From: Pan Li
This patch would like to refactor the all signed SAT_* patterns for
the saturated value. Aka, overflow to INT_MAX when > 0 and downflow
to INT_MIN when < 0. Thus, we can remove sorts of duplicated expression
in different patterns.
The below test suites are passed for this patch.
*
From: Pan Li
This patch would like to refactor the all signed SAT_ADD patterns,
aka:
* Extract type check outside.
* Re-arrange the related match pattern forms together.
The below test suites are passed for this patch.
* The rv64gcv fully regression test.
* The x86 bootstrap test.
* The x86 full
From: Pan Li
This patch would like to refactor the all signed SAT_TRUNC patterns,
aka:
* Extract type check outside.
* Re-arrange the related match pattern forms together.
The below test suites are passed for this patch.
* The rv64gcv fully regression test.
* The x86 bootstrap test.
* The x86 fu
From: Pan Li
Given the SAT_* patterns are grouped for each alu and signed or not,
add leading comments to indicate the beginning of the pattern.
gcc/ChangeLog:
* match.pd: Update comments for sat_* pattern.
Signed-off-by: Pan Li
---
gcc/match.pd | 6 ++
1 file changed, 6 insertio
On Wed, 11 Dec 2024 at 17:54, Richard Earnshaw (lists)
wrote:
>
> On 07/11/2024 09:18, Christophe Lyon wrote:
> > This patch series re-implements the store_scatter and load_gather
> > intrinsincs using the new framework, similarly to previous series.
> >
> > A few points worth mentioning:
> >
> >
Hi!
I believe we need to clear padding bits even in empty types when using
zero initialization,
https://eel.is/c++draft/dcl.init.general#6.2
doesn't have an exception for empty types.
I came to this when playing with an optimization for PR116416 to improve
tree-ssa/pr78687.C testcase back.
Initia
From: Daniel King
gcc/ada/ChangeLog:
* init.c (__gnat_error_handler): Handle SIGPROT
(__gnat_install_handler): Install SIGPROT handler
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/init.c | 79 ++
1 file changed
From: Ronan Desplanques
This fixes an omission in the recent change that was made to file lookup
for External_Initialization.
gcc/ada/ChangeLog:
* doc/gnat_rm/gnat_language_extensions.rst: Update
External_Initialization section.
* gnat_rm.texi: Regenerate.
* gnat
From: Johannes Kliemann
On targets with preallocated task stacks the minimum stack size is
defined as a constant in System.Parameters. When adding preallocated
tasks to the expanded code the compiler does not have direct access to
that value. Instead generate the expression
Max (Task_Size, Minimu
From: Eric Botcazou
This just moves a couple of checks done in conjunction with the predicate
Aggr_Assignment_OK_For_Backend into its body and adds a couple of comments.
No functional changes.
gcc/ada/ChangeLog:
* exp_aggr.adb (Aggr_Assignment_OK_For_Backend): Add Target formal
From: Piotr Trojanek
Code cleanup.
gcc/ada/ChangeLog:
* par-ch5.adb (Test_Statement_Required): Fix comment.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/par-ch5.adb | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/ada/par-ch5.adb b/gcc/ada/pa
From: Ronan Desplanques
The clauses in section 3.5 of the reference manual were moved around
along the different Ada versions, which caused some comments in our
source code to go out of date. This patch updates the references in
those comments.
gcc/ada/ChangeLog:
* libgnat/a-tifiio.adb:
From: Eric Botcazou
The in-place expansion has been historically disabled for them, but there
does not seem to be any good reason left for this. However, this requires
a small trick in order for the expanded code not to be flagged as using the
object uninitialized by the code generator.
gcc/ada
From: Piotr Trojanek
Originally loop parameter specification only occurred in loops, but now
it also occurs in quantified expressions. This patch guards against
flagging non-loop nodes as null loop statements. This was causing
internal compiler errors that were only visible with switch -gnatdk,
w
From: Piotr Trojanek
Code cleanup; semantics is unaffected.
gcc/ada/ChangeLog:
* sem_ch5.adb (Analyze_Loop_Parameter_Specification): Move call
to Comes_From_Source to the outer if-statement.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/sem_ch5.adb | 17
From: Javier Miranda
gcc/ada/ChangeLog:
* sem_res.adb (Is_Ambiguous_Operand): Add missing decoration of
the operand when it is labeled overloaded but has just one
interpretation.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/sem_res.adb | 3 +--
1 fil
From: Ronan Desplanques
Access parameters are not allowed in specifications of task entries.
Before this patch, the compiler failed to detect that case in accept
statements that were not directly in their task body's scopes. This
patch fixes this issue.
gcc/ada/ChangeLog:
* sem_ch3.adb
From: Eric Botcazou
The pragma is consistenly rejected for the sizes of nonstatic array types
because Eval_Attribute does not evaluate it even if it is known.
gcc/ada/ChangeLog:
* sem_attr.adb (Eval_Attribute): Treat the various size attributes
like Component_Size for nonstatic
Hi!
As the following testcase shows, -fsanitize=builtin instruments the
builtins in the ubsan pass which is done shortly after going into
SSA, but if optimizations optimize the builtins away before that,
nothing is instrumented. Now, I think it is just fine if the
result of the builtins isn't use
r15-6128-gfa878dc8c45fa3 missed the regeneration of the URL doc map, so
regenerate it here to make the buildbots happy.
gcc/ChangeLog:
* config/i386/i386.opt.urls: Regenerate.
---
Committed as obvious, as the buildbots kept emailing and thought it better
to get it out of the way.
gcc/co
On 10/11/2024 10:02, Torbjörn SVENSSON wrote:
Ok for trunk, releases/gcc-12, releases/gcc-13 and releases/gcc-14?
--
In version 6-2017-q1-update of the "GNU Arm Embedded Toolchain" build,
there are 2 pop instructions. In version 7-2018-q2-update, the next
version that still have a binary build
Hi,
the uabd and sabd optab names in the docs are missing a 3 suffix (for
their three arguments). This patch adds it.
Should be obvious enough so going push it in some days unless there
are complaints.
Regards
Robin
gcc/ChangeLog:
* doc/md.texi: Add "3" suffix.
---
gcc/doc/md.texi |
Although this constraint is not currently used for any instructions, it is very
useful for custom instructions. Additionally, some new standard extensions
(not yet upstream), such as `Zilsd` and `Zclsd`, are potential users of this
constraint. Therefore, I believe there is sufficient justification
Updated the patch according to comments. OK for trunk?
Thanks,
Feng
---
gcc/
PR tree-optimization/90594
* tree-scalar-evolution.cc (get_scev_final_value): New function.
(apply_scev_final_value_replacement): Likewise.
(final_value_replacement_loop): Call new function
On Thu, 12 Dec 2024 15:07:35 +0100
Richard Biener wrote:
> On Wed, Dec 11, 2024 at 4:19?PM James K. Lowden
> wrote:
> >
> > I think the term of art is "ping"?
> >
> > If GCC needs something from me to proceed with this, please tell me
> > what it is.
>
> I think we're waiting on the rest of the
Don't use 'c' modifier with a symbolic address. The riscv target doesn't
accept it as a CONSTANT_ADDRESS_P and doesn't support 'c' with SYMBOL_REF.
* c-c++-common/toplevel-asm-1.c: Remove 'c' from %3 and %4.
---
gcc/testsuite/c-c++-common/toplevel-asm-1.c | 2 +-
1 file changed, 1 insert
On Thu, Dec 12, 2024 at 06:05:31PM +0100, Andreas Schwab wrote:
> Don't use 'c' modifier with a symbolic address. The riscv target doesn't
> accept it as a CONSTANT_ADDRESS_P and doesn't support 'c' with SYMBOL_REF.
>
> * c-c++-common/toplevel-asm-1.c: Remove 'c' from %3 and %4.
The intent
From: Bob Duff
The compiler crashes on an assignment statement of the form
"X.Y := new T;", where X.Y is an expanded name (i.e. not a record
component or similar) and T is a type containing tasks.
gcc/ada/ChangeLog:
* exp_util.adb (Build_Task_Image_Decls):
Deal properly with the
From: Viljar Indus
Expanding a function call that returns a controlled type
on the left-hand side of an assignment should be avoided.
Otherwise we will miss the diagnostic for
trying to assign something to a non-variable element.
gcc/ada/ChangeLog:
* exp_ch6.adb (Expand_Ctrl_Function_Ca
From: Eric Botcazou
This mechanism is the only producer of N_Compound_Statement in the expanded
code and parks the statements generated for the in-place initialization of
objects by an aggregate, so that they can be moved to the freeze point if
there is an address aspect/clause, or even cancelled
From: Eric Botcazou
The address passed to the routine attaching a controlled object to the
finalization master must be that of its dope vector for an object whose
nominal subtype is an unconstrained array type, but this is not the case
when this subtype has a private declaration.
gcc/ada/ChangeL
From: Eric Botcazou
The mechanim deferring the expansion of record aggregates nested in other
aggregates with intermediate conditional expressions is disabled in the
case where they contain self-references, because of a technical limitation
in the replacements done by Build_Record_Aggr_Code. Thi
From: Ronan Desplanques
Before this patch, External_Initialization looked for files in all
directories of the source search path, which led to inconsistencies in
some cases. This patch restricts the file lookup so the argument is
interpreted as relative to the current source file's directory only
From: Ronan Desplanques
This patches fixes a couple of details that were wrong in the
documentation comment for System.Val_Util.Scan_Sign.
gcc/ada/ChangeLog:
* libgnat/s-valuti.ads (Scan_Sign): Fix documentation comment.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada
From: Daniel King
This allows CHERI exceptions to be raised from C code in the runtime.
gcc/ada/ChangeLog:
* libgnat/i-cheri-exceptions.ads: Export CHERI exception IDs.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/libgnat/i-cheri-exceptions.ads | 30 +++
From: Eric Botcazou
This extends the processing done for the Address aspect to other delayed
aspects. The External_Name aspect is also reclassified as a representation
aspect and the three representation aspects External_Name, Link_Name and
Linker_Section are moved from the Always_Delay to the R
From: Ronan Desplanques
Before this patch, the machinery to generate validity checks got
confused in some situations involving private views of types, and ended
up generating incorrect conversions from floating point types to integer
types. This patch fixes this.
gcc/ada/ChangeLog:
* ex
From: Ronan Desplanques
Before this patch, the argument to the External_Initialization aspect
had to be a string literal. This patch extends the possibilities so that
any static string is accepted.
A new helper function, Is_OK_Static_Expression_Of_Type, is introduced,
and in addition to the main
From: Ronan Desplanques
This patch slightly widens the set of filenames that the compiler
considers predefined. That makes it possible to build the GNAT runtime
using only the file mapping facilities of the compiler, without having
to rename files.
gcc/ada/ChangeLog:
* fname.adb (Is_Pre
From: Eric Botcazou
The in-place expansion has been historically disabled for them, but there
does not seem to be any good reason left for this.
gcc/ada/ChangeLog:
* exp_aggr.adb (Expand_Array_Aggregate): Do not exclude aggregates
of bit-packed array types in allocators from in-
From: Eric Botcazou
They comprise using a nonnull accesss type for the indirect expansion to
avoid useless checks, smplifying the expansion of if expressions whose
condition is known at compile time to avoid an N_Expression_With_Actions,
using the indirect expansion for them in the indefinite cas
From: Ronan Desplanques
GNAT implements a format with trailing '*' signs for the Image attribute
of NaN, +inf and -inf. It was probably always intended to be the same
length as the image of 1.0, but one '*' was actually missing. This patch
fixes this.
gcc/ada/ChangeLog:
* libgnat/s-imag
From: Ronan Desplanques
A recently fixed bug caused an infinite loop when assertions were not
checked. With assertions checked, the symptom was just an internal
error caused by an assertion failure. This patch makes it so that if
another bug ever causes the same condition to fail, there will neve
From: Sebastian Poeplau
gcc/ada/ChangeLog:
* doc/gnat_ugn/building_executable_programs_with_gnat.rst: Move
the LLVM chapter one level up.
* gnat_ugn.texi: Regenerate.
Tested on x86_64-pc-linux-gnu, committed on master.
---
.../building_executable_programs_with_gnat.rst
Hi!
"replacement is succeeded" doesn't look correct to me (though, sure, I'm
not native English speaker), this patch drops the is.
2024-12-12 Jakub Jelinek
* gimple-crc-optimization.cc (crc_optimization::optimize_crc_loop):
Comment spelling fix, is succeeded -> succeeded.
---
Hi!
As we don't have a SRA fix for PR117971, I thought I'd try to improve
it using an optimization during gimplification.
This is about the tree-ssa/pr78687.C testcase, which is a variant with
struct option_1
{
void *a, *b, *c, *d, *e;
};
struct option_2
{
};
variants. Since the PR116416
From: Viljar Indus
gcc/ada/ChangeLog:
* errout.adb (Write_JSON_Location): Avoid going through
symbolic links when printing the full name.
Tested on x86_64-pc-linux-gnu, committed on master.
---
gcc/ada/errout.adb | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --
ср, 11 дек. 2024 г. в 16:56, Georg-Johann Lay :
>
> This patch adds __flashx as a new named address space that allocates
> objects in .progmemx.data. The handling is mostly the same or similar
> to that of 24-bit space __memx, except that the output routines are
> simpler and more efficient. Load
On 09/12/2024 15:05, Christophe Lyon wrote:
Changes v1->v2:
- Keep MAX_TUPLE_SIZE=0 and update accesses to acle_vector_types
accordingly.
- implement arm_array_mode in patch 4/4 instead of 2/4 to avoid
temporary regressions when running the testsuite at patch 2/4 (helps
future bisects)
Hi,
in PR117353 and PR117878 we expand a const vector during reload. For
this we use an unpredicated left shift. Normally an insn like this is
split but as we introduce it late and cannot create pseudos anymore
it remains unpredicated and is not recognized by the vsetvl pass (where
we expect all
Tamar Christina writes:
> diff --git a/gcc/config/aarch64/aarch64-sve-builtins.cc
> b/gcc/config/aarch64/aarch64-sve-builtins.cc
> index
> 0fec1cd439e729dca495aac4dea054a25ede20a7..e6c2bdeb00681848a838009c833cfe3271a94049
> 100644
> --- a/gcc/config/aarch64/aarch64-sve-builtins.cc
> +++ b/gcc/c
On 2024-12-04 12:42, Richard Earnshaw (lists) wrote:
On 21/11/2024 19:01, Torbjörn SVENSSON wrote:
Ok for trunk and releases/gcc-14?
--
On Cortex-M4, the code generated is:
cmp r0, r1
ittene
lslne r0, r0, r1
asrne r0, r0, #1
moveq r0, r1
add
On Wed, Dec 11, 2024 at 6:32 PM Alexandre Oliva wrote:
>
> On Dec 11, 2024, Richard Biener wrote:
>
> > I think These 0, 0 args are supposed to indicate Maximum extent of the
> > resulting Access
>
> Thanks, that looks much better indeed.
>
>
> A bootstrap on aarch64-linux-gnu revealed that somet
On Thu, 12 Dec 2024 at 14:24, Giuseppe D'Angelo
wrote:
>
> Hi,
>
> On 12/12/2024 01:04, Jonathan Wakely wrote:
> >> I'll prepare a patch to do that,
> > Et voila:
> > https://gcc.gnu.org/pipermail/gcc-patches/2024-December/671432.html
>
> Thanks! All done, new patch is attached.
Thanks. Do you ha
Tamar Christina writes:
> Hi All,
>
> These tests are x86 specific and shouldn't be run for aarch64.
>
> gcc/testsuite/ChangeLog:
>
> PR target/96342
> * c-c++-common/gomp/declare-variant-14.c: Make i?86 and x86_64 target
> only test.
> * gfortran.dg/gomp/declare-variant-14
On 2024-12-04 12:57, Richard Earnshaw (lists) wrote:
On 21/11/2024 14:24, Torbjörn SVENSSON wrote:
Update test cases to use -mcpu=unset/-march=unset feature introduced in
r15-3606-g7d6c6a0d15c.
gcc/testsuite/ChangeLog:
* gcc.target/arm/bfloat16_scalar_1_1.c: Use effective-target
Rename those constraint since we want define other constraint start with
`c`, those constraints are internal and undocumented, so it's fine to
rename.
gcc/ChangeLog:
* config/riscv/constraints.md (c01): Rename to...
(k01): ...this.
(c02): Rename to...
(k02): ...thi
gcc/ChangeLog:
* config/riscv/constraints.md (cr): New.
(cf): New.
* config/riscv/riscv.h (reg_class): Add RVC_GR_REGS and
RVC_FP_REGS.
(REG_CLASS_NAMES): Ditto.
(REG_CLASS_CONTENTS): Ditto.
* doc/md.texi: Document cr and cf constraint.
Tamar Christina writes:
> Hi All,
>
> This patch adds support for vector constructor from two partial SVE vectors
> into
> a full SVE vector. It also implements support for the standard vec_init obtab
> to
> do this.
>
> gcc/ChangeLog:
>
> PR target/96342
> * config/aarch64/aarch64-s
> On 6 Dec 2024, at 08:41, Jennifer Schmitz wrote:
>
>
>
>> On 5 Dec 2024, at 20:07, Richard Sandiford wrote:
>>
>> External email: Use caution opening links or attachments
>>
>>
>> Jennifer Schmitz writes:
On 5 Dec 2024, at 11:44, Richard Biener wrote:
External email: U
On 2024-12-12 12:26, Richard Earnshaw (lists) wrote:
On 10/11/2024 13:38, Torbjörn SVENSSON wrote:
Hi Richard,
I'm not sure if I'm doing something wrong here, or if it was an oversight
when doing the update in r12-8108-g62082d278d1.
Anyway, the commit message suggest that it's only the const
Here is a purposal that using N for printing register encoding number,
so let rename the existing internal operand modifier `N` to `n`.
gcc/ChangeLog:
* config/riscv/corev.md (*cv_branch): Update modifier.
(*branch): Ditto.
* config/riscv/riscv.cc (riscv_print_operand): Up
The modifier `N`, to print the raw encoding of a register. This is used
when using `.insn , `, where the user wants to pass
a value to the instruction in a known register, but where the
instruction doesn't follow the existing instruction formats, so the
assembly parser is not expecting a register n
On 14/11/2024 10:41, Christophe Lyon wrote:
Remove floating-point condition from mve_vec_extract_sext_internal and
mve_vec_extract_zext_internal, since the MVE_2 iterator does not
include any FP mode.
gcc/ChangeLog:
* config/arm/mve.md (mve_vec_extract_sext_internal): Fix
condit
On Wed, Dec 11, 2024 at 4:19 PM James K. Lowden
wrote:
>
> I think the term of art is "ping"?
>
> If GCC needs something from me to proceed with this, please tell me what
> it is.
I think we're waiting on the rest of the patches adding the Cobol
frontend at this point.
Richard.
> --jkl
>
> On T
On Wed, Dec 11, 2024 at 11:39:51AM -0500, James K. Lowden wrote:
> On Thu, 12 Dec 2024 15:07:35 +0100
> Richard Biener wrote:
>
> > On Wed, Dec 11, 2024 at 4:19?PM James K. Lowden
> > wrote:
> > >
> > > I think the term of art is "ping"?
> > >
> > > If GCC needs something from me to proceed with
On Thu, Dec 12, 2024 at 4:05 PM James K. Lowden
wrote:
>
> On Thu, 12 Dec 2024 15:07:35 +0100
> Richard Biener wrote:
>
> > On Wed, Dec 11, 2024 at 4:19?PM James K. Lowden
> > wrote:
> > >
> > > I think the term of art is "ping"?
> > >
> > > If GCC needs something from me to proceed with this, p
Thanks for doing this Tamar,
> On 11 Dec 2024, at 10:54, Tamar Christina wrote:
>
>> -Original Message-
>> From: Richard Sandiford
>> Sent: Wednesday, December 11, 2024 9:50 AM
>> To: Tamar Christina
>> Cc: gcc-patches@gcc.gnu.org; nd ; Richard Earnshaw
>> ; ktkac...@gcc.gnu.org
>> Sub
Hi,
On 12/12/2024 01:04, Jonathan Wakely wrote:
I'll prepare a patch to do that,
Et voila:
https://gcc.gnu.org/pipermail/gcc-patches/2024-December/671432.html
Thanks! All done, new patch is attached.
These mem-initializers are in the wrong order (we had an existing
constructor with the sam
* rampz_rtx et al. were missing MEM_VOLATILE_P. This is needed because
avr_emit_cpymemhi is setting RAMPZ explicitly with an own insn.
* avr_out_cpymem was missing a final RAMPZ = 0 on EBI devices.
This only affects the __flash1 ... __flash5 spaces since the other ASes
use different routines,
On 2024-12-04 12:46, Richard Earnshaw (lists) wrote:
On 21/11/2024 17:23, Torbjörn SVENSSON wrote:
I'm not sure how to verify that adding the parameter won't destroy the test.
I've tried to repoduce the ICE on old Arm builds of arm-none-eabi, but none of
them ICE. I suppose it should be safe
On 2024-12-04 13:03, Richard Earnshaw (lists) wrote:
On 21/11/2024 14:24, Torbjörn SVENSSON wrote:
Update test cases to use -mcpu=unset/-march=unset feature introduced in
r15-3606-g7d6c6a0d15c.
gcc/testsuite/ChangeLog:
* gcc.dg/pr41574.c: Added option "-mcpu=unset".
* gcc.dg/
On 14/11/2024 10:42, Christophe Lyon wrote:
V2DF is not supported by MVE, so remove it from the only iterator
which contains it.
gcc/ChangeLog:
* config/arm/iterators.md (MVE_vecs): Remove V2DF.
---
gcc/config/arm/iterators.md | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
di
Hi all,
attached patch improves analysis of cycles in derived types, i.e. type
dependencies ala:
type(T)
type(T2), allocatable :: c
end type
type(T2)
type(T), allocatable :: t
end type
are now detected and deallocating an object that is of any of the types
now no longer crashes the compiler
On 2024-12-12 12:02, Richard Earnshaw (lists) wrote:
On 10/11/2024 10:02, Torbjörn SVENSSON wrote:
Ok for trunk, releases/gcc-12, releases/gcc-13 and releases/gcc-14?
--
In version 6-2017-q1-update of the "GNU Arm Embedded Toolchain" build,
there are 2 pop instructions. In version 7-2018-q2
On Wed, 2024-11-27 at 15:18:39 +, Richard Biener wrote:
> I'm not sure it will work this way together with the output_die hunk,
> instead
> assemblers likely expect all this to happen close to the actual label
> emission, so I suggest to only split out the visibiltiy/globalizing fancy
> and emi
On 12/12/24 03:53, Sam James wrote:
r15-6128-gfa878dc8c45fa3 missed the regeneration of the URL doc map, so
regenerate it here to make the buildbots happy.
I apologize for this breakage. :-( Can someone explain how I can
detect this problem *before* submitting patches, and how to fix it, so
On Sun, 2024-11-17 at 19:15:04 +, Jan Hubicka wrote:
>
> I would suggest renaming produce_asm to produce_symbol_asm
> and making produce_asm wrapper which passes fn=NULL and output_order=-1,
> so we do not have odd parameters everywhere in streaming code.
>
> OK with this change.
> Honza
Ap
> Am 12.12.2024 um 11:28 schrieb Jakub Jelinek :
>
> Hi!
>
> As the following testcase shows, -fsanitize=builtin instruments the
> builtins in the ubsan pass which is done shortly after going into
> SSA, but if optimizations optimize the builtins away before that,
> nothing is instrumented.
This fixes a precondition failure triggered when the Eigenvalues routine of
Ada.Numerics.Generic_Real_Arrays is instantiated with -gnata, beause it calls
Sort_Eigensystem on an empty vector.
Tested on x86-64/Linux, applied on all active branches.
2024-12-12 Eric Botcazou
PR ada/117
This simple patch improves the diagnostic and prepares for some future work.
Note: has_device_addr for C is permitted but pointless, for C++ it
requires some follow-up work to be useful.
For C, a hint that 'need_device_addr' it not valid has been added and
for 'has_device_addr' a middle-end warnin
This patch set implements the proposal from riscv-c-api-doc[1].
It adds two constraints and one modifier with the goal of improving the user
experience for `.insn`, making it easier for users to experiment with new ISA
extensions.
A quick summary of this patch set:
- Add R constraint for even-odd
On 2024-12-04 12:53, Richard Earnshaw (lists) wrote:
On 21/11/2024 14:24, Torbjörn SVENSSON wrote:
Update test cases to use -mcpu=unset/-march=unset feature introduced in
r15-3606-g7d6c6a0d15c.
gcc/testsuite/ChangeLog:
* g++.dg/other/pr56184.C: Use effective-target
arm_arch_
On 2024-12-04 12:41, Richard Earnshaw (lists) wrote:
On 22/11/2024 09:37, Torbjörn SVENSSON wrote:
Changes since v1:
- Rewrote the padding instructions in the macro to instead write to volatile
memory. This ensures that every expansion of the base macro is exactly 2
bytes.
If the `GO(
Ping^2
Thanks,
Tejas.
On 11/4/24 10:06 AM, Tejas Belagod wrote:
Ping.
Thanks,
Tejas.
On 10/18/24 11:59 AM, Tejas Belagod wrote:
Hi Jakub,
Just wanted to add that I'm sorry for the delay in respinning the
patchset - I was caught up with another piece of work. Thanks for the
reviews so far
For the avr.cc part, the __flashx named address space has been approved,
which means that the natural choice in the target hook would be
ADDR_SPACE_FLASHX instead of ADDR_SPACE_MEMX:
+/* Implement `TARGET_ADDR_SPACE_FOR_ARTIFICIAL_RODATA'. */
+
+static addr_space_t
+avr_addr_space_for_artificia
On 10/11/2024 13:38, Torbjörn SVENSSON wrote:
Hi Richard,
I'm not sure if I'm doing something wrong here, or if it was an oversight
when doing the update in r12-8108-g62082d278d1.
Anyway, the commit message suggest that it's only the constant that is of
interrest, so I updated the test to only c
On 12/12/2024 13:47, Torbjorn SVENSSON wrote:
On 2024-12-12 12:02, Richard Earnshaw (lists) wrote:
On 10/11/2024 10:02, Torbjörn SVENSSON wrote:
Ok for trunk, releases/gcc-12, releases/gcc-13 and releases/gcc-14?
--
In version 6-2017-q1-update of the "GNU Arm Embedded Toolchain" build,
ther
On 2024-12-04 19:54, Richard Earnshaw (lists) wrote:
On 21/11/2024 14:24, Torbjörn SVENSSON wrote:
Update test case to use -mcpu=unset/-march=unset feature introduced in
r15-3606-g7d6c6a0d15c.
gcc/testsuite/ChangeLog:
* gcc.target/arm/lto/pr96939_0.c: Use effective-target
ar
On 2024-12-04 19:52, Richard Earnshaw (lists) wrote:
On 21/11/2024 14:24, Torbjörn SVENSSON wrote:
The test case gcc.target/arm/its.c was created together with restriction
of IT blocks for Cortex-M7. As the test case fails on all tunes that
does not match Cortex-M7, explicitly test it for Cor
> gcc/ChangeLog:
>
> * Makefile.in: Add lto-ltrans-cache.o.
> * lto-wrapper.cc: Use ltrans cache.
> * lto-ltrans-cache.cc: New file.
> * lto-ltrans-cache.h: New file.
OK.
> +
> +/* Computes checksum for given file, returns NULL_CHECKSUM if not
> + possible.
> + */
As a f
On 12/12/2024 13:36, Torbjorn SVENSSON wrote:
On 2024-12-12 12:26, Richard Earnshaw (lists) wrote:
On 10/11/2024 13:38, Torbjörn SVENSSON wrote:
Hi Richard,
I'm not sure if I'm doing something wrong here, or if it was an
oversight
when doing the update in r12-8108-g62082d278d1.
Anyway, the
We currently ICE upon the following valid (under -Wno-vla) code
=== cut here ===
void f(int c) {
constexpr int r = 4;
[&](auto) { int t[r * c]; }(0);
}
=== cut here ===
The problem is that when parsing the lambda body, and more specifically
the multiplication, we mark the lambda as LAMBDA_EXP
Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk?
-- >8 --
This ICE started with the recent prvalue optimization (r15-6052). In
cp_fold_r we have:
if (tree &init = TARGET_EXPR_INITIAL (stmt))
{
cp_walk_tree (&init, cp_fold_r, data, NULL);
// ...
Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look OK for
trunk? The older regression does not seem worth fixing.
-- >8 --
In the first testcase we're overeagerly diagnosing qualified name lookup
failure for f from the current instantiation B::C ahead of time
because we (correctly
Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look
OK for trunk/14?
This fixes the testcase in the PR but doesn't thoroughly fix the
underlying issue since if we replace fnPtr with e.g. a constexpr variable
so that the callee is truly potentially constant then the ICE reappears
(due
Bootstrapped and regtested on x86_64-pc-linux-gnu, does this look
OK for trunk and perhpas 14?
-- >8 --
When we encounter an unexpected (likely templated) tree code during
constexpr evaluation we currently ICE even in release mode. But it
seems more user-friendly to just gracefully treat the exp
Am Montag, dem 09.12.2024 um 16:20 + schrieb Qing Zhao:
>
> > On Dec 7, 2024, at 03:57, Martin Uecker wrote:
> >
> > Am Freitag, dem 06.12.2024 um 16:13 + schrieb Qing Zhao:
> > >
> > > > On Dec 6, 2024, at 10:56, Martin Uecker wrote:
> > > >
> > > > Am Freitag, dem 06.12.2024 um 14:1
Each lambda that can be converted to a plain function pointer has a
thunk generated for it, which invokes the body of the lambda function.
When a section attribute is added to a lambda function, it only applies
to the body of the lambda function, and not the thunk. When a lambda is
only ever used
The list of -Wsuggest-attribute= variants was out of date in the option
summary (and getting too long to fit on one line), and an index entry was
missing for -Wsuggest-attribute=returns_nonnull.
gcc/c-family/ChangeLog
PR c/115532
* c.opt.urls: Regenerated.
gcc/ChangeLog
PR
1 - 100 of 112 matches
Mail list logo
