--===============5587472905563124861==
Content-Type: multipart/alternative; boundary="00000000000042c78505ba470086"

--00000000000042c78505ba470086
Content-Type: text/plain; charset="UTF-8"

On Sunday, January 31, 2021, Matthew Miller <mattdm@fedoraproject.org>
wrote:

> On my Intel i7 laptop, xxhash is a small but clear performance win over
> crc32c:
>
>     $ ./hash-speedtest  10000000
>     Block size:     4096
>     Iterations:     10000000
>     Implementation: builtin
>
>         NULL-NOP: cycles:   1372543560, c/i      137
>      NULL-MEMCPY: cycles:   2844174884, c/i      284
>           CRC32C: cycles:   9673117404, c/i      967
>           XXHASH: cycles:   7129819594, c/i      712
>           SHA256: cycles: 649914613520, c/i    64991
>          BLAKE2b: cycles: 153513008046, c/i    15351
>
>
> And I'm given to understand that this is even more the case on newer CPUs.
>
> Plus, it's 64 bit instead of 32 bit. The 256-bit algorithms are obviously
> much, much slower and probably not right for a default, but should we
> consider making xxhash the default for Fedora Linux systems with btrfs?
>
>
Comparing the hash algorithms in isolation doesn't mean much - does it make
a difference if you try various file system workloads with the different
algorithms?



> --
> Matthew Miller
> <mattdm@fedoraproject.org>
> Fedora Project Leader
> _______________________________________________
> devel mailing list -- devel@lists.fedoraproject.org
> To unsubscribe send an email to devel-leave@lists.fedoraproject.org
> Fedora Code of Conduct: https://docs.fedoraproject.
> org/en-US/project/code-of-conduct/
> List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
> List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.
> fedoraproject.org
>

--00000000000042c78505ba470086
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<br><br>On Sunday, January 31, 2021, Matthew Miller &lt;<a href=3D"mailto:m=
attdm@fedoraproject.org">mattdm@fedoraproject.org</a>&gt; wrote:<br><blockq=
uote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc =
solid;padding-left:1ex">On my Intel i7 laptop, xxhash is a small but clear =
performance win over<br>
crc32c:<br>
<br>
=C2=A0 =C2=A0 $ ./hash-speedtest=C2=A0 10000000<br>
=C2=A0 =C2=A0 Block size:=C2=A0 =C2=A0 =C2=A04096<br>
=C2=A0 =C2=A0 Iterations:=C2=A0 =C2=A0 =C2=A010000000<br>
=C2=A0 =C2=A0 Implementation: builtin<br>
<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 NULL-NOP: cycles:=C2=A0 =C2=A01372543560, c/i=
=C2=A0 =C2=A0 =C2=A0 137<br>
=C2=A0 =C2=A0 =C2=A0NULL-MEMCPY: cycles:=C2=A0 =C2=A02844174884, c/i=C2=A0 =
=C2=A0 =C2=A0 284<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 CRC32C: cycles:=C2=A0 =C2=A09673117404, =
c/i=C2=A0 =C2=A0 =C2=A0 967<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 XXHASH: cycles:=C2=A0 =C2=A07129819594, =
c/i=C2=A0 =C2=A0 =C2=A0 712<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 SHA256: cycles: 649914613520, c/i=C2=A0 =
=C2=A0 64991<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0BLAKE2b: cycles: 153513008046, c/i=C2=A0 =
=C2=A0 15351<br>
<br>
<br>
And I&#39;m given to understand that this is even more the case on newer CP=
Us.<br>
<br>
Plus, it&#39;s 64 bit instead of 32 bit. The 256-bit algorithms are obvious=
ly<br>
much, much slower and probably not right for a default, but should we<br>
consider making xxhash the default for Fedora Linux systems with btrfs?<br>
<br></blockquote><div><br></div><div>Comparing the hash algorithms in isola=
tion doesn&#39;t mean much - does it make a difference if you try various f=
ile system workloads with the different algorithms?</div><div><br></div><di=
v>=C2=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex">
-- <br>
Matthew Miller<br>
&lt;<a href=3D"mailto:mattdm@fedoraproject.org">mattdm@fedoraproject.org</a=
>&gt;<br>
Fedora Project Leader<br>
______________________________<wbr>_________________<br>
devel mailing list -- <a href=3D"mailto:devel@lists.fedoraproject.org">deve=
l@lists.fedoraproject.org</a><br>
To unsubscribe send an email to <a href=3D"mailto:devel-leave@lists.fedorap=
roject.org">devel-leave@lists.<wbr>fedoraproject.org</a><br>
Fedora Code of Conduct: <a href=3D"https://docs.fedoraproject.org/en-US/pro=
ject/code-of-conduct/" target=3D"_blank">https://docs.fedoraproject.<wbr>or=
g/en-US/project/code-of-<wbr>conduct/</a><br>
List Guidelines: <a href=3D"https://fedoraproject.org/wiki/Mailing_list_gui=
delines" target=3D"_blank">https://fedoraproject.org/<wbr>wiki/Mailing_list=
_guidelines</a><br>
List Archives: <a href=3D"https://lists.fedoraproject.org/archives/list/dev=
el@lists.fedoraproject.org" target=3D"_blank">https://lists.fedoraproject.<=
wbr>org/archives/list/devel@lists.<wbr>fedoraproject.org</a><br>
</blockquote>

--00000000000042c78505ba470086--

--===============5587472905563124861==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: base64
Content-Disposition: inline

X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZGV2ZWwgbWFp
bGluZyBsaXN0IC0tIGRldmVsQGxpc3RzLmZlZG9yYXByb2plY3Qub3JnClRvIHVuc3Vic2NyaWJl
IHNlbmQgYW4gZW1haWwgdG8gZGV2ZWwtbGVhdmVAbGlzdHMuZmVkb3JhcHJvamVjdC5vcmcKRmVk
b3JhIENvZGUgb2YgQ29uZHVjdDogaHR0cHM6Ly9kb2NzLmZlZG9yYXByb2plY3Qub3JnL2VuLVVT
L3Byb2plY3QvY29kZS1vZi1jb25kdWN0LwpMaXN0IEd1aWRlbGluZXM6IGh0dHBzOi8vZmVkb3Jh
cHJvamVjdC5vcmcvd2lraS9NYWlsaW5nX2xpc3RfZ3VpZGVsaW5lcwpMaXN0IEFyY2hpdmVzOiBo
dHRwczovL2xpc3RzLmZlZG9yYXByb2plY3Qub3JnL2FyY2hpdmVzL2xpc3QvZGV2ZWxAbGlzdHMu
ZmVkb3JhcHJvamVjdC5vcmcK

--===============5587472905563124861==--