WebAssembly's `memory.fill` and `memory.copy` instructions trap if the pointers are out of bounds, even if the length is zero. This is different from LLVM, which expects that it can call `memcpy` on arbitrary invalid pointers if the length is zero. To avoid spurious traps, branch around `memory.fill` and `memory.copy` when the length is zero. --------- Co-authored-by: Heejin Ahn <aheejin@gmail.com>
These commits are causing a test-suite build failure on AIX. Revert for now for time to investigate. https://lab.llvm.org/buildbot/#/builders/214/builds/5779/steps/9/logs/stdio This reverts commit bd87a2449da0c82e63cebdf9c131c54a5472e3a7 and 4c72266830ffa332ebb7cf1d3bbd6c56d001fa0f.
This function was added for ARM targets, but aligning global/stack pointer arguments passed to memcpy/memmove/memset can improve code size and performance for all targets that don't have fast unaligned accesses. This adds a generic implementation that adjusts the alignment to pointer size if unaligned accesses are slow. Review D134168 suggests that this significantly improves performance on synthetic benchmarks such as Dhrystone on RV32 as it avoids memcpy() calls. Reviewed By: efriedma Differential Revision: https://reviews.llvm.org/D134282
Accompanying https://reviews.llvm.org/D125728, this updates LLVM Codegen's "generic" CPU to enable the same new features. Differential Revision: https://reviews.llvm.org/D125729
This reverts commit 11afbf396e10e1b1e91a5991e2aec1916e29a910. There are 10 tests still failing after follow-up fix b5d0bf9b9853, this should get the following bots back to green: - https://lab.llvm.org/buildbot/#/builders/183/builds/8194 - https://lab.llvm.org/buildbot/#/builders/186/builds/9491 - https://lab.llvm.org/buildbot/#/builders/214/builds/3908 - https://lab.llvm.org/buildbot/#/builders/93/builds/11740 - https://lab.llvm.org/buildbot/#/builders/231/builds/4200 - https://lab.llvm.org/buildbot/#/builders/121/builds/24519 - https://lab.llvm.org/buildbot/#/builders/230/builds/4466 - https://lab.llvm.org/buildbot/#/builders/94/builds/11639 - https://lab.llvm.org/buildbot/#/builders/45/builds/9325 - https://lab.llvm.org/buildbot/#/builders/124/builds/5219 - https://lab.llvm.org/buildbot/#/builders/67/builds/8623 - https://lab.llvm.org/buildbot/#/builders/123/builds/13836 - https://lab.llvm.org/buildbot/#/builders/109/builds/49355 - https://lab.llvm.org/buildbot/#/builders/58/builds/27751 - https://lab.llvm.org/buildbot/#/builders/117/builds/9922 - https://lab.llvm.org/buildbot/#/builders/16/builds/37012 - https://lab.llvm.org/buildbot/#/builders/104/builds/9490 - https://lab.llvm.org/buildbot/#/builders/42/builds/7725 - https://lab.llvm.org/buildbot/#/builders/196/builds/20077 - https://lab.llvm.org/buildbot/#/builders/3/builds/15217 - https://lab.llvm.org/buildbot/#/builders/6/builds/15251 - https://lab.llvm.org/buildbot/#/builders/9/builds/15247 - https://lab.llvm.org/buildbot/#/builders/36/builds/26487 - https://lab.llvm.org/buildbot/#/builders/54/builds/2474 - https://lab.llvm.org/buildbot/#/builders/74/builds/14536 - https://lab.llvm.org/buildbot/#/builders/5/builds/28555
Accompanying https://reviews.llvm.org/D125728, this updates LLVM Codegen's "generic" CPU to enable the same new features. Differential Revision: https://reviews.llvm.org/D125729
The data layout strings do not have any effect on llc tests and will become misleadingly out of date as we continue to update the canonical data layout, so remove them from the tests. Differential Revision: https://reviews.llvm.org/D105842
Summary: Large slowdowns were observed in Rust due to many small, constant sized copies in conjunction with poorly-optimized memory.copy implementations. Since memory.copy cannot be expected to be inlined efficiently by engines at this time, stop using it for the smallest copies. We continue to lower all memcpy intrinsics to memory.copy, though. Reviewers: aheejin, alexcrichton Subscribers: dschuff, sbc100, jgravelle-google, hiraditya, JDevlieghere, sunfish, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D67639 llvm-svn: 372275
Summary: Uses the named operands tablegen feature to look up the indices of offset, address, and p2align operands for all load and store instructions. This replaces brittle, incorrect logic for identifying loads and store when eliminating frame indices, which previously crashed on bulk-memory ops. It also cleans up the SetP2Alignment pass. Reviewers: aheejin, dschuff Subscribers: sbc100, jgravelle-google, hiraditya, sunfish, jfb, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D59007 llvm-svn: 355770