misleadingly indented code. Yes, the code is badly indented, but to my
eyes the code is suffici
Hi,
I've backported the patch "Fix oacc kernels default mapping for scalars"
from trunk to gomp-4_0-branch.
Committed to gomp-4_0-branch.
Thanks,
- Tom
backport: "Fix oacc kernels default mapping for scalars"
2015-12-02 Tom de Vries
backport from trunk:
* gimplify.c (enum gimplify_omp_
On 12/09/2015 05:58 PM, David Malcolm wrote:
On Wed, 2015-11-04 at 14:56 +0100, Bernd Schmidt wrote:
This seems like fairly low impact but also low cost, so I'm fine with it
in principle. I wonder whether the length of the marker is the same
across all versions of patch (and VC tools)?
It's h
Richard Biener wrote:
> On Fri, Dec 4, 2015 at 8:59 PM, Sebastian Paul Pop wrote:
> > I would highly recommend updating the required version of ISL to isl-0.15:
> > that would simplify the existing code, removing a lot of code under "#ifdef
> > old ISL",
> > and allow us to fully transition to sch
On Wed, Dec 09, 2015 at 04:53:37PM +0100, Tobias Burnus wrote:
>
> Build and regtested on x86-64-gnu-linux.
> OK for the trunk?
>
OK.
--
Steve
On December 9, 2015 6:45:42 PM GMT+01:00, Sebastian Pop
wrote:
>Richard Biener wrote:
>> On Fri, Dec 4, 2015 at 8:59 PM, Sebastian Paul Pop
> wrote:
>> > I would highly recommend updating the required version of ISL to
>isl-0.15:
>> > that would simplify the existing code, removing a lot of code
---
gcc/graphite-optimize-isl.c | 8
1 file changed, 8 insertions(+)
diff --git a/gcc/graphite-optimize-isl.c b/gcc/graphite-optimize-isl.c
index f90fcfd..50f2b3c 100644
--- a/gcc/graphite-optimize-isl.c
+++ b/gcc/graphite-optimize-isl.c
@@ -50,6 +50,7 @@ along with GCC; see the file COP
we used to add the access functions in the wrong order, Fortran style, leading
to unprofitable interchanges.
---
gcc/graphite-sese-to-poly.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/graphite-sese-to-poly.c b/gcc/graphite-sese-to-poly.c
index 887c212..480c552 100644
---
gcc/graphite-optimize-isl.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/graphite-optimize-isl.c b/gcc/graphite-optimize-isl.c
index 8727e39..f90fcfd 100644
--- a/gcc/graphite-optimize-isl.c
+++ b/gcc/graphite-optimize-isl.c
@@ -426,7 +426,7 @@ optimize_isl (scop_p s
This is a MIPS patch to make mips_output_division obey the
-fno-delayed-branch flag. Right now, with mips1 and -mcheck-zero-division,
the division instruction is put into the bne delay slot even when
-fno-delayed-branch is specified. This change uses a similar strategy
to MIPS16 where we do the d
On 20 November 2015 at 22:28, Jeff Law wrote:
>>> That was the overall plan and he posted a patch for that. But that patch
>>> didn't do the due diligence to verify that once the shortening code was
>>> made
>>> "pure" that we didn't regress on the quality of the code we generated.
>>
>>
>> I tho
On 09/12/15 15:34, Claudiu Zissulescu wrote:
Well, it seems to me that we prefer to disable optimizations when talking about
debug related information (see PR target/60598 git-svn-id:
svn+ssh://gcc.gnu.org/svn/gcc/trunk@208749 138bc75d-0d04-0410-961f-82ee72b054a4
commit).
Actually, unwind in
On 12/09/2015 11:16 AM, Kyrill Tkachov wrote:
On 09/12/15 17:02, Kyrill Tkachov wrote:
On 09/12/15 16:59, Evandro Menezes wrote:
On 12/09/2015 10:52 AM, Kyrill Tkachov wrote:
Hi Evandro,
On 08/12/15 21:35, Evandro Menezes wrote:
Emit square root using the Newton series
2015-12-03 Evan
Hello,
I have created a git-only branch "gomp-nvptx" for development of OpenMP
offloading on NVPTX. Changelogs on the branch have the corresponding suffix,
and I'll use it to tag emails with patches for the branch.
The branch currently has 41 patch on top of yesterday's trunk. Some of
those hav
On Wed, Dec 9, 2015 at 6:05 AM, Richard Biener
wrote:
> On Tue, Dec 8, 2015 at 5:22 PM, H.J. Lu wrote:
>> On Mon, Nov 23, 2015 at 12:53 PM, H.J. Lu wrote:
>>> On Mon, Nov 23, 2015 at 1:57 AM, Richard Biener
>>> wrote:
On Sat, Nov 21, 2015 at 12:46 AM, H.J. Lu wrote:
> On Fri, Nov 20,
On 06.12.15 23:54, Andreas Tobler wrote:
Hi,
I'm going to commit this patch to trunk, 5.4 and 4.9 branch if there are
no objections.
The redefinition of WCHAR_TYPE for PowerPC64 is wrong since its
beginning. My fault.
We use the definition from freebsd.h.
Committed to trunk. But I left the
1 - 100 of 128 matches
Mail list logo
