]> git.ipfire.org Git - thirdparty/gcc.git/commitdiff
Daily bump.
authorGCC Administrator <gccadmin@gcc.gnu.org>
Fri, 7 Feb 2025 00:18:53 +0000 (00:18 +0000)
committerGCC Administrator <gccadmin@gcc.gnu.org>
Fri, 7 Feb 2025 00:18:53 +0000 (00:18 +0000)
gcc/ChangeLog
gcc/DATESTAMP
gcc/fortran/ChangeLog
gcc/testsuite/ChangeLog
libgfortran/ChangeLog

index 89b8a4d6ad154725bec590bbbe7d36d238283215..095de11a11eec2d9e6880337cea5ffd98fcbd122 100644 (file)
@@ -1,3 +1,76 @@
+2025-02-06  Craig Blackmore  <craig.blackmore@embecosm.com>
+
+       * config/riscv/riscv.md: Move UNSPEC_SSP_SET and UNSPEC_SSP_TEST
+       to unspec enum.
+
+2025-02-06  Georg-Johann Lay  <avr@gjlay.de>
+
+       * config/avr/avr.opt.urls: Add mcvt.
+
+2025-02-06  Tamar Christina  <tamar.christina@arm.com>
+
+       PR tree-optimization/118756
+       * tree-ssa-loop-ivopts.cc (contain_complex_addr_expr): Remove.
+
+2025-02-06  Georg-Johann Lay  <avr@gjlay.de>
+
+       PR target/118764
+       * config/avr/avr.opt (-mcvt): New target option.
+       * config/avr/avr-arch.h (AVR_CVT): New enum value.
+       * config/avr/avr-mcus.def: Add AVR_CVT flag for devices that
+       support it.
+       * config/avr/avr.cc (avr_handle_isr_attribute) [TARGET_CVT]: Issue
+       an error when a vector number larger that 3 is used.
+       * config/avr/gen-avr-mmcu-specs.cc (McuInfo.have_cvt): New property.
+       (print_mcu) <*avrlibc_startfile>: Use crt<mcu>-cvt.o depending
+       on -mcvt (or issue an error when the device doesn't support a CVT).
+       * doc/invoke.texi (AVR Options): Document -mcvt.
+
+2025-02-06  Jakub Jelinek  <jakub@redhat.com>
+
+       PR rtl-optimization/117506
+       * loop-iv.cc (get_biv_step_1): For {ZERO,SIGN}_EXTEND
+       of PLUS apply {ZERO,SIGN}_EXTEND to op1.
+
+2025-02-06  Georg-Johann Lay  <avr@gjlay.de>
+
+       PR target/118768
+       * config/avr/genmultilib.awk: Parse the AVR_MCU lines in
+       a more robust way w.r.t. white spaces.
+
+2025-02-06  Lulu Cheng  <chenglulu@loongson.cn>
+
+       PR target/118561
+       * config/loongarch/loongarch-builtins.cc
+       (loongarch_expand_builtin_lsx_test_branch):
+       NULL_RTX will not be returned when an error is detected.
+       (loongarch_expand_builtin): Likewise.
+
+2025-02-06  Richard Sandiford  <richard.sandiford@arm.com>
+
+       PR tree-optimization/110449
+       * tree-ssa-loop-manip.h (insert_iv_increment): Declare.
+       * tree-ssa-loop-manip.cc (insert_iv_increment): New function,
+       split out from...
+       (create_iv): ...here and generalized to gimple_seqs.
+       * tree-vect-loop.cc (vectorizable_induction): Use
+       standard_iv_increment_position and insert_iv_increment
+       to insert the IV increment.
+
+2025-02-06  Richard Biener  <rguenther@suse.de>
+
+       PR rtl-optimization/117922
+       * fold-mem-offsets.cc (pass_fold_mem_offsets::execute):
+       Do nothing for a highly connected CFG.
+
+2025-02-06  Richard Biener  <rguenther@suse.de>
+
+       PR tree-optimization/118749
+       * tree-vect-data-refs.cc (vector_alignment_reachable_p): Pass
+       in the vectorization factor, when that cannot maintain
+       the DRs target alignment do not claim we can reach that
+       by peeling.
+
 2025-02-05  Jeff Law  <jlaw@ventanamicro.com>
 
        * config/bfin/bfin.md (abssi): Disable pattern.
index 8817645f579fd1d6a9bbaf526446e2bcdd5fea47..293639e313776c694793bd1bde48527ebf3e7d9d 100644 (file)
@@ -1 +1 @@
-20250206
+20250207
index f61107593a15f2f89e4d44ce671edb0ec602e113..cc15c1653ab962750ec4a86b6b981ffc6ea12263 100644 (file)
@@ -1,3 +1,11 @@
+2025-02-06  Paul Thomas  <pault@gcc.gnu.org>
+
+       PR fortran/118750
+       * resolve.cc (resolve_assoc_var): If the target expression has
+       a rank, do not use gfc_expression_rank, since it will return 0
+       if the function is elemental. Resolution will have produced the
+       correct rank.
+
 2025-02-05  Tobias Burnus  <tburnus@baylibre.com>
 
        PR fortran/118740
index bd6dc3c324751d24d8bc0995f5b55247afd4401f..6c5febac1a1d338d7d4e2c3846f7eb97ebafd73d 100644 (file)
@@ -1,3 +1,55 @@
+2025-02-06  Jeff Law  <jlaw@ventanamicro.com>
+
+       * gcc.target/riscv/rvv/autovec/madd-split2-1.c: Update expected
+       output.
+
+2025-02-06  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
+
+       PR libfortran/114618
+       * gfortran.dg/pr114618.f90: New test.
+
+2025-02-06  Jakub Jelinek  <jakub@redhat.com>
+
+       PR c++/110345
+       * g++.dg/cpp0x/attr-no_unique_address1.C: New test.
+
+2025-02-06  Jakub Jelinek  <jakub@redhat.com>
+
+       PR c++/110345
+       * g++.dg/cpp0x/attr-noreturn1.C: New test.
+
+2025-02-06  Jakub Jelinek  <jakub@redhat.com>
+
+       PR c++/110345
+       * g++.dg/cpp0x/attr-nodiscard1.C: New test.
+
+2025-02-06  Paul Thomas  <pault@gcc.gnu.org>
+
+       PR fortran/118750
+       * gfortran.dg/associate_72.f90: New test.
+
+2025-02-06  Jakub Jelinek  <jakub@redhat.com>
+
+       PR rtl-optimization/117506
+       * gcc.dg/pr117506.c: New test.
+       * gcc.target/riscv/pr117506.c: New test.
+
+2025-02-06  Lulu Cheng  <chenglulu@loongson.cn>
+
+       PR target/118561
+       * gcc.target/loongarch/pr118561.c: New test.
+
+2025-02-06  Richard Sandiford  <richard.sandiford@arm.com>
+
+       PR tree-optimization/110449
+       * gcc.target/aarch64/pr110449.c: Expect an increment by 8.0,
+       but test that there is no MOV.
+
+2025-02-06  Richard Biener  <rguenther@suse.de>
+
+       PR tree-optimization/118749
+       * gcc.dg/vect/pr118749.c: New testcase.
+
 2025-02-05  Simon Martin  <simon@nasilyan.com>
 
        PR c++/118319
index f65163e0fb6f605f00d33e240f3ac2a8dbaec67c..5237848418f52d7886a2250600ca3e3093b96b99 100644 (file)
@@ -1,3 +1,11 @@
+2025-02-06  Jerry DeLisle  <jvdelisle@gcc.gnu.org>
+
+       PR libfortran/114618
+       * io/transfer.c (formatted_transfer_scalar_write): Change name
+       of vriable 'pos' to 'tab_pos' to improve clarity. Add new
+       variable next_pos when calculating the maximum position.
+       Update the calculation of pending spaces.
+
 2025-02-04  Thomas Koenig  <tkoenig@gcc.gnu.org>
 
        * Makefile.am: Add files for unsigned exponentiation.