Linked by Eugenia Loli on Mon 28th Apr 2003 15:48 UTC
Permalink for comment
To read all comments associated with this story, please click here.
To read all comments associated with this story, please click here.
News
Linked by Thom Holwerda on 06/19/13 22:28 UTC
Linked by Thom Holwerda on 06/18/13 22:33 UTC
Linked by Anonymous on 06/18/13 22:26 UTC
Linked by Thom Holwerda on 06/18/13 22:25 UTC
Linked by Thom Holwerda on 06/18/13 17:45 UTC
Linked by Thom Holwerda on 06/18/13 17:32 UTC, submitted by poundsmack
Linked by Thom Holwerda on 06/17/13 17:58 UTC
Linked by Thom Holwerda on 06/17/13 17:52 UTC
Linked by Thom Holwerda on 06/14/13 21:03 UTC
Linked by Thom Holwerda on 06/14/13 20:46 UTC
More News »
Sponsored Links



However, FreeBSD 5.0's locking granularity is much finer than in Linux.
That really isn't true at all. Both Linux 2.5 and FBSD 5.0 continue to use a global lock although the former currently uses it far less than the latter (is seems this is largely because work to push Giant down hasn't progressed very far yet.)
Linux has a very ecclectic VM implementation, especially in 2.6, utilizing new, untested, and untuned technologies almost exclusively. This lack of testing and maturity was what lead to the VM switch in the 2.4 series.
That isn't a mistake which'll happen again as organisations like OSDL have been performing extensive regression testing and benchmarking all throughout the 2.5 development process.
Linux's VM has to stretch a lot further than FreeBSD's - with Linux being used on everything from swapless embedded devices to SSI boxes with half a terrabyte of RAM there are a lot of cases to cover. A /lot/ of work has been done on the VM in 2.5 and current indications are extremely good.
FreeBSD's VM is very robust within the <=4GiB PC/Server segment, but it hasn't been tuned (or even designed) to cope with situations outside of this bracket. cf. the PAE work.
It's also not fair to claim that the technologies used in 2.5's VM are untested or untuned - the work done by OSDL, IBM and various other vendors combined with the leasons of the early 2.4 debacle mean that it's far from untuned and untested.