LLVM Weekly - #669, October 26th 2026
LLVM Weekly - #669, October 26th 2026
=====================================
If you prefer, you can read a HTML version of this email at
<https://llvmweekly.org/issue/669>.
Welcome to the six hundred and sixty-ninth issue of LLVM Weekly, a weekly newsletter (published every Monday) covering developments in LLVM, Clang, and related projects. LLVM Weekly is brought to you by [Alex Bradbury](https://www.linkedin.com/in/alex-bradbury/). Subscribe to future issues at <https://llvmweekly.org> and pass it on to anyone else you think may be interested. Please send any tips or feedback via email: <[email protected]>, or Mastodon: [@[email protected]](https://fosstodon.org/@llvmweekly) / [@[email protected]](https://fosstodon.org/@asb), or Bluesky: [@llvmweekly.org](https://bsky.app/profile/llvmweekly.org) / [@asbradbury.org](https://bsky.app/profile/asbradbury.org).
## News and articles from around the web and events
The calls for presentations for the [LLVM FOSDEM 2027 devroom](https://discourse.llvm.org/t/fosdem-2027-llvm-devroom-call-for-presentations/92597) and the [MLIR FOSDEM 2027 devroom](https://discourse.llvm.org/t/fosdem-2027-mlir-devroom-call-for-presentations/92599) are now up. In both cases, the deadline for submissions is the 22nd November. FOSDEM will take place on 30th and 31st January.
The LLVM Bay Area Monthly Meetup is [taking place today, Monday 26th October](https://discourse.llvm.org/t/llvm-bay-area-monthly-meetup-mon-oct-26-26-6pm/91953).
According to the [LLVM Calendar](https://calendar.google.com/calendar/u/0/[email protected]) in the coming week there will be the following. It's the LLVM Dev Meeting this week (sorry I can't make it this time!) so it's possible some of the scheduled meetings wont take place.
* Office hours with the following hosts: Amara Emerson.
* Online sync-ups on the following topics: Flang, modules, libc++,
LLVM/Offload, HLSL, LLVM memory safety working group, reflection.
* For more details see the [LLVM
calendar](https://calendar.google.com/calendar/u/0/[email protected]),
getting involved documentation on [online sync
ups](https://llvm.org/docs/GettingInvolved.html#online-sync-ups) and [office
hours](https://llvm.org/docs/GettingInvolved.html#office-hours).
## On the forums
* Grigory Pastukhov made an [LLD proposal on mixing small and large code
models](https://discourse.llvm.org/t/rfc-lld-mixing-small-and-large-code-models-demote-large-sections-referenced-from-small-code/92553)
"When small code references a large section through a 32-bit relocation, we
propose to demote that section to a small one. Demotion clears
SHF_X86_64_LARGE and drops the l prefix from the section name, so the linker
places it together with the small sections. This makes it practical to link
translation units (TUs) compiled with the small and large code models into
the same binary."
* Stephen Tozer posted a [public service announcement on planned LLVM debug
info API
changes](https://discourse.llvm.org/t/psa-planned-llvm-debug-info-api-changes/92551).
* Paul Osmialowski would like to [introduce LLVM's mathvec (libc/src/mathvec)
library as another veclib functions
provider](https://discourse.llvm.org/t/rfc-mathvec-introduce-the-llvms-mathvec-library-as-yet-another-veclib-functions-provider/92591).
* Alexander Yermolovich suggests [declaring LLVM's optimisation remark kinds
in TableGen and adding a `!remark` metadata kind that attaches a remark to
the instruction it
describes](https://discourse.llvm.org/t/rfc-tablegen-declared-optimization-remarks-carried-to-codegen-as-remark-metadata/92594).
* A new update on Flang development [is now
available](https://discourse.llvm.org/t/flang-liaison-report-to-j3/68468/14).
* Wontae Choi and Rohan Jacob-Rao started an RFC discussion on [a spatial
safety mode for
C++](https://discourse.llvm.org/t/rfc-spatial-safety-mode-for-c-fspatial-safety/92577)
(`-fspatial-safety`). It would build on top of existing technologies such as
the `-Wunsafe-buffer-usage` checker, UBSan, and libc++ hardening.
* Henry Cheang (of Nintendo) posted an RFC about [introducing Nintendo target
triples to Clang and
LLVM](https://discourse.llvm.org/t/rfc-proposal-to-add-nintendo-target-to-clang-and-llvm/92585).
* LLVM Foundation board meeting minutes for September [were
shared](https://discourse.llvm.org/t/board-meeting-minutes-sep-4-2026/92555).
* LLVM 23.1.4 [was
released](https://discourse.llvm.org/t/llvm-23-1-4-released/92564).
* Matthew Nagy wrote an RFC on [compressing TypeSanitizer shadow memory using
descriptor
indices](https://discourse.llvm.org/t/rfc-compress-typesanitizer-shadow-memory-using-descriptor-indices/92565).
i.e. storing 32-bit indices into lookup tables of type descriptors rather
than 64-bit pointers directly to those descriptors.
* Aaron Ballman and the Clang area team initiated a discussion on [library
layering of Parser and
Sema](https://discourse.llvm.org/t/library-layering-of-parser-and-sema/92573)
noting that for a range of reasons (including language standards committees
increasingly specifying semantics in the form of language rewriting rules) it
may be worth considering if it still makes sense for Parser and Sema to
remain separate. A design discussion will be hosted on this topic and
interested people are requested to indicate availability to help find the
best slot.
* Xiaoyang Gong is seeking input on [a flexible multi-pass algorithm for
branch-range
extension](https://discourse.llvm.org/t/lld-mach-o-rfc-a-flexible-multipass-algorithm-for-branch-range-extension/92527).
This would be used for resolving out-of-range Arm64 branches in LLD's Mach-O
port.
* Raphael Isemann started threads on [the situation with flaky LLDB
tests](https://discourse.llvm.org/t/the-lldb-flakey-test-situation/92550)
and [removing LLDB's GUI
mode](https://discourse.llvm.org/t/rfc-removing-gui-mode/92570).
* Yingwei Zheng shares that [the llvm-ub-aware-interpreter now has an online
'llubi playground' using a Wasm
build](https://discourse.llvm.org/t/rfc-upstreaming-llvm-ub-aware-interpreter/89645/15).
## LLVM commits
* LLVM docs "landing your changes" was updated to reflect the current
workflow.
[3b4061c](https://github.com/llvm/llvm-project/commit/3b4061ce1974).
* The llvm-readobj `--call-graph-info` output format was documented.
[4aecd4c](https://github.com/llvm/llvm-project/commit/4aecd4c9767f).
* The transition to TableGen-based command line option definitions continues.
[6ec4fff](https://github.com/llvm/llvm-project/commit/6ec4fffa79fd).
* LiveDebugVariables was fixed to avoid quadratic rescans of inserted
`DBG_VALUE`s.
[3cfa4fc](https://github.com/llvm/llvm-project/commit/3cfa4fcffa81).
* MachineCombiner support was added for LoongArch.
[59cdc9a](https://github.com/llvm/llvm-project/commit/59cdc9a5e41f).
* Function filtering support was added to `opt-bisect`.
[849bf7e](https://github.com/llvm/llvm-project/commit/849bf7ea0a23).
* The exception model is now read from the module flag.
[64c7493](https://github.com/llvm/llvm-project/commit/64c74930fef2),
[e7ffa95](https://github.com/llvm/llvm-project/commit/e7ffa953f8db).
## Clang commits
* Loop hint lowering was implemented in ClangIR.
[d48c176](https://github.com/llvm/llvm-project/commit/d48c176a87ce).
* The `aarch64-unknown-uefi` target was added.
[91a6410](https://github.com/llvm/llvm-project/commit/91a641049eed).
* Initial Xtensa ABI handling was implemented.
[ebeb26d](https://github.com/llvm/llvm-project/commit/ebeb26d53eaf).
* The C defect report status page was updated.
[87cefb0](https://github.com/llvm/llvm-project/commit/87cefb0b69ac).
## Other project commits
* As part of supporting address sanitizer on AIX, shadow memory is now
supported.
[52f8774](https://github.com/llvm/llvm-project/commit/52f8774eb9e4).
* The LoopTripCountVersioning pass was added to Flang, which will version an
innermost `fir.do_loop` whose trip count is a runtime value likely to equal
a compile-time constant N.
[7bb03d8](https://github.com/llvm/llvm-project/commit/7bb03d8afce3).
* `popen` and `pclose` were implemented in LLVM's libc.
[c759b57](https://github.com/llvm/llvm-project/commit/c759b57a74de).
* LLDB data formatters were added for some libcxx data types.
[62141d5](https://github.com/llvm/llvm-project/commit/62141d57be96).
* LLD's identical code folding (ICF) was sped up.
[6ceff48](https://github.com/llvm/llvm-project/commit/6ceff48b8184),
[86c2b09](https://github.com/llvm/llvm-project/commit/86c2b0936681).