Re: AVX-512 for LAME: code to try, if you have the hardware
Alexander Leidinger via Lame-dev <[email protected]> Sun, 26 Jul 2026 00:02:21 +0200
| Newsgroups | gmane.comp.audio.mp3.lame |
|---|---|
| Organization | No organization, this is a private message. |
| Message-ID | <[email protected]> |
QW0gMjAyNi0wNy0yNSAyMjo1Nywgc2NocmllYiBUb2JpYXMgRGFtaXNjaCB2aWEgTGFtZS1kZXY6 Cj4gSGV5LAo+IAo+IEkgaGFwcGVuIHRvIGJlIHRoZSBsdWNreSBvd25lciBvZiBhIFplbiA0IG1h Y2hpbmUuCj4gCj4gQ1BVOiBBTUQgUnl6ZW4gOSA3OTUwWDNEICgzMikgQCA1LjdHSHoKPiBPUzog TGludXggTWludCB0cml4aWUvc2lkICh6ZW5hKSB4ODZfNjQKPiAKPiBEaWQgU3RhZ2UgMS4gQWZ0 ZXIgaW5zdGFsbGluZyBsaWJjbW9ja2EtZGV2LCBsaWJtcGcxMjMtZGV2IGl0IGRpZCAKPiBhY3R1 YWxseQo+IGNvbmZpZ3VyZSBhbmQgY29tcGlsZSwgaW5jbHVkaW5nIHRoZSBsaW5lcyB5b3Ugd3Jv dGUgdG8gbG9vayBmb3IsCj4gYW5kIG1ha2UgY2hlY2sgcHJvZHVjZWQ6Cj4gCj4gUEFTUzogdGVz dF9zYW5pdHkKPiBQQVNTOiB0ZXN0X2dldF9hdWRpb19haWZmCj4gUEFTUzogdGVzdF9wYXJzZV9w YXRobGVuCj4gUEFTUzogdGVzdF9tZXJnZV9hcmd2Cj4gUEFTUzogdGVzdF9pZDN0YWcKPiBQQVNT OiB0ZXN0X3BhcnNlX2lkMwo+IFBBU1M6IHRlc3RfZW5jb2RlX25vbmZpbml0ZQo+IFBBU1M6IHRl c3RfZW5jb2RlX21vbm9faW50ZXJsZWF2ZWQKPiBQQVNTOiB0ZXN0X3RvdGFsZnJhbWVzCj4gUEFT UzogdGVzdF9zZXRfZ2V0Cj4gUEFTUzogdGVzdF9jaG9vc2VfdGFibGVfdmVjdG9yCj4gUEFTUzog dGVzdF9xdWFudGl6ZV9saW5lc192ZWN0b3IKPiBQQVNTOiB0ZXN0X2NhbGNfc2ZiX25vaXNlX3Zl Y3Rvcgo+ID09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PT0KPiBUZXN0c3VpdGUgc3VtbWFyeSBmb3IgbGFtZSA0 LjEKPiA9PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09Cj4gIyBUT1RBTDogMTMKPiAjIFBBU1M6ICAxMwo+ICMg U0tJUDogIDAKPiAjIFhGQUlMOiAwCj4gIyBGQUlMOiAgMAo+ICMgWFBBU1M6IDAKPiAjIEVSUk9S OiAwCj4gPT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09PT09 PT09PT09PT09PT09PT09PT09PT09PT09PQoKVGhpcyBpcyB2ZXJ5IHVzZWZ1bCBvdXRwdXQhIEFs bCB3b3JrcyBhcyBleHBlY3RlZC4gVGhlIG9ubHkgbWlzc2luZyBpbmZvIAphcmUgdGhlIHBlcmZv cm1hbmNlIGZpZ3VyZXMuCgo+IEZvciAxYiBJIGhhZCB0byBhY3R1YWxseSBzdXBwbHkgYSB3YXYg ZmlsZSB0byBwcm9kdWNlIG1lYW5pZ2Z1bCBvdXRwdXQuCj4gVGhlbiBpdCBkaWQgc2hvdzoKPiBD UFUgZmVhdHVyZXM6IE1NWCwgU1NFLCBTU0UyLCBBVlgyLCBBVlgtNTEyCj4gQnV0IGRpZCBOT1Qg c2hvdyBBVlgtNTEyIHVuZGVyICJ2ZWN0b3Igcm91dGluZXMiOgo+IHZlY3RvciByb3V0aW5lczog QVZYMgo+IAo+IFRoaXMgZGlzY291cmFnZWQgbWUgZnJvbSBwcm9jZWVkaW5nIGZ1cnRoZXIuCgpQ bGVhc2UgYWRkCiAgI2luY2x1ZGUgInZlY3Rvci9sYW1lX2ludHJpbi5oIgppbiB1dGlsLmMgYW5k IHJlY29tcGlsZS4KClRoZW4gdGhlIHBlcmYgdGVzdCAoaXQgc2hvdWxkIHNob3cgdGhlIEFWWC01 MTIgaW4gdGhlIHZlY3RvciByb3V0aW5lcyAKbm93KS4KCj4gSSBob3BlIHRoaXMgdGVzdCBpcyBz b21laG93IGhlbHBmdWwgZm9yIHlvdSBhbnl3YXlzIQoKVmVyeSEKClRoYW5rcywKQWxleGFuZGVy LgoKPiBDaGVlcnMsCj4gICAgICAgICAgVG9iaWFzCj4gCj4gCj4gCj4gQW0gMjUuMDcuMjYgdW0g MjE6MTMgc2NocmllYiBBbGV4YW5kZXIgTGVpZGluZ2VyIHZpYSBMYW1lLWRldjoKPj4gSGksCj4+ IAo+PiBUaGVyZSBpcyBkZW1hbmQgZm9yIEFWWC01MTIgb3B0aW1pemF0aW9ucy4gVGhlIEFWWDIg d29yayBpbiA0LjEgCj4+IGFuc3dlcmVkIGhhbGYgb2YgdGhhdDsgdGhpcyBpcyBhbiBhdHRlbXB0 IGF0IHRoZSBvdGhlciBoYWxmLCBhbmQgaXQgCj4+IG5lZWRzIHBlb3BsZSB3aXRoIHRoZSByaWdo dCBoYXJkd2FyZSwgYmVjYXVzZSB3ZSBkbyBub3QgaGF2ZSBhbnkuCj4+IAo+PiBUaGVyZSBpcyBh IHRlc3Qga2l0IGF0Cj4+ICDCoMKgwqAgaHR0cHM6Ly93aWtpLkxlaWRpbmdlci5uZXQvbGFtZS1h dng1MTItdGVzdGtpdC50YXIuZ3ogKH44NzBrYikKPj4gSXQgcGF0Y2hlcyB0cnVuayByNjY2MiBh bmQgYWRkcyBhbiBBVlgtNTEyIHRpZXIgdG8gdGhlIHZlY3RvciBsYWRkZXIgCj4+IHRoYXQgYWxy ZWFkeSBwaWNrcyBiZXR3ZWVuIHNjYWxhciwgU1NFMiBhbmQgQVZYMiBhdCBydW4gdGltZS4gVHdv IAo+PiBrZXJuZWxzIGluIHRoZSBmaXJzdCBzdGFnZSAtIHRoZSBIdWZmbWFuIHRhYmxlIHNlYXJj aCBhbmQgdGhlIHhyXigzLzQpIAo+PiBxdWFudGl6YXRpb24gbG9vcCAtIGFuZCB0aGUgVkJSIGJh bmQtbm9pc2Ugcm91dGluZSBpbiBhIHNlY29uZC4KPj4gCj4+IFRoZSBxdWVzdGlvbiBpcyBvbmx5 IHRoaXM6Cj4+IAo+PiAgwqDCoMKgIElzIGl0IGFjdHVhbGx5IGZhc3Rlciwgb24gcmVhbCBoYXJk d2FyZSwgZW5kIHRvIGVuZD8KPj4gCj4+IFdlIGNhbm5vdCBhbnN3ZXIgaXQuIE5vIG1hY2hpbmUg aGVyZSBoYXMgQVZYLTUxMiwgYW5kIHdlIGRpZCBub3QgdXNlIAo+PiBhbiBlbXVsYXRvciwgc28g LSBhbmQgdGhpcyBpcyB0aGUgcGFydCB3b3J0aCByZWFkaW5nIHR3aWNlIC0KPj4gCj4+ICDCoMKg wqAgVEhJUyBDT0RFIEhBUyBORVZFUiBCRUVOIEVYRUNVVEVELiBOT1QgT05DRSwgQU5ZV0hFUkUu Cj4+IAo+PiBJdCBoYXMgYmVlbiBjb21waWxlZCBieSBmb3VyIGNvbXBpbGVycywgZGlzYXNzZW1i bGVkIGFuZCByZWFkIAo+PiBpbnN0cnVjdGlvbiBieSBpbnN0cnVjdGlvbiwgY2hlY2tlZCBhZ2Fp bnN0IHRoZSBzY2FsYXIgY29kZSBpdCAKPj4gcmVwbGFjZXMsIGFuZCB3aXJlZCBpbnRvIHRoZSB1 bml0IHRlc3Qgc3VpdGUuIEJ1dCB0aGUgZmlyc3QgdGltZSB0aGVzZSAKPj4gaW5zdHJ1Y3Rpb25z IGFjdHVhbGx5IHJ1biBvbiBhIHByb2Nlc3NvciB3aWxsIGJlIG9uIHNvbWVvbmUgZWxzZSdzIAo+ PiBtYWNoaW5lLiBBIGNyYXNoIG9yIGEgZmFpbGluZyB0ZXN0IGlzIGEgdXNlZnVsIHJlc3VsdCBh bmQgd2Ugd291bGQgCj4+IHJhdGhlciBoZWFyIGFib3V0IGl0IHRoYW4gbm90OyBwbGVhc2UgZG8g bm90IHJ1biBpdCBvbiBhbnl0aGluZyB5b3UgCj4+IGNhcmUgYWJvdXQuCj4+IAo+PiBSZWxldmFu dCBoYXJkd2FyZSBpcyBBTUQgWmVuIDQgb3IgWmVuIDUsIG9yIEludGVsIFhlb24gZnJvbSBTa3ls YWtlLVNQIAo+PiBvbndhcmQuIEN1cnJlbnQgSW50ZWwgZGVza3RvcCBwYXJ0cyBoYXZlIG5vIEFW WC01MTIgYXQgYWxsLCBhbmQgd2lsbCAKPj4gbm90IHVudGlsIE5vdmEgTGFrZS4gSWYgeW91IGhh dmUgb25lIG9mIHRoZSBhYm92ZSBhbmQgdHdlbnR5IG1pbnV0ZXMsIAo+PiB0aGUgUkVBRE1FIGlu IHRoZSB0YXJiYWxsIHdhbGtzIHRocm91Z2ggaXQgaW4gc3RhZ2VzOyB0aGUgZmlyc3Qgc3RhZ2Ug Cj4+IGFsb25lIGlzIGEgdXNlZnVsIGFuc3dlciwgYW5kICJubyBkaWZmZXJlbmNlIG9uIFplbiA1 IiBpcyBhcyB1c2VmdWwgYXMgCj4+IGEgc3BlZWR1cC4KPj4gCj4+IFdoYXQgaXMgYWxyZWFkeSBk b25lLCBzbyBub2JvZHkgcmVwZWF0cyBpdDogYnVpbGRzIGNsZWFuIHdpdGggMCAKPj4gd2Fybmlu Z3Mgb24gZ2NjIDEzLCBjbGFuZyAxOCwgY2xhbmcgMjEgYW5kIE1TVkMgMTkuNTEsIG9uIExpbnV4 LCAKPj4gRnJlZUJTRCBhbmQgV2luZG93czsgYXBwbGllcyB0byBhIGZyZXNoIHI2NjYyIGNoZWNr b3V0IGFuZCBidWlsZHMgd2l0aCAKPj4gYXV0b3Rvb2xzIHJlbW92ZWQgZnJvbSBQQVRIICh0aGUg cmVnZW5lcmF0ZWQgYnVpbGQgZmlsZXMgc2hpcCBpbiB0aGUgCj4+IGtpdCwgc28gYXV0b2NvbmYv YXV0b21ha2UgYXJlIG5vdCBuZWVkZWQpOyBhbmQgb24gYSBtYWNoaW5lIHdpdGhvdXQgCj4+IEFW WC01MTIgdGhlIG91dHB1dCBpcyBieXRlLWlkZW50aWNhbCB0byB0cnVuayBhY3Jvc3MgNTE5IGVu Y29kZXIgCj4+IG9wdGlvbiBzZXRzLCBzbyB0aGUgbmV3IHJ1bmcgZG9lcyBub3QgZGlzdHVyYiB0 aGUgb25lcyBiZWxvdyBpdC4KPj4gCj4+IFR3byB0aGluZ3MgYWJvdXQgaG93IGl0IGlzIGJ1aWx0 Ogo+PiAKPj4gLSBJdCBpcyBjb21waWxlZCBvbiBhbHBoYSB2ZXJzaW9ucyBvbmx5LCBhbmQgbGVh dmVzIHRoZSBidWlsZCBieSAKPj4gaXRzZWxmIHdoZW4KPj4gIMKgIHRoZSB2ZXJzaW9uIHR5cGUg bW92ZXMgb24uIFRoZXJlIGlzIG5vIGNvbmZpZ3VyZSBzd2l0Y2gsIAo+PiBkZWxpYmVyYXRlbHk6 IGEKPj4gIMKgIHRpZXIgeW91IGhhdmUgdG8gb3B0IGludG8gaXMgYSB0aWVyIG5vYm9keSBtZWFz dXJlcy4gV2l0aGluIGFuIAo+PiBhbHBoYSBidWlsZAo+PiAgwqAgaXQgYmVoYXZlcyBsaWtlIHRo ZSBvdGhlcnMgLSB3aGF0IHRoZSBjb21waWxlciBjYW4gYnVpbGQgaXMgYnVpbHQsIAo+PiB0aGUg Q1BVCj4+ICDCoCBkZWNpZGVzIGF0IHJ1biB0aW1lLCBhbmQgLS1ub2FzbSBhdng1MTIgdHVybnMg aXQgb2ZmLgo+PiAKPj4gLSBUaGUgZmlyc3Qgc3RhZ2UgY2Fubm90IGNoYW5nZSB0aGUgZW5jb2Rl ZCBvdXRwdXQsIGFuZCB0aGUga2l0IGNoZWNrcyAKPj4gdGhhdC4KPj4gIMKgIFRoZSBzZWNvbmQg c3RhZ2UgaXMgbW9yZSBpbnRlcmVzdGluZzogdGhlIGJhbmQtbm9pc2Ugcm91dGluZSBlbmRzIAo+ PiBpbiBhIHN1bQo+PiAgwqAgb2Ygc3F1YXJlcywgYW5kIHN1bW1pbmcgc2l4dGVlbiB2YWx1ZXMg YXQgb25jZSBtZWFucyBzdW1taW5nIHRoZW0gCj4+IGluIGEKPj4gIMKgIGRpZmZlcmVudCBvcmRl ciwgd2hpY2ggY2hhbmdlcyB0aGUgbGFzdCBiaXQgYW5kIHRoZXJlZm9yZSB0aGUgCj4+IGJpdHN0 cmVhbS4KPj4gIMKgIFNvIGl0IHNoaXBzIGluIHR3byBmb3JtcyAtIG9uZSB0aGF0IGtlZXBzIHRo ZSBleGlzdGluZyBvcmRlciBhbmQgaXMKPj4gIMKgIGJpdC1leGFjdCwgYW5kIG9uZSB0aGF0IGRv ZXMgbm90IGFuZCBpcyBmYXN0ZXIuIEhvdyBtdWNoIGZhc3RlciBpcyAKPj4gZXhhY3RseQo+PiAg wqAgd2hhdCB3ZSBkbyBub3Qga25vdywgYW5kIGl0IGRlY2lkZXMgd2hldGhlciB0aGUgcXVlc3Rp b24gaXMgd29ydGggCj4+IGhhdmluZy4KPj4gCj4+IE5vdGhpbmcgaGVyZSBpcyBjb21taXR0ZWQu IElmIHRoZSBudW1iZXJzIHNheSBpdCBpcyBub3Qgd29ydGggaXQsIHRoYXQgCj4+IGlzIGEgcmVz dWx0IGFuZCB0aGUgdGllciBkb2VzIG5vdCBnbyBpbi4KPj4gCj4+IFBsZWFzZSByZXBvcnQgYmFj azogQ1BVIG1vZGVsLCBjb21waWxlciBhbmQgdmVyc2lvbiwgT1MsIHRoZSB0d28gCj4+IHRpbWlu Z3MsIGFuZCBhbnl0aGluZyB0aGF0IGJyb2tlLgo+PiAKPj4gQnllLAo+PiBBbGV4YW5kZXIuCj4+ IAo+IAo+IAo+IAo+IF9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fCj4gTGFtZS1kZXYgbWFpbGluZyBsaXN0Cj4gTGFtZS1kZXZAbGlzdHMuc291cmNlZm9yZ2Uu bmV0Cj4gaHR0cHM6Ly9saXN0cy5zb3VyY2Vmb3JnZS5uZXQvbGlzdHMvbGlzdGluZm8vbGFtZS1k ZXYKCi0tIApodHRwOi8vd3d3LkxlaWRpbmdlci5uZXQgQWxleGFuZGVyQExlaWRpbmdlci5uZXQ6 IFBHUCAweDhGMzE4MzBGOUYyNzcyQkYKaHR0cDovL3d3dy5GcmVlQlNELm9yZyAgICBuZXRjaGls ZEBGcmVlQlNELm9yZyAgOiBQR1AgMHg4RjMxODMwRjlGMjc3MkJGCgpfX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fXwpMYW1lLWRldiBtYWlsaW5nIGxpc3QKTGFt ZS1kZXZAbGlzdHMuc291cmNlZm9yZ2UubmV0Cmh0dHBzOi8vbGlzdHMuc291cmNlZm9yZ2UubmV0 L2xpc3RzL2xpc3RpbmZvL2xhbWUtZGV2Cg==