27 messages in org.freebsd.freebsd-hackersRe: How does swap work address spacew...| From | Sent On | Attachments |
|---|---|---|
| Daniel Lang | Jul 5, 2002 4:35 am | |
| David Schultz | Jul 5, 2002 6:34 am | |
| David Schultz | Jul 5, 2002 6:38 am | |
| Terry Lambert | Jul 5, 2002 7:33 am | |
| Daniel Lang | Jul 5, 2002 4:41 pm | |
| Darren Pilgrim | Jul 5, 2002 5:57 pm | |
| Bernd Walter | Jul 5, 2002 7:06 pm | |
| Terry Lambert | Jul 5, 2002 11:12 pm | |
| Matthew Dillon | Jul 6, 2002 1:10 am | |
| Darren Pilgrim | Jul 6, 2002 2:36 pm | |
| Matthew D. Fuller | Jul 6, 2002 2:51 pm | |
| Bernd Walter | Jul 6, 2002 3:06 pm | |
| Darren Pilgrim | Jul 6, 2002 3:42 pm | |
| Terry Lambert | Jul 6, 2002 4:00 pm | |
| Bernd Walter | Jul 6, 2002 4:04 pm | |
| Bernd Walter | Jul 6, 2002 4:13 pm | |
| Matthew Dillon | Jul 6, 2002 4:33 pm | |
| Matthew Dillon | Jul 6, 2002 4:42 pm | |
| Bernd Walter | Jul 6, 2002 5:09 pm | |
| Matthew Dillon | Jul 6, 2002 5:33 pm | |
| Bernd Walter | Jul 6, 2002 5:42 pm | |
| Matthew Dillon | Jul 6, 2002 5:46 pm | |
| Ian Dowse | Jul 6, 2002 5:49 pm | |
| Bernd Walter | Jul 6, 2002 6:21 pm | |
| Darren Pilgrim | Jul 6, 2002 10:01 pm | |
| Matthew Dillon | Jul 6, 2002 10:09 pm | |
| Darren Pilgrim | Jul 6, 2002 10:43 pm |
| Subject: | Re: How does swap work address spacewise? | |
|---|---|---|
| From: | Darren Pilgrim (dm...@pantherdragon.org) | |
| Date: | Jul 6, 2002 10:01:52 pm | |
| List: | org.freebsd.freebsd-hackers | |
Matthew Dillon wrote:
The nominal limit for swap space is around 14 GB due to limitations in available KVM. There are three major limiting factors in the kernel:
* The swap bitmap eats 2 bits per page of swap. The bitmap is sized to handle NSWAP (default 4) x size_of_largest_swap_partition.
Is NSWAP tied to the NSWAPDEV kernel option, or is it the actual number of active swap devices? If the prior, is setting NSWAPDEV to the actual number of swap devices a useful for improving memory usage? Is NSWAPDEV just a compile-time tunable, or is there a sysctl to do the same thing?
* The system has to keep track of pages that are swapped out. The system reserves 8 x <physical_pages_in_system> worth of KVM to keep track of swapped out pages. A machine with 4G of ram reserves enough KVA to hold 32GB worth of swapped out pages.
* The system limits the size of the above zone to VM_SWZONE_SIZE_MAX, which is typically around 70 MB of KVM (enough to hold 14 GB worth of swap mappings).
So the nominal limit is around 14 GB on a 32 bit architecture. With tuning this limit can be bumped up, but the practical limit is going to be around 60GB unless you give the kernel more KVA (reducing the amount of VM a user process can access).
Can VM_SWZONE_SIZE_MAX be tuned down as well, or does the kernel already handle this efficiently enough to keep it at a minimum useful value sized relative to PHYS?
To Unsubscribe: send mail to majo...@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message