Path: csiph.com!weretis.net!feeder9.news.weretis.net!news.misty.com!news.iecc.com!.POSTED.news.iecc.com!nerds-end From: "Steven G. Kargl" Newsgroups: comp.compilers Subject: Re: Fully integrating the Flang Fortran compiler with standard MLIR Date: Thu, 03 Oct 2024 04:23:14 -0000 Organization: Compilers Central Sender: johnl%iecc.com Approved: comp.compilers@iecc.com Message-ID: <24-10-001@comp.compilers> References: <24-09-008@comp.compilers> MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Injection-Info: gal.iecc.com; posting-host="news.iecc.com:2001:470:1f07:1126:0:676f:7373:6970"; logging-data="59431"; mail-complaints-to="abuse@iecc.com" Keywords: optimize, comment Posted-Date: 03 Oct 2024 12:18:57 EDT X-submission-address: compilers@iecc.com X-moderator-address: compilers-request@iecc.com X-FAQ-and-archives: http://compilers.iecc.com Xref: csiph.com comp.compilers:3600 On Mon, 30 Sep 2024 18:10:06 +0100, John R Levine wrote: > The authors modified Flang, the LLVM Fortran compiler, to use MLIR rather > than its private IR, and found that the code got better, but still not as > good as the Cray compiler. > The paper, like many involving comparisons/benchmarks, is IMHO flawed. The author does not give the options used with the various compilers. The author also used a 3 year old version of gfortran from 2021 and compares it to 2023 versions of flang. I cannot find the release date for Cray Fortran 15.0.0. PS: The Fortran compiler available from LLVM github is named flang-new. There is no compiler named flang. -- steve [In case it wasn't clear, I'm posting links to these draft papers because they look interesting, not necessarily because I think they are brilliant and insightful. Asssessing those is for all you guys. -John]