2026-03-26 Paul Thomas <pault@gcc.gnu.org>
gcc/fortran
PR fortran/115315
* decl.cc (insert_parameter_exprs): Make strcmp condition more
concise.
(gfc_get_pdt_instance): Use gf_replace_expr where possible and
use return value of gfc_simplify_expr. Correct error in which
params->expr was being simplified instead of c2->initializer.
* expr.cc (gfc_simplify_expr): If the substring 'start' value
is less than zero, it is clearly out of range and so return
false.
gcc/testsuite/
PR fortran/115315
* gfortran.dg/pdt_90.f03: New test.
TCL 9 removed case command which has been deprecated before.
https://core.tcl-lang.org/tcl/wiki?name=Migrating+scripts+to+Tcl+9
says "Replace case with switch" and switch has been working in TCL 8
just fine.
2026-03-26 Jakub Jelinek <jakub@redhat.com>
* lib/mike-gcc.exp (postbase): Use switch instead of case for
TCL 9 compatibility.
* lib/mike-g++.exp (postbase): Likewise.
This patch adds create_asm_partitions to cache partitioning
to prevent issues with non-renameable symbols (while partition
joining) and static asm symbols (while partition splitting).
All other relevant partitionings use create_asm_partitions.
This was not used in cache partitioning, because toplevel asm
could be in principle special handled in cache partitioning
with marginally better results, but I never implemented it.
lto/124289
gcc/lto/ChangeLog:
* lto-partition.cc (enum map1to1_content): New.
(map_1_to_1): Use map1to1_content.
(lto_1_to_1_map): Likewise.
(create_asm_partitions): Likewise.
(lto_max_map): Likewise.
(lto_cache_map): Use create_asm_partitions.
gcc/testsuite/ChangeLog:
* gcc.dg/lto/toplevel-extended-asm-2_0.c: Add padding to asm label.
* gcc.dg/lto/toplevel-extended-asm-2_1.c: Add padding to asm label.
In substitute7.C we have
template <typename T, auto ... Vs>
constexpr auto construct_from = T{Vs...}; // #1
struct Inner {};
struct Outer { Inner m; };
constexpr auto r = substitute(^^construct_from,
{ ^^Outer, ^^construct_from<Inner> });
which crashes because the auto in #1 hadn't been deduced when we
called eval_can_substitute -> lookup_template_variable when processing
the substitute call. We can call mark_used to resolve this, because
it has:
if (undeduced_auto_decl (decl))
maybe_instantiate_decl (decl);
and to make the json-parser.C test work, we have to do something
similar in eval_substitute, otherwise we crash due to another
undeduced auto.
PR c++/123613
gcc/cp/ChangeLog:
* reflect.cc (get_reflection): Call mark_used.
(eval_can_substitute): Don't resolve_nondeduced_context here.
(eval_substitute): Call lookup_and_finish_template_variable instead
of lookup_template_variable and finish_template_variable.
gcc/testsuite/ChangeLog:
* g++.dg/reflect/json-parser.C: New test.
* g++.dg/reflect/substitute7.C: New test.
* g++.dg/reflect/test.json: New test.
Reviewed-by: Jason Merrill <jason@redhat.com>
PR fortran/124567
gcc/fortran/ChangeLog:
* interface.cc (gfc_check_dummy_characteristics): Split shape check
into a separate check for rank and a check for shape, taking into
account a corner case where the ambiguity between deferred shape
and assumed shape has not been fully resolved at the time of
checking.
gcc/testsuite/ChangeLog:
* gfortran.dg/pr124567.f90: New test.
* gfortran.dg/proc_decl_30.f90: Likewise.
The following fixes the issue of two open gcov files at the
same time, one from -ftest-coverage, opened/closed by
coverage_init/finish and one from -fauto-profile, attempted
to be opened by pass_ipa_auto_profile. The solution is
to open the coverage files only during pass_ipa_tree_profile.
PR gcov-profile/121074
* coverage.h (coverage_init_file): Declare.
(coverage_finish_file): Likewise.
* coverage.cc (coverage_init_file): New function, split
out actual file opening and writing from ...
(coverage_init): ... here.
(coverage_finish_file): Likewise for file closing, from ...
(coverage_finish): ... here.
* tree-profile.cc (tree_profiling): Call coverage_init_file
and coverage_finish_file here.
When we fixup a noreturn call during inlining we can end up eliding
debug stmts, so we have to make sure to not re-instantiate SSA
operands on those later when processing all debug stmts.
PR tree-optimization/124627
* tree-inline.cc (copy_debug_stmts): Only copy debug
stmts that are still in the IL.
xtreme-header-8.C fails on VxWorks because various TU-local gthr
functions defined as static inline are referenced from libstdc++
symbols with global visibility.
C++ modules require those functions to be non-static inline, and other
gthr implementations adopt always_inline in C++.
Follow this practice in gthr-vxworks.h as well.
for libgcc/ChangeLog
* config/gthr-vxworks.h (__GTHREAD_ALWAYS_INLINE,
__GTHREAD_INLINE): Copy from gthr-posix.h. Replace static
inline with __GTHREAD_INLINE.
The test (presumably by mistake) calls setjmp with a sigjmp_buf.
Because of the sigsetjmp requirement, I'm replacing the setjmp call
with sigsetjmp, though the comments in the test explicitly mention
setjmp.
for gcc/testsuite/ChangeLog
* gcc.misc-tests/gcov-32.c: Replace mismatched setjmp call
with sigsetjmp.
The expected macro debug information is not issued with
-gstrict-dwarf, and ports such as vxworks default to that. Allow
non-strict dwarf for the test.
pr115066.c needs the same problem adjustment.
for gcc/testsuite/ChangeLog
* gcc.dg/pr111409.c: Allow non-strict dwarf.
* gcc.dg/pr115066.c: Likewise.
ARM test epilog-1.c needs -mno-long-calls to match the expected asm
outputs on targets that default to -mlong-calls.
for gcc/testsuite/ChangeLog
* gcc.target/arm/epilog-1.c: Add -mno-long-calls.
The loop invokes undefined behavior: the flags array runs 0..31, but
so does the iterator i, thus the accesses to pairs therein run 0..63.
If we're lucky we find all zeros and don't mess with anything else,
but if we find any nonzero out-of-range flags, we'll end up scribbling
out of range onto x, that runs 0..15.
Surely the iterator wasn't meant to go up to 32. If it goes up to 16,
as proposed herein, we only access flags in range, and since its
second half is all zeros, we only access x in range as well.
for gcc/testsuite/ChangeLog
* gcc.dg/vect/slp-mask-store-1.c (foo): Fix iteration range.
30_threads/condition_variable_any/stop_token/wait_on.cc's
test_wait_until occasionally fails on vxworks under very high load, in
a way that suggests wait_until times out before the main thread
requests it to stop. Extend the timeouts to make more room for the
stop request.
for libstdc++-v3/ChangeLog
* testsuite/30_threads/condition_variable_any/stop_token/wait_on.cc
(test_wait_until): Extend the timeout for a stop request.
The test calls sysconf but it doesn't require it as it should.
vect-early-break_130.c needs that treatment as well.
for gcc/testsuite/ChangeLog
* gcc.dg/vect/vect-early-break_109-pr113588.c: Require sysconf.
* gcc.dg/vect/vect-early-break_130.c: Likewise.
The first loop in main gets stores "vectorized" on powerpc64 into
full-word stores, even without any vector instruction support, so the
test's expectation of no loop vectorization is not met.
for gcc/testsuite/ChangeLog
* gcc.dg/tree-ssa/gen-vect-11c.c: xfail the test for no
vectorization on powerpc*-*-* && lp64.
The test needs to require sysconf for _SC_PAGESIZE, and it shouldn't
override the vect.exp-set vect-enabling options by using dg-options.
for gcc/testsuite/ChangeLog
* gcc.dg/vect/vect-early-break_137-pr121190.c: Require sysconf.
Don't override vect.exp options.
VxWorks doesn't support __cxa_thread_atexit, so XFAIL the test that
relies on it for correct destruction order of thread-local objects.
for gcc/testsuite/ChangeLog
* g++.dg/tls/thread_local-order2.C: XFAIL on VxWorks.
ARM requires C++ constructors to return the 'this' pointer. Setting
that up disables warnings at the ctor location for Wuse_after_free.
Unfortunately, the nowarn_spec_t mapping maps both Wuse_after_free and
Wdeprecated_declaration to the same NW_OTHER catchall group, so we
don't get deprecation warnings for constructors, and libstdc++'s
20_util/pair/cons/99957.cc fails.
We may want to consider also explicitly mapping Wdeprecated and
Wdeprecated_declaration to another group, but this minimal change
appears to be enough.
for gcc/ChangeLog
* gcc-diagnostic-spec.cc (nowarn_spec_t::nowarn_spec_t): Map
OPT_Wuse_after_free like OPT_Wuse_after_free_.
2026-03-25 Paul Thomas <pault@gcc.gnu.org>
gcc/fortran
PR fortran/119273
* trans-array.cc (expr_contains_impure_fcn): New function.
(gfc_expr_contains_impure_fcn): New function calling above fcn.
(array_bound_check_elemental): Add indexse pre-block to se pre.
Warn if the index expression contains a function not declared
to be pure.
gcc/testsuite/
PR fortran/119273
* gfortran.dg/pr119273.f90: New test.
The gcc.target/i386/pr61599-1.c test has been FAILing on Solaris/x86
with the native assembler for a long time:
FAIL: gcc.target/i386/pr61599-1.c (test for excess errors)
UNRESOLVED: gcc.target/i386/pr61599-1.c compilation failed to produce executable
Excess errors:
ld: fatal: relocation error: R_AMD64_PC32: file gcc/amd64/crtbegin.o: symbol completed.0: value 0xc01004b9 does not fit
ld: fatal: relocation error: R_AMD64_PC32: file gcc/amd64/crtbegin.o: symbol completed.0: value 0xc01004a7 does not fit
Looking closer, pr61599-1.o is 3.1 GB large with the native assembler.
Comparing the section headers between as and gas, one sees the difference:
With as there's
Section Header[3]: sh_name: .lbss.a
sh_addr: 0 sh_flags: [ SHF_WRITE SHF_ALLOC ]
sh_size: 0x40000000 sh_type: [ SHT_PROGBITS ]
while gas creates
Section Header[5]: sh_name: .lbss.a
sh_addr: 0 sh_flags: [ SHF_WRITE SHF_ALLOC SHF_AMD6
4_LARGE ]
sh_size: 0x40000000 sh_type: [ SHT_NOBITS ]
The use of SHT_PROGBITS by as is due to gcc not emitting @nobits for the
.lbss sections, so as uses the default section type:
.section .lbss.a,"aw"
For .bss, we get this instead:
.section .bss.a,"aw",@nobits
as usually doesn't care about section names, but relies on the compiler
to correctly emit section types and flags.
This happens because x86_64_elf_section_type_flags first calls
default_section_type_flags, which doesn't know about .lbss and sets
SECTION_NOTYPE. When default_elf_asm_named_section later emits the
.section directive, @nobits is omitted even though SECTION_BSS has been
added by then.
To work around this, this patch clears SECTION_NOTYPE when SECTION_BSS
is set.
Similarly, the current assembler output relies on the assembler to set
the SHF_X86_64_LARGE section flag based on the section name. While gcc
emits the necessary flag letter ('l' with gas, 'h' with Solaris as) for
plain .lbss sections, it fails to do so for .lbss.*.
This patch adresses this by moving the gas definition of
MACH_DEP_SECTION_ASM_FLAG to i386/x86_64.h so it's picked up by all
64-bit x86 targets. It also extends ix86_in_large_data_p to also
consider .lbss.* etc. sections as large data sections, not just .lbss,
matching x86_64_elf_section_type_flags.
pr61599-1.o is identical without and with the section letter and section
type specifiec.
Bootstrapped without regressions on i386-pc-solaris2.11,
amd64-pc-solaris2.11, x86_64-pc-linux-gnu, and i686-pc-linux-gnu.
2026-03-14 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
gcc:
* config/i386/i386.cc (ix86_in_large_data_p): Check for .lbss etc.
(x86_64_elf_section_type_flags): Clear SECTION_NOTYPE for .lbss etc.
* config/i386/x86-64.h (MACH_DEP_SECTION_ASM_FLAG): Define.
* config/i386/sol2.h [HAVE_SOLARIS_AS]
(MACH_DEP_SECTION_ASM_FLAG): Redefine as 'h'.
gcc/testsuite:
* gcc.target/i386/pr61599-1.c (dg-options): Add -save-temps.
(scan-assembler-times): Check for @nobits.
This patch uses next_speculative_call_target in get_next_speculative_id
As the specualtive_id should be unique only within on speculative id
block.
gcc/ChangeLog:
* cgraph.cc (cgraph_edge::get_next_speculative_id): Use
next_speculative_call_target in get_next_speculative_id.
Signed-off-by: Kugan Vivekanandarajah <kvivekananda@nvidia.com>
It was assumed that elimination in LRA does not generate new reload
insns. In the testcase the elimination of SFP subreg generates reload
insn of the subreg and this insn is added at end of RTL code. The insn
is also skipped for necessary processing. This results in ICE. The patch
checks creation of reloads insn during elimination, insert them in the right
place, and add them for later processing.
gcc/ChangeLog:
PR rtl-optimization/124572
* lra-eliminations.cc (lra_eliminate): Push new reload insns for
eliminations in insns.
gcc/testsuite/ChangeLog:
PR rtl-optimization/124572
* gcc.target/aarch64/pr124572.c: New.
The g++.dg/coroutines/pr110872.C test XPASSes on many 32-bit targets:
XPASS: g++.dg/coroutines/pr110872.C -std=c++26 (internal compiler error)
Darwin/x86 is the exception where the test ICEs for both 32 and 64-bit.
Tested on i386-pc-solaris2.11, sparc-sun-solaris2.11,
x86_64-apple-darwin17.7.0, and x86_64-pc-linux-gnu.
2026-03-04 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
gcc/testsuite:
PR testsuite/124326
* g++.dg/coroutines/pr110872.C (dg-ice): Restrict to 64-bit
targets or Darwin.
The g++.dg/warn/Wstringop-overflow-6.C test FAILs on Solaris:
FAIL: g++.dg/warn/Wstringop-overflow-6.C -std=gnu++11 (test for excess errors)
with
Excess errors:
/usr/include/iso/math_c99.h:361: warning: ignoring '#pragma does_not_read_global_data ' [-Wunknown-pragmas]
and
/usr/include/errno.h:93: warning: ignoring '#pragma no_side_effect ' [-Wunknown-pragmas]
Those pragmas are specific to the Studio compilers but unrelated to the
testcase, so this patch disables the warnings on Solaris.
Tested on i386-pc-solaris2.11, sparc-sun-solaris2.11, and
x86_64-pc-linux-gnu.
2026-03-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
gcc/testsuite:
PR testsuite/123135
* g++.dg/warn/Wstringop-overflow-6.C (dg-additional-options): Add
-Wno-unknown-pragmas on Solaris.
The libgomp.fortran/uses_allocators-7.f90 test has been UNRESOLVED from
the beginning:
UNRESOLVED: libgomp.fortran/uses_allocators-7.f90 -O compilation failed to produce executable
The compilation is expected to fail, so this must be changed into a
compile test.
Tested on i386-pc-solaris2.11.
2026-03-13 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
libgomp:
PR libgomp/123177
* testsuite/libgomp.fortran/uses_allocators-7.f90: Change to
compile test.
(dg-message): Adjust line numbers.
(dg-bogus): Likewise.
Without -fno-short-enums, some targets might have a different size for
enums than regular `int`, resulting in unpredictable name mangling.
gcc/testsuite/ChangeLog:
* g++.dg/reflect/mangle1.C: Add -fno-short-enums.
Signed-off-by: Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
In the 'RTL SSA Access Lists' subsection of
GCC's documentation, an example of code with
an 'if'...'else' structure is given. The
given list of the full list of accesses is
said to include "use of the ebb4's R phi
definition of R by B". That cannot be true
because B is in the first substatement of the
'if' statement, whereas ebb4 is the second
substatement. First and second substatements
are mutually exclusive: there is no path to
B that goes through the degenerate phi node
of ebb4.
It looks as though D was intended, not B.
gcc/ChangeLog:
* doc/rtl.texi: Fix example of the SSA form.
Previously code for the error message about VMAT_ELEMENTWISE and
VMAT_STRIDED_SLP could never be reached because it was dominated by
the logic for the unsupported access type for masked load.
This change reverses the two checks so that the more specific case
comes first.
gcc/ChangeLog:
* tree-vect-stmts.cc (vectorizable_load): Swap order of failure
message checks.
When structure types are completed and are variably-modified, we
need to make sure that C_TYPE_VARIABLY_MODIFIED is updated also
for derived types. For derived types which are not updated and
remain TYPE_STRUCTURAL_EQUALITY_P, we do recursion when checking
the bit instead. This change then fixes also a weird corner
case when forming composite types of mutually recursively
defined types.
We do not update other structure or union type that end up
with variably modified pointers, as it is not clear this is
needed and also needs further analysis.
PR c/123424
gcc/c/ChangeLog:
* c-decl.cc (c_update_variably_modified): New function.
(finish_struct): Call c_update_variably_modified.
* c-tree.h (c_variably_modified_p): Recurse if necessary.
* c-typeck.cc (c_verify_type): Update.
(c_set_type_bits): Use c_variably_modified_p.
(mark_decl_used): Dito.
(build_function_call_vec): Dito.
(c_build_qualified_type): Dito.
* c-objc-common.cc (c_var_p): Dito.
gcc/testsuite/ChangeLog:
* gcc.dg/gnu23-tag-composite-7.c: New test.
* gcc.dg/gnu23-varmod-3.c: New test.
* gcc.dg/pr123424.c: New test.
This is an old regression present on all active branches for a quite unusual
setup where a generic package is declared and instantiated in a subunit, and
which comes from a thinko in the Has_Body.Find_Body routine of Sem_Elab.
gcc/ada/
PR ada/124607
* sem_elab.adb (Has_Body.Find_Body): Fix thinko.
gcc/testsuite/
* gnat.dg/generic_inst20.ads, gnat.dg/generic_inst20.adb: New test.
* gnat.dg/generic_inst20-sub.adb: New helper.
This comes from a confusion in the mapping maintained between formal and
actual parameters of the instantiation caused by the equivalent mapping
maintained for the instantiation of the formal package. The change just
removes the offending lines, which do not seem to serve any useful purpose.
gcc/ada/
PR ada/124606
* sem_ch12.adb (Find_Actual_Type): Rename formal parameter.
(Map_Formal_Package_Entities): Do not register base types.
gcc/testsuite/
* gnat.dg/generic_inst19.adb: New test.
The following fixes the reported compile-time regression after the
PRE change to properly track reference expressions in r16-8128-g5f1024922d3f67.
The issue is that the VN reference lookup code performs alias walks and
those are made quadratic for the transition from one VUSE to another
when that involves multiple translation steps. The error was to make
the VUSE part of the expression, but it is really part of the value.
So the following patch reverts this change, fixing the issue.
PR tree-optimization/124599
* tree-ssa-pre.cc (compute_avail): Make the VUSE in the
PRE reference expression part of the value again.
(phi_translate_1): Likewise.
Andreas mentioned in the PR that {gas,gnu_ld}_flag vars should be only
ever used in configure.ac before . config.gcc and I agree with that.
config.gcc initializes the {gas,gnu_ld} vars to {gas,gnu_ld}_flag, then
for some triplets has gas=yes and/or gnu_ld=yes and in some rare cases
(for targets which basically require people to use --with{,out}-gnu-{as,ld})
x$gas = xyes/x$gnu_ld = xyes/x$gas != xyes/x$gnu_ld != xyes tests.
So, the following patch uses just what comes up from config.gcc from that
point onward (so, the user requested setting (which can be now empty if
unknown) possibly overridden by triplet). With the exception of solaris
(which does what it did before PR124547, i.e. always probe linker and
assembler) and the fallback case of checking gas/ld version if still unset
(but that in this patch only affects targets which don't override it
by triplet, i.e. targets which typically have a choice between GNU binutils
and some proprietary ones).
2026-03-23 Jakub Jelinek <jakub@redhat.com>
PR bootstrap/124547
* configure.ac: After config.gcc inclusion use or set only gas instead
of gas_flag and gnu_ld instead of gnu_ld_flag.
* config.gcc: Document possibility of gas_flag or gnu_ld_flag being
empty.
* acinclude.m4 (gcc_GAS_FLAGS): Use gas instead of gas_flag.
(gcc_AC_INITFINI_ARRAY): Likewise. Use gnu_ld instead of gnu_ld_flag.
* configure: Regenerate.
Large structs passed by value as function parameters can be replaced by
const references to improve performance.
PR fortran/124259
gcc/fortran/ChangeLog:
* target-memory.cc (gfc_merge_initializers): Change struct
gfc_typespec argument into a const reference.
* target-memory.h (gfc_merge_initializers): Adjust prototype.
* trans.cc (gfc_finalize_tree_expr): Change struct symbol_attribute
argument into a const reference.
* trans.h (gfc_finalize_tree_expr): Adjust prototype.
Use a compile-time usability probe for the process-shared pthread API
required by caf_shmem and only build libcaf_shmem when that probe succeeds.
caf_shmem needs process-shared pthread primitives that are visible through
the default pthread headers. AX_PTHREAD already established the baseline
pthread flags; this additional check only verifies that the specific
process-shared API surface is usable for caf/shmem/thread_support.c.
PR fortran/124512
libgfortran/ChangeLog:
* Makefile.am: Adjust the dependencies.
* Makefile.in: Regenerate.
* configure: Regenerate.
* configure.ac: Add new check.
Signed-off-by: Christopher Albert <albert@tugraz.at>
This adds the testcase for the eh edge removal with va_end folding.
PR target/124597
gcc/testsuite/ChangeLog:
* g++.target/pr124597.C: New test.
Signed-off-by: Andrew Pinski <andrew.pinski@oss.qualcomm.com>
