Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang |

[ Thread Index | Date Index | More lists.tuxfamily.org/eigen Archives ]

*To*: eigen@xxxxxxxxxxxxxxxxxxx*Subject*: Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang*From*: Edward Lam <edward@xxxxxxxxxx>*Date*: Thu, 8 Feb 2018 16:19:10 -0500*Dkim-signature*: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sidefx.com; s=google; h=subject:to:references:from:message-id:date:user-agent:mime-version :in-reply-to:content-language:content-transfer-encoding; bh=2KZzaoCx4k9q102Obo3Ju2B80I99kSLaR7cknwYSaMM=; b=cRYLLgMYXDcjO01RJpE7zL+GCSgjnxvfeoc2xZcSNS9Xn9O2guOCTBx4qHRTCNtatW VaErpA7wSxPtSltmTEh2JvJwDtupKHlKZu2+pMkvdkzk0FBNVxfhQdNrDFdki39bZrKg ZJmGN/Aof76Rkc3GrrQJsFyY1JtO6Y9x9hHQgwx5D8Z0Znpmx+kFEQQkdFVTCnI+N8J3 EfZ7NtMI/GSTCZpFRpExnqpcABYCB4Xb9K5GGPNHuZ7VgpgJdochH1GHtbd0AN8Ta2O6 bzxuRabEjwLk5j5yNqp3sVaA6FP4NOutCobOJxqwjpoPInPvG5NTqnSTLoW2g4vf6Ky1 VqCw==

Hi Patrik, On 2/8/2018 3:08 PM, Patrik Huber wrote: >

I think this is incorrect though. I thin /fp:fast is not needed for MSVC togenerate FMA code. Also, gcc and clang can generate FMA code without -ffast-math(which I guess is sort-of equivalent to /fp:fast).

========= //foo.cpp // // Test with: cl /Fa /O2 /arch:AVX2 /fp:fast foo.cpp // Generates foo.exe and foo.asm float mul_add(float a, float b, float c) { return a*b + c; } int main() { return 0; } ========= Best regards, -Edward

**Follow-Ups**:**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang***From:*Patrik Huber

**References**:**[eigen] Matrix multiplication much slower on MSVC than on g++/clang***From:*Patrik Huber

**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang***From:*Gael Guennebaud

**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang***From:*Edward Lam

**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang***From:*Christoph Hertzberg

**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang***From:*Edward Lam

**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang***From:*Edward Lam

**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang***From:*Patrik Huber

**Messages sorted by:**[ date | thread ]- Prev by Date:
**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang** - Next by Date:
**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang** - Previous by thread:
**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang** - Next by thread:
**Re: [eigen] Matrix multiplication much slower on MSVC than on g++/clang**

Mail converted by MHonArc 2.6.19+ | http://listengine.tuxfamily.org/ |