AVX512 code now committed, some more testing needed
Alexander Leidinger via Lame-dev <[email protected]> Thu, 30 Jul 2026 22:27:42 +0200
| Newsgroups | gmane.comp.audio.mp3.lame |
|---|---|
| Organization | No organization, this is a private message. |
| Message-ID | <[email protected]> |
SGksCgpkdWUgdG8gYW4gb3ZlcnNpZ2h0IGluIHRoZSBjb2RlLCBzb21lIG9mIHRoZSBBVlg1MTIg Y29kZSB3YXMgbm90IHRlc3RlZC4gClRoaXMgbWF5IG1ha2UgYSBkaWZmZXJlbmNlIGZvciBib3Ro IENCUiBhbmQgVkJSLiBDb3VsZCB5b3UgcGxlYXNlIHJlLXJ1biAKc29tZSB0ZXN0cyB3aXRoIHRo ZSBtb3N0IHJldmVudCBjb2RlIGluIFNWTiAoYXQgbGVhc3QgcjY3MjApPwoKUGxlYXNlIG1ha2Ug c3VyZSB5b3UgaGF2ZSBhIGNsZWFuIHRyZWUgYW5kIGNvbmZpZ3VyZSBlaXRoZXIgd2l0aCAKIi0t ZGlzYWJsZS1zaGFyZWQiLCBvciB1c2UgdGhlIExEX0xJQlJBUllfUEFUSCBzZXR0aW5nIGluIHRo ZSBleGFtcGxlcyAKYmVsb3cgKHRvIG1ha2Ugc3VyZSB0aGUgY29ycmVjdCBsaWJtcDNsYW1lIGlz IGZvdW5kLCBhbmQgbm90IGFuIGFscmVhZHkgCmluc3RhbGxlZCBvbmUpLgoKVG8gY2hlY2tvdXQg ZnJvbSBTVk46Cgpzdm4gY2hlY2tvdXQgLXIgNjcyMCBodHRwczovL3N2bi5jb2RlLnNmLm5ldC9w L2xhbWUvc3ZuL3RydW5rL2xhbWUgbGFtZSAKJiYgY2QgbGFtZSAmJiAuL2NvbmZpZ3VyZSAmJiBt YWtlIC1qOAoKQ29uZmlybSB0aGUgYnVpbGQgaGFzIHRoZSB0aWVyIGFuZCB0aGUgcHJvY2Vzc29y IGNhbiBydW4gaXQuIFRoZSAKLS12ZXJib3NlIGhlYWRlciBtdXN0IHNheSBBVlgtNTEyOyBpZiBp dCBkb2VzIG5vdCwgbm90aGluZyBiZWxvdyAKbWVhc3VyZXMgYW55dGhpbmc6CkxEX0xJQlJBUllf UEFUSD0kUFdEL2xpYm1wM2xhbWUvLmxpYnMgLi9mcm9udGVuZC9sYW1lIC0tdmVyYm9zZSAtLXZl Y3RvciAKYXZ4NTEyIC1WOSB0ZXN0Y2FzZS53YXYgL2Rldi9udWxsIDI+JjEgfCBoZWFkIC0yMAoK UXVpY2sgdmFsaWRhdGlvbiAodXNlIHlvdXIgb3duIDQ0LjEgS0h6IDE2IGJpdCBzdGVyZW8gaW5w dXQpOgoKZXhwb3J0IExEX0xJQlJBUllfUEFUSD0kUFdEL2xpYm1wM2xhbWUvLmxpYnMKSU49JEhP TUUvYXVkaW8vYWxidW0ud2F2Cgpmb3IgcmF0ZSBpbiAiLWIgMzIwIiAiLVYgMiI7IGRvCiAgIHRh Zz0kKGVjaG8gIiRyYXRlIiB8IHRyIC1kICcgJykKICAgZWNobyAiPT0gYXZ4MiAkcmF0ZSA9PSIK ICAgLi9mcm9udGVuZC9sYW1lIC0tdmVyYm9zZSAtLXZlY3RvciBhdngyICRyYXRlICIkSU4iIG91 dC1hdngyJHRhZy5tcDMgCjI+JjEgfAogICAgIGdyZXAgLUUgJ14oZXhwZXJpbWVudGFsfHZlY3Rv ciByb3V0aW5lc3xDUFUgZmVhdHVyZXMpOicKICAgZWNobyAiPT0gYXZ4NTEyICRyYXRlID09Igog ICAuL2Zyb250ZW5kL2xhbWUgLS12ZXJib3NlIC0tdmVjdG9yIGF2eDUxMiAkcmF0ZSAiJElOIiAK b3V0LWF2eDUxMiR0YWcubXAzIDI+JjEgfAogICAgIGdyZXAgLUUgJ14oZXhwZXJpbWVudGFsfHZl Y3RvciByb3V0aW5lc3xDUFUgZmVhdHVyZXMpOicKICAgZWNobyAiPT0gYXZ4NTEyK3RhYmxlICRy YXRlID09IgogICBlbnYgTEFNRV9BVlg1MTJfQ0hPT1NFX1RBQkxFPTEgXAogICAgIC4vZnJvbnRl bmQvbGFtZSAtLXZlcmJvc2UgLS12ZWN0b3IgYXZ4NTEyICRyYXRlICIkSU4iIApvdXQtdGFibGUk dGFnLm1wMyAyPiYxIHwKICAgICBncmVwIC1FICdeKGV4cGVyaW1lbnRhbHx2ZWN0b3Igcm91dGlu ZXN8Q1BVIGZlYXR1cmVzKTonCiAgIGVjaG8gIj09IGF2eDUxMitib3RoICRyYXRlID09IgogICBl bnYgTEFNRV9BVlg1MTJfQ0hPT1NFX1RBQkxFPTEgTEFNRV9BVlg1MTJfU0ZCX05PSVNFPTEgXAog ICAgIC4vZnJvbnRlbmQvbGFtZSAtLXZlcmJvc2UgLS12ZWN0b3IgYXZ4NTEyICRyYXRlICIkSU4i IApvdXQtYm90aCR0YWcubXAzIDI+JjEgfAogICAgIGdyZXAgLUUgJ14oZXhwZXJpbWVudGFsfHZl Y3RvciByb3V0aW5lc3xDUFUgZmVhdHVyZXMpOicKZG9uZQoKVHdvIHRoaW5ncyB0byByZWFkIGlu IHRoYXQgb3V0cHV0LCBhbmQgbmVpdGhlciBpcyBvcHRpb25hbDoKCnZlY3RvciByb3V0aW5lczog bXVzdCBzYXkgYXZ4MiBpbiB0aGUgZmlyc3QgYXJtIGFuZCBhdng1MTIgaW4gdGhlIG90aGVyIAp0 aHJlZS4gQW4gYXJtIHRoYXQgc2lsZW50bHkgZmVsbCBiYWNrIHRvIGEgbmFycm93ZXIgdGllciBp cyBhIAptZWFzdXJlbWVudCBvZiB0aGUgcmVmZXJlbmNlIHVuZGVyIGFub3RoZXIgbmFtZS4KZXhw ZXJpbWVudGFsOiBtdXN0IGFwcGVhciBvbmNlIGluIHRoZSB0aGlyZCBhcm0sIHR3aWNlIGluIHRo ZSBmb3VydGgsIAphbmQgbm90IGF0IGFsbCBpbiB0aGUgZmlyc3QgdHdvLiBUaGF0IGxpbmUgaXMg dGhlIG9ubHkgZXZpZGVuY2UgdGhlIApzd2l0Y2ggcmVhY2hlZCB0aGUgZGlzcGF0Y2ggc2l0ZSBy YXRoZXIgdGhhbiBtZXJlbHkgYmVpbmcgc2V0IGluIHRoZSAKZW52aXJvbm1lbnQuCgpmb3IgdGFn IGluIC1iMzIwIC1WMjsgZG8KICAgY21wIG91dC1hdngyJHRhZy5tcDMgb3V0LWF2eDUxMiR0YWcu bXAzICYmCiAgIGNtcCBvdXQtYXZ4MiR0YWcubXAzIG91dC10YWJsZSR0YWcubXAzICYmCiAgIGNt cCBvdXQtYXZ4MiR0YWcubXAzIG91dC1ib3RoJHRhZy5tcDMgJiYgZWNobyAiJHRhZyBpZGVudGlj YWwiCmRvbmUKCklmIGEgZGlmZmVyZW5jZSBpcyBzaG93biBpbiB0aGUgY29tcGFyZSBsb29wLCBk byBub3QgcnVuIHRoZSBiZW5jaG1hcmssIApzb21ldGhpbmcgd2VudCB3cm9uZy4KClRoZSAiZXhw ZXJpbWVudGFsOiIgbGluZXMgYXJlIHRoZSBwb2ludCBvZiBwcmludGluZyB0aGVtOiBhbiBhcm0g d2hvc2UgCmhlYWRlciBkb2VzIG5vdCBuYW1lIHRoZSBzd2l0Y2ggZGlkIG5vdCB0YWtlIHRoYXQg cGF0aCwgYW5kIGl0cyB0aW1lIGlzIAphIG1lYXN1cmVtZW50IG9mIHRoZSBiYXNlbGluZSB1bmRl ciBhbm90aGVyIG5hbWUuIHRhYmxlIG11c3QgcHJpbnQgb25lIApzdWNoIGxpbmUsIGJvdGggbXVz dCBwcmludCB0d28sIGJhc2VsaW5lIG5vbmUuCgpUaGUgcmVhbCB0ZXN0cy4KClRoaXMgbmVlZHMg aHlwZXJmaW5lIGluc3RhbGxlZCAoZS5nLiAiYXB0IGluc3RhbGwgaHlwZXJmaW5lIiksIHdpdGgg eW91ciAKb3duIDQ0LjEga0h6IDE2IGJpdCBzdGVyZW8gZmlsZSwgc2V2ZXJhbCBtaW51dGVzIGlu IGxlbmd0aDoKCmV4cG9ydCBMRF9MSUJSQVJZX1BBVEg9JFBXRC9saWJtcDNsYW1lLy5saWJzCklO PSRIT01FL2F1ZGlvL2FsYnVtLndhdgpMPSRQV0QvZnJvbnRlbmQvbGFtZQoKIyBmb3J3YXJkIG9y ZGVyOiBhdngyLCBhdng1MTIsICt0YWJsZSwgK2JvdGgKZm9yIHJhdGUgaW4gIi1iIDMyMCIgIi1W IDIiOyBkbwogICB0YWc9JChlY2hvICIkcmF0ZSIgfCB0ciAtZCAnICcpCiAgIHRhc2tzZXQgLWMg MiBoeXBlcmZpbmUgLS13YXJtdXAgMSAtLXJ1bnMgMjUgLS1zdHlsZSBiYXNpYyBcCiAgICAgLS1l eHBvcnQtanNvbiBmd2QkdGFnLmpzb24gXAogICAgIC1uICJhdngyICRyYXRlIiBcCiAgICAgICAi JEwgLVMgLS12ZWN0b3IgYXZ4MiAkcmF0ZSAkSU4gL2Rldi9udWxsIiBcCiAgICAgLW4gImF2eDUx MiAkcmF0ZSIgXAogICAgICAgIiRMIC1TIC0tdmVjdG9yIGF2eDUxMiAkcmF0ZSAkSU4gL2Rldi9u dWxsIiBcCiAgICAgLW4gImF2eDUxMit0YWJsZSAkcmF0ZSIgXAogICAgICAgImVudiBMQU1FX0FW WDUxMl9DSE9PU0VfVEFCTEU9MSAkTCAtUyAtLXZlY3RvciBhdng1MTIgJHJhdGUgJElOIAovZGV2 L251bGwiIFwKICAgICAtbiAiYXZ4NTEyK2JvdGggJHJhdGUiIFwKICAgICAgICJlbnYgTEFNRV9B Vlg1MTJfQ0hPT1NFX1RBQkxFPTEgTEFNRV9BVlg1MTJfU0ZCX05PSVNFPTEgJEwgLVMgCi0tdmVj dG9yIGF2eDUxMiAkcmF0ZSAkSU4gL2Rldi9udWxsIgpkb25lCgojIHJldmVyc2VkIG9yZGVyOiBz YW1lIGZvdXIgYXJtcywgb3Bwb3NpdGUgcG9zaXRpb25zCmZvciByYXRlIGluICItYiAzMjAiICIt ViAyIjsgZG8KICAgdGFnPSQoZWNobyAiJHJhdGUiIHwgdHIgLWQgJyAnKQogICB0YXNrc2V0IC1j IDIgaHlwZXJmaW5lIC0td2FybXVwIDEgLS1ydW5zIDI1IC0tc3R5bGUgYmFzaWMgXAogICAgIC0t ZXhwb3J0LWpzb24gcmV2JHRhZy5qc29uIFwKICAgICAtbiAiYXZ4NTEyK2JvdGggJHJhdGUiIFwK ICAgICAgICJlbnYgTEFNRV9BVlg1MTJfQ0hPT1NFX1RBQkxFPTEgTEFNRV9BVlg1MTJfU0ZCX05P SVNFPTEgJEwgLVMgCi0tdmVjdG9yIGF2eDUxMiAkcmF0ZSAkSU4gL2Rldi9udWxsIiBcCiAgICAg LW4gImF2eDUxMit0YWJsZSAkcmF0ZSIgXAogICAgICAgImVudiBMQU1FX0FWWDUxMl9DSE9PU0Vf VEFCTEU9MSAkTCAtUyAtLXZlY3RvciBhdng1MTIgJHJhdGUgJElOIAovZGV2L251bGwiIFwKICAg ICAtbiAiYXZ4NTEyICRyYXRlIiBcCiAgICAgICAiJEwgLVMgLS12ZWN0b3IgYXZ4NTEyICRyYXRl ICRJTiAvZGV2L251bGwiIFwKICAgICAtbiAiYXZ4MiAkcmF0ZSIgXAogICAgICAgIiRMIC1TIC0t dmVjdG9yIGF2eDIgJHJhdGUgJElOIC9kZXYvbnVsbCIKZG9uZQoKInRhc2tzZXQgLWMgMiIgcGlu cyB0aGUgcnVuIHRvIG9uZSBjb3JlIGFuZCBpcyB0aGUgc2luZ2xlIG1vc3QgZWZmZWN0aXZlIAp0 aGluZyB0aGF0IGNhbiBiZSBkb25lIHRvIHRoZXNlIG51bWJlcnM7IG9uIEZyZWVCU0QgaXQgaXMg Y3B1c2V0IC1sIDEsIAphbmQgb24gYSBtYWNoaW5lIHdoZXJlIHBpbm5pbmcgaXMgYXdrd2FyZCwg ZHJvcCBpdCBhbmQgc2F5IHNvIHJhdGhlciAKdGhhbiBza2lwcGluZyB0aGUgcnVuLgoKRm91ciBp bnZvY2F0aW9ucywgZm91ciBhcm1zLCAyNSBydW5zIGVhY2ggcGVyIG9yZGVyaW5nIOKAlCA0MDAg ZW5jb2Rlcy4gCkV4cGVjdCBpdCB0byB0YWtlIGEgd2hpbGUuIFRoYXQgaXMgdGhlIHNhbWUgcHJv dG9jb2wgdGhlIFplbiA1IGFuZCBJY2UgCkxha2Ugcm91bmRzIHVzZWQsIHdoaWNoIGlzIHdoYXQg bWFrZXMgdGhpcyByb3VuZCBjb21wYXJhYmxlIHRvIHRoZW0uCgpIb3cgdG8gcmVhZCBpdC4gUGVy Y2VudGFnZXMgYXJlIGFnYWluc3QgdGhlIGF2eDIgYXJtLCBpbiBib3RoIG1vZGVzLiBBbiAKYXJt IGlzIGZhc3RlciBpZiBpdCBpcyBmYXN0ZXIgaW4gYm90aCBvcmRlcmluZ3MgYnkgYWJvdXQgdGhl IHNhbWUgCmFtb3VudDsgYW4gZWZmZWN0IHRoYXQgYXBwZWFycyBmb3J3YXJkIGFuZCB2YW5pc2hl cyByZXZlcnNlZCBpcyB0aGUgCm1hY2hpbmUgZHJpZnRpbmcsIG5vdCB0aGUgY29kZS4gaHlwZXJm aW5lIHByaW50cyBpdHMgb3duIHJlbGF0aXZlIApzdW1tYXJ5IGF0IHRoZSBlbmQgb2YgZWFjaCBp bnZvY2F0aW9uLCBhbmQgaXQgdGFrZXMgdGhlIGZhc3Rlc3QgY29tbWFuZCAKYXMgdGhlIHJlZmVy ZW5jZSDigJQgc28gZm9yIGEgZmlndXJlIGFnYWluc3QgQVZYMiBzcGVjaWZpY2FsbHksIHRha2Ug aXQgCmZyb20gdGhlIEpTT04sIHdoaWNoIGNhcnJpZXMgZXZlcnkgaW5kaXZpZHVhbCBydW4gYW5k IGlzIGFsc28gd2hhdCAKYWxsb3dzIHRoZSB0d28gb3JkZXJpbmdzIHRvIGJlIHBvb2xlZC4KCldo YXQgdG8gcmVwb3J0IGJhY2s6ClByb2Nlc3NvciBhbmQgaXRzIGV4YWN0IG1vZGVsLCBjb21waWxl ciBhbmQgdmVyc2lvbiwgb3BlcmF0aW5nIHN5c3RlbSwgCmFuZCB3aGV0aGVyIHRoZSBtYWNoaW5l IHdhcyBvbiBtYWlucyBwb3dlci4gVGhlbiB0aGUgZm91ciBoeXBlcmZpbmUgCnN1bW1hcmllcywg YW5kIG1vc3QgaW1wb3J0YW50bHkgdGhlIEpTT04gZmlsZXMgKG1heWJlIHNlbmQgdGhlbSAKY29t cHJlc3NlZCB0byBtZSBkaXJlY3RseSkg4oCUIHRoZXkgY2FycnkgdGhlIGluZGl2aWR1YWwgcnVu cywgYW5kIHBvb2xpbmcgCnRoZSB0d28gb3JkZXJpbmdzIG5lZWRzIHRob3NlIHJhdGhlciB0aGFu IHRoZSBtZWFucy4gV2hldGhlciBjbXAgc2FpZCAKaWRlbnRpY2FsLCBmb3IgYm90aCBtb2Rlcy4g QW5kIGEgb25lLWxpbmUgImV4cGVyaW1lbnRhbDoiIHRyYW5zY3JpcHQgZm9yIAp0aGUgdGFibGUg YW5kIGJvdGggYXJtcywgcGFzdGVkIHZlcmJhdGltOiBpdCBpcyB0aGUgb25seSBldmlkZW5jZSB0 aGUgCmFybSB3YXMgcmVhbC4KCk5lZ2F0aXZlIGFuZCBwb3NpdGl2ZSByZXN1bHRzIGFyZSB0aGUg dXNlZnVsIGtpbmQgaGVyZS4KCkJ5ZSwKQWxleGFuZGVyLgoKLS0gCmh0dHA6Ly93d3cuTGVpZGlu Z2VyLm5ldCBBbGV4YW5kZXJATGVpZGluZ2VyLm5ldDogUEdQIDB4OEYzMTgzMEY5RjI3NzJCRgpo dHRwOi8vd3d3LkZyZWVCU0Qub3JnICAgIG5ldGNoaWxkQEZyZWVCU0Qub3JnICA6IFBHUCAweDhG MzE4MzBGOUYyNzcyQkYKCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fCkxhbWUtZGV2IG1haWxpbmcgbGlzdApMYW1lLWRldkBsaXN0cy5zb3VyY2Vmb3JnZS5u ZXQKaHR0cHM6Ly9saXN0cy5zb3VyY2Vmb3JnZS5uZXQvbGlzdHMvbGlzdGluZm8vbGFtZS1kZXYK