Hi!
5 further backports to 7 branch, bootstrapped/regtested on x86_64-linux and
i686-linux, committed to branch.
Jakub
2017-12-22 Jakub Jelinek
Backported from mainline
2017-12-19 Jakub Jelinek
PR ipa/82801
PR ipa/83346
* ipa-inline.c (flatt
Hi,
I would like to again ping this patch:
https://gcc.gnu.org/ml/gcc-patches/2017-11/msg01451.html
It has been reviewed (with thanks) by David Malcolm[1] and
Martin Sebor[2]. Their concerns are addressed in the latest
revision of the patch:
https://gcc.gnu.org/ml/gcc-patches/2017-12/msg00544.h
This is a follow up patch for pr83488 to fix an error in setting
OPTION_MASK_ISA_AVX512VNNI_SET and OPTION_MASK_ISA_AVX512F_SET bits.
There were both set in ix86_isa_flags2 while being defined in
different ISA sets. Additionally move OPTION_MASK_ISA_AVX512VNNI_SET
to ix86_isa_flags as it can be use
On Fri, Dec 22, 2017 at 11:35 AM, Tsimbalist, Igor V
wrote:
> This is a follow up patch for pr83488 to fix an error in setting
>
> OPTION_MASK_ISA_AVX512VNNI_SET and OPTION_MASK_ISA_AVX512F_SET bits.
>
> There were both set in ix86_isa_flags2 while being defined in
>
> different ISA sets. Addition
On Fri, Dec 22, 2017 at 11:41:27AM +0100, Uros Bizjak wrote:
> > Ok for trunk?
>
> OK.
On top of this patch and the reason why we needed to move avx512vnni
from flags2 to flags instead of just:
case OPT_mavx512vnni:
if (value)
{
opts->x_ix86_isa_flags2 |= OPTION_MA
Hi Janne,
I think, in general, our current system where
-std=gnu means roughly "everything we have implemented so far, except
legacy extensions that break standards conformance" is pretty good.
The problem that I see with implementing this change is stack space,
especially the traditionally li
Hello, Julia,
On 12 Nov 12:51, Koval, Julia wrote:
> Hi, this patch enables AVX512BITALG and AVX512VPOPCNTDQ instructions from
> https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensions-programming-reference.pdf.
> Ok for trunk?
OK for trunk. I've check
Hello Jakub,
On 22 Dec 12:05, Jakub Jelinek wrote:
> On Fri, Dec 22, 2017 at 11:41:27AM +0100, Uros Bizjak wrote:
> > > Ok for trunk?
> >
> > OK.
>
> On top of this patch and the reason why we needed to move avx512vnni
> from flags2 to flags instead of just:
> case OPT_mavx512vnni:
>
On Fri, Dec 22, 2017 at 03:38:03PM +0300, Kirill Yukhin wrote:
> Hello, Julia,
> On 12 Nov 12:51, Koval, Julia wrote:
> > Hi, this patch enables AVX512BITALG and AVX512VPOPCNTDQ instructions from
> > https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensio
Hi!
The recent change to clear TREE_SIDE_EFFECTS on STATEMENT_LIST containing
DEBUG_BEGIN_STMTs and a single other statement without TREE_SIDE_EFFECTS on
it breaks the C stmt expr handling. The problem is that it assumes if
TREE_SIDE_EFFECTS is clear on a STATEMENT_LIST then that means the
STATEM
Hi!
Starting with GCC6 we emit for C (C++ is correct) incorrect DW_AT_decl_*
location for structs which are forward declared, then have some incomplete
var defined with it and finally defined.
The problem is that we process incomplete_vars before actually updating
DECL_SOURCE_LOCATION (TYPE_STUB_
Hi.
In order to fix the PR, I save temporary expression to a SSA_NAME that
is then used in the gcall.
Patch can bootstrap on ppc64le-redhat-linux and survives regression tests.
Ready to be installed?
Martin
gcc/ChangeLog:
2017-12-22 Martin Liska
PR tree-optimization/83552
This patch to the Go frontend by Cherry Zhang brings the Go frontend's
escape analysis mostly in line with gc compiler.
Changes include:
- In the gc compiler, the variable expression is represented with the
variable node itself (ONAME). It is the same node used in the AST for
multiple var express
This patch by Cherry Zhang fixes the Go frontend to not propagate
address-taken of a slice element to the slice.
Array_index_expression may be used for indexing/slicing array or
slice. If a slice element is address taken, the slice itself is not
necessarily address taken. Only propagate address-ta
On Fri, Dec 22, 2017 at 04:18:05PM +0100, Martin Liška wrote:
> In order to fix the PR, I save temporary expression to a SSA_NAME that
> is then used in the gcall.
You need to do that only if (!is_gimple_val (arg1_len)).
Can you please emit the additional stmt only if that isn't true?
> --- a/gcc
On Fri, 22 Dec 2017, Jakub Jelinek wrote:
> Hi!
>
> The recent change to clear TREE_SIDE_EFFECTS on STATEMENT_LIST containing
> DEBUG_BEGIN_STMTs and a single other statement without TREE_SIDE_EFFECTS on
> it breaks the C stmt expr handling. The problem is that it assumes if
> TREE_SIDE_EFFECTS
On Fri, 22 Dec 2017, Jakub Jelinek wrote:
> Hi!
>
> Starting with GCC6 we emit for C (C++ is correct) incorrect DW_AT_decl_*
> location for structs which are forward declared, then have some incomplete
> var defined with it and finally defined.
>
> The problem is that we process incomplete_vars
On Fri, Dec 22, 2017 at 03:38:03PM +0300, Kirill Yukhin wrote:
> Hello, Julia,
> On 12 Nov 12:51, Koval, Julia wrote:
> > Hi, this patch enables AVX512BITALG and AVX512VPOPCNTDQ instructions from
> > https://software.intel.com/sites/default/files/managed/c5/15/architecture-instruction-set-extensio
On Thu, 2017-12-21 at 00:00 -0500, Jason Merrill wrote:
> On Wed, Dec 20, 2017 at 12:33 PM, David Malcolm
> wrote:
> > On Tue, 2017-12-19 at 14:55 -0500, Jason Merrill wrote:
> > > On 12/17/2017 11:29 AM, David Malcolm wrote:
> > > > On Mon, 2017-12-11 at 18:45 -0500, Jason Merrill wrote:
> > > >
On Fri, Dec 22, 2017 at 07:40:11PM +0100, Jakub Jelinek wrote:
> Starting bootstrap/regtest on x86_64-linux and i686-linux right now, ok for
> trunk if it passes?
Bootstrapped/regtested successfully on both.
Jakub
This breaks gcc.dg/tls/opt-3.c, gcc.dg/tls/pr47715-3.c and
gcc.dg/tls/struct-1.c on m68k:
/daten/aranym/gcc/gcc-20171222/gcc/testsuite/gcc.dg/tls/opt-3.c:11:3: internal
compiler error: in add_args_size_note, at rtlanal.c:2379
0xae7aa9 add_args_size_note(rtx_insn*, poly_int<1u, l
With a value of 85 GCC has a CPU performance degradation of 11%,
reverting PRED_LOOP_EXIT to 92 this degradation disappear.
Those values where measured by running c-ray ray-tracer that is a
floating point benchmark that runs out of L1 cache.
Signed-off-by: David Esparza
---
contrib/analyze_brpro
Hi,
in this error recovery issue cp_check_const_attributes and more
generally cplus_decl_attributes have lots of troubles handling the
error_mark_node returned by cp_parser_std_attribute_spec_seq, as called
by cp_parser_direct_declarator. I fiddled quite a bit with these parsing
facilities to
23 matches
Mail list logo
