Denys Vlasenko 205042c07a libbb/sha1: in unrolled x86-64 code, pass initial W[] in registers, not on stack
This can be faster on some CPUs.
On Skylake, evidently load latency from L1 (or store-to-load
forwarding in LSU) is fast enough to completely hide
memory reference latencies here.

function                                             old     new   delta
sha1_process_block64                                3495    3514     +19

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
2022-01-25 17:21:45 +01:00
..
2021-06-22 15:28:34 +02:00
2020-12-14 18:25:28 +01:00
2020-12-17 11:22:44 +01:00
2021-02-03 20:43:07 +01:00
2022-01-08 00:41:09 +01:00
2022-01-08 00:41:09 +01:00
2022-01-08 01:25:23 +01:00
2020-10-01 03:07:22 +02:00
2022-01-18 00:36:42 +01:00
2021-02-03 20:52:40 +01:00
2018-07-17 15:04:17 +02:00
2021-09-17 00:47:23 +02:00
2022-01-18 00:36:42 +01:00
2020-12-17 20:53:43 +01:00

Please see the LICENSE file for copyright information (GPLv2)

libbb is BusyBox's utility library.  All of this stuff used to be stuffed into
a single file named utility.c.  When I split utility.c to create libbb, some of
the very oldest stuff ended up without their original copyright and licensing
information (which is now lost in the mists of time).  If you see something
that you wrote that is mis-attributed, do let me know so we can fix that up.

	Erik Andersen
	<andersen@codepoet.org>