From: "Jason A. Donenfeld" Subject: Re: [PATCH] poly1305: generic C can be faster on chips with slow unaligned access Date: Thu, 3 Nov 2016 23:20:08 +0100 Message-ID: References: <20161103004934.GA30775@gondor.apana.org.au> <20161103.130852.1456848512897088071.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 Cc: Herbert Xu , Martin Willi , LKML , linux-crypto@vger.kernel.org, WireGuard mailing list To: David Miller Return-path: In-Reply-To: <20161103.130852.1456848512897088071.davem@davemloft.net> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: wireguard-bounces@lists.zx2c4.com Sender: "WireGuard" List-Id: linux-crypto.vger.kernel.org SGkgRGF2aWQsCgpPbiBUaHUsIE5vdiAzLCAyMDE2IGF0IDY6MDggUE0sIERhdmlkIE1pbGxlciA8 ZGF2ZW1AZGF2ZW1sb2Z0Lm5ldD4gd3JvdGU6Cj4gSW4gYW55IGV2ZW50IG5vIHBpZWNlIG9mIGNv ZGUgc2hvdWxkIGJlIGRvaW5nIDMyLWJpdCB3b3JkIHJlYWRzIGZyb20KPiBhZGRyZXNzZXMgbGlr ZSAieCArIDMiIHdpdGhvdXQsIGF0IGEgdmVyeSBtaW5pbXVtLCBnb2luZyB0aHJvdWdoIHRoZQo+ IGtlcm5lbCB1bmFsaWduZWQgYWNjZXNzIGhhbmRsZXJzLgoKRXhjZWxsZW50IHBvaW50LiBJbiBv dGhlcndvcmRzLAoKICAgIGN0eC0+clswXSA9IChsZTMyX3RvX2NwdXZwKGtleSArICAwKSA+PiAw KSAmIDB4M2ZmZmZmZjsKICAgIGN0eC0+clsxXSA9IChsZTMyX3RvX2NwdXZwKGtleSArICAzKSA+ PiAyKSAmIDB4M2ZmZmYwMzsKICAgIGN0eC0+clsyXSA9IChsZTMyX3RvX2NwdXZwKGtleSArICA2 KSA+PiA0KSAmIDB4M2ZmYzBmZjsKICAgIGN0eC0+clszXSA9IChsZTMyX3RvX2NwdXZwKGtleSAr ICA5KSA+PiA2KSAmIDB4M2YwM2ZmZjsKICAgIGN0eC0+cls0XSA9IChsZTMyX3RvX2NwdXZwKGtl eSArIDEyKSA+PiA4KSAmIDB4MDBmZmZmZjsKCnNob3VsZCBjaGFuZ2UgdG86CgogICAgY3R4LT5y WzBdID0gKGxlMzJfdG9fY3B1dnAoa2V5ICsgIDApID4+IDApICYgMHgzZmZmZmZmOwogICAgY3R4 LT5yWzFdID0gKGdldF91bmFsaWduZWRfbGUzMihrZXkgKyAgMykgPj4gMikgJiAweDNmZmZmMDM7 CiAgICBjdHgtPnJbMl0gPSAoZ2V0X3VuYWxpZ25lZF9sZTMyKGtleSArICA2KSA+PiA0KSAmIDB4 M2ZmYzBmZjsKICAgIGN0eC0+clszXSA9IChnZXRfdW5hbGlnbmVkX2xlMzIoa2V5ICsgIDkpID4+ IDYpICYgMHgzZjAzZmZmOwogICAgY3R4LT5yWzRdID0gKGxlMzJfdG9fY3B1dnAoa2V5ICsgMTIp ID4+IDgpICYgMHgwMGZmZmZmOwoKPiBXZSBrbm93IGV4cGxpY2l0bHkgdGhhdCB0aGVzZSBvZmZz ZXRzIHdpbGwgbm90IGJlIDMyLWJpdCBhbGlnbmVkLCBzbwo+IGl0IGlzIHJlcXVpcmVkIHRoYXQg d2UgdXNlIHRoZSBoZWxwZXJzLCBvciBhbHRlcm5hdGl2ZWx5IGRvIHRoaW5ncyB0bwo+IGF2b2lk IHRoZXNlIHVuYWxpZ25lZCBhY2Nlc3NlcyBzdWNoIGFzIHVzaW5nIHRlbXBvcmFyeSBzdG9yYWdl IHdoZW4KPiB0aGUgSEFWRV9FRkZJQ0lFTlRfVU5BTElHTkVEX0FDQ0VTUyBrY29uZmlnIHZhbHVl IGlzIG5vdCBzZXQuCgpTbyB0aGUgcXVlc3Rpb24gaXM6IGlzIHRoZSBjbGV2ZXIgYXZvaWRhbmNl IG9mIHVuYWxpZ25lZCBhY2Nlc3NlcyBvZgp0aGUgb3JpZ2luYWwgcGF0Y2ggZmFzdGVyIG9yIHNs b3dlciB0aGFuIGNoYW5naW5nIHRoZSB1bmFsaWduZWQKYWNjZXNzZXMgdG8gdXNlIHRoZSBoZWxw ZXIgZnVuY3Rpb24/CgpJJ3ZlIHB1dCBhIGxpdHRsZSB0ZXN0IGhhcm5lc3MgdG9nZXRoZXIgZm9y IHBsYXlpbmcgd2l0aCB0aGlzOgoKICAgICQgZ2l0IGNsb25lIGdpdDovL2dpdC56eDJjNC5jb20v cG9seWJlbmNoCiAgICAkIGNkIHBvbHliZW5jaAogICAgJCBtYWtlIHJ1bgoKVG8gdGVzdCB3aXRo IG9uZSBtZXRob2QsIGRvIGFzIG5vcm1hbC4gVG8gdGVzdCB3aXRoIHRoZSBvdGhlciwgcmVtb3Zl CiIjZGVmaW5lIFVTRV9GSVJTVF9NRVRIT0QiIGZyb20gdGhlIHNvdXJjZSBjb2RlLgoKQFJlbsOp OiBkbyB5b3UgdGhpbmsgeW91IGNvdWxkIHJldGVzdCBvbiB5b3VyIE1JUFMzMnIyIGhhcmR3YXJl IGFuZApyZXBvcnQgYmFjayB3aGljaCBpcyBmYXN0ZXI/CgpBbmQgaWYgYW55Ym9keSBlbHNlIGhh cyBvdGhlciBoYXJkd2FyZSBhbmQgd291bGQgbGlrZSB0byB0cnksIHRoaXMKY291bGQgYmUgbmlj ZS4KClJlZ2FyZHMsCkphc29uCl9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fCldpcmVHdWFyZCBtYWlsaW5nIGxpc3QKV2lyZUd1YXJkQGxpc3RzLnp4MmM0LmNv bQpodHRwOi8vbGlzdHMuengyYzQuY29tL21haWxtYW4vbGlzdGluZm8vd2lyZWd1YXJkCg==