On вторник, 3 септември 2019 г. 14:54:19 EEST Richard Biener wrote:
> 2019-09-02 Richard Biener
>
> PR rtl-optimization/36262
> * postreload-gcse.c: Include intl.h and gcse.h.
> (insert_expr_in_table): Insert at the head of cur_expr->avail_occr
> to avoid linear
On Wed, Sep 4, 2019 at 9:44 AM Hongtao Liu wrote:
>
> On Wed, Sep 4, 2019 at 12:50 AM Uros Bizjak wrote:
> >
> > On Tue, Sep 3, 2019 at 1:33 PM Richard Biener
> > wrote:
> >
> > > > > Note:
> > > > > Removing limit of cost would introduce lots of regressions in
> > > > > SPEC2017 as follow
> >
Built and regression tested on x86_64-*-freebsd.
The patch restores an improved error message for a malformed
type-spec. See pr91660_1.f90 for code demonstrating problem.
While here, improve the error messages for other malformed
type-specs. Sett pr91660_2.f90.
OK to commit?
2019-09-04
This patch to the Go frontend defaults to -fno-partial-inlining. This
is to work around PR 91663: a case in which a function is split for
partial inlining, and then the two functions are inlined together into
a single function, leading to a confusing stack trace. This doesn't
work for Go since
This is a version of Cesar's patch to rewrite OpenACC reference reductions
(e.g. Fortran function arguments) to use temporary scalar variables. This
is necessary at present to avoid stack-slot clashes between multiple
workers on AMD GCN.
Julian
ChangeLog
gcc/
* gimplify.c
This patch enables middle-end worker partitioning and multiple workers
on AMD GCN.
Julian
ChangeLog
gcc/
* config/gcn/gcn.c (gcn_goacc_validate_dims): Remove
no-flag_worker-partitioning assertion.
(TARGET_GOACC_WORKER_PARTITIONING): Define target hook to true.
This patch implements worker-partitioning support in the middle end,
by rewriting gimple. The OpenACC execution model requires that code can
run in either "worker single" mode where only a single worker per gang
is active, or "worker partitioned" mode, where multiple workers per gang
are active.
This patch adjusts some tests after the splitting of the oaccdevlow pass
into three passes.
Julian
ChangeLog
gcc/testsuite/
* c-c++-common/goacc/classify-kernels-unparallelized.c,
c-c++-common/goacc/classify-kernels.c,
c-c++-common/goacc/classify-parallel.c,
This patch adds support for rewriting variables marked up with the "oacc
gangprivate" attributes in a target-dependent way in the oaccdevlow pass
of the offload compiler.
This behaviour is controlled by a new target hook,
TARGET_GOACC_ADJUST_GANGPRIVATE_DECL. This is conceptually similar to
the
This patch renames the TARGET_GOACC_ADJUST_PROPAGATION_RECORD
hook introduced in the GCN backend by a previous merge to
TARGET_GOACC_CREATE_PROPAGATION_RECORD, and removes a FIXME relating to
missing worker-partitioning support.
Julian
ChangeLog
gcc/
* config/gcn/gcn-protos.h
This patch series provides support for worker partitioning in the middle
end. The OpenACC device-lowering pass (oaccdevlow) is split into three
passes: the first assigns parallelism levels to loops, the second (new)
part rewrites basic blocks to implement a neutering/broadcasting scheme
for the
Arnaud,
> > This change (not posted to `gcc-patches' for some reason) has caused a
> > regression in the form of a build failure with the `riscv-linux-gnu'
> > target (and for the record the `x86_64-linux-gnu' build/host running GCC
> > 8.3.0):
> >
> > ali.adb:34:28: warning: use clause for
This patch fixes a bug with the lib-13.f90 test -- an asynchronous
compute region inside a synchronous data region leads to a data race
copying out/unmapping target data.
This test failed intermittently for AMD GCN. I will apply to the
openacc-gcc-9-branch shortly.
Thanks,
Julian
ChangeLog
Thank you for the thorough review!
On Tue, Sep 3, 2019 at 9:31 AM Jonathan Wakely wrote:
> It looks like__adl_begin and __adl_end should be guarded by #ifdef
> _GLIBCXX_P1394 because they're not needed otherwise.
I'm absolutely going to need it for the bit data structures (and
in
On 04/09/19 18:21 +0100, Mike Crowe wrote:
On Wednesday 04 September 2019 at 17:57:45 +0100, Mike Crowe wrote:
On Wednesday 04 September 2019 at 17:14:30 +0100, Jonathan Wakely wrote:
> On 04/09/19 15:49 +0100, Mike Crowe wrote:
> > On Wednesday 04 September 2019 at 14:39:35 +0100, Jonathan
Vtable verification currently does not work properly with LTO. Fixing
this will be non-trivial, and I currently do not have the time do
this. Until this can be fixed, we should not allow users to specify
both vtable verification and link-time optimization. The attached
patch checks to see if
On 9/4/19 3:59 PM, Caroline Tice via gcc-patches wrote:
> Vtable verification currently does not work properly with LTO. Fixing
> this will be non-trivial, and I currently do not have the time do
> this. Until this can be fixed, we should not allow users to specify
> both vtable verification and
While testing some other buffer overflow improvements I was
reminded that GCC doesn't detect even some basic instances
of overflowing struct members involving strcpy calls with
non-constant strings, with or without _FORTIFY_SOURCE. For
example:
struct S { char a[4]; void (*pf)(void); };
On Wed, Sep 04, 2019 at 03:09:26PM +0300, Janne Blomqvist wrote:
> Currently, when a new thread needs to use the RANDOM_NUMBER intrinsic,
> the per-thread PRNG state is initialized by copying the master state
> and then jumping forwards N*2**128 entries in the stream so that the
> PRNG streams for
On Wed, Sep 04, 2019 at 04:00:24PM -0400, Jason Merrill wrote:
> On 9/3/19 4:08 PM, Marek Polacek wrote:
> > First constinit bug report (yay?). The problem is that while I made sure
> > that constinit variable templates work as they should, I clearly neglected
> > ordinary static variables
On Wed, 4 Sep 2019 at 16:16, Kyrill Tkachov wrote:
>
> Hi Christophe,
>
> On 5/15/19 1:39 PM, Christophe Lyon wrote:
> > Support additional relocations: TLS_GD32_FDPIC, TLS_LDM32_FDPIC, and
> > TLS_IE32_FDPIC.
> >
> > We do not support the GNU2 TLS dialect.
> >
> > 2019-XX-XX Christophe Lyon
>
On 9/3/19 4:08 PM, Marek Polacek wrote:
First constinit bug report (yay?). The problem is that while I made sure
that constinit variable templates work as they should, I clearly neglected
ordinary static variables declared constinit in function templates.
Hmm, right, locals don't get
On Tue, 3 Sep 2019 at 10:40, Richard Sandiford
wrote:
>
> Christophe Lyon writes:
> > @@ -3485,6 +3485,14 @@ arm_option_override (void)
> >if (flag_pic && TARGET_VXWORKS_RTP)
> > arm_pic_register = 9;
> >
> > + /* If in FDPIC mode then force arm_pic_register to be r9. */
> > + if
On 9/3/19 8:01 PM, Marek Polacek wrote:
On Tue, Sep 03, 2019 at 08:52:54PM -0400, Marek Polacek wrote:
As I recently threatened, it's time we let -fdeduce-init-list go.
This patch removes the implementation while keeping the option for
backward compatibility.
Bootstrapped/regtested on
Attached patch fixes PR 32413 by forcing secondary memory for QI and
HImode moves between SSE and integer registers. This works together
with (existing) ix86_secondary_memory_needed_mode, where QI and HImode
secondary memory mode is extended to SImode.
The previous patch that fixe PR 32413 forced
Hi, Jeff
Em seg, 29 de abr de 2019 às 18:22, Jeff Law escreveu:
>
> On 1/22/19 12:31 PM, Rafael Tsuha wrote:
> > This patch simplifies the expression sinh (x) / cosh (x) to tanh (x).
> > This rule is mathematically valid.
> >
> > There's a slight difference in the result when applying this
> >
On Tue, Sep 03, 2019 at 06:33:26PM -0500, Segher Boessenkool wrote:
> On Tue, Sep 03, 2019 at 07:20:13PM -0400, Michael Meissner wrote:
> > On Tue, Sep 03, 2019 at 05:56:03PM -0500, Segher Boessenkool wrote:
> > > Hi!
> > >
> > > On Mon, Aug 26, 2019 at 05:43:41PM -0400, Michael Meissner wrote:
>
On Wed, 4 Sep 2019 at 01:22, Jeff Law wrote:
>
> On 8/28/19 8:55 PM, Prathamesh Kulkarni wrote:
> > Hi,
> > This is a rebased patch on trunk for PR78736. The last time, it got
> > stuck, because of warning issues with libgfortran, for which I filed
> > PR91593. The patch relegates the warning to
On Wed, Sep 04, 2019 at 10:08:38PM +0530, Prathamesh Kulkarni wrote:
> On Tue, 9 May 2017 at 18:59, Jakub Jelinek wrote:
> >
> > On Tue, May 09, 2017 at 06:55:12PM +0530, Prathamesh Kulkarni wrote:
> > > ping https://gcc.gnu.org/ml/gcc-patches/2017-05/msg00135.html
> >
> > The libgomp patch is ok
On Tue, 9 May 2017 at 18:59, Jakub Jelinek wrote:
>
> On Tue, May 09, 2017 at 06:55:12PM +0530, Prathamesh Kulkarni wrote:
> > ping https://gcc.gnu.org/ml/gcc-patches/2017-05/msg00135.html
>
> The libgomp patch is ok provided the warning is added.
> Though, should it be in -Wall and not say just
On 04/09/19 15:49 +0100, Mike Crowe wrote:
On Wednesday 04 September 2019 at 14:39:35 +0100, Jonathan Wakely wrote:
On 15/07/19 17:47 +0100, Mike Crowe wrote:
> The pthread_cond_clockwait function was recently added[1] to glibc, and is
> due to be released in glibc 2.30. If this function is
On 04/09/2019 16:00, Bernd Edlinger wrote:
On 9/4/19 4:14 PM, Richard Earnshaw (lists) wrote:
On 04/09/2019 14:28, Bernd Edlinger wrote:
On 9/4/19 2:53 PM, Richard Earnshaw (lists) wrote:
Index: gcc/testsuite/gcc.target/arm/unaligned-argument-2.c
On 8/15/19 6:42 PM, Carroll, Paul wrote:
> This is a proposed patch to libiberty, with an accompanying patch to GCC.
> The purpose of this patch is to make it possible for Windows-hosted
> toolchains to have the ability to control whether Canonicalized
> filenames are converted to all lower-case.
On 8/23/19 10:50 AM, Andre Vieira (lists) wrote:
> I am not sure whether this is the right thing to do, we might want to
> enable this only for certain targets and if so, I also don't know how to
> tackle the testing issue where all the tests that check the number of
> times a dump mentions
On 8/23/19 10:50 AM, Andre Vieira (lists) wrote:
> Hi,
>
> This patch is an improvement on my last RFC. As you pointed out, we can
> do the vectorization analysis of the epilogues before doing the
> transformation, using the same approach as used by openmp simd. I have
> not yet incorporated
On 9/4/19 3:21 PM, Kyrill Tkachov wrote:
On 9/3/19 4:00 PM, Jeff Law wrote:
> On 9/3/19 8:01 AM, Kyrill Tkachov wrote:
>> Hi all,
>>
>> On 9/5/18 12:48 PM, a...@codesourcery.com wrote:
>>> At present, pointers passed to builtin functions, including atomic
>>> operators,
>>> are stripped of
On 9/4/19 3:18 AM, Richard Sandiford wrote:
> This patch uses the workaround Richi suggested in the PR: make
> discover_nonconstant_array_refs mark the source of an IFN_LOAD_LANES
> call and the destination of an IFN_STORE_LANES call as addressable,
> so that they don't end up being REG rtxes
On 9/4/19 4:14 PM, Richard Earnshaw (lists) wrote:
> On 04/09/2019 14:28, Bernd Edlinger wrote:
>> On 9/4/19 2:53 PM, Richard Earnshaw (lists) wrote:
>>> Index: gcc/testsuite/gcc.target/arm/unaligned-argument-2.c
>>> ===
>>> ---
On 9/4/19 2:05 AM, Martin Liška wrote:
> On 9/3/19 6:37 PM, Jeff Law wrote:
>> On 8/30/19 2:55 AM, Martin Liška wrote:
>>> PING^1
>>>
>>> On 8/13/19 9:49 AM, Martin Liska wrote:
Hi.
I'm sending format independent changes to mklog that should
improve the script. It addresses
On 9/4/19 3:35 AM, Richard Sandiford wrote:
> Jeff Law writes:
>> On 9/3/19 2:16 AM, Mihailo Stojanovic wrote:
>>> From: Mihailo Stojanovic
>>>
>>> Hi everybody,
>>>
>>> This fixes the MSA implementation on big-endian targets which is
>>> essentially broken for things like SUBREG handling and
On Wednesday 04 September 2019 at 14:39:35 +0100, Jonathan Wakely wrote:
> On 15/07/19 17:47 +0100, Mike Crowe wrote:
> > The pthread_cond_clockwait function was recently added[1] to glibc, and is
> > due to be released in glibc 2.30. If this function is available in the C
> > library it can be
On 9/3/19 4:00 PM, Jeff Law wrote:
On 9/3/19 8:01 AM, Kyrill Tkachov wrote:
Hi all,
On 9/5/18 12:48 PM, a...@codesourcery.com wrote:
At present, pointers passed to builtin functions, including atomic
operators,
are stripped of their address space properties. This doesn't seem to be
Hi Christophe,
On 5/15/19 1:39 PM, Christophe Lyon wrote:
2019-XX-XX Christophe Lyon
Mickaël Guêné
libgcc/
* unwind-arm-common.inc (ARM_SET_R7_RT_SIGRETURN)
(THUMB2_SET_R7_RT_SIGRETURN, FDPIC_LDR_R12_WITH_FUNCDESC)
(FDPIC_LDR_R9_WITH_GOT,
Hi Christophe,
On 5/15/19 1:39 PM, Christophe Lyon wrote:
Support additional relocations: TLS_GD32_FDPIC, TLS_LDM32_FDPIC, and
TLS_IE32_FDPIC.
We do not support the GNU2 TLS dialect.
2019-XX-XX Christophe Lyon
Mickaël Guêné
gcc/
* config/arm/arm.c (tls_reloc): Add
On 04/09/2019 14:28, Bernd Edlinger wrote:
On 9/4/19 2:53 PM, Richard Earnshaw (lists) wrote:
On 15/08/2019 20:47, Bernd Edlinger wrote:
On 8/15/19 6:29 PM, Richard Biener wrote:
Please split it into the parts for the PR and parts making the
asserts not trigger.
Yes, will do.
Okay,
On Sep 04 2019, "Maciej W. Rozycki" wrote:
> This means that no released GCC version can be used to build trunk GNAT,
> which would provide a clean upgrade path, and there is only a narrow
> window between r272860 and r272975 inclusive that lets one move on to GNAT
> 10. I find this rather
Maciej,
> This change (not posted to `gcc-patches' for some reason) has caused a
> regression in the form of a build failure with the `riscv-linux-gnu'
> target (and for the record the `x86_64-linux-gnu' build/host running GCC
> 8.3.0):
>
> ali.adb:34:28: warning: use clause for package
Hi,
On Wed, 3 Jul 2019, pmdero...@gcc.gnu.org wrote:
> Author: pmderodat
> Date: Wed Jul 3 08:14:57 2019
> New Revision: 272976
>
> URL: https://gcc.gnu.org/viewcvs?rev=272976=gcc=rev
> Log:
> [Ada] ABE checks v3.0, foundations of Elaboration order v4.0
>
>
> --
On 15/07/19 17:47 +0100, Mike Crowe wrote:
The pthread_cond_clockwait function was recently added[1] to glibc, and is
due to be released in glibc 2.30. If this function is available in the C
library it can be used it to fix
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=41861 by supporting
On 9/4/19 2:53 PM, Richard Earnshaw (lists) wrote:
> On 15/08/2019 20:47, Bernd Edlinger wrote:
>> On 8/15/19 6:29 PM, Richard Biener wrote:
>
> Please split it into the parts for the PR and parts making the
> asserts not trigger.
>
Yes, will do.
>>
>> Okay, here is
Hello,
On 3.9.19. 21:16, Jeff Law wrote:
On 9/3/19 2:16 AM, Mihailo Stojanovic wrote:
From: Mihailo Stojanovic
Hi everybody,
This fixes the MSA implementation on big-endian targets which is
essentially broken for things like SUBREG handling and calling
convention for vector types. It
I've backported the fix for PR81800 to GCC8 and GCC9:
PR81800 is about the lrint inline giving spurious FE_INEXACT exceptions.
The previous change for PR81800 didn't fix this: when lrint is disabled
in the backend, the midend will simply use llrint. This actually makes
things worse since llrint
On 9/4/19 7:20 AM, Martin Liška wrote:
On 9/4/19 11:22 AM, Jonathan Wakely wrote:
The point of the warning was to see if users complain. Three weeks
isn't a very long time to see if users are going to complain :-)
I see :) That said, I'm going to install the patch close to the end
of
On 15/08/2019 20:47, Bernd Edlinger wrote:
On 8/15/19 6:29 PM, Richard Biener wrote:
Please split it into the parts for the PR and parts making the
asserts not trigger.
Yes, will do.
Okay, here is the rest of the PR 89544 fix,
actually just an optimization, making the larger stack
Currently, when a new thread needs to use the RANDOM_NUMBER intrinsic,
the per-thread PRNG state is initialized by copying the master state
and then jumping forwards N*2**128 entries in the stream so that the
PRNG streams for different threads don't alias each other, where N is
the number of
On 9/4/19 11:22 AM, Jonathan Wakely wrote:
> On Wed, 4 Sep 2019 at 09:13, Martin Liška wrote:
>>
>> On 7/9/19 3:00 PM, Martin Liška wrote:
>>> Great. Then I'm sending patch that does the functionality removal.
>>
>> Hi.
>>
>> The GCC 9.2.0 is released and the version contains a deprecation
>>
Some minor changes.
Feng
diff --git a/gcc/ipa-fnsummary.c b/gcc/ipa-fnsummary.c
index 278bf606661..89a6b8bf061 100644
--- a/gcc/ipa-fnsummary.c
+++ b/gcc/ipa-fnsummary.c
@@ -296,9 +296,9 @@ set_hint_predicate (predicate **p, predicate new_predicate)
}
-/* Compute what conditions may or
>> + if (unmodified_parm_or_parm_agg_item (fbi, stmt, expr, index_p,
>> + size_p, aggpos))
>> + {
>> + if (param_ops_p)
>> + {
>> + /* Find use of parameter, add a convert operation to describe
>> + result type,
Jeff Law writes:
> On 9/3/19 2:16 AM, Mihailo Stojanovic wrote:
>> From: Mihailo Stojanovic
>>
>> Hi everybody,
>>
>> This fixes the MSA implementation on big-endian targets which is
>> essentially broken for things like SUBREG handling and calling
>> convention for vector types. It borrows
On Wed, 4 Sep 2019 at 09:13, Martin Liška wrote:
>
> On 7/9/19 3:00 PM, Martin Liška wrote:
> > Great. Then I'm sending patch that does the functionality removal.
>
> Hi.
>
> The GCC 9.2.0 is released and the version contains a deprecation
> warning for -frepo.
>
> Is it the right time now to
On Wed, Aug 28, 2019 at 3:43 AM Steve Kargl
wrote:
>
> The attach patch was built and tested on i586-*-freebsd.
> It includes a check for ALLOCATED with no arguments.
> OK to commit?
Ok.
--
Janne Blomqvist
This patch uses the workaround Richi suggested in the PR: make
discover_nonconstant_array_refs mark the source of an IFN_LOAD_LANES
call and the destination of an IFN_STORE_LANES call as addressable,
so that they don't end up being REG rtxes during expansion.
I had to move the
On 7/9/19 3:00 PM, Martin Liška wrote:
Great. Then I'm sending patch that does the functionality removal.
Hi.
The GCC 9.2.0 is released and the version contains a deprecation
warning for -frepo.
Is it the right time now to remove it from trunk?
Thanks,
Martin
On 9/3/19 6:37 PM, Jeff Law wrote:
On 8/30/19 2:55 AM, Martin Liška wrote:
PING^1
On 8/13/19 9:49 AM, Martin Liska wrote:
Hi.
I'm sending format independent changes to mklog that should
improve the script. It addresses couple of improvement
mentioned here:
64 matches
Mail list logo
