Alex Bradbury via llvm-dev
2018-Jun-11 20:51 UTC
[llvm-dev] LLVM Weekly - #232, June 11th 2018
LLVM Weekly - #232, June 11th 2018 ================================= If you prefer, you can read a HTML version of this email at <http://llvmweekly.org/issue/232>. Welcome to the two hundred and thirty-second 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 <http://llvmweekly.org> and pass it on to anyone else you think may be interested. Please send any tips or feedback to <asb at asbradbury.org>, or @llvmweekly or @asbradbury on Twitter. ## News and articles from around the web The ISO C++ Committee meeting recently took place in Rapperswil, Switzerland. Several committee members have shared a [detailed write-up](https://old.reddit.com/r/cpp/comments/8prqzm/2018_rapperswil_iso_c_committee_trip_report/) on Reddit. The next LLVM Social in Shanghai has been [announced](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123775.html) for July 1st. [Zig](https://ziglang.org/) is an interesting LLVM-based language which has been mentioned in LLVM Weekly previously, and aims to be a better C. The author is moving to working full-time on Zig and is [seeking donations](https://andrewkelley.me/post/full-time-zig.html). ## On the mailing lists * Johannes Doerfert proposes [abstract parallel IR optimisations](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123841.html) in a new RFC. * Hans Wennborg has proposed the [schedule for the 7.0.0 release](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123843.html). This would see the branch taking place on 1st August with RC1 soon after, RC2 coming on the 22nd August, and the final tag on 5th September. * Graham Hunter has shared an [updated RFC](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123780.html) on supporting SIMD instruction sets with variable vector lengths (e.g. SVE). There is also a sub-thread discussing support for the RISC-V vector extension. * Tanya Lattner is [looking](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123781.html) for booth volunteers for the Grace Hopper Conference in Houston in September. * Brian Gesiak [queries](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123769.html) the current state of LLD support for Mach-O. The thread received a number of responses from active LLD developers. * Robin Eklind has created a [BNF grammar for LLVM IR assembly](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123851.html). * Chijun Sima has shared an RFC on [a new dominator tree updater API](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123883.html). * Fedor Sergeev has authored an RFC on [a new pass execution instrumentation interface](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123835.html). This aims to allow the easy implementation of facilities such as `-print-after`, `-opt-bisect-limit`, etc. Importantly, it aims to be compatible with the new pass manager. * Masaki Arai is looking for feedback on an [RFC to port MachinePipeliner to AArch64+SVE](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123884.html). * Nicolai Hähnle is [drawing attention](http://lists.llvm.org/pipermail/llvm-dev/2018-June/123915.html) to recent work which generalizes the "searchable tables" backend of TableGen. ## LLVM commits * An initial set of optimisation remarks has been added to the shrink-wrapping pass. [r333972](https://reviews.llvm.org/rL333972). * The SETCCE SelectionDAG node has been removed in favour of SETCCCARRY. [r333939](https://reviews.llvm.org/rL333939). * The semantics of NAME in TableGen have been streamlined. [r333900](https://reviews.llvm.org/rL333900). * Documentation was added for masked vector expanding load and compressing store intrinsics. [r334075](https://reviews.llvm.org/rL334075). * The 'li' pseudo instruction is now supported by the RISC-V MC assembly parser. [r334203](https://reviews.llvm.org/rL334203). ## Clang commits * Implementation of clang-doc continues with the introduction of the 'reducer' portion of the frontend framework. [r333932](https://reviews.llvm.org/rL333932). * Bittest intrinsics have been implemented generically for non-x86 platforms. [r334059](https://reviews.llvm.org/rL334059). ## Other project commits * libFuzzer gained an initial implementation of `-data_flow_trace`. [r334058](https://reviews.llvm.org/rL334058). * LLD now supports multi-GOT for MIPS. This allows large applications such as LLVM+Clang to be linked. [r334390](https://reviews.llvm.org/rL334390).