Date: Wed, 13 Oct 2004 17:07:53 -0700 From: <Venkatesh.Babu@nokia.com> To: <freebsd-smp@freebsd.org> Cc: Venkatesh.Babu@nokia.com Subject: Pointer to 4.8 SMP implementation Message-ID: <F252EE46074AB645B4F44947EEAD572701254563@mvebe001.americas.nokia.com>
next in thread | raw e-mail | index | archive | help
Hi
I have newly joined to this mailing list. Pardon me for bothering =
with my silly questions.
I was not able to get this information from the official website. I know =
this can be determined by looking at the source code, but it is going to =
be too much of a work. I have this option as a last resort.
I know that FreeBSD 5.x supports fine grained locking and 4.x just =
supports coarse grained / giant kernel lock. I would like to know the =
capabilities of SMP implementation in 4.x (say 4.8). I am interested to =
know -
- are the interrupts shared between the multiprocessors?
- what algorithm is used to decide which processor services which =
interrupt ?
- does it distributes the interrupts based on the processor load ?
- does it distributes the software interrupt processing also ?
- how the user level processes are scheduled to run on other processors =
?
- can the user process be run on a particular processor ?=20
- how the processor affinity is taken into consideration ?
Can any of you give some pointers to the SMP capability list in FreeBSD =
4.x ?
It would be great if you can also point me to some design / =
implementation of SMP architecture in FreeBSD 4.x.=20
Thanks a lot,
VBabu
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?F252EE46074AB645B4F44947EEAD572701254563>