Hi Segher,
Thanks for the review!
on 2020/12/15 上午2:51, Segher Boessenkool wrote:
> Hi!
>
> On Wed, Dec 02, 2020 at 05:44:24PM +0800, Kewen.Lin wrote:
>> This patch is to use paradoxical subreg instead of
>> zero_extend for promoting QI/HI to SI/DI when we
>> want to construct one vector with
On Mon, 6 Jan 2020, Jeff Law wrote:
>> p.s. You know anything about Denis Chertykov? He used to reply to avr
>> patches within hours, but my latest patches (after a long period where I
>> didn't propose any patches) where all approved by you; not a single mail
>> from Denis. So I am concerned
A simple redirect; pushed.
Gerald
gcc/ChangeLog:
2020-12-15 Gerald Pfeifer
* doc/invoke.texi (Instrumentation Options): Update link to
KernelAddressSanitizer.
---
gcc/doc/invoke.texi | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/doc/invoke.texi
The webmasters for arm.com appear to be a little bored (and like
ever longer URLs); at least they put a redirect in place.
Pushed.
Gerald
---
htdocs/gcc-10/changes.html | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/htdocs/gcc-10/changes.html
A straightforward redirect resolved.
Gerald
---
htdocs/extensions.html | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/htdocs/extensions.html b/htdocs/extensions.html
index 76a89e1a..2e929a5b 100644
--- a/htdocs/extensions.html
+++ b/htdocs/extensions.html
@@ -30,7 +30,7 @@
2020-12-14 Max Filippov
gcc/
* config/xtensa/predicates.md (addsubx_operand): Change accepted
values from 2/4/8 to 1..3.
* config/xtensa/xtensa.md (*addx, *subx): Change RTL pattern
to use 'ashift' instead of 'mult'. Update operands[3] value.
gcc/testsuite/
This patch to the Go testsuite driver retains any characters at the
end of an ERROR line in a test. This is needed for at least one
upcoming test. Bootstrapped and ran Go testsuite on
x86_64-pc-linux-gnu. Committed to mainline.
Ian
* go.test/go-test.exp (errchk): Retain any characters at the
On 12/14/20 10:34 PM, Ian Lance Taylor wrote:
On Mon, Dec 14, 2020 at 7:14 PM Nikhil Benesch wrote:
Also godump now emits a dummy `type _u?pad128_t struct {}` entry,
so we just suppress that and conditionally add it back.
I don't understand this bit. Why are we seeing an empty struct
On Mon, Dec 14, 2020 at 7:14 PM Nikhil Benesch wrote:
>
> On 12/14/20 5:30 AM, Rainer Orth wrote:
> > with the revised godump.c patch and this one for mk*sysinfo.sh, I still
> > get failures on all of Solaris 11.3/x86, 11.4/x86, and 11.4/SPARC
> > (didn't try 11.3/SPARC):
> >
> > * Solaris
On 12/14/20 3:47 PM, Patrick Palka wrote:
On Mon, 14 Dec 2020, Jason Merrill wrote:
On 12/14/20 1:07 PM, Patrick Palka wrote:
During satisfaction that's entered through the three-parameter version
of satisfy_declaration_constraints, current_function_decl gets set to
the dependent
On 12/14/20 5:30 AM, Rainer Orth wrote:
with the revised godump.c patch and this one for mk*sysinfo.sh, I still
get failures on all of Solaris 11.3/x86, 11.4/x86, and 11.4/SPARC
(didn't try 11.3/SPARC):
* Solaris 11.3/x86 and 11.4/x86:
runtime_sysinfo.go:5995:6: error: redefinition of
On Tue, 15 Dec 2020 at 04:01, Jeff Law wrote:
>
>
> On 12/14/20 12:10 PM, Dimitar Dimitrov wrote:
> > On четвъртък, 10 декември 2020 г. 10:24:50 EET Richard Biener wrote:
> >> On Thu, Dec 10, 2020 at 6:42 AM Dimitar Dimitrov
> wrote:
> >>> On сряда, 9 декември 2020 г. 15:12:49 EET abebeos via
On Mon, Dec 14, 2020 at 5:06 PM Jeff Law wrote:
>
>
>
> On 12/8/20 5:51 AM, H.J. Lu wrote:
> > When definitions marked with used attribute and unmarked definitions are
> > placed in the section with the same name, switch to a new section if the
> > SECTION_RETAIN bit doesn't match.
> >
> > gcc/
>
On 12/14/20 12:10 PM, Dimitar Dimitrov wrote:
> On четвъртък, 10 декември 2020 г. 10:24:50 EET Richard Biener wrote:
>> On Thu, Dec 10, 2020 at 6:42 AM Dimitar Dimitrov wrote:
>>> On сряда, 9 декември 2020 г. 15:12:49 EET abebeos via Gcc-patches wrote:
Essence:
I need a
On 12/8/20 5:51 AM, H.J. Lu wrote:
> When definitions marked with used attribute and unmarked definitions are
> placed in the section with the same name, switch to a new section if the
> SECTION_RETAIN bit doesn't match.
>
> gcc/
>
> PR target/98146
> * output.h (switch_to_section):
Jonathan Wakely writes:
>>Hmm, yes, you pointed me to __frexpieee128 a few months ago, but for
>>some reason I either didn't switch to using it, or lost a patch when
>>squashing and rebasing branches. Hopefully I just forgot to change it,
>>but I'll double check to make sure I haven't left any
On Tue, Dec 15, 2020 at 12:29:45AM +0100, Gerald Pfeifer wrote:
> On Mon, 14 Dec 2020, Segher Boessenkool wrote:
> >> + powerpc*le-*-*)
> >> + tm_file="${tm_file} rs6000/sysv4le.h" ;;
> >> + esac
> > What an obvious filename ;-)
>
> ;-)
>
> > The patch looks fine. Okay for trunk.
Yes, there is, thanks for noticing that!
Fixed patch attached.
On 20-12-15 00:37:02, Gerald Pfeifer wrote:
> On Mon, 14 Dec 2020, Piotr Kubaj via Gcc-patches wrote:
> > --- gcc/configure.ac.orig 2020-12-14 15:22:23 UTC
> > +++ gcc/configure.ac
> > @@ -5874,6 +5874,8 @@ case
On Mon, 7 Dec 2020, Jozef Lawrynowicz wrote:
> Committed as obvious.
Thank you!
+
+For ELF targets that support the GNU or FreeBSD OSABIs, the
+used attribute will now save the symbol declaration it is
+applied to from linker garbage collection.
+
+To support this behavior,
On Mon, 14 Dec 2020, Matthew Malcomson via Gcc-patches wrote:
> gcc-11/changes: Document new Hardware-assisted AddressSanitizer.
>
> I have put it in the "general" section rather than AArch64 since the
> feature does add a general framework, so I believe the news is
> interesting for people
On Mon, 14 Dec 2020, Piotr Kubaj via Gcc-patches wrote:
> --- gcc/configure.ac.orig 2020-12-14 15:22:23 UTC
> +++ gcc/configure.ac
> @@ -5874,6 +5874,8 @@ case "$target:$tm_file" in
> *-*-linux*)
> emul_name="-melf64ppc"
> ;;
>+ *le-*-freebsd*)
>+
On Mon, 14 Dec 2020, Segher Boessenkool wrote:
>> +powerpc*le-*-*)
>> +tm_file="${tm_file} rs6000/sysv4le.h" ;;
>> +esac
> What an obvious filename ;-)
;-)
> The patch looks fine. Okay for trunk. Also okay for any backports you
> want (after waiting to see if it
Hi!
On Thu, Dec 03, 2020 at 10:57:56PM -0500, Michael Meissner wrote:
> --- a/gcc/testsuite/gcc.target/powerpc/pr70117.c
> +++ b/gcc/testsuite/gcc.target/powerpc/pr70117.c
> @@ -1,5 +1,6 @@
> -/* { dg-do run { target { powerpc*-*-linux* powerpc*-*-darwin*
> powerpc*-*-aix* rs6000-*-* } } } */
>
On 12/14/20 2:05 PM, Martin Sebor wrote:
> On 12/13/20 10:23 AM, Jeff Law wrote:
>>
>>
>> On 12/8/20 1:46 PM, Martin Sebor wrote:
>>> PR 98160 reports an ICE in pretty printer code called from the newly
>>> added -Wmismatched-new-delete. The ICE is just a simple oversight,
>>> but much more
Fix instruction length for MMA insns.
Prefixed instructions should not have their length explicitly set to '8'. The
function get_attr_length() will adjust the length appropriately based on the
value of the "prefixed" attribute.
Bootstrap/regtest on powerpc64le (Power8/Power10) with no new
On 14/12/20 11:08 am, Jonathan Wakely wrote:
On Mon, 14 Dec 2020, 06:51 François Dumont via Libstdc++,
mailto:libstdc%2b...@gcc.gnu.org>> wrote:
On 13/12/20 11:17 pm, Jonathan Wakely wrote:
> On 13/12/20 15:52 +0100, François Dumont via Libstdc++ wrote:
>> Some tests are XPASS
In manually applying a name-lookup patch, I managed to transpose two
calls. That caused a test failure. Fixed thusly.
gcc/cp/
* name-lookup.c (do_namespace_alias): Set originating module
before pushing.
pushing to trunk
--
Nathan Sidwell
diff --git
On 2020-12-14 11:23 a.m., Andrea Corallo wrote:
Vladimir Makarov writes:
On 2020-12-10 10:12 a.m., Andrea Corallo wrote:
Hi all,
following discussion on PR97092 I'd like to submit the following patch
with a fix plus associated testcase.
With this patch applied mode is recomputed at each
On 12/13/20 10:23 AM, Jeff Law wrote:
On 12/8/20 1:46 PM, Martin Sebor wrote:
PR 98160 reports an ICE in pretty printer code called from the newly
added -Wmismatched-new-delete. The ICE is just a simple oversight,
but much more interesting is the warning issued for the test case.
It
On Thu, Dec 03, 2020 at 11:06:12PM -0500, Michael Meissner wrote:
> +proc check_effective_target_ppc_long_double_ibm { } {
> +return [check_cached_effective_target ppc_long_double_ibm {
> + int main()
> + {
> + #if !defined(_ARCH_PPC) || !defined(__LONG_DOUBLE_IBM128__)
> +
On Mon, 14 Dec 2020, Jason Merrill wrote:
> On 12/14/20 1:07 PM, Patrick Palka wrote:
> > During satisfaction that's entered through the three-parameter version
> > of satisfy_declaration_constraints, current_function_decl gets set to
> > the dependent DECL_TEMPLATE_RESULT for sake of access
On Mon, 14 Dec 2020, Jason Merrill wrote:
> On 12/14/20 1:07 PM, Patrick Palka wrote:
> > This makes tracking of potentially unstable satisfaction results more
> > precise by recording the specific types for which completion failed
> > during satisfaction. We now recompute a satisfaction result
On 12/14/20 9:37 AM, Richard Sandiford wrote:
> Jeff Law writes:
>> On 11/13/20 1:20 AM, Richard Sandiford via Gcc-patches wrote:
>>> This patch adds some classes for gathering the list of registers
>>> and memory that are read and written by an instruction, along
>>> with various properties
On 12/14/20 1:07 PM, Patrick Palka wrote:
This makes tracking of potentially unstable satisfaction results more
precise by recording the specific types for which completion failed
during satisfaction. We now recompute a satisfaction result only if one
of these types has been completed since the
On четвъртък, 10 декември 2020 г. 10:24:50 EET Richard Biener wrote:
> On Thu, Dec 10, 2020 at 6:42 AM Dimitar Dimitrov wrote:
> > On сряда, 9 декември 2020 г. 15:12:49 EET abebeos via Gcc-patches wrote:
> > > Essence:
> > >
> > > I need a confirmation that the testsuite setup as presented in:
>
On Mon, Dec 14, 2020 at 07:44:12PM +0100, Tobias Burnus wrote:
> Regarding this discussion, see also the comments from an LLVM reviewer
> at https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97827 * — to keep them
> at one place, I won't quote the discussion here.
I think the LLVM people must be
On 12/14/20 1:07 PM, Patrick Palka wrote:
During satisfaction that's entered through the three-parameter version
of satisfy_declaration_constraints, current_function_decl gets set to
the dependent DECL_TEMPLATE_RESULT for sake of access checking. This
makes the predicate in_template_function
On Wed, Dec 02, 2020 at 04:32:47PM -0600, will schmidt wrote:
> On Wed, 2020-12-02 at 17:44 +0800, Kewen.Lin via Gcc-patches wrote:
> > * gcc.target/powerpc/pr96933-1.c: Adjusted to check no rlwinm.
> > * gcc.target/powerpc/pr96933-2.c: Likewise.
>
> Ok. (I'd hope a few more extend
On петък, 11 декември 2020 г. 19:00:35 EET abebeos wrote:
> After "digesting" a bit more your review, I need to thank you for opening
> my eyes re "cherrypick" suggestion and... the missing g++ summaries. I need
> to update my setup to provide the g++ test-deltas, too. Note that in my
>
Hi!
On Wed, Dec 02, 2020 at 05:44:24PM +0800, Kewen.Lin wrote:
> This patch is to use paradoxical subreg instead of
> zero_extend for promoting QI/HI to SI/DI when we
> want to construct one vector with these modes.
> Since we do the gpr->vsx movement and vector merge
> or pack later, the high
Regarding this discussion, see also the comments from an LLVM reviewer
at https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97827 * — to keep them
at one place, I won't quote the discussion here.
It would be great if this could be solved one way or the other...
Tobias
* for the LLVM side, see also
This makes tracking of potentially unstable satisfaction results more
precise by recording the specific types for which completion failed
during satisfaction. We now recompute a satisfaction result only if one
of these types has been completed since the last time we computed the
satisfaction
During satisfaction that's entered through the three-parameter version
of satisfy_declaration_constraints, current_function_decl gets set to
the dependent DECL_TEMPLATE_RESULT for sake of access checking. This
makes the predicate in_template_function return true during satisfaction
from this
On Mon, Dec 14, 2020 at 05:29:30PM +0100, Piotr Kubaj wrote:
> On 20-12-14 10:22:32, Segher Boessenkool wrote:
> > On Mon, Dec 14, 2020 at 03:57:27PM +0100, Piotr Kubaj wrote:
> > > > It is both, actually (-mcpu= implies -mtune=)
> > > Yes, but -mtune doesn't imply -mcpu. If I set up only -mtune,
Hi!
On Mon, Dec 14, 2020 at 04:35:38PM +0100, Piotr Kubaj wrote:
> this patch implements support for powerpc64le architecture on FreeBSD. Since
> we don't have powerpcle (32-bit), I did not add support for powerpcle here.
> This remains to be changed if there is powerpcle support in the future.
Jeff Law writes:
> On 11/13/20 1:20 AM, Richard Sandiford via Gcc-patches wrote:
>> This patch adds some classes for gathering the list of registers
>> and memory that are read and written by an instruction, along
>> with various properties about the accesses. In some ways it's
>> similar to the
On 20-12-14 10:22:32, Segher Boessenkool wrote:
> On Mon, Dec 14, 2020 at 03:57:27PM +0100, Piotr Kubaj wrote:
> > > It is both, actually (-mcpu= implies -mtune=)
> > Yes, but -mtune doesn't imply -mcpu. If I set up only -mtune, -mcpu is the
> > generic one (ppc970 for BE)
>
> But that is not
On Mon, Dec 14, 2020 at 03:57:27PM +0100, Piotr Kubaj wrote:
> > It is both, actually (-mcpu= implies -mtune=)
> Yes, but -mtune doesn't imply -mcpu. If I set up only -mtune, -mcpu is the
> generic one (ppc970 for BE)
But that is not what the patch does?
> > > Linux also does similarly in
Vladimir Makarov writes:
> On 2020-12-10 10:12 a.m., Andrea Corallo wrote:
>> Hi all,
>>
>> following discussion on PR97092 I'd like to submit the following patch
>> with a fix plus associated testcase.
>>
>> With this patch applied mode is recomputed at each iteration while
>> looping across
Tamar Christina writes:
> Hi Richard,
>
> Do you object to me splitting off complex add and addressing your remaining
> feedback later when the rewrite of mul and fma are done.
No, sounds good to me.
Thanks,
Richard
Hello.
The patch fixes intermediate format for situations where we have
a nested function or a lambda function. In that case, line number
was wrong.
Patch can bootstrap on x86_64-linux-gnu and survives regression tests.
I'll install the patch if there are no objections.
Martin
gcc/ChangeLog:
The patch updates comments in gcov.c file.
Pushed to master.
gcc/ChangeLog:
* gcov.c (output_json_intermediate_file): Update comments.
---
gcc/gcov.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/gcc/gcov.c b/gcc/gcov.c
index 36938bd0fce..daa1266db86 100644
The first adjustment is to the documentation contained in the units,
which explains the implementation choices. The second adjustment is
to the size selection mechanism, which is better done on base types.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
*
This replaces the remaining uses of Current_Input with Current_In and
those of Current_Output with Current_Out for the sake of consistency.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* libgnat/a-tifiio.adb (Get): Replace Current_Input with Current_In.
*
Recent improvements to sem_eval.adb (Fold_Shift) do not take into
account that these operations should be performed on the base type.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* sem_eval.adb (Fold_Shift): Compute values using the base type.diff --git
Attribute Scaling has two parameters but only one of them has been
resolved.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* sem_attr.adb (Analyze_Attribute): Resolve second parameter of
attribute Scaling just like it is resolved for a similar
attribute
This fixes an issue in Remove_Side_Effects, which insists that a renaming
be handled by the back-end whereas Evaluation_Required says that it should
be handled by the front-end (renamings involving Volatile_Full_Access must
be fully expanded by the front-end).
This also reinstates an earlier fix
Analysis of attribute Truncation was first calling Resolve and then
Set_Etype, while all the similar attributes had these calls in the
reverse order. This inconsistency was present since the support for
Truncation was added, but it had no semantics implications.
Tested on x86_64-pc-linux-gnu,
This patch fixes an issue in the compiler whereby a class-wide allocator
for a type declared within a generic formal package was not treated the
same as a formal type with respect to accessibility checks - leading to
spurious accessibility errors.
Tested on x86_64-pc-linux-gnu, committed on trunk
This reimplements the low-level primitives of System.Fat_Gen, i.e.
Copy_Sign, Decompose and Scaling more efficiently, by using the same
direct approach as the existing implementation of Valid; only the
normalization of denormalized numbers and the gradual underflow are
still left to the hardware.
This removes the direct declaration of the smallest denormalized number
and replaces it with an overlay of the equivalent integer constant,
because such a direct declaration cannot be translated into C90.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* libgnat/s-fatgen.adb:
This removes the use clause for System.Unsigned_Type and replaces it
with a localized use type clause for Long_Long_Unsigned.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* libgnat/s-fatgen.adb: Remove use clause for
System.Unsigned_Types.
(Scaling): Add
This commit implements AI12-0398-1/03, which adds the ability to specify
aspects on discriminant specifications, extended return object
declarations and entry index specifications.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* par-ch3.adb (P_Discriminant_Part_Opt): Parse
After the recent implementation work for the Default_Initial_Condition
aspect there were still some cases where DIC checks weren't performed
properly with respect to calls to Initialize in the case of array and
record components of controlled types. That is now corrected. There was
also an issue
Global variable CV_Cache, where values known at compilation time are
stored, was initialization both by elaboration and by
Sem_Eval.Initialize, which is called from Gnatdrv1 and required by the
GNSA (GNAT Semantic Analyzer).
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
*
Implement non-symbolic tracebacks as a first step towards symbolic.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* tracebak.c: Add a section for ARM Linux.diff --git a/gcc/ada/tracebak.c b/gcc/ada/tracebak.c
--- a/gcc/ada/tracebak.c
+++ b/gcc/ada/tracebak.c
@@ -309,6
The condition for adding a dummy constituent is now simpler (and seems
more robust) and the dummy constituent is Any_Id (which exists precisely
for avoiding cascaded errors). Finally, an assertion guards against
returning with an ill-formed refinement.
Cleanup only; the tool behaviour is
The first issue is that the attribute leaks into a multiplication or
a division explicitly done in another fixed-point type before being
passed to the attribute. The second issue is that the attribute is
ignored for simple conversions from another fixed-point type.
Tested on x86_64-pc-linux-gnu,
This patch fixes a bug that causes the compiler to crash if a
renaming-as-body renames a function that is an instance of a generic
function that returns a generic formal type, and the actual type is
controlled, and the instantiation is nested within a procedure.
Tested on x86_64-pc-linux-gnu,
When analysis of pragma Depends/Refined_Depends returns early because of
a malformed pragma, it marks the pragma as analyzed to avoid repeated
work. This is now done consistently for all early returns.
Cleanup only; the GNAT and GNATprove behaviours are not really affected
by this patch.
Tested
In the Depends/Refined_Depends contracts we now accept constant objects
as the dependency outputs if they are of an access-to-variable type, but
not when they are of an access-to-constant or access-to-subprogram
types.
This needs to be implemented in two places: for checking the Depends
contract
Besides the improved commentary, the removal of superfluous subtypes and
other small tweaks throughout to make it more readable, this replaces the
use of T'Mantissa by the more appropriate T'Machine_Mantissa in the body
of System.Fat_Gen.Valid and also removes an unnecessary trick used to
In GNATprove mode, a call inside a Subprogram_Variant should lead to the
called subprogram being marked as not always inlined. This was not
always the case, now fixed.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* inline.adb (Cannot_Inline): Add No_Info parameter to
The routine is now used both for decimal and for ordinary fixed-point types.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* libgnat/s-imguti.ads (Set_Decimal_Digits): Adjust documentation.diff --git a/gcc/ada/libgnat/s-imguti.ads b/gcc/ada/libgnat/s-imguti.ads
---
The code for pragma Assertion_Policy has been recently updated to
include Subprogram_Variant and change the kind of
Default_Initial_Condition, but the comment was not modified.
Tested on x86_64-pc-linux-gnu, committed on trunk
gcc/ada/
* sa_messages.ads: Reference Subprogram_Variant in
Hi Richard,
Do you object to me splitting off complex add and addressing your remaining
feedback later when the rewrite of mul and fma are done.
Thanks,
Tamar
From: Richard Sandiford
Sent: Monday, December 14, 2020 1:30 PM
To: Tamar Christina
Cc:
Hello,
this patch implements support for powerpc64le architecture on FreeBSD. Since we
don't have powerpcle (32-bit), I did not add support for powerpcle here. This
remains to be changed if there is powerpcle support in the future.
Patch implements similar endian detection to what linux64.h
On Mon, Dec 14, 2020 at 9:44 AM Rainer Orth
wrote:
>
> Hi David,
>
> > On Fri, Dec 4, 2020 at 5:35 AM Rainer Orth
> > wrote:
> >
> >> On AIX 7.2, there are changes like
> >>
> >> -PASS: g++.dg/gomp/tls-5.C -std=c++2a scan-assembler-symbol-section
> >> symbol ^_?ir$ (found ir) has section
Hi,
On 2/21/20 8:35 AM, Richard Biener wrote:
>
> IIRC this definitely clashes with Alex work to overhaul
> -auxdir/-dumpdir queued for GCC 11 where some of the above
> is improved.
>
> So whatever we do we should do it for GCC 11 after Alex patches
> land.
>
> Richard.
>
Okay, I think this
For deferred macros we also need a new field on the macro itself, so
that the module machinery can determine the macro was imported. Also
the documentation for the hashnode's deferred field was incomplete.
libcpp/
* include/cpplib.h (struct cpp_macro): Add imported_p field.
Hi All,
Similarly to UNKNOWN permutes, TOP needs to be excluded from being considered
for blends because it produces no permute to check.
Bootstrapped Regtested on aarch64-none-linux-gnu, x86_64-pc-linux-gnu
and no issues.
Committed under the GCC obvious rules.
Thanks,
Tamar
gcc/ChangeLog:
On Mon, Dec 14, 2020 at 7:09 AM Uros Bizjak wrote:
>
> On Mon, Dec 14, 2020 at 2:13 PM Jakub Jelinek wrote:
> >
> > Hi!
> >
> > If somebody has -march=x86-64-v2 (or -v3 or -v4) in $CFLAGS, $CXXFLAGS etc.,
> > then -m32 or -mabi=ms stops working.
> > What is worse, if one configures gcc
On Mon, Dec 14, 2020 at 2:13 PM Jakub Jelinek wrote:
>
> Hi!
>
> If somebody has -march=x86-64-v2 (or -v3 or -v4) in $CFLAGS, $CXXFLAGS etc.,
> then -m32 or -mabi=ms stops working.
> What is worse, if one configures gcc --with-arch-32=x86-64-v2 (or -v3 or -v4),
> then -mabi=ms stops working.
>
>
On 20-12-13 09:48:35, Segher Boessenkool wrote:
> Hi!
>
> On Sun, Dec 13, 2020 at 03:34:57PM +0100, Piotr Kubaj wrote:
> > this is only default tuning (-mtune, not -mcpu).
>
> It is both, actually (-mcpu= implies -mtune=)
Yes, but -mtune doesn't imply -mcpu. If I set up only -mtune, -mcpu is the
Hi David,
> On Fri, Dec 4, 2020 at 5:35 AM Rainer Orth
> wrote:
>
>> On AIX 7.2, there are changes like
>>
>> -PASS: g++.dg/gomp/tls-5.C -std=c++2a scan-assembler-symbol-section
>> symbol ^_?ir$ (found ir) has section ^\\.tbss|\\[TL\\] (found
>> _tls5.tls_[TL],4)
>> +PASS: g++.dg/gomp/tls-5.C
Hi Iain,
>> * Next, I ran the tests on Darwin 11 and found two failing tests:
[...]
>> The other failing test is
>>
>> FAIL: g++.dg/gomp/tls-5.C -std=c++14 scan-assembler-symbol-section
>> symbol ^_?_ZGR2ir_\$ (symbol not found) has section
>> ^.tdata|[TL]
>> FAIL:
On 12/14/20 7:19 AM, Ilya Leoshkevich wrote:
On Fri, 2020-12-11 at 07:51 -0500, Nathan Sidwell wrote:
I'm pretty sure this is wrong. I think the test_text in
inclhack.def
should be a pre-fixed string that the testsuite presumably checks is
converted.
You're right; I've added your change
Hi Ian,
Yes it is.
Thanks,
Clément
From: Ian Lance Taylor
Sent: Friday, December 11, 2020 6:58 PM
To: CHIGOT, CLEMENT
Cc: gcc-patches@gcc.gnu.org ; David Edelsohn
; gofrontend-dev
Subject: Re: [PATCH] gcc: handle double quotes in symbol name during
On 12/11/20 3:39 PM, Jason Merrill wrote:
On 12/2/20 9:19 AM, Nathan Sidwell wrote:
This is modified from the original patch. firstly it contains a few
more necessary bits. But mainly I renamed the entities from
+extern tree make_binding_vec (tree, unsigned clusters);
...
Rearranging slightly…
> @@ -708,6 +713,10 @@ (define_c_enum "unspec"
> UNSPEC_FCMLA90 ; Used in aarch64-simd.md.
> UNSPEC_FCMLA180 ; Used in aarch64-simd.md.
> UNSPEC_FCMLA270 ; Used in aarch64-simd.md.
> +UNSPEC_FCMUL ; Used in aarch64-simd.md.
> +UNSPEC_FCMUL180
Hi!
If somebody has -march=x86-64-v2 (or -v3 or -v4) in $CFLAGS, $CXXFLAGS etc.,
then -m32 or -mabi=ms stops working.
What is worse, if one configures gcc --with-arch-32=x86-64-v2 (or -v3 or -v4),
then -mabi=ms stops working.
I think that is a nightmare user experience. It is ok that
Przemyslaw Wirkus writes:
> Hi,
>
> Recent 'support SVE comparisons for unpacked integers' patch extends
> operands of define_expands from SVE_FULL to SVE_ALL. This causes an ICE
> hence this PR patch.
>
> This patch adds this relaxation for:
> + reduc__scal_ and
> + arch64_pred_reduc__
> in
gcc-11/changes: Document new Hardware-assisted AddressSanitizer.
I have put it in the "general" section rather than AArch64 since the feature
does add a general framework, so I believe the news is interesting for people
interesting in architectures other than AArch64 that may want to implement
On Fri, 2020-12-11 at 07:51 -0500, Nathan Sidwell wrote:
>
> I'm pretty sure this is wrong. I think the test_text in
> inclhack.def
> should be a pre-fixed string that the testsuite presumably checks is
> converted.
You're right; I've added your change from the Bugzilla and updated the
On 12/14/20 12:43 PM, Alejandro Colomar wrote:
Reported-by: Heinrich Schuchardt
Signed-off-by: Alejandro Colomar
Reviewed-by: Heinrich Schuchardt
---
man2/cacheflush.2 | 24
1 file changed, 24 insertions(+)
diff --git a/man2/cacheflush.2 b/man2/cacheflush.2
Reported-by: Heinrich Schuchardt
Signed-off-by: Alejandro Colomar
---
man2/cacheflush.2 | 24
1 file changed, 24 insertions(+)
diff --git a/man2/cacheflush.2 b/man2/cacheflush.2
index aba625721..7a2eed506 100644
--- a/man2/cacheflush.2
+++ b/man2/cacheflush.2
@@ -86,6
Hi all,
I've committed the attached patch as obvious.
This is to set `fill-column' to 80 in c-mode (Emacs default it to 70) so
now M-q does the right thing. I think is very handy especially in
comments.
Question: should we update the copyright manually for this file or have
it updated by
This patch fixes the section type conflict that I have been seeing. I have
tested this patch using :
Build: x86_64
Host: x86_64
Target: aarch64-none-linux-gnu
On 08/12/2020, 12:52, "Gcc-patches on behalf of H.J. Lu via Gcc-patches"
wrote:
When definitions marked with used attribute and
> -Original Message-
> From: Tamar Christina
> Sent: 10 December 2020 17:00
> To: gcc-patches@gcc.gnu.org
> Cc: nd ; Richard Earnshaw ;
> Marcus Shawcroft ; Kyrylo Tkachov
> ; Richard Sandiford
>
> Subject: [PATCH]AArch64: Add NEON, SVE and SVE2 RTL patterns for
> Complex Addition,
Ping.
On 2020/12/4 10:15 PM, Chung-Lin Tang wrote:
Hi Jakub,
this is a new version of the structure element mapping patch for OpenMP 5.0
requirement
changes.
This one uses the approach you've outlined in your concept patch [1], basically
to
use more special REFCOUNT_* values to mark them,
1 - 100 of 115 matches
Mail list logo
