It might be worth adding that AVX has been around for a long time - it has been supported by Intel and AMD CPUs since 2011 (see https://en.wikipedia.org/wiki/Advanced_Vector_Extensions). AVX2 came to CPU generations in 2013. You might want to check or guess how many people really run an older CPU. If it is a fairly compute-heavy application, chances are that users won't have much fun with it anyway on older CPUs.
That being said, in particular the Sandy Bridge (e.g. i5-25X0K, i7-2700K) and Ivy Bridge (e.g. i5-3550) were extremely popular CPUs and are probably still widely used. I myself have a i5-3550 and it runs everything perfectly, so I don't have a real reason to upgrade even that 7 year old CPU. So I would not go as far as assuming that the majority of your user's CPUs would support AVX2 - but it might be true for AVX. One useful data point: Check the Steam Hardware Survey https://store.steampowered.com/hwsurvey/, scroll down to "Other Settings". According to that, as of Aug 2020, 93% of Steam users have CPUs supporting AVX, and 77% AVX2. This is likely biased towards gaming computers out there, but should be fairly representative still and I doubt you'll find better data. Best wishes, Patrik On Sat, 19 Sep 2020 at 22:30, Sripathi, Vamsi <[email protected]> wrote: > Another option to consider is to use a Compiler that supports function > multi-versioning based on ISA. For e.g., Intel Compiler has -ax flag – > https://software.intel.com/content/www/us/en/develop/documentation/cpp-compiler-developer-guide-and-reference/top/compiler-reference/compiler-options/compiler-option-details/code-generation-options/ax-qax.html#ax-qax > > > > GCC seems to support this through attribute directives - > https://lwn.net/Articles/691932/ > > > > -Vamsi > > > > *From:* Rob McDonald <[email protected]> > *Sent:* Friday, September 18, 2020 10:19 AM > *To:* [email protected] > *Subject:* Re: [eigen] Vectorization for general use > > > > Thanks for everyone's responses and links. Very helpful. > > > > This seems like it is quite a thorny issue... It really makes using these > advanced features fairly challenging. > > > > I'm not sure that it is practical for me to separate out a shared library > to be selectively loaded (vs. just separate executables). Although the > algorithms may be somewhat contained, the data structures can have quite > wide reach. It isn't obvious how to separate what needs to be compiled > with these flags and what does not (particularly since we didn't design for > this from the start). This is also a case where Eigen being a header-only > library is a bit of a drawback. If it was a traditional compiled library, > it would likely be easier to draw the line at eigen_sse.so, eigen_avx1.so > or whatever. > > > > My project builds with CMake, which isn't very friendly at using different > toolchains for different parts of the project -- or compiling the same part > multiple times. It is possible, but not particularly pretty. > > > > Rob > > > > > > On Thu, Sep 17, 2020 at 11:09 PM William Tambellini <[email protected]> > wrote: > > A solution : > > - do all the math/algos outside the main, in a dynamic libs (.so, > .dll, ...) > - build multiple dyn libs for the ISA you care about (sse.so, avx1.so, > avx2.so, avx512.so, ... ) > - dynamic loading the right lib from the main according to the > features of the current running deployed cpu: ( > https://github.com/google/cpu_features) > - calling your api in the lib from the main to let the backends run > the algo with the best optim > > Now, I have the feeling that the long term solution would be for eigen to > do a minimum of JIT. Example: oneDNN with asmjit : > https://github.com/asmjit/asmjit > > Kind > > W. > > > > <https://www.sdl.com/> > > *Share your* > *feedback with us* <https://www.surveymonkey.com/r/PYF190816> > > > ------------------------------ > > *From:* Edward Lam <[email protected]> > *Sent:* Thursday, September 17, 2020 9:24 PM > *To:* [email protected] <[email protected]> > *Subject:* Re: [eigen] Vectorization for general use > > > > Offhand, I wonder if you could put main() in its own source file and > compile it without any vectorization compiler options, and have that call > your real main() renamed in a different source file that does have > vectorization compiler options enabled. Then your new main() could do CPUID > checks (eg. https://stackoverflow.com/a/4823889 ) and bail out > gracefully. You will of course need to ensure that the CPUID checks are > accurate for your compiler options, which may present its own challenges. > > > > Cheers, > > -Edward > > > > On Thu, Sep 17, 2020 at 10:52 PM Rob McDonald <[email protected]> > wrote: > > I maintain an open source program that uses Eigen. The vast majority of > my users do not compile the program, instead downloading a pre-compiled > binary from our website. About 80% are on Windows, 10% on Mac and 10% on > Linux. I only provide X86 builds, 32 and 64-bit on Windows, 64-bit only on > Mac and Linux. We may eliminate the 32-bit Windows build soon. > > > > Historically, I have compiled with no special flags enabling vectorization > options for the CPU. I would like to pursue this as I expect it will > unlock some nice performance gains. However, I'd like to keep things > simple and compatible for users. > > > > What happens when someone runs a program compiled with vectorization when > their CPU does not support it? If it fails, how graceful is the failure? > > > > Is there a standard approach to identify the capabilities of a given > machine? I could add that to my program and survey users before making a > change... Would such code still run on a machine that was in the process > of failing due to not having support for the built in vectorization? I.e. > if it is crashing, can we send a message as to why we're going down? > > > > Is there a graceful way to support multiple options? > > > > Any tips from other broad use applications is greatly appreciated. > > > > Rob > > > > > > > > Click here > <https://www.mailcontrol.com/sr/IDXDiOSqylnGX2PQPOmvUhe0y89-yNqhZAviLmkDXL06gGw831_8qiYaAxJOEWVK7LHzKdJh-eoDMGoTToeXlw==> > to report this email as spam. > >
