Big Throughput Boost & Lower Latency With New Patch For Linux Checksum Function

[phoronix]

Phoronix: Big Throughput Boost & Lower Latency With New Patch For Linux Checksum Function

Queued up ahead of the Linux 6.5 cycle kicking off in about one month is a new Linux x86 optimization patch for further tuning csum_partial, the function used within the kernel for calculating 32-bit checksums on blocks of data. Much lower latency and higher throughput can be observed with the newly-optimized csum_partial on the latest Intel/AMD processors...

https://www.phoronix.com/news/Linux-6.5-csum_partial-Optimize

[avis]

Michael

This looks worth giving light to, a very serious data corruption issue in the XFS filesystem in kernel 6.3: https://bugzilla.redhat.com/show_bug.cgi?id=2208553

[MorrisS.]

What about 16bit and 2bit units?

[Weasel]

Michael

This looks worth giving light to, a very serious data corruption issue in the XFS filesystem in kernel 6.3: https://bugzilla.redhat.com/show_bug.cgi?id=2208553

As if Michael will ever post something bad about GNOME or Red Hat, accidentally "forgets" it and will forget your post too.

[avis]

As if Michael will ever post something bad about GNOME or Red Hat, accidentally "forgets" it and will forget your post too.

He definitely hates me only I've no idea why. I don't think I've done less to improve Linux than him. Sometimes I'm extremely sarcastic, pedantic and nit-picky but that's who I am. I'm not overtly aggressive or insulting towards people.

Edit: it's on the front page. You are welcome, Michael.

[intelfx]

<...> I don't think I've done less to improve Linux than him. <...>

You definitely don't lack in hubris, lol. Sadly there's nothing to back it up.

[avis]

You definitely don't lack in hubris, lol. Sadly there's nothing to back it up.

As a bug reporter:

• Over a dozen closed errors in the kernel.
• Over a dozen closed errors in GCC.
• Several dozen closed errors in KDE and XFCE.
• Several dozen closed errors in Wine.
• Pushed the issue of horrible memory handling in Linux which resulted first in the adoption of earlyoomd, and later on in systemd-oomd being created. The issue since then has been completely resolved.
• And close to a hundred closed issues in a dozen lesser Open Source projects.

In addition to that I've pushed many other issues to no avail but for some reasons a ton of people find them valid. E.g this one. And what about today? Ah, nothing. It's not a good day if someone doesn't sh!t on me here on Phoronix. I know, I know.

Sadly you know nothing about me and throw baseless accusations like there's no tomorrow. BTW what have you done other than hiding under a made up name on Phoronix?

I need to log off. I'm not welcome here. As ridiculous as it sounds I'm getting shat on by people who primarily do nothing, expect others to work for them for free and have the guts to accuse Linux contributors of "showing off".

[dreamcat4]

Michael

This looks worth giving light to, a very serious data corruption issue in the XFS filesystem in kernel 6.3: https://bugzilla.redhat.com/show_bug.cgi?id=2208553

Avis, thank you for reporting this issue. I appreciated the heads up. Would have preferred if Michael at least gave some mention that the issue was reported on the forums. But regardless - appreciated many thanks.

I can fairly say that I do respect [at least some of] your opinions. We don't always see eye to eye, but that is often only because you seem to want to be dismissing some fewer certain things because well... you may not be affected by or see yourself. To be personally aware of. Which is entirely understandable - it just does not mean other users don't actually encounter those raised issues themselves. For the different set of priorities or use cases that they are operating under. And in all other respects I can usually accept the witty sarcasm about such viewpoint on its face. As being no reason to get upset about. It may sometimes provoke a reaction / response. Or generate some arguments. But does not really bother me, as i do also equally appreciate your participation and contributions more widely than focussing purely any such specific disagreements. Thanks again for raising awareness for those XFS kernel bug. (for which have already commented on it's dedicated news thread). And have a good day.

But Michael should still have given some brief link. Some little response or credit / passing reference. Was my feelings.

[intelfx]

Sadly you know nothing about me and throw baseless accusations like there's no tomorrow. BTW what have you done other than hiding under a made up name on Phoronix?

You can look it up. Let's rather talk about what I did not do — for example, never got one banned for behaving in a way not acceptable of a spoiled toddler, much less of a reasonable adult who you are claiming to be. Who's hiding under a made-up name, huh birdie?

[GreenReaper]

I presume Avis is his middle name.