GNU Tools Weekly News Week 57 (September 27, 2026) Note there will be no newsletter next week due to the cauldron.
Release updates for GNU toolchain: * GDB 18.1.0 was released * https://inbox.sourceware.org/gdb/[email protected]/ GNU toolchain conference reminders: * GNU Tools Cauldron 2026 * This Fri-Sun, October 2-4 (Prague, Czechia) * https://gcc.gnu.org/wiki/cauldron2026 * https://conf.gnu-tools-cauldron.org/prg26/cfp * Toolchains Track at LPC 2026, October 7th (Prague, Czechia) * https://lpc.events/event/20/abstracts/ General/big GNU toolchain news (including sourceware news): * GCC easy issue to tackle of the week: * Missed optimization in math expression: optimize float / in some cases * https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83715 * Hint, this is like previous one: https://gcc.gnu.org/PR126745; though slightly more involved due to float rules * Reach out to Andrea Pinski <[email protected]> for mentoring on this issue. * Previous ones are listed on https://gcc.gnu.org/wiki/EasyIssuesOfTheWeek GCC commits: * https://gcc.gnu.org/r17-4503 ... r17-4705 - 203 commits * match: Fold redundant integer-float-integer conversions * https://gcc.gnu.org/r17-4503-gd0e28326d4254e * match: fold extract-bitfield-and-shift expressions * https://gcc.gnu.org/r17-4511-g885ad73e630ca2 * aarch64: Improve isnan, isfinite, isinf and isnormal and fpclassify * https://gcc.gnu.org/r17-4523-ge8fb8ba77cf7a1 * https://gcc.gnu.org/r17-4524-g52330debcdf1fc * https://gcc.gnu.org/r17-4525-g5cc9ef64063c45 * match: fold abs(x) comparisons with x * https://gcc.gnu.org/r17-4549-g59c8c4bbac4657 * Improve libstdc++ rege * https://gcc.gnu.org/r17-4552 https://gcc.gnu.org/r17-4553 * https://gcc.gnu.org/r17-4554 https://gcc.gnu.org/r17-4555 * AArch64: Turn off outline atomics with -mcmodel=large * https://gcc.gnu.org/r17-4655-gee5bf81d7816b0 * simplify-rtx: [SU]MUL_HIGHPART RTL simplifications * https://gcc.gnu.org/r17-4656-gd382f953805cb0 * C++26 P3769R1: Clarification of placement new deallocation as a DR * https://gcc.gnu.org/r17-4658-geed9f52920aa36 * range-ops: derive operand equality from a zero MINUS_EXPR * https://gcc.gnu.org/r17-4663-gbddea69ee406a0 * match.pd: Fold (a != 0) | (popcount(a) == n) and related patterns * https://gcc.gnu.org/r17-4693-g56efa4a739786b * vect: Look through promotions for SAT_TRUNC * https://gcc.gnu.org/r17-4697-g19378ecd385156 * middle-end: fold VEC_UNPACK_LO/_HI over poly constants * https://gcc.gnu.org/r17-4593-g1743eb860b739c GCC discussion: * GCC bugzilla stats * 123 new issues filed * 63 issues closed binutils/gdb commits: * LoongArch: Enable shared library and PIE support for loongarch*-elf * https://sourceware.org/cgit/binutils-gdb/commit/?id=196286d25ccc652dc086ae173d1cfa836bed0618 * RISC-V: Enable shared library and PIE support for riscv*-elf * https://sourceware.org/cgit/binutils-gdb/commit/?id=b73d23cccf5568321cabaad3a8ec5cddafdef4ec * RISC-V: Add optional register argument to tail pseudo * https://sourceware.org/cgit/binutils-gdb/commit/?id=441953ac9d0b2902be92906817c68feadccd190f gdb discussion: * binutils discussion: * glibc commits: * math: Sync lgamma, acosf, exp10m1f, and tgamma with CORE-MATH * https://sourceware.org/cgit/glibc/log/?id=90156a65a6318b2e1a5ef44501f50fe87db1b601&qt=range&q=90156a65a6318b2e1a5ef44501f50fe87db1b601...1555d46b08dba3a9abbcd956237108b3072f4290%5E * AArch64: Add SVE2 memchr * https://sourceware.org/cgit/glibc/commit/?id=3edeb81021feff11a3d0e934d68446f6a67dd1a3 * glibc discussion: * segfault after reallocation (Was: malloc: Change mmap chunk layout) * https://inbox.sourceware.org/libc-alpha/armngyyT1otFqpSq@end/
