Patience is a virtue
|
Author | Content |
---|---|
montezuma May 29, 2010 9:49 AM EDT |
This is a development kernel not a final release. Calling it a benchmark disaster is premature and silly imho. Linus does not want to release a dud and given how gross the drop in performance is I'm sure there is a rather simple explanation for all this. Shouting hysterically about how this is unacceptable in a mature project is hype at its worst. THIS IS A DEVELOPMENT RELEASE GUYS. |
gus3 May 29, 2010 10:59 AM EDT |
True enough, but the nearly continuous testing Phoronix does on the Linux kernel shows clearly that someone committed code that caused a severe performance drop. It affects several test types, so it's in the core, not in some driver module. This is all in public, on a (*ahem*) very popular project. With so many eyes, and computers, watching every move, it is foolish to think something like this would go unnoticed. Once the specific commit causing the performance hit is found, it may be obvious why it was necessary; it may also be obvious that it was a brain-dead thing to do, and it will be promptly reverted. (The actual article: http://www.phoronix.com/scan.php?page=article&item=linux_263... ) |
Posting in this forum is limited to members of the group: [ForumMods, SITEADMINS, MEMBERS.]
Becoming a member of LXer is easy and free. Join Us!