[PATCH] D71469: [AArch64] Add sq(r)dmulh_lane(q) LLVM IR intrinsics

sanwou01 marked 2 inline comments as done. sanwou01 added a comment. Thanks Eli. > The "trick" is something like the following? > [...] Yeah, that's exactly right. Your assessment of the options (dedicated pass, "volatile") matches our thinking as well. I'll update the commit message to make

[PATCH] D71469: [AArch64] Add sq(r)dmulh_lane(q) LLVM IR intrinsics

efriedma added a comment. > This makes it impossible to do a neat trick when using NEON intrinsics: one > can load a number of constants using a single vector load, which are then > repeatedly used to multiply whole vectors by one of the constants. This trick > is used for a nice performance

[PATCH] D71469: [AArch64] Add sq(r)dmulh_lane(q) LLVM IR intrinsics

sanwou01 added a comment. ping? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D71469/new/ https://reviews.llvm.org/D71469 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D71469: [AArch64] Add sq(r)dmulh_lane(q) LLVM IR intrinsics

sanwou01 created this revision. Herald added subscribers: llvm-commits, cfe-commits, jdoerfert, hiraditya, kristof.beyls. Herald added projects: clang, LLVM. sanwou01 added reviewers: SjoerdMeijer, dmgreen, t.p.northover. Currently, sqdmulh_lane and friends from the ACLE (implemented in