Hi:
Currently for VCOND_EXPR, integer mask operation is only available
for 512-bit vector, but since mask register is related to isa not
vector size, under avx512f we can also have 128/256-bit vector
condition move. My local tests show there's no boost frequency penalty
for using integer mask reg
This patch implements deprecate POD for the C++ Standard Library,
bringing libstdc++ that much closer to 2020 conformance 🎉!
Hilariously, a small bug in the [[deprecated]] warning message was
found while implementing this patch, which drove me a bit insane for a
good 10 minutes until I realized wh
Hi,
When running tests on a slightly-revised version of the
worker-partitioning patch series previously posted starting here:
https://gcc.gnu.org/ml/gcc-patches/2019-11/msg01475.html
I discovered that Kwok's patch to optimise GCN register usage
(r278301) was causing one of the new tests to fail:
This defines operator<=> as a non-member function template and does not
alter operator==. This contradicts the changes made by P1614R2, which
specify both as hidden friends, but that specification of operator<=> is
broken and the subject of a soon-to-be-published LWG issue.
* include/bits
* include/bits/stl_pair.h (pair): Remove stray Doxygen closing marker.
I added this stray marker a few months ago by mistake.
Committed to trunk.
commit ba39ad7d7e29e1459ad1d816dc84753ef7503f51
Author: Jonathan Wakely
Date: Tue Dec 3 11:50:31 2019 +
libstdc++: Fix Doxygen ma
On Nov 14, 2019, Alexandre Oliva wrote:
> In order to address this, I propose we add an internal option (not for
> the driver), -dumpbase-ext, that names the extension to be discarded
> from dumpbase to form aux output names.
Here's a WIP patch that implements much of the desired semantics.
I'm
After allocating a new chunk of memory hash_table::expand() copy-
assigns elements from the current array over the newly allocated
elements without constructing them.
Similarly, after destroying existing elements, hash_table::
empty_slow() assigns a new value to them. This bug was
introduced in
Hello
I've made an implementation of P0829 and tested it with gcc for avr.
I've included the patch and I believe it might need some modification.
Should I rather just maintain my own fork until P0829 is a bit
further? (It is currently being split into smaller proposals).
I have not included further
gimple_call_fndecl (and gimple_call_addr_fndecl) fail to identify functions
for non-trivial uses of function pointers.
This patch eliminates most uses of these functions from the analyzer in
favor of resolving the rvalue for the fn ptr in the region_model, so that
with the patch it can e.g. detect
On 8/5/19 4:30 PM, Jason Merrill wrote:
On Mon, Aug 5, 2019 at 5:50 PM Martin Sebor wrote:
On 8/5/19 1:25 PM, Jason Merrill wrote:
On 8/1/19 7:35 PM, Martin Sebor wrote:
On 8/1/19 12:09 PM, Jason Merrill wrote:
On 7/22/19 12:34 PM, Martin Sebor wrote:
Ping: https://gcc.gnu.org/ml/gcc-patch
On Tue, Dec 03, 2019 at 03:37:41PM -0500, Jason Merrill wrote:
> On 11/15/19 7:28 AM, Jakub Jelinek wrote:
> > + loc = LOCATION_LOCUS (loc);
> ...
> > + entry.loc
> > += linemap_resolve_location (line_table, loc, LRK_MACRO_EXPANSION_POINT,
> > + &map);
>
> You don't
On Tue, Dec 03, 2019 at 01:20:04PM -0500, Michael Meissner wrote:
> On Tue, Nov 26, 2019 at 01:20:20PM -0600, Segher Boessenkool wrote:
> > > I needed to add a new constraint (em) in addition to new predicate
> > > functions.
> > > I discovered that with the predicate function alone, the register
On 11/15/19 7:28 AM, Jakub Jelinek wrote:
+ loc = LOCATION_LOCUS (loc);
...
+ entry.loc
+= linemap_resolve_location (line_table, loc, LRK_MACRO_EXPANSION_POINT,
+ &map);
You don't need LOCATION_LOCUS if you're calling
linemap_resolve_location. LGTM with t
On 12/3/19 7:39 AM, Richard Sandiford wrote:
Jason Merrill writes:
On 11/29/19 5:59 AM, Richard Sandiford wrote:
Ping
Richard Sandiford writes:
This is the C++ equivalent of r277950, which prevented the
use of the GNU vector extensions with SVE vector types for C.
[https://gcc.gnu.org/viewc
On Tue, Dec 03, 2019 at 01:11:55PM -0500, Michael Meissner wrote:
> On Fri, Nov 22, 2019 at 08:11:16PM -0600, Segher Boessenkool wrote:
> > On Thu, Nov 14, 2019 at 05:56:50PM -0500, Michael Meissner wrote:
> > > * lib/target-supports.exp
> > > (check_effective_target_powerpc_future_ok): Do not
On Tue, Dec 03, 2019 at 12:57:24PM -0500, Michael Meissner wrote:
> No, the change for num_insns_constant_gpr could not go in until the support
> for
> PLI went in (patch V6 #1).
Well, I lost track. So your version 7 to 9 patches do *not* replace the
v6 patches? Or does "V" mean something else?
On 12/2/19 5:22 PM, Paolo Carlini wrote:
Hi,
On 02/12/19 19:58, Jason Merrill wrote:
On 11/29/19 8:08 AM, Paolo Carlini wrote:
Hi,
a few more rather straightforward uses for cp_expr_loc_or_input_loc.
Additionally, while working on the latter, I noticed that, compared
to say, gcc-7, lately t
On 12/3/19 3:42 AM, Jakub Jelinek wrote:
Hi!
The following testcase shows that during constexpr evaluation we didn't
handle TARGET_EXPR_CLEANUP at all (which was probably fine when there
weren't constexpr dtors). My understanding is that TARGET_EXPR cleanups
should be queued and evaluated only
On 12/3/19 1:40 PM, Jason Merrill wrote:
On 12/3/19 12:49 PM, Marek Polacek wrote:
On Tue, Dec 03, 2019 at 02:07:02AM -0500, Jason Merrill wrote:
On 12/2/19 5:09 PM, Marek Polacek wrote:
On Mon, Dec 02, 2019 at 12:09:17PM -0500, Jason Merrill wrote:
On 12/1/19 8:09 PM, Marek Polacek wrote:
+
On 12/3/19 12:49 PM, Marek Polacek wrote:
On Tue, Dec 03, 2019 at 02:07:02AM -0500, Jason Merrill wrote:
On 12/2/19 5:09 PM, Marek Polacek wrote:
On Mon, Dec 02, 2019 at 12:09:17PM -0500, Jason Merrill wrote:
On 12/1/19 8:09 PM, Marek Polacek wrote:
+|| (skip_empty
+
Andi Kleen writes:
Ping!
> Andi Kleen writes:
>
> Ping!
>
>> From: Andi Kleen
>>
>> [v4: Rebased on current tree. Avoid some redundant log statements
>> for locals and a few other fixes. Fix some comments. Improve
>> documentation. Did some studies on the debug information quality,
>> see bel
Hi,
while looking into Firefox inlining dumps I noticed that we often do not
inline because we think function calls comdat local while the comdat group
itself has been dissolved.
Bootstrapped/regtested x86_64-linux, comitted.
* cgraph.c (cgraph_node::verify_node): Check that calls_comdat_
On Tue, Nov 26, 2019 at 01:20:20PM -0600, Segher Boessenkool wrote:
> Hi!
>
> On Thu, Nov 14, 2019 at 06:09:09PM -0500, Michael Meissner wrote:
> > In this case, the current code re-uses the temporary for calculating the
> > offset
> > of the element to load up the address of the vector, losing t
On Fri, Nov 22, 2019 at 08:11:16PM -0600, Segher Boessenkool wrote:
> On Thu, Nov 14, 2019 at 05:56:50PM -0500, Michael Meissner wrote:
> > * lib/target-supports.exp
> > (check_effective_target_powerpc_future_ok): Do not require 64-bit
> > or Linux support before doing the test. Use a
On Tue, Dec 03, 2019 at 10:33:48PM +0900, Oleg Endo wrote:
> On Mon, 2019-11-25 at 16:47 -0600, Segher Boessenkool wrote:
> >
> > > > - sh (that's sh4-linux):
> > > >
> > > > /home/segher/src/kernel/net/ipv4/af_inet.c: In function
> > > > 'snmp_get_cpu_field':
> > > > /home/segher/src/kernel/net
On Mon, 2 Dec 2019, Ian Lance Taylor wrote:
> > gcc/testsuite/
> > * lib/gcc-defs.exp (gcc-set-multilib-library-path): Use
> > `-print-file-name=' to determine the multilib root directory.
> > Use `remote_exec host' rather than `exec' to invoke the
> > compi
Hi!
On Tue, Dec 03, 2019 at 05:38:41PM +0800, Kewen.Lin wrote:
> PR92760 exposed one issue that VECTOR_UNIT_NONE_P (V2DImode) is true on Power7
> then we won't return it as preferred_simd_mode but ISA 2.06 (Power7) does
> introduce partial support on vector doubleword (very limitted) and more bas
On Mon, Nov 25, 2019 at 06:49:49PM -0600, Segher Boessenkool wrote:
> On Mon, Nov 25, 2019 at 05:17:08PM -0500, Michael Meissner wrote:
> > On Fri, Nov 22, 2019 at 06:20:52PM -0600, Segher Boessenkool wrote:
> > > > (define_split
> > > >[(set (match_operand:SI 0 "gpc_reg_operand")
> > > >
On 12/3/19 1:45 PM, Wilco Dijkstra wrote:
Hi,
Part 2, split off from
https://gcc.gnu.org/ml/gcc-patches/2019-11/msg00399.html
To enable cores to use the correct max_cond_insns setting, use the
core-specific
tuning when a CPU/tune is selected unless -mrestrict-it is explicitly set.
On Cor
On Tue, Dec 03, 2019 at 02:07:02AM -0500, Jason Merrill wrote:
> On 12/2/19 5:09 PM, Marek Polacek wrote:
> > On Mon, Dec 02, 2019 at 12:09:17PM -0500, Jason Merrill wrote:
> > > On 12/1/19 8:09 PM, Marek Polacek wrote:
> > > > +|| (skip_empty
> > > > +&& is_really_empty
On 12/3/19 4:16 PM, Thomas Schwinge wrote:
On 2019-11-15T20:11:29+0100, Tobias Burnus wrote:
* Make no_create.c effective by adding 'has_firstprivate = true;' to
target.c.*
(* If one tries to access c or e in the no_create-3.{c,f90} run-time
test case, plugin-nvidia rightly complains (illegal m
On Tue, Dec 03, 2019 at 11:52:13AM -0500, David Malcolm wrote:
> > > Our plugin "interface" as such is very broad.
> >
> > Just to sneak in here I don't like exposing our current plugin "non-
> > API"
> > more. In fact I'd just build the analyzer into GCC with maybe an
> > option to disable its b
On Wed, 2019-11-20 at 11:18 +0100, Richard Biener wrote:
> On Tue, Nov 19, 2019 at 11:02 PM David Malcolm
> wrote:
> > > > The checker is implemented as a GCC plugin.
> > > >
> > > > The patch kit adds support for "in-tree" plugins i.e. GCC
> > > > plugins
> > > > that
> > > > would live in the G
The problem here is that one gets two symbols - one inside the block and
one outside and they do not really agree whether one has a function or a
variable – which later gives an ICE. As sym->module was "(intrinsic)"
and FL_VARIABLE, one was running into an assert.
The problem is that when reso
I've just committed the attached.
The gcn vcondu patterns had accidentally omitted the proper variants for
floating point modes. I believe this is intentional for the comparison
mode, but not for the data mode.
This omission caused testcase gcc.dg/vect/pr65947-10.c to ICE. That
testcase no
On 12/3/19 9:38 AM, Marek Polacek wrote:
On Tue, Dec 03, 2019 at 02:01:24AM -0500, Jason Merrill wrote:
On 12/2/19 7:31 PM, Marek Polacek wrote:
@@ -1967,8 +1978,23 @@ expand_default_init (tree binfo, tree true_exp, tree
exp, tree init, int flags,
tree ctor_name = (true_exp == exp
On Tue, 3 Dec 2019 15:06:48 +0100
Thomas Schwinge wrote:
> We probably can't/shouldn't change 'amdgcn' in the target triplet now,
> but as far as I'm concerned, it's not too late to change
> 'gcc/config/gcn' etc., but I guess that won't happen: too much
> effort. (And then, I don't feel too stro
Hi Thomas,
Quick version: The attached patch seems to work, kind of, but fails at
run time with:
libgomp: Trying to map into device [0x407218..0x40721c) object when
[0x407210..0x40721c) is already mapped
This marks the common-block decl but not the common-block vars as
'device resident' (alia
Hi!
On Fri, Nov 15, 2019 at 01:28:17PM +0100, Jakub Jelinek wrote:
> On Thu, Nov 14, 2019 at 08:34:26PM +0100, Jakub Jelinek wrote:
> > The following WIP patch implements __builtin_source_location (),
> > which returns const void pointer to a std::source_location::__impl
> > struct that is require
Hi!
Jakub, please note question below.
On 2019-11-15T20:11:29+0100, Tobias Burnus wrote:
> updated version. Changes:
> * Incorporate Thomas's changes
> * Add no_create clause to newly added 'acc serial'
> * Renamed (G)OMP_MAP_NO_ALLOC to (G)OMP_MAP_IF_PRESENT as proposed
> * Make no_create.c eff
Hi Richard,
> But what uses CMP_BRANCH after the patch? It looked like you renamed
> all existing uses and didn't add any new ones.
My next patch will be adding uses of it now I've done some benchmarking
to decide when to turn it on.
>> +Â && reg_referenced_p (SET_DEST (curr_set), PATTERN (
Regarding the acc_device_, I want to observe that there is no
fundamental reason that one cannot have multiple names which resolve to
the same constant.
Thus, one could add acc_device_radeon while keeping acc_device_gcn.
Whether this makes sense or causes even more confusion is another questi
Hi Thomas,
On 03.12.19 13:14, Thomas Schwinge wrote:
> You once had this patch separate, but then merged into the upstream
> submission of 'acc_get_property'; let's please keep this separate.
>
> With changes as indicated below, please commit this to trunk [...]
Ok, I have committed the patch as
On Mon, Dec 2, 2019 at 5:33 PM Richard Sandiford
wrote:
>
> Even EXACT_DIV_EXPR doesn't distribute across addition for wrapping
> types, so in general we can't fold EXACT_DIV_EXPRs of POLY_INT_CSTs
> at compile time. This was causing an ICE when trying to gimplify the
> element size field in an A
On Mon, Dec 2, 2019 at 5:30 PM Richard Sandiford
wrote:
>
> [finally getting back to this]
>
> Richard Biener writes:
> > On Fri, Nov 8, 2019 at 10:40 AM Richard Sandiford
> > wrote:
> >>
> >> If SVE code is written for a specific vector length, it might load from
> >> or store to fixed-sized ob
The PR shows I added the bail-out prematurely.
Bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2019-12-03 Richard Biener
PR tree-optimization/92758
* tree-ssa-forwprop.c (simplify_vector_constructor): Restore
operation on uniform vectors.
Ind
Hi!
On 2019-12-03T14:20:13+, Julian Brown wrote:
> On Tue, 3 Dec 2019 10:32:57 +0100
> Thomas Schwinge wrote:
>> On 2019-12-02T14:50:42+, Julian Brown
>> wrote:
>> > On Mon, 2 Dec 2019 15:43:29 +0100
>> > Thomas Schwinge wrote:
>> >
>> >> > --- a/libgomp/openacc.h
>> >> > +++ b/libgo
On Tue, 3 Dec 2019, Jan Hubicka wrote:
> Hi,
> input_functions ends with building virtual SSA which is unnecesary
> excercise when function is only loaded to be inlined, compared by
> ipa-icf or its profile merged.
>
> This patch moves the SSA update later before we start working on the
> functio
On Tue, Dec 03, 2019 at 02:01:24AM -0500, Jason Merrill wrote:
> On 12/2/19 7:31 PM, Marek Polacek wrote:
> > @@ -1967,8 +1978,23 @@ expand_default_init (tree binfo, tree true_exp, tree
> > exp, tree init, int flags,
> > tree ctor_name = (true_exp == exp
> > ? complete_
On Tue, 3 Dec 2019 10:32:57 +0100
Thomas Schwinge wrote:
> Hi!
>
> On 2019-12-02T14:50:42+, Julian Brown
> wrote:
> > On Mon, 2 Dec 2019 15:43:29 +0100
> > Thomas Schwinge wrote:
> >
> >> > --- a/libgomp/openacc.h
> >> > +++ b/libgomp/openacc.h
> >> > @@ -55,6 +55,7 @@ typedef enum acc_d
Hi,
input_functions ends with building virtual SSA which is unnecesary
excercise when function is only loaded to be inlined, compared by
ipa-icf or its profile merged.
This patch moves the SSA update later before we start working on the
function body.
lto-bootstrapped/regtested x86_64-linux, OK?
Hi!
On 2019-12-03T13:13:33+, Andrew Stubbs wrote:
> On 02/12/2019 14:43, Thomas Schwinge wrote:
>> On 2019-11-12T13:29:13+, Andrew Stubbs wrote:
>>> --- a/include/gomp-constants.h
>>> +++ b/include/gomp-constants.h
>>> @@ -174,6 +174,7 @@ enum gomp_map_kind
>>> #define GOMP_DEVICE_NVID
Hi,
Part 2, split off from https://gcc.gnu.org/ml/gcc-patches/2019-11/msg00399.html
To enable cores to use the correct max_cond_insns setting, use the core-specific
tuning when a CPU/tune is selected unless -mrestrict-it is explicitly set.
On Cortex-A57 this gives 1.1% performance gain on SPECIN
On Mon, 2019-11-25 at 16:47 -0600, Segher Boessenkool wrote:
>
> > > - sh (that's sh4-linux):
> > >
> > > /home/segher/src/kernel/net/ipv4/af_inet.c: In function
> > > 'snmp_get_cpu_field':
> > > /home/segher/src/kernel/net/ipv4/af_inet.c:1638:1: error: unable to find
> > > a register to spill
Hi!
On 2019-12-03T12:56:49+, Andrew Stubbs wrote:
> On 02/12/2019 14:19, Thomas Schwinge wrote:
>> Generally, I'm in favor if you'd consider such a thing (that in principle
>> is just a copy/adapt of the existing cases) as obvious to commit (even
>> more so with your "amdgcn port" maintainer
Hi Kyrill,
> Hmm, I'm not too confident on that. I'd support such a change for the
> generic arm_cortex_tune, definitely, and the Armv8-a based ones, but I
> don't think the argument is as strong for Cortex-A7, Cortex-A8, Cortex-A9.
>
> So let's make the change for the Armv8-A-based cores now. I
On 02/12/2019 14:43, Thomas Schwinge wrote:
Hi!
On 2019-11-12T13:29:13+, Andrew Stubbs wrote:
--- a/include/gomp-constants.h
+++ b/include/gomp-constants.h
@@ -174,6 +174,7 @@ enum gomp_map_kind
#define GOMP_DEVICE_NVIDIA_PTX5
#define GOMP_DEVICE_INTEL_MIC 6
#
On 02/12/2019 14:19, Thomas Schwinge wrote:
Generally, I'm in favor if you'd consider such a thing (that in principle
is just a copy/adapt of the existing cases) as obvious to commit (even
more so with your "amdgcn port" maintainer hat on), especially so given
that this has been/is blocking you,
Jason Merrill writes:
> On 11/29/19 5:59 AM, Richard Sandiford wrote:
>> Ping
>>
>> Richard Sandiford writes:
>>> This is the C++ equivalent of r277950, which prevented the
>>> use of the GNU vector extensions with SVE vector types for C.
>>> [https://gcc.gnu.org/viewcvs/gcc?view=revision&revisi
Hi Frederik!
You once had this patch separate, but then merged into the upstream
submission of 'acc_get_property'; let's please keep this separate.
With changes as indicated below, please commit this to trunk (without the
three hunks related to 'acc_get_property', of course; these will then go
in
On Tue, 3 Dec 2019, Richard Biener wrote:
> On Mon, 2 Dec 2019, Richard Biener wrote:
>
> > On December 2, 2019 4:27:47 PM GMT+01:00, Alexander Monakov
> > wrote:
> > >On Mon, 2 Dec 2019, Richard Biener wrote:
> > >
> > >> +typedef long long v4di __attribute__((vector_size(32)));
> > >> +struct
Hi Jakub,
On 08.11.19 07:41, Harwath, Frederik wrote:
> On 06.11.19 14:00, Jakub Jelinek wrote:
> [...]
>> I'm not sure it is a good idea to use a TREE_LIST in this case, vec would be
>> more natural, wouldn't it.
>
> Yes.
>
> [...]
>> If gimplifier is not the right spot, then use a splay tree +
On Fri, Nov 15, 2019 at 09:55:37AM +, Szabolcs Nagy wrote:
> Musl does not support initial-exec tls in dynamically loaded shared
> libraries.
>
> libgomp/ChangeLog:
>
> 2019-11-15 Szabolcs Nagy
>
> * configure.tgt: Avoid IE tls on *-*-musl*.
>
> libitm/ChangeLog:
>
> 2019-11-15 S
On 20/11/2019 14:42, Szabolcs Nagy wrote:
> On 15/11/2019 09:55, Szabolcs Nagy wrote:
>> Musl does not support initial-exec tls in dynamically loaded shared
>> libraries.
>
> ping.
ping.
>>
>> libgomp/ChangeLog:
>>
>> 2019-11-15 Szabolcs Nagy
>>
>> * configure.tgt: Avoid IE tls on *-*-mu
On 12/2/19 9:27 PM, Joseph Myers wrote:
> On Mon, 2 Dec 2019, Jeff Law wrote:
>
>>> 2019-11-13 Stam Markianos-Wright
>>>
>>> * real.c (struct arm_bfloat_half_format,
>>> encode_arm_bfloat_half, decode_arm_bfloat_half): New.
>>> * real.h (arm_bfloat_half_format): New.
>
This streamlines the encoding used in the debug info for record types with
components of variable size.
Tested on x86_64-suse-linux, applied on the mainline.
2019-12-03 Eric Botcazou
* gcc-interface/utils.c (potential_alignment_gap): Delete.
(rest_of_record_type_compilation)
On 12/2/19 7:15 PM, Matthias Klose wrote:
GCC 10 comes with a new lto-dump texi file, but the man page isn't built and
installed. Fix with the attached patch. Ok to install?
Matthias
Hello.
Thank you for the patch. I would consider it as obvious.
Martin
This fixes a bug in the encoding used in the debug info generated for record
types with variant part for targets where the alignment is capped to 16 bits
like the m68k (unless -malign-int is specified).
Tested on m68k-elf and x86_64-suse-linux, applied on the mainline.
2019-12-03 Eric Botcazo
This adjusts the return part of the mechanism used to pass In Out or Out
parameters on 64-bit targets to avoid generating problematic paradoxical
subregs with floating-point mode.
Tested on x86_64-suse-linux, applied on the mainline.
2019-12-03 Eric Botcazou
* gcc-interface/decl.c
On Tue, 3 Dec 2019, Richard Sandiford wrote:
> Jakub Jelinek writes:
> > Hi!
> >
> > As discussed in the PR, we can't optimize e.g.
> > int a = t - 1;
> > int b = a * v;
> > return b + v;
> > into return t * v; for signed non-wrapv arithmetics. This can be done
> > by the match.pd (A * B)
On 02/12/19 16:23 +, Jonathan Wakely wrote:
On 15/10/19 18:57 +0100, Mike Crowe wrote:
glibc v2.30 added the pthread_mutex_clocklock,
pthread_rwlock_clockrdlock and pthread_rwlock_clockwrlock
functions. These accept CLOCK_MONOTONIC, so they can be used to
implement proper steady_clock suppor
Jakub Jelinek writes:
> Hi!
>
> As discussed in the PR, we can't optimize e.g.
> int a = t - 1;
> int b = a * v;
> return b + v;
> into return t * v; for signed non-wrapv arithmetics. This can be done
> by the match.pd (A * B) +- A -> (B +- 1) * A or
> A +- (A * B) -> (1 +- B) * A canonical
Hi,
PR92760 exposed one issue that VECTOR_UNIT_NONE_P (V2DImode) is true on Power7
then we won't return it as preferred_simd_mode but ISA 2.06 (Power7) does
introduce partial support on vector doubleword (very limitted) and more basic
support origins from ISA 2.07 (Power8) though. To make vector
Hi!
On 2019-12-02T14:50:42+, Julian Brown wrote:
> On Mon, 2 Dec 2019 15:43:29 +0100
> Thomas Schwinge wrote:
>
>> > --- a/libgomp/openacc.h
>> > +++ b/libgomp/openacc.h
>> > @@ -55,6 +55,7 @@ typedef enum acc_device_t {
>> >/* acc_device_host_nonshm = 3 removed. */
>> >acc_device_n
On 03/12/19 09:11 +0100, Christophe Lyon wrote:
On Mon, 16 Sep 2019 at 04:34, Tom Honermann wrote:
A revised patch is attached that modifies the tests for deleted ostream
inserters to require C++2a. This is required by the revision of patch
2/4 that adds proper preprocessor conditionals to th
Hi Tobias!
On 2019-11-29T18:47:12+0100, Tobias Burnus wrote:
> On 11/28/19 6:02 PM, Thomas Schwinge wrote:
> [Test case which uses common blocks in device_resident.]
>> If you'd like to, please commit that, to document the status quo. (I
>> have not reviewed.)
>
> Did so as r278845 with a slight
The following fixes value-numbering being confused by the CLOBBERs
that appear in Skia vector constructors like
:
v ={v} {CLOBBER};
MEM[(struct Vec *)&v] ={v} {CLOBBER};
MEM[(struct Vec *)&v] ={v} {CLOBBER};
MEM[(struct Vec *)&v].val = 1;
MEM[(struct Vec *)&v + 4B] ={v} {CLOBBER};
On Tue, 3 Dec 2019, Jakub Jelinek wrote:
> Hi!
>
> The following patch extends the improvements Marc did to the
> (A +- CST1) +- CST2 -> A +- CST3
> match.pd simplification some time ago to the other two patterns,
> in particular handle the case when the inner subtraction is done in a
> different
I have now installed https://gcc.gnu.org/gcc-10/porting_to.html – which
is linked from https://gcc.gnu.org/gcc-10/changes.html (first paragraph
is now uncommented).
Comments? Omissions, additional suggestions for either page?
Cheers,
Tobias
On 12/2/19 1:48 PM, Tobias Burnus wrote:
Revised ve
Hi!
The following patch extends the improvements Marc did to the
(A +- CST1) +- CST2 -> A +- CST3
match.pd simplification some time ago to the other two patterns,
in particular handle the case when the inner subtraction is done in a
different, but nop_convert compatible, type from the outer +/-.
Hi!
The following testcase shows that during constexpr evaluation we didn't
handle TARGET_EXPR_CLEANUP at all (which was probably fine when there
weren't constexpr dtors). My understanding is that TARGET_EXPR cleanups
should be queued and evaluated only at the end of CLEANUP_POINT_EXPR, so
that i
On Mon, 16 Sep 2019 at 04:34, Tom Honermann wrote:
>
> A revised patch is attached that modifies the tests for deleted ostream
> inserters to require C++2a. This is required by the revision of patch
> 2/4 that adds proper preprocessor conditionals to the definitions.
>
> Tom.
>
> On 9/15/19 3:40
82 matches
Mail list logo
