Re: AVX-512 for LAME: code to try, if you have the hardware
Tobias Damisch via Lame-dev <[email protected]> Sat, 25 Jul 2026 22:57:45 +0200
| Newsgroups | gmane.comp.audio.mp3.lame |
|---|---|
| Message-ID | <[email protected]> |
SGV5LAoKSSBoYXBwZW4gdG8gYmUgdGhlIGx1Y2t5IG93bmVyIG9mIGEgWmVuIDQgbWFjaGluZS4K CkNQVTogQU1EIFJ5emVuIDkgNzk1MFgzRCAoMzIpIEAgNS43R0h6Ck9TOiBMaW51eCBNaW50IHRy aXhpZS9zaWQgKHplbmEpIHg4Nl82NAoKRGlkIFN0YWdlIDEuIEFmdGVyIGluc3RhbGxpbmcgbGli Y21vY2thLWRldiwgbGlibXBnMTIzLWRldiBpdCBkaWQgYWN0dWFsbHkKY29uZmlndXJlIGFuZCBj b21waWxlLCBpbmNsdWRpbmcgdGhlIGxpbmVzIHlvdSB3cm90ZSB0byBsb29rIGZvciwKYW5kIG1h a2UgY2hlY2sgcHJvZHVjZWQ6CgpQQVNTOiB0ZXN0X3Nhbml0eQpQQVNTOiB0ZXN0X2dldF9hdWRp b19haWZmClBBU1M6IHRlc3RfcGFyc2VfcGF0aGxlbgpQQVNTOiB0ZXN0X21lcmdlX2FyZ3YKUEFT UzogdGVzdF9pZDN0YWcKUEFTUzogdGVzdF9wYXJzZV9pZDMKUEFTUzogdGVzdF9lbmNvZGVfbm9u ZmluaXRlClBBU1M6IHRlc3RfZW5jb2RlX21vbm9faW50ZXJsZWF2ZWQKUEFTUzogdGVzdF90b3Rh bGZyYW1lcwpQQVNTOiB0ZXN0X3NldF9nZXQKUEFTUzogdGVzdF9jaG9vc2VfdGFibGVfdmVjdG9y ClBBU1M6IHRlc3RfcXVhbnRpemVfbGluZXNfdmVjdG9yClBBU1M6IHRlc3RfY2FsY19zZmJfbm9p c2VfdmVjdG9yCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT0KVGVzdHN1aXRlIHN1bW1hcnkgZm9yIGxhbWUg NC4xCj09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT0KIyBUT1RBTDogMTMKIyBQQVNTOiAgMTMKIyBTS0lQOiAg MAojIFhGQUlMOiAwCiMgRkFJTDogIDAKIyBYUEFTUzogMAojIEVSUk9SOiAwCj09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT0KCkZvciAxYiBJIGhhZCB0byBhY3R1YWxseSBzdXBwbHkgYSB3YXYgZmlsZSB0byBw cm9kdWNlIG1lYW5pZ2Z1bCBvdXRwdXQuClRoZW4gaXQgZGlkIHNob3c6CkNQVSBmZWF0dXJlczog TU1YLCBTU0UsIFNTRTIsIEFWWDIsIEFWWC01MTIKQnV0IGRpZCBOT1Qgc2hvdyBBVlgtNTEyIHVu ZGVyICJ2ZWN0b3Igcm91dGluZXMiOgp2ZWN0b3Igcm91dGluZXM6IEFWWDIKClRoaXMgZGlzY291 cmFnZWQgbWUgZnJvbSBwcm9jZWVkaW5nIGZ1cnRoZXIuCkkgaG9wZSB0aGlzIHRlc3QgaXMgc29t ZWhvdyBoZWxwZnVsIGZvciB5b3UgYW55d2F5cyEKCkNoZWVycywKICAgICAgICAgIFRvYmlhcwoK CgpBbSAyNS4wNy4yNiB1bSAyMToxMyBzY2hyaWViIEFsZXhhbmRlciBMZWlkaW5nZXIgdmlhIExh bWUtZGV2Ogo+IEhpLAo+IAo+IFRoZXJlIGlzIGRlbWFuZCBmb3IgQVZYLTUxMiBvcHRpbWl6YXRp b25zLiBUaGUgQVZYMiB3b3JrIGluIDQuMSBhbnN3ZXJlZCAKPiBoYWxmIG9mIHRoYXQ7IHRoaXMg aXMgYW4gYXR0ZW1wdCBhdCB0aGUgb3RoZXIgaGFsZiwgYW5kIGl0IG5lZWRzIHBlb3BsZSAKPiB3 aXRoIHRoZSByaWdodCBoYXJkd2FyZSwgYmVjYXVzZSB3ZSBkbyBub3QgaGF2ZSBhbnkuCj4gCj4g VGhlcmUgaXMgYSB0ZXN0IGtpdCBhdAo+ICDCoMKgwqAgaHR0cHM6Ly93aWtpLkxlaWRpbmdlci5u ZXQvbGFtZS1hdng1MTItdGVzdGtpdC50YXIuZ3ogKH44NzBrYikKPiBJdCBwYXRjaGVzIHRydW5r IHI2NjYyIGFuZCBhZGRzIGFuIEFWWC01MTIgdGllciB0byB0aGUgdmVjdG9yIGxhZGRlciAKPiB0 aGF0IGFscmVhZHkgcGlja3MgYmV0d2VlbiBzY2FsYXIsIFNTRTIgYW5kIEFWWDIgYXQgcnVuIHRp bWUuIFR3byAKPiBrZXJuZWxzIGluIHRoZSBmaXJzdCBzdGFnZSAtIHRoZSBIdWZmbWFuIHRhYmxl IHNlYXJjaCBhbmQgdGhlIHhyXigzLzQpIAo+IHF1YW50aXphdGlvbiBsb29wIC0gYW5kIHRoZSBW QlIgYmFuZC1ub2lzZSByb3V0aW5lIGluIGEgc2Vjb25kLgo+IAo+IFRoZSBxdWVzdGlvbiBpcyBv bmx5IHRoaXM6Cj4gCj4gIMKgwqDCoCBJcyBpdCBhY3R1YWxseSBmYXN0ZXIsIG9uIHJlYWwgaGFy ZHdhcmUsIGVuZCB0byBlbmQ/Cj4gCj4gV2UgY2Fubm90IGFuc3dlciBpdC4gTm8gbWFjaGluZSBo ZXJlIGhhcyBBVlgtNTEyLCBhbmQgd2UgZGlkIG5vdCB1c2UgYW4gCj4gZW11bGF0b3IsIHNvIC0g YW5kIHRoaXMgaXMgdGhlIHBhcnQgd29ydGggcmVhZGluZyB0d2ljZSAtCj4gCj4gIMKgwqDCoCBU SElTIENPREUgSEFTIE5FVkVSIEJFRU4gRVhFQ1VURUQuIE5PVCBPTkNFLCBBTllXSEVSRS4KPiAK PiBJdCBoYXMgYmVlbiBjb21waWxlZCBieSBmb3VyIGNvbXBpbGVycywgZGlzYXNzZW1ibGVkIGFu ZCByZWFkIAo+IGluc3RydWN0aW9uIGJ5IGluc3RydWN0aW9uLCBjaGVja2VkIGFnYWluc3QgdGhl IHNjYWxhciBjb2RlIGl0IHJlcGxhY2VzLCAKPiBhbmQgd2lyZWQgaW50byB0aGUgdW5pdCB0ZXN0 IHN1aXRlLiBCdXQgdGhlIGZpcnN0IHRpbWUgdGhlc2UgCj4gaW5zdHJ1Y3Rpb25zIGFjdHVhbGx5 IHJ1biBvbiBhIHByb2Nlc3NvciB3aWxsIGJlIG9uIHNvbWVvbmUgZWxzZSdzIAo+IG1hY2hpbmUu IEEgY3Jhc2ggb3IgYSBmYWlsaW5nIHRlc3QgaXMgYSB1c2VmdWwgcmVzdWx0IGFuZCB3ZSB3b3Vs ZCAKPiByYXRoZXIgaGVhciBhYm91dCBpdCB0aGFuIG5vdDsgcGxlYXNlIGRvIG5vdCBydW4gaXQg b24gYW55dGhpbmcgeW91IGNhcmUgCj4gYWJvdXQuCj4gCj4gUmVsZXZhbnQgaGFyZHdhcmUgaXMg QU1EIFplbiA0IG9yIFplbiA1LCBvciBJbnRlbCBYZW9uIGZyb20gU2t5bGFrZS1TUCAKPiBvbndh cmQuIEN1cnJlbnQgSW50ZWwgZGVza3RvcCBwYXJ0cyBoYXZlIG5vIEFWWC01MTIgYXQgYWxsLCBh bmQgd2lsbCBub3QgCj4gdW50aWwgTm92YSBMYWtlLiBJZiB5b3UgaGF2ZSBvbmUgb2YgdGhlIGFi b3ZlIGFuZCB0d2VudHkgbWludXRlcywgdGhlIAo+IFJFQURNRSBpbiB0aGUgdGFyYmFsbCB3YWxr cyB0aHJvdWdoIGl0IGluIHN0YWdlczsgdGhlIGZpcnN0IHN0YWdlIGFsb25lIAo+IGlzIGEgdXNl ZnVsIGFuc3dlciwgYW5kICJubyBkaWZmZXJlbmNlIG9uIFplbiA1IiBpcyBhcyB1c2VmdWwgYXMg YSBzcGVlZHVwLgo+IAo+IFdoYXQgaXMgYWxyZWFkeSBkb25lLCBzbyBub2JvZHkgcmVwZWF0cyBp dDogYnVpbGRzIGNsZWFuIHdpdGggMCB3YXJuaW5ncyAKPiBvbiBnY2MgMTMsIGNsYW5nIDE4LCBj bGFuZyAyMSBhbmQgTVNWQyAxOS41MSwgb24gTGludXgsIEZyZWVCU0QgYW5kIAo+IFdpbmRvd3M7 IGFwcGxpZXMgdG8gYSBmcmVzaCByNjY2MiBjaGVja291dCBhbmQgYnVpbGRzIHdpdGggYXV0b3Rv b2xzIAo+IHJlbW92ZWQgZnJvbSBQQVRIICh0aGUgcmVnZW5lcmF0ZWQgYnVpbGQgZmlsZXMgc2hp cCBpbiB0aGUga2l0LCBzbyAKPiBhdXRvY29uZi9hdXRvbWFrZSBhcmUgbm90IG5lZWRlZCk7IGFu ZCBvbiBhIG1hY2hpbmUgd2l0aG91dCBBVlgtNTEyIHRoZSAKPiBvdXRwdXQgaXMgYnl0ZS1pZGVu dGljYWwgdG8gdHJ1bmsgYWNyb3NzIDUxOSBlbmNvZGVyIG9wdGlvbiBzZXRzLCBzbyB0aGUgCj4g bmV3IHJ1bmcgZG9lcyBub3QgZGlzdHVyYiB0aGUgb25lcyBiZWxvdyBpdC4KPiAKPiBUd28gdGhp bmdzIGFib3V0IGhvdyBpdCBpcyBidWlsdDoKPiAKPiAtIEl0IGlzIGNvbXBpbGVkIG9uIGFscGhh IHZlcnNpb25zIG9ubHksIGFuZCBsZWF2ZXMgdGhlIGJ1aWxkIGJ5IGl0c2VsZiAKPiB3aGVuCj4g IMKgIHRoZSB2ZXJzaW9uIHR5cGUgbW92ZXMgb24uIFRoZXJlIGlzIG5vIGNvbmZpZ3VyZSBzd2l0 Y2gsIGRlbGliZXJhdGVseTogYQo+ICDCoCB0aWVyIHlvdSBoYXZlIHRvIG9wdCBpbnRvIGlzIGEg dGllciBub2JvZHkgbWVhc3VyZXMuIFdpdGhpbiBhbiBhbHBoYSAKPiBidWlsZAo+ICDCoCBpdCBi ZWhhdmVzIGxpa2UgdGhlIG90aGVycyAtIHdoYXQgdGhlIGNvbXBpbGVyIGNhbiBidWlsZCBpcyBi dWlsdCwgCj4gdGhlIENQVQo+ICDCoCBkZWNpZGVzIGF0IHJ1biB0aW1lLCBhbmQgLS1ub2FzbSBh dng1MTIgdHVybnMgaXQgb2ZmLgo+IAo+IC0gVGhlIGZpcnN0IHN0YWdlIGNhbm5vdCBjaGFuZ2Ug dGhlIGVuY29kZWQgb3V0cHV0LCBhbmQgdGhlIGtpdCBjaGVja3MgCj4gdGhhdC4KPiAgwqAgVGhl IHNlY29uZCBzdGFnZSBpcyBtb3JlIGludGVyZXN0aW5nOiB0aGUgYmFuZC1ub2lzZSByb3V0aW5l IGVuZHMgaW4gCj4gYSBzdW0KPiAgwqAgb2Ygc3F1YXJlcywgYW5kIHN1bW1pbmcgc2l4dGVlbiB2 YWx1ZXMgYXQgb25jZSBtZWFucyBzdW1taW5nIHRoZW0gaW4gYQo+ICDCoCBkaWZmZXJlbnQgb3Jk ZXIsIHdoaWNoIGNoYW5nZXMgdGhlIGxhc3QgYml0IGFuZCB0aGVyZWZvcmUgdGhlIGJpdHN0cmVh bS4KPiAgwqAgU28gaXQgc2hpcHMgaW4gdHdvIGZvcm1zIC0gb25lIHRoYXQga2VlcHMgdGhlIGV4 aXN0aW5nIG9yZGVyIGFuZCBpcwo+ICDCoCBiaXQtZXhhY3QsIGFuZCBvbmUgdGhhdCBkb2VzIG5v dCBhbmQgaXMgZmFzdGVyLiBIb3cgbXVjaCBmYXN0ZXIgaXMgCj4gZXhhY3RseQo+ICDCoCB3aGF0 IHdlIGRvIG5vdCBrbm93LCBhbmQgaXQgZGVjaWRlcyB3aGV0aGVyIHRoZSBxdWVzdGlvbiBpcyB3 b3J0aCAKPiBoYXZpbmcuCj4gCj4gTm90aGluZyBoZXJlIGlzIGNvbW1pdHRlZC4gSWYgdGhlIG51 bWJlcnMgc2F5IGl0IGlzIG5vdCB3b3J0aCBpdCwgdGhhdCAKPiBpcyBhIHJlc3VsdCBhbmQgdGhl IHRpZXIgZG9lcyBub3QgZ28gaW4uCj4gCj4gUGxlYXNlIHJlcG9ydCBiYWNrOiBDUFUgbW9kZWws IGNvbXBpbGVyIGFuZCB2ZXJzaW9uLCBPUywgdGhlIHR3byAKPiB0aW1pbmdzLCBhbmQgYW55dGhp bmcgdGhhdCBicm9rZS4KPiAKPiBCeWUsCj4gQWxleGFuZGVyLgo+IAoKCgpfX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXwpMYW1lLWRldiBtYWlsaW5nIGxpc3QK TGFtZS1kZXZAbGlzdHMuc291cmNlZm9yZ2UubmV0Cmh0dHBzOi8vbGlzdHMuc291cmNlZm9yZ2Uu bmV0L2xpc3RzL2xpc3RpbmZvL2xhbWUtZGV2Cg==