On Mon, Sep 14, 2020 at 09:30:44PM -0400, Marek Polacek via Gcc-patches wrote:
> --- a/gcc/c/c-tree.h
> +++ b/gcc/c/c-tree.h
> @@ -147,6 +147,11 @@ struct c_expr
> etc), so we stash a copy here. */
>source_range src_range;
>
> + /* True iff the sizeof expression was enclosed in parent
On Tue, Sep 15, 2020 at 10:49:46AM +0930, Alan Modra wrote:
> Implement more two insn constants.
And tests. rot_cst1 checks the values generated, rot_cst2 checks
instruction count.
* gcc.target/powerpc/rot_cst.h,
* gcc.target/powerpc/rot_cst1.c,
* gcc.target/powerpc/rot_c
On Mon, Sep 14, 2020 at 5:19 AM Feng Xue OS via Gcc-patches
wrote:
>
> Thanks,
> Feng
>
>
> From: Feng Xue OS
> Sent: Thursday, September 3, 2020 5:29 PM
> To: Richard Biener; gcc-patches@gcc.gnu.org
> Subject: Re: [PATCH 2/2 V3] Simplify plusminus-mult-wi
Hi!
These patterns printed bogus <>s around the {1to16} and similar strings.
Fixed thusly, bootstrapped/regtested on x86_64-linux and i686-linux,
committed to trunk as obvious. Will do backports momentarily.
2020-09-15 Jakub Jelinek
PR target/97028
* config/i386/sse.md (*mul
Hi Hans,
on 2020/9/6 上午10:47, Hans-Peter Nilsson wrote:
> On Tue, 1 Sep 2020, Bin.Cheng via Gcc-patches wrote:
>>> Great idea! With explicitly specified -funroll-loops, it's bootstrapped
>>> but the regression testing did show one failure (the only one):
>>>
>>> PASS->FAIL: gcc.dg/sms-4.c scan-
Hi,
Gentle ping this:
https://gcc.gnu.org/pipermail/gcc-patches/2020-May/546698.html
BR,
Kewen
on 2020/8/31 下午1:49, Kewen.Lin via Gcc-patches wrote:
> Hi,
>
> I'd like to gentle ping this since IVOPTs part is already to land.
>
> https://gcc.gnu.org/pipermail/gcc-patches/2020-May/546698.html
Hi!
The following testcase is miscompiled (in particular the a and i
initialization). The problem is that build_special_member_call due to
the immediate constructors (but not evaluated in constant expression mode)
doesn't create a CALL_EXPR, but returns a TARGET_EXPR with CONSTRUCTOR
as the initi
gcc/ChangeLog:
* config/csky/csky.md (CSKY_NPARM_FREGS): New.
(call_value_internal_vs/d): New.
(untyped_call): New.
* config/csky/csky.h (TARGET_SINGLE_FPU): New.
(TARGET_DOUBLE_FPU): New.
(FUNCTION_VARG_REGNO_P): New.
(CSKY_VREG_MODE_P): New
On Tue, Sep 15, 2020 at 4:59 AM Hongtao Liu wrote:
>
> Hi:
> This patch would avoid spill gprs to mask registers for non-AVX512
> micro-architecture and fix regression in PR96744.
>
> Bootstrap is ok, regression test for i386/x86-64 backend is ok.
> No big performance impact on SPEC2017.
>
>
Hi,
the cleanup_all_empty_eh function was originally doing a post-order traversal
of the EH region tree to optimize it:
/* Do a post-order traversal of the EH region tree. Examine each
post_landing_pad block and see if we can eliminate it as empty. */
That's sensible since the worker funct
OG10 = devel/omp/gcc-10
Committed backport plus two cherry picks:
a93cc852103 [PATCH] OpenMP/Fortran: Fix (re)mapping of allocatable/pointer
arrays [PR96668]
e524656359b Merge remote-tracking branch 'origin/releases/gcc-10' into
devel/omp/gcc-10
f73772df64c Daily bump.
0ea1b39e8e
On Tue, Sep 15, 2020 at 10:36:20AM +0200, Eric Botcazou wrote:
> This was recently changed to use another order and this trivially breaks
> testcases with nested regions like the attached one. So the attached patch
> restores the post-order traversal and it also contains a small tweak to the
>
Qing Zhao writes:
>> On Sep 14, 2020, at 2:20 PM, Richard Sandiford
>> wrote:
>>
>> Qing Zhao mailto:qing.z...@oracle.com>> writes:
On Sep 14, 2020, at 11:33 AM, Richard Sandiford
wrote:
Qing Zhao writes:
>> Like I mentioned earlier though, passes that run after
gcc/ChangeLog:
* genemit.c (main): Print 'split line'.
* Makefile.in (insn-emit.c): Define split count and file
---
gcc/Makefile.in | 19 +
gcc/genemit.c | 104 +---
2 files changed, 83 insertions(+), 40 deletions(-)
diff --
Hi,
Ok & Thanks, I will do my best for it :)
It’s fixed in patch v7.
Jojo
在 2020年9月11日 +0800 PM11:29，Segher Boessenkool ，写道：
> Hi!
>
> On Fri, Sep 11, 2020 at 03:26:17PM +0800, Jojo R wrote:
> > +#define printf_include() do { \
>
> Don't use macros please, use a function?
>
> And
>> This patch is to handle simplification of plusminus-mult-with-convert
>> expression
>> as ((T) X) +- ((T) Y), in which at least one of (X, Y) is result of
>> multiplication.
>> This is done in forwprop pass. We try to transform it to (T) (X +- Y), and
>> resort
>> to gimple-matcher to fold (X
Add a rule (T)(A) +- (T)(B) -> (T)(A +- B), which works only when (A +- B)
could be folded to a simple value. By this rule, a plusminus-mult-with-convert
expression could be handed over to the rule (A * C) +- (B * C) -> (A +- B).
Bootstrapped/regtested on x86_64-linux and aarch64-linux.
Feng
---
On Tue, Sep 15, 2020 at 12:14 PM Feng Xue OS
wrote:
>
> Add a rule (T)(A) +- (T)(B) -> (T)(A +- B), which works only when (A +- B)
> could be folded to a simple value. By this rule, a plusminus-mult-with-convert
> expression could be handed over to the rule (A * C) +- (B * C) -> (A +- B).
Please
> So it breaks PR93199 again?
Indeed, although there is no regression in the testsuite AFAICS. I guess that
we can do the new walk before and not instead of the post-order traversal.
Revised patch attached, same ChangeLog.
--
Eric Botcazoudiff --git a/gcc/tree-eh.c b/gcc/tree-eh.c
index 4246d
Segher Boessenkool writes:
> On Mon, Sep 14, 2020 at 05:33:33PM +0100, Richard Sandiford wrote:
>> > However, for the cases on Power as Segher mentioned, there are also some
>> > scratch registers used for
>> > Other purpose, not sure whether we can correctly generate zeroing in
>> > middle-end
On Tue, Sep 15, 2020 at 1:29 PM Eric Botcazou wrote:
>
> > So it breaks PR93199 again?
>
> Indeed, although there is no regression in the testsuite AFAICS.
Yeah, too big of a testcase ...
> I guess that
> we can do the new walk before and not instead of the post-order traversal.
>
> Revised pat
>> Add a rule (T)(A) +- (T)(B) -> (T)(A +- B), which works only when (A +- B)
>> could be folded to a simple value. By this rule, a
>> plusminus-mult-with-convert
>> expression could be handed over to the rule (A * C) +- (B * C) -> (A +- B).
>
>Please use INTEGRAL_TYPE_P () instead of TREE_CODE ==
Hi,
As suggested by PR93334 comments, this patch adds an interface identifying
output dependence which can be skipped in terms of reordering and skip it in
loop distribution. It also adds a new test case. Any comment?
Thanks,
bin
0001-Skip-output-dependence-if-values-stored-are-bytewise.patch
D
On Tue, Sep 15, 2020 at 2:25 PM Feng Xue OS wrote:
>
> >> Add a rule (T)(A) +- (T)(B) -> (T)(A +- B), which works only when (A +- B)
> >> could be folded to a simple value. By this rule, a
> >> plusminus-mult-with-convert
> >> expression could be handed over to the rule (A * C) +- (B * C) -> (A +
The following allows more BB vectorization by generally building leafs
from scalars rather than giving up. Note this is only a first step
towards this and as can be seen with the exception for node splitting
it is generally hard to get this heuristic sound. I've added variants
of the bb-slp-48.c
Hi Sebastian,
This patch implements missing intrinsics.
I'm okay with this being applied to the GCC 8 branch as these intrinsics have
been defined in ACLE for a long time.
It is arguably a bug that they've been missing from GCC8.
Their implementation is fairly self-contained we haven't had any bu
commit 1bcb4c4faa4bd6b1c917c75b100d618faf9e628c
Author: Richard Sandiford
Date: Wed Oct 2 07:37:10 2019 +
[LRA] Don't make eliminable registers live (PR91957)
didn't make eliminable registers live which breaks
register void *cur_pro asm("reg");
where "reg" is an eliminable register.
On Tue, Sep 15, 2020 at 08:32:54AM +0200, Richard Biener wrote:
> On Tue, Sep 15, 2020 at 12:06 AM Segher Boessenkool
> wrote:
> >
> > After the previous patch we are left with an unreachable BB. This will
> > ICE if either we have -fschedule-fusion, or we do not have peephole2.
> >
> > This fixe
Hi!
On Thu, Sep 10, 2020 at 04:58:03PM -0500, Peter Bergner wrote:
> Generating arbitrary 64-bit constants on POWER can take up to 5 instructions.
> However, some special constants can be generated in fewer instructions.
> One special class of constants we don't handle, is constants that have one
Hi!
On Tue, Sep 15, 2020 at 02:23:16PM +0930, Alan Modra wrote:
> On Thu, Sep 10, 2020 at 04:58:03PM -0500, Peter Bergner via Gcc-patches wrote:
> > +unsigned long
> > +test0 (void)
> > +{
> > + return 0x0000UL;
> > +}
> > +
> > +unsigned long
> > +test1 (void)
> > +{
> > + return
Since some time we're only using this argument to communicate from
vect_build_slp_tree_1 to vect_get_and_check_slp_defs. This makes
the direction of information flow clear.
Bootstrap / regtest running on x86_64-unknown-linux-gnu.
2020-09-15 Richard Biener
* tree-vect-slp.c (vect_get_
On 14/09/2020 20:04, Wilco Dijkstra wrote:
> Hi Richard,
>
>> On 14/09/2020 15:19, Wilco Dijkstra wrote:
>>> The --with-cpu/--with-arch configure option processing not only checks
>>> valid arguments
>>> but also sets TARGET_CPU_DEFAULT with a CPU and extension bitmask. This
>>> isn't used
>>>
commit e95395926a84a2406faefe0995295d199d595440
Author: Uros Bizjak
Date: Thu Jun 18 20:12:48 2020 +0200
i386: Fix mode of ZERO_EXTRACT RTXes, remove ext_register_operand predicate.
caused
FAIL: gcc.target/i386/pr78904-4a.c scan-assembler [ \t]movb[\t ]+%.h, t
when compiled with --target
Thanks for looking at this.
"H.J. Lu" writes:
> commit 1bcb4c4faa4bd6b1c917c75b100d618faf9e628c
> Author: Richard Sandiford
> Date: Wed Oct 2 07:37:10 2019 +
>
> [LRA] Don't make eliminable registers live (PR91957)
>
> didn't make eliminable registers live which breaks
>
> register voi
On Tue, Sep 15, 2020 at 7:44 AM Richard Sandiford
wrote:
>
> Thanks for looking at this.
>
> "H.J. Lu" writes:
> > commit 1bcb4c4faa4bd6b1c917c75b100d618faf9e628c
> > Author: Richard Sandiford
> > Date: Wed Oct 2 07:37:10 2019 +
> >
> > [LRA] Don't make eliminable registers live (PR919
On 9/14/20 6:47 PM, Tobias Burnus wrote:
This patch cause run-time fails for
g++ -fopenmp libgomp/testsuite/libgomp.c++/udr-13.C
The follow-up fix does not help.
Namely, in udr-3.C:115:
115 if (t.s != 11 || v.v != 9 || q != 0 || d != 3.0) abort ();
(gdb) p t.s
oops, I forgot t
Over the last couple of months quite a few warnings about uninitialized
variables were raised while building GCC. A reason why these warnings
show up on S/390 only is due to the aggressive inlining settings here.
Some of these warnings (2c832ffedf0, b776bdca932, 2786c0221b6,
1657178f59b) could be
> On Sep 15, 2020, at 4:11 AM, Richard Sandiford
> wrote:
>
> Qing Zhao mailto:qing.z...@oracle.com>> writes:
>>> On Sep 14, 2020, at 2:20 PM, Richard Sandiford
>>> wrote:
>>>
>>> Qing Zhao mailto:qing.z...@oracle.com>> writes:
> On Sep 14, 2020, at 11:33 AM, Richard Sandiford
>
I am seeing this unused parameter 'opts' error when building for this
configuration:
Build: arm-none-linux-gnueabihf
Host: arm-none-linux-gnueabihf
Target: arm-none-linux-gnueabihf
In function 'void arm_option_restore(gcc_options*, gcc_options*,
cl_target_option*)':
/src/gcc/gcc/config/arm/arm.c
On 9/14/20 7:17 PM, Marek Polacek via Gcc-patches wrote:
> On Mon, Sep 14, 2020 at 11:13:18AM -0400, Jason Merrill via Gcc-patches wrote:
>> On Mon, Jun 29, 2020 at 1:25 PM Martin Liška wrote:
>>> On 6/29/20 4:57 PM, Marek Polacek wrote:
On Mon, Jun 29, 2020 at 09:51:57AM +0200, Martin Liška
> rs6000_is_valid_shift_mask handles this already (but it requires you to
> pass in the shift needed). rs6000_is_valid_mask will handle it.
> rs6000_is_valid_and_mask does not get a shift count parameter, so cannot
> use rldic currently.
After talking with you off line, I changed to using rs6000_
Hi Ke Wen,
On Tue, Sep 15, 2020 at 02:40:38PM +0800, Kewen.Lin wrote:
> >>* config/rs6000/rs6000-p8swap.c (insn_rtx_pair_t): New type.
> >
> > Please don't do that. The "first" and "second" are completely
> > meaningless. Also, keeping it separate arrays can very well result in
> > better
This fixes an ICE when trying to copy a legacy value_range containing a
symbolic to a multi-range:
min = make_ssa_name (type);
max = build_int_cst (type, 55);
value_range vv (min, max);
int_range<2> vr = vv;
This doesn't affect anything currently, as we don't hav
On Tue, Sep 15, 2020 at 08:51:09AM +0200, Richard Biener wrote:
> On Tue, Sep 15, 2020 at 5:56 AM luoxhu wrote:
> > > u[n % 4] = i;
> > >
> > > I guess. Is the % 4 mandated by the vec_insert semantics btw?
(As an aside -- please use "& 3" instead: that works fine if n is signed
as well, but
On Tue, Sep 15, 2020 at 12:48:59AM +0200, Tobias Burnus wrote:
> + bool has_nullptr;
> + size_t j;
> + for (j = 0; j < n->tgt->list_count; j++)
> + if (n->tgt->list[j].key == n)
> + {
> + has_nullptr = n->tgt->list[j].has_null_
The attached patch updates the manual to mention that Wuninitialized
and -Wmaybe-uninitialized are issued for both auto and allocated
objects, as well as for passing pointers to uninitialized objects
to const-qualified parameters. Both of these features are GCC 11
enhancements.
Martin
Document -
On Linux/x86_64,
c9de716a59c873859df3b3e1fbb993200fce5a73 is the first bad commit
commit c9de716a59c873859df3b3e1fbb993200fce5a73
Author: Richard Biener
Date: Tue Sep 15 14:35:40 2020 +0200
Allow more BB vectorization
caused
FAIL: g++.dg/vect/slp-pr87105.cc -std=c++14 scan-tree-dump sl
On Tue, 2020-09-15 at 10:49 +0930, Alan Modra via Gcc-patches wrote:
> The existing "case AND" in this function is not sufficient for
> optabs.c:avoid_expensive_constant usage, where the AND is passed in
> outer_code.
>
> * config/rs6000/rs6000.c (rs6000_rtx_costs): Move costing for
>
On Tue, 2020-09-15 at 10:49 +0930, Alan Modra via Gcc-patches wrote:
> This patch series fixes a number of issues in rs6000_rtx_costs, the
> aim being to provide costing somewhat closer to reality. Probably
> the
> most important patch of the series is patch 4, which just adds a
> comment. Withou
Hello, Mike,
On Sep 11, 2020, Michael Meissner via Gcc-patches
wrote:
> +case SFmode:
> +case DFmode:
gcc110 (ppc64) in the build farm didn't like this. The bootstrap
compiler barfs on these expressions, because of some constexpr issue I
haven't really looked into.
I'm testing this p
On Thu, Sep 10, 2020 at 01:45:50PM +0200, Jakub Jelinek wrote:
> On Thu, Sep 10, 2020 at 01:16:57PM +0200, Jakub Jelinek via Gcc-patches wrote:
> > As for the test assembly, I'd say we should take
> > #define F void foo (void) {}
> > F
> > compile it with
> > gcc -S -O2 -g1 -dA -gno-as-loc-support
On Mon, Sep 14, 2020 at 10:07:31PM -0500, Qing Zhao wrote:
> > On Sep 14, 2020, at 6:09 PM, Segher Boessenkool
> > wrote:
> >> Gadget 1:
> >>
> >> mov rax, value
> >> syscall
> >> ret
> >
> > No, just
> >
> > mov rax,59
> > syscall
> >
> > (no ret necessary!)
>
> But for ROP, a typical gad
The -Wuninitialized/-Wmaybe-uninitialized enhancement to warn when
a pointer or reference to an uninitialized object is passed to
a const-qualified function argument tries to avoid triggering for
objects of empty types. However, the suppression is incomplete
and lets the warning trigger in some c
On Linux/x86_64,
8f0d743c2dee6afae5c6f861b0642b7b112a4a70 is the first bad commit
commit 8f0d743c2dee6afae5c6f861b0642b7b112a4a70
Author: Feng Xue
Date: Mon Aug 17 23:00:35 2020 +0800
tree-optimization/94234 - add plusminus-with-convert pattern
caused
FAIL: gcc.dg/tree-ssa/20030807-10.c
On Linux/x86_64,
8f0d743c2dee6afae5c6f861b0642b7b112a4a70 is the first bad commit
commit 8f0d743c2dee6afae5c6f861b0642b7b112a4a70
Author: Feng Xue
Date: Mon Aug 17 23:00:35 2020 +0800
tree-optimization/94234 - add plusminus-with-convert pattern
caused
FAIL: gcc.dg/ifcvt-3.c scan-rtl-dump
Dear Tobias,
I can see that you want a proper fix.
However, after having looked at all these comments about the powerpc
situation, I do not really think I'd want to ever touch that stuff.
It's clearly beyond my capabilities and ressources.
I do feel responsible for the regression introduced by m
On Tue, Sep 15, 2020 at 12:46:00PM +0100, Richard Sandiford wrote:
> Segher Boessenkool writes:
> > On Mon, Sep 14, 2020 at 05:33:33PM +0100, Richard Sandiford wrote:
> >> > However, for the cases on Power as Segher mentioned, there are also some
> >> > scratch registers used for
> >> > Other pur
On Tue, Sep 15, 2020 at 10:11:41AM +0100, Richard Sandiford wrote:
> Qing Zhao writes:
> >> On Sep 14, 2020, at 2:20 PM, Richard Sandiford
> >> wrote:
(Putting correct info in DF, inserting the new insns in pro_and_epi).
But, scheduling runs *after* that, and then you need to prevent the
insert
Overflowing the size of a dynamic allocation (e.g., malloc or VLA)
can lead to a subsequent buffer overflow corrupting the heap or
stack. The attached patch diagnoses a subset of these cases where
the overflow/wraparound is still detectable.
Besides regtesting GCC on x86_64-linux I also verified
On 9/15/20 7:03 PM, Jakub Jelinek wrote:
On Tue, Sep 15, 2020 at 12:48:59AM +0200, Tobias Burnus wrote:
+ bool has_nullptr;
+ size_t j;
+ for (j = 0; j < n->tgt->list_count; j++)
+if (n->tgt->list[j].key == n)
+ {
+has_nullptr
On 9/15/20 1:38 PM, Alexandre Oliva wrote:
>> +case SFmode:
>> +case DFmode:
>
> gcc110 (ppc64) in the build farm didn't like this. The bootstrap
> compiler barfs on these expressions, because of some constexpr issue I
> haven't really looked into.
>
> I'm testing this patch. I'll check
On Tue, Sep 15, 2020 at 10:48:37AM -0500, Peter Bergner wrote:
> > rs6000_is_valid_shift_mask handles this already (but it requires you to
> > pass in the shift needed). rs6000_is_valid_mask will handle it.
> > rs6000_is_valid_and_mask does not get a shift count parameter, so cannot
> > use rldic
On 9/15/20 11:57 AM, Aldy Hernandez wrote:
This fixes an ICE when trying to copy a legacy value_range containing
a symbolic to a multi-range:
min = make_ssa_name (type);
max = build_int_cst (type, 55);
value_range vv (min, max);
int_range<2> vr = vv;
This doesn't affect anythin
On Tue, Sep 15, 2020 at 03:38:05PM -0300, Alexandre Oliva wrote:
> On Sep 11, 2020, Michael Meissner via Gcc-patches
> wrote:
>
> > +case SFmode:
> > +case DFmode:
>
> gcc110 (ppc64) in the build farm didn't like this. The bootstrap
> compiler barfs on these expressions, because of som
On Tue, Sep 15, 2020 at 09:04:41AM +0200, Jakub Jelinek via Gcc-patches wrote:
> On Mon, Sep 14, 2020 at 09:30:44PM -0400, Marek Polacek via Gcc-patches wrote:
> > --- a/gcc/c/c-tree.h
> > +++ b/gcc/c/c-tree.h
> > @@ -147,6 +147,11 @@ struct c_expr
> > etc), so we stash a copy here. */
> >
As stated in the PR, the Fortran 2018 standard removed the restriction
prohibiting ELEMENTAL RECURSIVE procedures. Adjust the relevant check.
Regtested on x86_64-pc-linux-gnu.
OK for master?
Thanks,
Harald
PR fortran/97036 - [F2018] Allow ELEMENTAL RECURSIVE procedure prefix
gcc/fortran/Chan
On Wed, Aug 26, 2020 at 10:46:37PM -0400, Michael Meissner wrote:
> diff --git a/gcc/config/rs6000/predicates.md b/gcc/config/rs6000/predicates.md
> index 2709e46f7e5..60b45601e9b 100644
> --- a/gcc/config/rs6000/predicates.md
> +++ b/gcc/config/rs6000/predicates.md
> @@ -1225,6 +1225,11 @@ (define
PR analyzer/96650 reports an assertion failure when merging the
intersection of two sets of constraints, due to the resulting
constraints being infeasible.
It turns out that the two input sets were each infeasible if
transitivity were considered, but -fanalyzer-transitivity was off.
However for th
On Tue, Sep 15, 2020 at 10:49:39AM +0930, Alan Modra wrote:
> rldimi is generated by rs6000_emit_set_long_const when the high and
> low 32 bits of a 64-bit constant are equal.
>
> * config/rs6000/rs6000.c (num_insns_constant_gpr): Count rldimi
> constants correctly.
Wow, did I miss th
> On Sep 15, 2020, at 2:41 PM, Segher Boessenkool
> wrote:
>
> On Tue, Sep 15, 2020 at 10:11:41AM +0100, Richard Sandiford wrote:
>> Qing Zhao writes:
On Sep 14, 2020, at 2:20 PM, Richard Sandiford
wrote:
> (Putting correct info in DF, inserting the new insns in pro_and_epi).
>
On Tue, Sep 15, 2020 at 10:49:40AM +0930, Alan Modra wrote:
> These functions do behave a little differently for SImode, so the
> mode should be passed.
>
> * config/rs6000/rs6000.c (rs6000_rtx_costs): Pass mode to
> reg_or_add_cint_operand and reg_or_sub_cint_operand.
Okay for trunk.
On Wed, 9 Sep 2020, Martin Sebor via Gcc-patches wrote:
> Joseph, do you have any concerns with or comments on the most
> recent patch or is it okay as is?
>
> https://gcc.gnu.org/pipermail/gcc-patches/2020-August/552266.html
I'm not yet convinced by the logic for extracting an array bound from
On Tue, Sep 15, 2020 at 05:31:48PM -0500, Qing Zhao wrote:
> > But, scheduling runs *after* that, and then you need to prevent the
> > inserted (zeroing) insns from moving -- if you don't, the code after
> > some zeroing can be used as gadget! You want to always have all
> > zeroing insns after *a
On Wed, 9 Sep 2020, Martin Sebor via Gcc-patches wrote:
> Ping: https://gcc.gnu.org/pipermail/gcc-patches/2020-August/552500.html
>
> Aldy provided a bunch of comments on this patch but I'm still looking
> for a formal approval.
This patch is OK.
> > Some testing revealed that the code has diff
On Linux/x86_64,
8f0d743c2dee6afae5c6f861b0642b7b112a4a70 is the first bad commit
commit 8f0d743c2dee6afae5c6f861b0642b7b112a4a70
Author: Feng Xue
Date: Mon Aug 17 23:00:35 2020 +0800
tree-optimization/94234 - add plusminus-with-convert pattern
caused
FAIL: gcc.dg/tree-ssa/20030807-10.c
On Linux/x86_64,
8f0d743c2dee6afae5c6f861b0642b7b112a4a70 is the first bad commit
commit 8f0d743c2dee6afae5c6f861b0642b7b112a4a70
Author: Feng Xue
Date: Mon Aug 17 23:00:35 2020 +0800
tree-optimization/94234 - add plusminus-with-convert pattern
caused
FAIL: gcc.dg/ifcvt-3.c scan-rtl-dump
This libgo patch by Than McIntosh adds a few more explicit references
to enumeration constants (RUSAGE_SELF, DT_UNKNOWN) in sysinfo.c to
insure that their hosting enums are emitted into DWARF, when using a
clang host compiler during the gollvm build. Bootstrapped and ran Go
testsuite on x86_64-pc-
> On Sep 15, 2020, at 6:09 PM, Segher Boessenkool
> wrote:
>
> On Tue, Sep 15, 2020 at 05:31:48PM -0500, Qing Zhao wrote:
>>> But, scheduling runs *after* that, and then you need to prevent the
>>> inserted (zeroing) insns from moving -- if you don't, the code after
>>> some zeroing can be us
The description in rs6000-builtin.def provides for a builtin named
__builtin_altivec_xst_len_r. However, it is hand-defined in
altivec_init_builtins as __builtin_xst_len_r, against the usual naming
practice. Fix that.
Bootstrapped and tested on powerpc64le-unknown-linux-gnu with no
regressions;
It looks good to me, pushed it to trunck.
Thanks,
Cooper
On 9/15/20 4:08 PM, Jojo R wrote:
gcc/ChangeLog:
* config/csky/csky.md (CSKY_NPARM_FREGS): New.
(call_value_internal_vs/d): New.
(untyped_call): New.
* config/csky/csky.h (TARGET_SINGLE_FPU): New.
Hi:
Rtx cost of sse_to_integer would be used by pass_stv as a
measurement for the scalar-to-vector transformation. As
https://gcc.gnu.org/pipermail/gcc-patches/2019-August/528839.html
indicates, movement between sse regs and gprs should be much expensive
than movement inside gprs(which is 2 as de
Hi all,
This patch sets the default value to 16 for parameter
max_unrolled_average_calls which could be used to restict calls in loop
when unrolling. This default value(16) is a big number which keeps
current behavior for almost all cases.
Bootstrap and regtest pass on powerpc64le. Is this ok f
gcc/ChangeLog:
* config/csky/csky-linux-elf.h (GLIBC_DYNAMIC_LINKER): Use mfloat-abi.
---
gcc/config/csky/csky-linux-elf.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/config/csky/csky-linux-elf.h b/gcc/config/csky/csky-linux-elf.h
index 9a57dd04..cf587ae 10064
Thanks Kyrill for your review.
I committed the patches to the gcc-8 branch:
https://gcc.gnu.org/git/?p=gcc.git;a=commitdiff;h=2c55e6caa9432b2c1f081cb3aeddd36abec03233
https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=a4004f62d60ada3a20dbf30146ca461047a575cc
and to the gcc-9 branch:
https://gcc.gnu.or
Hi Segher,
Thanks for your suggestions!
>> + for (unsigned i = 0; i < and_insns.length (); ++i)
>
> "i++" is used more often, is more traditional.
>
Updated.
>> --- /dev/null
>> +++ b/gcc/testsuite/gcc.target/powerpc/pr97019.c
>> @@ -0,0 +1,82 @@
>> +/* This issue can only exist on littl
On 2020/9/15 14:51, Richard Biener wrote:
>> I only see VAR_DECL and PARM_DECL, is there any function to check the tree
>> variable is global? I added DECL_REGISTER, but the RTL still expands to
>> stack:
>
> is_global_var () or alternatively !auto_var_in_fn_p (), I think doing
> IFN_SET onl
Hi:
If -mavx implies -mxsave, then -mno-xsave should imply -mno-avx.
Current status is -mno-avx implies -mno-xsave which should be wrong.
Bootstrap is ok, Regression test is ok for i386/x86 backend.
Ok for trunk?
gcc/ChangeLog
* common/config/i386/i386-common.c
(OPTION_MASK_ISA_A
Hi!
As the testcases show, if we have something like:
MEM [&b + 8B] = {};
MEM[(short *) &b] = 5;
_5 = *x_4(D);
MEM [&b + 2B] = _5;
MEM[(char *)&b + 16B] = 88;
MEM[(int *)&b + 20B] = 1;
then in sort_by_bitpos the stores are almost like in the given order,
except the first store is aft
On Wed, 16 Sep 2020, Jakub Jelinek wrote:
> Hi!
>
> As the testcases show, if we have something like:
> MEM [&b + 8B] = {};
> MEM[(short *) &b] = 5;
> _5 = *x_4(D);
> MEM [&b + 2B] = _5;
> MEM[(char *)&b + 16B] = 88;
> MEM[(int *)&b + 20B] = 1;
> then in sort_by_bitpos the stores ar
89 matches
Mail list logo
