diff options
author | Nick Sayer <nsayer@FreeBSD.org> | 2000-12-29 13:52:39 +0000 |
---|---|---|
committer | Nick Sayer <nsayer@FreeBSD.org> | 2000-12-29 13:52:39 +0000 |
commit | 26f0edd0b7cdbf1acbe52420c3c5fcebd0fbad40 (patch) | |
tree | 2472269a8d759da1958f3431bd7ba1373191f1ec /emulators | |
parent | Update to 2000.12.28 - fake version to go with the cyrus-imapd-2.0.9 (diff) |
Add sections about netgraph bridging and the restore-from-disk
performance issue.
Notes
Notes:
svn path=/head/; revision=36436
Diffstat (limited to 'emulators')
-rw-r--r-- | emulators/vmware2/files/Hints.FreeBSD | 55 | ||||
-rw-r--r-- | emulators/vmware3/files/Hints.FreeBSD | 55 |
2 files changed, 110 insertions, 0 deletions
diff --git a/emulators/vmware2/files/Hints.FreeBSD b/emulators/vmware2/files/Hints.FreeBSD index e59a95b810c3..9197a20849ba 100644 --- a/emulators/vmware2/files/Hints.FreeBSD +++ b/emulators/vmware2/files/Hints.FreeBSD @@ -108,6 +108,61 @@ in /tmp and passes it to some external program; you'll see it actually creates a file in /compat/linux/tmp when the external program searches /tmp literally. +- There is a bug you may wish to work around. It is unclear (at least to +me) if this is a bug in VMware, the Linuxulator or the FreeBSD VM +system. + +Some background first: When you mmap a file, the syncer will attempt to +periodically synchronize the on-disk version of the file with the +changes being made in the mmap'd region. If you unlink an mmap'd file, +however, this no longer is necessary because if the machine were to +reboot fsck would throw the file away. The hint is given to the syncer +in the form of the MAP_NOSYNC flag to the mmap() syscall. + +The problem is that VMware somehow performs the operations differently +when resuming from a save-to-disk session than it does when booting. The +result is that the MAP_NOSYNC flag doesn't get set, which causes the +performance of a resumed session to be terrible compared to a new +session. + +I suspect that the problem is that when an open file is unlinked, the +unlink() call does not know to check to see if the file is mmap'ed and +if so to add the MAP_NOSYNC flag. This means that open(), unlink(), +mmap() sequences work, but open(), mmap(), unlink() sequences do not. +VMware probably does not unlink the save-to-disk file right away in case +the resume fails, which causes the difference in behavior compared to a +clean boot. + +The only workaround at present is to force all Linuxulator mmap() calls +to get MAP_NOSYNC. A patch against -current to do just this is available +at http://home.jp.FreeBSD.ORG/cgi-bin/showmail/FreeBSD-users-jp/55885 +(ignore the Japanese text if you can't read it. -stable users may need +to apply part of the patch to /sys/compat/linux/linux_mib.c). + +- If you configure vmware to use bridging, you must still specify the +"Host only" mode to the VMware configurator. It will still work just +like a bridged interface. If your bridged VMware guest is a DHCP client, +you may wish to fix its Ethernet address so as not to generate a new +lease every time you start VMware. To do so, add this line to your +guest's .cfg file: + +ethernet0.address = "00:50:56:1e:ad:bf" + +You can only change the last 5 hex digits, which MUST be unique (at least +within your LAN). + +Note that bridging only works on (real) Interfaces where both the 'set +promisc 1' and 'set autosrc 0' steps function. This means that the +interface must be capable of transmitting frames with other than its own +Ethernet address and receiving promiscuously. Most interfaces can, but +notably wi interfaces are among those that cannot. Note that promiscuous +mode is entered when the vmware.sh startup script is run, which may +cause increased interrupt loads on your machine if it's plugged into a +busy network. If you only run vmware infrequently, it may be better to +only manually run the vmware startup script (as root) just before you +start vmware and again (with the stop argument this time) when you're +finished. + - Don't miss the VMware FAQ available on the official site. http://www.vmware.com/products/productfaq.html diff --git a/emulators/vmware3/files/Hints.FreeBSD b/emulators/vmware3/files/Hints.FreeBSD index e59a95b810c3..9197a20849ba 100644 --- a/emulators/vmware3/files/Hints.FreeBSD +++ b/emulators/vmware3/files/Hints.FreeBSD @@ -108,6 +108,61 @@ in /tmp and passes it to some external program; you'll see it actually creates a file in /compat/linux/tmp when the external program searches /tmp literally. +- There is a bug you may wish to work around. It is unclear (at least to +me) if this is a bug in VMware, the Linuxulator or the FreeBSD VM +system. + +Some background first: When you mmap a file, the syncer will attempt to +periodically synchronize the on-disk version of the file with the +changes being made in the mmap'd region. If you unlink an mmap'd file, +however, this no longer is necessary because if the machine were to +reboot fsck would throw the file away. The hint is given to the syncer +in the form of the MAP_NOSYNC flag to the mmap() syscall. + +The problem is that VMware somehow performs the operations differently +when resuming from a save-to-disk session than it does when booting. The +result is that the MAP_NOSYNC flag doesn't get set, which causes the +performance of a resumed session to be terrible compared to a new +session. + +I suspect that the problem is that when an open file is unlinked, the +unlink() call does not know to check to see if the file is mmap'ed and +if so to add the MAP_NOSYNC flag. This means that open(), unlink(), +mmap() sequences work, but open(), mmap(), unlink() sequences do not. +VMware probably does not unlink the save-to-disk file right away in case +the resume fails, which causes the difference in behavior compared to a +clean boot. + +The only workaround at present is to force all Linuxulator mmap() calls +to get MAP_NOSYNC. A patch against -current to do just this is available +at http://home.jp.FreeBSD.ORG/cgi-bin/showmail/FreeBSD-users-jp/55885 +(ignore the Japanese text if you can't read it. -stable users may need +to apply part of the patch to /sys/compat/linux/linux_mib.c). + +- If you configure vmware to use bridging, you must still specify the +"Host only" mode to the VMware configurator. It will still work just +like a bridged interface. If your bridged VMware guest is a DHCP client, +you may wish to fix its Ethernet address so as not to generate a new +lease every time you start VMware. To do so, add this line to your +guest's .cfg file: + +ethernet0.address = "00:50:56:1e:ad:bf" + +You can only change the last 5 hex digits, which MUST be unique (at least +within your LAN). + +Note that bridging only works on (real) Interfaces where both the 'set +promisc 1' and 'set autosrc 0' steps function. This means that the +interface must be capable of transmitting frames with other than its own +Ethernet address and receiving promiscuously. Most interfaces can, but +notably wi interfaces are among those that cannot. Note that promiscuous +mode is entered when the vmware.sh startup script is run, which may +cause increased interrupt loads on your machine if it's plugged into a +busy network. If you only run vmware infrequently, it may be better to +only manually run the vmware startup script (as root) just before you +start vmware and again (with the stop argument this time) when you're +finished. + - Don't miss the VMware FAQ available on the official site. http://www.vmware.com/products/productfaq.html |