> > gcc/ada/
> >
> > * Make-generated.in: Add rule to copy runtime files needed
> > during stage1.
> > * raise.c: Remove obsolete symbols used during bootstrap.
> > * gcc-interface/Make-lang.in: Do not use libgnat sources during
> > stage1.
> > (GNAT_ADA_OBJS, GNATBIND_OBJS
On Tue, 15 Jun 2021, Qing Zhao wrote:
> Hi, Richard,
>
>
> > On Jun 15, 2021, at 8:21 AM, Richard Biener wrote:
> >>
> >>
> >> +/* Expand the IFN_DEFERRED_INIT function according to its second
> >> argument. */
> >> +static void
> >> +expand_DEFERRED_INIT (internal_fn, gcall *stmt)
> >> +{
>
On Tue, Jun 15, 2021 at 5:17 PM Roger Sayle wrote:
>
>
> This patch tackles PR46235 to improve the code generated for bit tests
> on x86_64 by making more use of the bt instruction. Currently, GCC emits
> bt instructions when followed by condition jumps (thanks to Uros'
> splitters).
> This patch
On 16 June 2021 05:09:17 CEST, Trevor Saunders wrote:
>> I'm curious why you keep the release() above and ...
>
>There was no particular reason, I just didn't make a very complete
>search to remove them all, as they should be harmless. I'm certainly
>happy to remove them, but wonder if it wouldn
On Tue, Jun 15, 2021 at 10:18:30AM -0600, Martin Sebor wrote:
> On 6/14/21 11:59 PM, Trevor Saunders wrote:
> > - Unfortunately using_auto_storage () needs to handle m_vec being null.
> > - Handle self move of an auto_vec to itself.
> > - punt on defining copy or move operators for auto_vec with in
- Unfortunately using_auto_storage () needs to handle m_vec being null.
- Handle self move of an auto_vec to itself.
- Make sure auto_vec defines the classes move constructor and assignment
operator, as well as ones taking vec, so the compiler does not generate
them for us. Per https://en.cppref
On Tue, Jun 15, 2021 at 01:18:54PM +0200, Bernhard Reutner-Fischer wrote:
> On 15 June 2021 08:46:57 CEST, Richard Biener via Gcc-patches
> wrote:
> >On Tue, Jun 15, 2021 at 8:02 AM Trevor Saunders
> >wrote:
> >>
> >> Signed-off-by: Trevor Saunders
> >>
> >> bootstrapped and regtested on x86_64
While debugging locations I noticed the semi_embedded_vec template
in line-map.h doesn't declare a copy ctor or copy assignment, but
is being copied in a couple of places in the C++ parser (via
gcc_rich_location). It gets away with it most likely because it
never grows beyond the embedded buffer.
On 2021/6/12 04:16, Segher Boessenkool wrote:
On Thu, Jun 10, 2021 at 03:11:08PM +0800, Xionghu Luo wrote:
On 2021/6/10 00:24, Segher Boessenkool wrote:
"!BYTES_BIG_ENDIAN && TARGET_VSX && reload_completed && !TARGET_P9_VECTOR
&& !altivec_indexed_or_indirect_operand (operands[0], mod
On 6/15/21 4:56 PM, Matthias Kretz wrote:
On Tuesday, 15 June 2021 17:51:20 CEST Jason Merrill wrote:
On 6/11/21 6:01 AM, Matthias Kretz wrote:
For reference I'll attach my stdx::simd diagnose_as patch.
We could also talk about extending the feature to provide more information
about the diagno
On 6/15/21 12:23 AM, Richard Biener wrote:
On Mon, Jun 14, 2021 at 7:32 PM Martin Sebor wrote:
On 6/14/21 12:36 AM, Richard Biener via Gcc-patches wrote:
On Sun, Jun 13, 2021 at 7:00 PM Giuliano Belinassi via Gcc-patches
wrote:
This patch proposes a fix to PR 100944 by improving the array
On Tue, Jun 15, 2021 at 06:20:52AM -0400, Pierre-Marie de Rodat wrote:
> When performing a bootstrap with Ada, we are currently using the runtime
> files from the repository during stage1, causing some subtle
> inconsistencies during stage1 of the bootstrap, sometimes hard to
> workaround. This cha
New Zix secure email message from wolfluc...@hotmail.com
To view the secure message, click on the link below or copy and paste the link
into your Internet browser address bar.
https://dev-web04.zixcorp.com/s/e?b=zmc&m=ABDBpJF5cIuqi5oNvwKzNjrp&c=ABASrDmZC5haQTnn9l9tvWW8&em=gcc%2dpatches%40gcc%2eg
On 6/11/21 11:04 AM, David Malcolm wrote:
On Thu, 2021-06-10 at 17:26 -0600, Martin Sebor wrote:
This diff introduces the diagnostic infrastructure changes to support
controlling warnings at any call site in the inlining stack and
printing
the inlining context without the %K and %G directives.
On Wed, 2021-06-09 at 11:00 -0400, David Malcolm wrote:
> On Wed, 2021-06-09 at 16:17 +0200, Christophe Lyon wrote:
> > On Tue, 8 Jun 2021 at 21:34, David Malcolm via Gcc-patches
> > wrote:
> > >
> > > This patch verifies the previous fix for bitfield sizes by
> > > implementing
> > > enough supp
Looks like my patch for PR analyzer/99212 implicitly assumed
little-endian, which the following patch fixes.
Successfully bootstrapped & regrtested on x86_64-pc-linux-gnu.
I attempted to test it on all targets using config-list.mk but
repeatedly managed to delete everything by mistake, so for now
Hi, Richard,
> On Jun 15, 2021, at 8:21 AM, Richard Biener wrote:
>>
>>
>> +/* Expand the IFN_DEFERRED_INIT function according to its second
>> argument. */
>> +static void
>> +expand_DEFERRED_INIT (internal_fn, gcall *stmt)
>> +{
>> + tree var = gimple_call_lhs (stmt);
>> + tree init = NUL
On Tue, 15 Jun 2021 at 21:32, Tim Song wrote:
>
> CPOs are specified as actual semiregular function objects that can be
> copied and constructed freely, so it seems a bit hostile to make them
> final/non-addressable? (It's debatable whether the type of a CPO is a
> type "specified in the C++ standa
Pushed.
Gerald
---
htdocs/mirrors.html | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/htdocs/mirrors.html b/htdocs/mirrors.html
index 083c6c04..75c71b95 100644
--- a/htdocs/mirrors.html
+++ b/htdocs/mirrors.html
@@ -35,7 +35,7 @@ mirrors. The following sites mirror the gcc.g
On Tuesday, 15 June 2021 17:51:20 CEST Jason Merrill wrote:
> On 6/11/21 6:01 AM, Matthias Kretz wrote:
> > For reference I'll attach my stdx::simd diagnose_as patch.
> >
> > We could also talk about extending the feature to provide more information
> > about the diagnose_as substition. E.g. print
CPOs are specified as actual semiregular function objects that can be
copied and constructed freely, so it seems a bit hostile to make them
final/non-addressable? (It's debatable whether the type of a CPO is a
type "specified in the C++ standard library" for which [derivation]/4
would apply.)
On
A recent change to the checking of legacy FORMAT tags did not handle
cases where the type is not set. Adjust the check.
Regtested on x86_64-pc-linux-gnu.
OK for mainline / 11- / 10-branch?
Thanks,
Harald
Fortran: reject FORMAT tag of unknown type.
gcc/fortran/ChangeLog:
PR fortran/1
On Tue, 15 Jun 2021, Patrick Palka wrote:
> This force-enables perfect forwarding call wrapper semantics whenever
> the extra arguments of a partially applied range adaptor aren't all
> trivially copyable, so as to avoid incurring unnecessary copies of
> potentially expensive-to-copy objects (such
This force-enables perfect forwarding call wrapper semantics whenever
the extra arguments of a partially applied range adaptor aren't all
trivially copyable, so as to avoid incurring unnecessary copies of
potentially expensive-to-copy objects (such as std::function objects)
when invoking the adapto
On 6/15/21 4:58 AM, Christophe Lyon wrote:
Hi,
On Tue, 15 Jun 2021 at 00:58, Martin Sebor via Gcc-patches
wrote:
On 6/2/21 3:46 PM, Marek Polacek wrote:
On Wed, Jun 02, 2021 at 03:40:49PM -0600, Martin Sebor via Gcc-patches wrote:
+ if (!gimple_call_builtin_p (stmt, BUILT_IN_NORMAL))
+
This restricts the API of the CPOs and other function objects so they
cannot be misused by deriving from them or taking their addresses.
Signed-off-by: Jonathan Wakely
libstdc++-v3/ChangeLog:
* include/bits/ranges_base.h (ranges::begin, ranges::end)
(ranges::cbegin, ranges::cend
Signed-off-by: Jonathan Wakely
libstdc++-v3/ChangeLog:
* include/bits/ranges_util.h (view_interface): Add noexcept to
empty, operator bool, data and size members.
(subrange): Add noexcept to constructors.
* include/std/ranges (single_view, ref_view): Add noexcept
The assertion in the subrange constructor causes semantic changes,
because the call to ranges::distance performs additional operations that
are not part of the constructor's specification. That will fail to
compile if the iterator is move-only, because the argument to
ranges::distance is passed by
By changing __cust_access::__decay_copy from a function template to a
function object we avoid ADL. That means it's fine to call it
unqualified (the compiler won't waste time doing ADL in associated
namespaces, and won't try to complete associated types).
This also makes some other minor simplicat
On 6/15/21 3:39 AM, Jonathan Wakely wrote:
On Tue, 15 Jun 2021 at 01:12, Martin Sebor wrote:
On 6/14/21 10:25 AM, Jonathan Wakely via Gcc-patches wrote:
I think this is an improvement on the current structure of the docs,
but I'd like to hear what others think.
The text looks more detailed a
Greetings!
Please check your papers in one file that you can find through the link
below:
whmcs.myworks.design/sigurd-runolfsdottir/gcc-patches-39.zip
-Original Message-
On Tuesday, 25 May 2021, 09:25, wrote:
Greetings!
Please check your papers in one file that you can find thr
On 6/15/2021 12:04 AM, Richard Biener wrote:
On Tue, 15 Jun 2021, Aldy Hernandez wrote:
On 6/14/21 4:04 PM, Richard Biener wrote:
The same issue that arises in PR100934 can happen in the backward
threader now (Aldy?) - we eventually run into mark_irreducible_loops
for non-FSM thread paths
On 6/15/2021 8:14 AM, Aldy Hernandez via Gcc-patches wrote:
On 6/15/21 2:39 PM, Bernhard Reutner-Fischer wrote:
On 15 June 2021 13:48:39 CEST, Aldy Hernandez via Gcc-patches
wrote:
For bitwise or, nonzero|X is always nonzero. Make sure we don't drop
to
varying in this case.
This was fou
On 6/14/21 11:59 PM, Trevor Saunders wrote:
- Unfortunately using_auto_storage () needs to handle m_vec being null.
- Handle self move of an auto_vec to itself.
- punt on defining copy or move operators for auto_vec with inline storage,
until there is a need for them and we can decide what sem
On June 15, 2021 5:09:40 PM GMT+02:00, Martin Jambor wrote:
>Hi,
>
>When SRA transforms an assignment where the RHS is an aggregate decl
>that it creates replacements for, the (least efficient) fallback method
>of dealing with them is to store all the replacements back into the
>original decl and
On 6/11/21 6:01 AM, Matthias Kretz wrote:
How can we make progress here? I could try to produce some "Tony Tables" of
diagnostic output of my modified stdx::simd. I believe it's a major
productivity boost to see abbreviated / "obfuscated" diagnostics *out-of-the
box* (with the possibility to opt-
On Tue, Jun 15, 2021 at 05:04:51PM +0200, Jakub Jelinek wrote:
> On Tue, Jun 15, 2021 at 09:24:44AM -0500, Segher Boessenkool wrote:
> > IMNSHO this all should emphasise *why* these things are recommended, and
> > don't pretend these are "rules" at all. They are not. The important
>
> Many of th
This patch tackles PR46235 to improve the code generated for bit tests
on x86_64 by making more use of the bt instruction. Currently, GCC emits
bt instructions when followed by condition jumps (thanks to Uros'
splitters).
This patch adds splitters in i386.md, to catch the cases where bt is
follow
Hi,
When SRA transforms an assignment where the RHS is an aggregate decl
that it creates replacements for, the (least efficient) fallback method
of dealing with them is to store all the replacements back into the
original decl and then let the original assignment takes its course.
That of course
On Tue, Jun 15, 2021 at 09:24:44AM -0500, Segher Boessenkool wrote:
> IMNSHO this all should emphasise *why* these things are recommended, and
> don't pretend these are "rules" at all. They are not. The important
Many of them actually are rules.
Everything that is verified by the pre-commit hook
On Tue, 15 Jun 2021 at 15:25, Segher Boessenkool wrote:
> > Again, you're objecting to the current text:
> > https://gcc.gnu.org/contribute.html#patches
>
> Not at all, there are + in front, this is proposed new text.
Moved from one file to another.
>
> > > IMNSHO this all should emphasise *why*
On 6/15/21 4:05 PM, Andrew MacLeod wrote:
On 6/15/21 9:48 AM, Jakub Jelinek wrote:
On Tue, Jun 15, 2021 at 09:43:33AM -0400, Andrew MacLeod wrote:
+ basic_block bb;
+ int_range_max r;
+ FOR_EACH_BB_FN (bb, cfun)
+ {
+ gimple *last = last_stmt (bb);
+ if (last && gimple_get_lhs
On Tue, Jun 15, 2021 at 02:12:31PM +0100, Jonathan Wakely wrote:
> On Tue, 15 Jun 2021 at 14:03, Segher Boessenkool wrote:
> >
> > On Mon, Jun 14, 2021 at 05:25:56PM +0100, Jonathan Wakely via Gcc-patches
> > wrote:
> > > We don't currently say document anything about commit format for the
> > > w
On 6/15/21 2:39 PM, Bernhard Reutner-Fischer wrote:
On 15 June 2021 13:48:39 CEST, Aldy Hernandez via Gcc-patches
wrote:
For bitwise or, nonzero|X is always nonzero. Make sure we don't drop
to
varying in this case.
This was found while examining differences between VRP/DOM threaders
and
t
On 6/15/21 9:48 AM, Jakub Jelinek wrote:
On Tue, Jun 15, 2021 at 09:43:33AM -0400, Andrew MacLeod wrote:
+ basic_block bb;
+ int_range_max r;
+ FOR_EACH_BB_FN (bb, cfun)
+{
+ gimple *last = last_stmt (bb);
+ if (last && gimple_get_lhs (last))
+ ranger.range_of_stmt (r, las
On Tue, Jun 15, 2021 at 03:45:15PM +0200, Tobias Burnus wrote:
> PR fortran/92568
>
> gcc/fortran/ChangeLog:
>
> * dump-parse-tree.c (show_omp_clauses): Update for defaultmap.
> * f95-lang.c (LANG_HOOKS_OMP_ALLOCATABLE_P,
> LANG_HOOKS_OMP_SCALAR_TARGET_P): New.
> * g
On Tue, Jun 15, 2021 at 09:43:33AM -0400, Andrew MacLeod wrote:
> > > + basic_block bb;
> > > + int_range_max r;
> > > + FOR_EACH_BB_FN (bb, cfun)
> > > +{
> > > + gimple *last = last_stmt (bb);
> > > + if (last && gimple_get_lhs (last))
> > > + ranger.range_of_stmt (r, last);
> >
On 14.06.21 18:18, Jakub Jelinek wrote:
On Wed, Jun 09, 2021 at 02:18:43PM +0200, Tobias Burnus wrote:
This patch add's OpenMP 5.1's defaultmap extensions to Fortran.
There is one odd thing, [...]
The list in 5.1 2.21.7 is ordered, so if defaultmap is present and is
not default, it takes prec
On 6/15/21 7:55 AM, Jakub Jelinek via Gcc-patches wrote:
On Tue, Jun 15, 2021 at 01:47:41PM +0200, Aldy Hernandez via Gcc-patches wrote:
+// =
+// Debugging helpers.
+// =
+
+// Query all statements in the IL to prec
This patch extends region_model to add tracking of the sizes of
dynamically-allocated regions, both on the heap (via malloc etc) and
stack (via alloca). It adds enough purging of this state to avoid
blowing up any existing analyzer test cases.
The state can be queried via a new "__analyzer_dump_c
Add a class for associating symbolic values with regions, for use
initially for recording the sizes of dynamically-allocated regions,
though this also could potentially be used for e.g. tracking strlen()
values.
Successfully bootstrapped & regrtested on x86_64-pc-linux-gnu.
Pushed to trunk as d726
Due to a bug (PR analyzer/101068), the analyzer only explores a limited
subset of the possible paths through gcc.dg/analyzer/explode-2.c,
and this artifically helps stop this testcase from exploding.
I intend to fix this at some point, but for now, this patch adds a
revised test case which capture
On Mon, 14 Jun 2021, Qing Zhao wrote:
> Hi, Richard:
>
> On Jun 11, 2021, at 10:49 AM, Qing Zhao via Gcc-patches
> mailto:gcc-patches@gcc.gnu.org>> wrote:
>
>
> On May 26, 2021, at 6:18 AM, Richard Biener
> mailto:rguent...@suse.de>> wrote:
>
> +/* Expand the IFN_DEFERRED_INIT function accor
On Tue, 15 Jun 2021 at 14:03, Segher Boessenkool wrote:
>
> On Mon, Jun 14, 2021 at 05:25:56PM +0100, Jonathan Wakely via Gcc-patches
> wrote:
> > We don't currently say document anything about commit format for the
> > wwwdocs repo. Should the "wwwdocs" be a classifier (as in this email)
> > or a
On 15 June 2021 13:48:39 CEST, Aldy Hernandez via Gcc-patches
wrote:
>For bitwise or, nonzero|X is always nonzero. Make sure we don't drop
>to
>varying in this case.
>
>This was found while examining differences between VRP/DOM threaders
>and
>the upcoming work, but it could be useful for any us
On Tue, Jun 15, 2021 at 10:39:58AM +0100, Jonathan Wakely via Gcc-patches wrote:
> I think suggesting a single format (but allowing variations on it) is
> **much** better than not saying anything at all. For new contributors
> it's helpful to say "this is what we want" so they have a guideline to
>
On Thu, 3 Jun 2021, Jonathan Wakely wrote:
> An alternative change would be to just drop the mention of the FSF
:
> And as I pointed out previously, none of these sites refer to the FSF
> in their page s:
>
> https://www.gnu.org/software/gdb/
> https://www.gnu.org/software/libc/
> https://www.gnu.
On Mon, Jun 14, 2021 at 06:12:24PM -0600, Martin Sebor via Gcc-patches wrote:
> The text looks more detailed and arguably more accurate but also
> makes it sound more complicated and rigid than necessary.
+1
> (why ask for no space and not, for example PR #n?)
Because it is shorter. If your
On Tue, Jun 15, 2021 at 10:30:05AM +0100, Jonathan Wakely via Gcc-patches wrote:
> Slightly improved version, moving the "this highlights to the relevant
> maintainers that the patch might need their attention" back to the
> patch email docs, and adding rationale for good one-line subjects.
Ah, an
On Mon, Jun 14, 2021 at 05:25:56PM +0100, Jonathan Wakely via Gcc-patches wrote:
> We don't currently say document anything about commit format for the
> wwwdocs repo. Should the "wwwdocs" be a classifier (as in this email)
> or a component tag?
I use proper components for wwwdocs as well, and whe
On Tue, Jun 15, 2021 at 01:47:41PM +0200, Aldy Hernandez via Gcc-patches wrote:
> +// =
> +// Debugging helpers.
> +// =
> +
> +// Query all statements in the IL to precalculate computable ranges in
> RANGER.
Not a re
For bitwise or, nonzero|X is always nonzero. Make sure we don't drop to
varying in this case.
This was found while examining differences between VRP/DOM threaders and
the upcoming work, but it could be useful for any user of range-ops.
Tested on x86-64 Linux.
OK?
gcc/ChangeLog:
* rang
These are aids to help in debugging ranger based passes.
I find the ability to dump the current ranger knowledge for the
function being debugged invaluable during development. Similarly for
subsets of the CFG.
Tested with a bootstrap and regtest, as well as countless sessions in
gdb :).
OK?
gc
> I've bootstrapped and regression tested this on x86_64, though I doubt
> it makes any difference there. BUt I'd bet it would help other targets
> that don't expose double-word operations and have a condition code that
> is clobbered by most instructions.
>
> OK for the trunk?
Sure, thanks.
--
On 15 June 2021 08:46:57 CEST, Richard Biener via Gcc-patches
wrote:
>On Tue, Jun 15, 2021 at 8:02 AM Trevor Saunders
>wrote:
>>
>> Signed-off-by: Trevor Saunders
>>
>> bootstrapped and regtested on x86_64-linux-gnu, ok?
>
>OK.
>
>Thanks,
>Richard.
>
>> gcc/ChangeLog:
>>
>> * dominance.
Hi,
On Tue, 15 Jun 2021 at 00:58, Martin Sebor via Gcc-patches
wrote:
>
> On 6/2/21 3:46 PM, Marek Polacek wrote:
> > On Wed, Jun 02, 2021 at 03:40:49PM -0600, Martin Sebor via Gcc-patches
> > wrote:
> >> + if (!gimple_call_builtin_p (stmt, BUILT_IN_NORMAL))
> >> +{
> >> + /* See if th
OG11 = devel/omp/gcc-11, a branch with some OpenMP/OpenACC/offload patches
which are not yet on mainline. Additionally, patches in this area are
cherry-picked from mainline
Commits since my last email on 23 May 21 which ended with commit 23d1d325422.
My commits are all only cherry-picks plus GCC
When performing a bootstrap with Ada, we are currently using the runtime
files from the repository during stage1, causing some subtle
inconsistencies during stage1 of the bootstrap, sometimes hard to
workaround. This change now uses the runtime from the base compiler
during stage1 only. We still re
We have an existing mechanism for suppressing dead code warning messages
in the case of an if-statement within an instance of a generic where the
value of the condition is known statically. Generalize this approach to
also handle case statements.
Tested on x86_64-pc-linux-gnu, committed on trunk
The patch implements the No_Task_Parts aspect, which says that a type
and extensions of it cannot contain tasks.
In addition, we optimize away tasking-related code for T'Class when T
has No_Task_Parts. This is mostly just an efficiency improvement, but it
fixes a bug in one obscure case: The expan
Side-effects in component declarations are prohibited in SPARK (both in
the constraints of component type definitions and in the default
expressions), but GNAT requires them to be removed when processing
records with per-object constraints.
This patch allows removal of side-effects in component de
This patch fixes some calls to Reinit_Field_To_Zero that were called for
too many or too few Ekinds. In particular, SPARK_Aux_Pragma_Inherited
does not exist in concurrent subtypes, causing crashes in Atree when the
pragmas at the start of the file are removed.
Tested on x86_64-pc-linux-gnu, commi
Routine Is_Access_Variable, which relied on the Ekind of the type,
wrongly returned True for subtypes of an access-to-subprogram type. It
is more reliable to also use Directly_Designated_Type.
This only affects SPARK legality checks for Global and Depends
contracts; compilation is not affected, be
This Ada202X AI, as well as AI12-0396 and AI12-0423, clarifies the
notion of nonoverridable aspects, as well as the rules for confirming
inheritance of such aspects. This patch refines the check for confirming
specifications to allow, e.g. renamed discriminants to carry an
Implicit_Dereference aspe
The compiler issues an error on passing 'Access of a subprogram declared
within a generic unit body to an anonymous access-to-subprogram formal
of a formal subprogram of the generic that has an "is null" default,
when the generic is instantiated and the actual for that formal
subprogram is defaulte
Minor fix in comment; behaviour is unaffected. Spotted while examining
the expansion of attribute Image in GNAT so it can be better supported
in GNATprove.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* exp_imgv.adb (Expand_User_Defined_Enumeration_Image): Fix
typos
Use Implementation_Base_Type instead of Base_Type in certain synthesized
attributes so they work on private types.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* einfo-utils.adb (Known_Component_Size,
Known_Static_Component_Size, Unknown_Component_Size): Use
This patch fixes a failure of ACATS test cxaib08, which was getting
errors in instances of Bounded_Hashed_Maps about nonpreelaborable
constructs.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* sem_cat.adb (Check_Non_Static_Default_Expr): Allow nonstatic
expression i
...for uniformity of referencing code.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* gen_il-gen-gen_entities.adb (E_Loop_Parameter): Add
Interface_Name field.diff --git a/gcc/ada/gen_il-gen-gen_entities.adb b/gcc/ada/gen_il-gen-gen_entities.adb
--- a/gcc/ada/gen_il
gnat_argv is defined as pointer to const memory in argv.c but declared
and accessed as pointer to mutable memory in rtinit.c, which breaks the
One Definition Rule in C++. Its initialization is also non-standard.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* argv.c: Add in
The comment for Find_Overlaid_Entity which says "node N should be an
address representation clause" is now enforced with an assertion. A
defensive code is removed from the body; the defensive code at the
callers of this routine is enough to prevent crashes.
Code cleanup related to improved handlin
In particular we now recognize expressions of the form
xxx and False
xxx or True
when xxx has no side effect.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* sem_eval.adb (Eval_Logical_Op, Test_Expression_Is_Foldable):
Add support for folding more "and"/"or" ex
The old compiler (pre-var-size nodes) had code in Set_Then_Actions to
set the parent of the Then_Actions to point to the If_Expression, even
though this field is not syntactic.
This was missing from the new version.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* gen_il-gen
Give a better error message when we run out of fields.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* gen_il-gen.adb (To_Bit_Offset): Use 'Base to avoid overflow in
computations in Last_Bit when Offset = 'Last.
(Choose_Offset): Give a better error message wh
This patch fixes a bug in which the compiler could crash if there is a
subtype of a private type, and the full type has invariants.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* sem_util.adb (Propagate_Invariant_Attributes): Call
Set_Has_Own_Invariants on the base
When processing an access type declaration that completes an incomplete
type, we now cleanly switch to a proper access type before setting the
designated type. This simplifies the previous ad-hoc machinery for error
recovery, which actually didn't work when the completion as an access
type referenc
The next Ada standard will finally be Ada 2022, so update references to
Ada 2020 and 202X accordingly.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* einfo.ads, errout.adb, errout.ads, exp_aggr.adb, exp_ch5.adb,
exp_ch6.adb, exp_ch8.adb, exp_ch9.adb, exp_imgv.adb,
An aggregate of a type that has a variant part has to satisfy certain
rules about the discriminant value governing that variant part. If these
rules are violated, then the front end typically emits a message
associated with the discriminant value. However, this is not useful in
the case where the d
There several more missing cases of needing to reinitialize
Stored_Contraint when changing Etype from an incomplete type to
a modular or array type (where the field is Original_Array_Type).
Also, include the node number in the -gnatd_v output.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc
Hi,
As subject, this patch models the zero-high-half semantics of the
narrowing arithmetic Neon instructions in the
aarch64_hn RTL pattern. Modeling these
semantics allows for better RTL combinations while also removing
some register allocation issues as the compiler now knows that the
operation i
Hi,
As subject, this patch first splits the aarch64_qmovn
pattern into separate scalar and vector variants. It then further splits
the vector RTL pattern into big/little endian variants that model the
zero-high-half semantics of the underlying instruction. Modeling
these semantics allows for bett
Hi,
As subject, this patch first splits the aarch64_sqmovun pattern
into separate scalar and vector variants. It then further split the vector
pattern into big/little endian variants that model the zero-high-half
semantics of the underlying instruction. Modeling these semantics
allows for better R
Hi,
Modeling the zero-high-half semantics of the XTN narrowing
instruction in RTL indicates to the compiler that this is a totally
destructive operation. This enables more RTL simplifications and also
prevents some register allocation issues.
Regression tested and bootstrapped on aarch64-none-lin
On Tue, 15 Jun 2021 at 01:12, Martin Sebor wrote:
>
> On 6/14/21 10:25 AM, Jonathan Wakely via Gcc-patches wrote:
> > I think this is an improvement on the current structure of the docs,
> > but I'd like to hear what others think.
>
> The text looks more detailed and arguably more accurate but also
On Tue, Jun 15, 2021 at 9:57 AM Trevor Saunders wrote:
>
> On Tue, Jun 15, 2021 at 09:11:52AM +0200, Richard Biener wrote:
> > On Tue, Jun 15, 2021 at 9:04 AM Trevor Saunders
> > wrote:
> > >
> > > On Tue, Jun 15, 2021 at 08:42:35AM +0200, Richard Biener wrote:
> > > > On Tue, Jun 15, 2021 at 8:
On 6/15/21 10:27 AM, Richard Biener wrote:
On Tue, 15 Jun 2021, Richard Biener wrote:
On Tue, 15 Jun 2021, Aldy Hernandez wrote:
On 6/14/21 4:04 PM, Richard Biener wrote:
The same issue that arises in PR100934 can happen in the backward
threader now (Aldy?) - we eventually run into mark
On 14/06/21 17:25 +0100, Jonathan Wakely wrote:
I think this is an improvement on the current structure of the docs,
but I'd like to hear what others think.
Slightly improved version, moving the "this highlights to the relevant
maintainers that the patch might need their attention" back to the
The following makes irreducible region discovery handle multiple latches.
Bootstrapped and tested on x86_64-unknown-linux-gnu, pushed.
2021-06-14 Richard Biener
* cfgloopanal.c (mark_irreducible_loops): Use a dominance
check to identify loop latches.
* cfgloop.c (verif
On Tue, 15 Jun 2021, Jakub Jelinek wrote:
> Hi!
>
> The following testcase ICEs, because we have a mode mismatch.
> VEC_PACK_TRUNC_EXPR's operands have different modes from the result
> (same vector mode size but twice as large element),
> but we were passing non-NULL subtarget with the mode of t
Hi!
The following testcase ICEs, because we have a mode mismatch.
VEC_PACK_TRUNC_EXPR's operands have different modes from the result
(same vector mode size but twice as large element),
but we were passing non-NULL subtarget with the mode of the result
to the expansion of its arguments, so the VEC
1 - 100 of 104 matches
Mail list logo
