[odc] Daily src changes for 2010-09-23
ODC
auto at squish.net
Fri Sep 24 07:00:01 BST 2010
OpenBSD src changes summary for 2010-09-23
==========================================
Makefile.cross bin/csh
distrib/miniroot distrib/sets
gnu/usr.sbin/sendmail sbin/disklabel
sbin/iked sbin/ipsecctl
share/man sys/arch/alpha/alpha
sys/arch/alpha/conf sys/arch/amd64/conf
sys/arch/arm/arm sys/arch/armish/conf
sys/arch/aviion/aviion sys/arch/gumstix/conf
sys/arch/hp300/hp300 sys/arch/hppa/conf
sys/arch/hppa/hppa sys/arch/hppa64/hppa64
sys/arch/i386/conf sys/arch/landisk/conf
sys/arch/landisk/landisk sys/arch/loongson/conf
sys/arch/loongson/dev sys/arch/loongson/loongson
sys/arch/luna88k/luna88k sys/arch/mac68k/mac68k
sys/arch/macppc/conf sys/arch/macppc/macppc
sys/arch/mvme68k/mvme68k sys/arch/mvme88k/mvme88k
sys/arch/mvmeppc/mvmeppc sys/arch/octeon/octeon
sys/arch/palm/conf sys/arch/sgi/sgi
sys/arch/socppc/conf sys/arch/socppc/socppc
sys/arch/sparc/dev sys/arch/sparc/sparc
sys/arch/sparc64/conf sys/arch/sparc64/dev
sys/arch/sparc64/sparc64 sys/arch/vax/vax
sys/arch/zaurus/conf sys/dev
sys/dev/ata sys/dev/isa
sys/dev/raidframe sys/dev/usb
sys/kern sys/miscfs/procfs
sys/msdosfs sys/net
sys/netatalk sys/netinet
sys/netinet6 sys/nnpfs
sys/scsi sys/sys
sys/ufs/ext2fs sys/ufs/ufs
usr.bin usr.bin/calendar
usr.bin/cvs usr.bin/mandoc
usr.bin/rcs usr.bin/ssh
usr.bin/systat usr.bin/tn3270
usr.sbin/npppd usr.sbin/snmpd
== Makefile.cross ==================================================== 01/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/Makefile.cross
Makefile.cross
> Remove tn3270 reference. (pirofti@)
== bin =============================================================== 02/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
csh
~ USD.doc/csh.2
> remove reference to msgs(1) (deraadt@)
== distrib =========================================================== 03/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
miniroot
~ install.sub
> Teach the installer how to process the new hw.disknames output.
> ok krw@ (jsing@)
sets
~ lists/base/mi ~ lists/etc/mi
~ lists/man/mi
> sync (deraadt@)
~ lists/base/mi ~ lists/man/mi
> sync (pirofti@)
== gnu =============================================================== 04/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/gnu
usr.sbin/sendmail
~ sendmail/mailq.8 ~ sendmail/newaliases.8
~ sendmail/sendmail.8
> use an EXIT STATUS section for these pages; (jmc@)
~ mailstats/mailstats.8 ~ praliases/praliases.1
> add EXIT STATUS here too; (jmc@)
~ sendmail/sendmail.8
> since msgs no longer exists, replace the example msgs alias with the one
> from vacation(8);
> ok millert (jmc@)
~ sendmail/sendmail.8
> NOTES->CAVEATS; ok millert (jmc@)
~ editmap/editmap.8 ~ mailstats/mailstats.8
~ makemap/makemap.8 ~ praliases/praliases.1
~ sendmail/mailq.8 ~ sendmail/newaliases.8
~ sendmail/sendmail.8 ~ smrsh/smrsh.8
> add RCSID and Mdocdate to these pages; ok millert (jmc@)
== sbin ============================================================== 05/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
disklabel
~ disklabel.c ~ editor.c
~ extern.h
> Print disklabel UIDs using printf/fprintf instead of using the hand rolled
> uid_print() function.
> ok krw@ (jsing@)
~ disklabel.c
> Correctly identify mount points when they are specified in /etc/fstab using
> a disklabel UID. Based on a diff from josh at elsasser.org. Resolves PR6471.
> ok krw@ (jsing@)
~ disklabel.8
> tweak the disklabel uid stuff; ok jsing (jmc@)
iked
~ iked.conf.5 ~ parse.y
~ pfkey.c
> support for aes-gcm
> OK reyk (mikeb@)
ipsecctl
~ ipsec.conf.5
> change description for AES-GMAC a bit.
> prompted by reyk (mikeb@)
== share ============================================================= 06/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/tun.4
> Eliminate mention of EPERM error condition from tun(4) man page. We
> just use the filesystem permissions now.
> Pointed out by stsp at . (matthew@)
~ man9/disk.9
> update struct disk; ok jsing (jmc@)
- man4/uyurex.4 ~ man4/Makefile
+ man4/utwitch.4
> rename uyurex(4) to utwitch(4).
> ok deraadt@ (yuo@)
~ man4/utwitch.4
> tweak previous; (jmc@)
~ man5/fstab.5
> document FSTAB_RQ; ok deraadt thib (jmc@)
~ man5/port-modules.5
> - first stab at documenting the haskell module
> feedback/ok kili@ (jasper@)
== sys =============================================================== 07/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/alpha
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/alpha/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/amd64/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/arm/arm
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/armish/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/aviion/aviion
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/gumstix/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/hp300/hp300
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/hppa/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/hppa/hppa
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/hppa64/hppa64
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/i386/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/landisk/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/landisk/landisk
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/loongson/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/loongson/dev
~ mainbus.c
> Move clock attachment later in the game. This is needed for cpu scaling.
> Okay miod at . (pirofti@)
arch/loongson/loongson
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/luna88k/luna88k
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/mac68k/mac68k
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/macppc/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/macppc/macppc
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/mvme68k/mvme68k
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/mvme88k/mvme88k
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/mvmeppc/mvmeppc
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/octeon/octeon
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/palm/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/sgi/sgi
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/socppc/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/socppc/socppc
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/sparc/dev
~ fd.c
> Introduce a DKF_NOLABELREAD flag, which is used by a device to prevent
> automatic reading of disklabel on attach.
> ok deraadt@ miod@ krw@ (jsing@)
arch/sparc/sparc
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/sparc64/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
arch/sparc64/dev
~ fd.c
> Introduce a DKF_NOLABELREAD flag, which is used by a device to prevent
> automatic reading of disklabel on attach.
> ok deraadt@ miod@ krw@ (jsing@)
arch/sparc64/sparc64
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/vax/vax
~ conf.c
> Add pppx(4) to the cdevs of the remaining archs.
> OK dlg@ (claudio@)
arch/zaurus/conf
~ GENERIC
> enable utwitch(4) (formely uyurex(4)) by default
> ok deraadt@ (yuo@)
dev
~ softraid.c ~ softraid_crypto.c
> The only sensible argument for VOP_* calls that take a struct proc pointer
> is
> curproc. A bunch of callers were passing in 0 (not even NULL, 0) as this
> pointer, which was fine until the called vnode function tried to do
> something with it. Typically, this code was then copy/pasted to various
> parts of the tree.
> Accept the facts of life and switch all of these over to passing curproc
> for now until the argument can be removed.
> Discovered by stsp trying to create a softraid on top of a vnd, which
> crashed with a NULL deref in vndioctl.
> softraid bits tested by mikeb and jsing. raidframe bits tested by pea,
> matthieu and naddy. The rest tested by at least thib, jsing and myself.
> ok thib@, jsing at . (oga@)
dev/ata
~ atascsi.h
> commands and bits for data set management. (dlg@)
~ atascsi.c
> translate WRITE SAME 16 into DSM TRIM. (dlg@)
dev/isa
~ fd.c
> Introduce a DKF_NOLABELREAD flag, which is used by a device to prevent
> automatic reading of disklabel on attach.
> ok deraadt@ miod@ krw@ (jsing@)
dev/raidframe
~ rf_disks.c ~ rf_openbsdkintf.c
> The only sensible argument for VOP_* calls that take a struct proc pointer
> is
> curproc. A bunch of callers were passing in 0 (not even NULL, 0) as this
> pointer, which was fine until the called vnode function tried to do
> something with it. Typically, this code was then copy/pasted to various
> parts of the tree.
> Accept the facts of life and switch all of these over to passing curproc
> for now until the argument can be removed.
> Discovered by stsp trying to create a softraid on top of a vnd, which
> crashed with a NULL deref in vndioctl.
> softraid bits tested by mikeb and jsing. raidframe bits tested by pea,
> matthieu and naddy. The rest tested by at least thib, jsing and myself.
> ok thib@, jsing at . (oga@)
dev/usb
~ usb.c
> rename usb_discover to usb_explore for consistency.
> OK krw (jakemsr@)
~ uhub.c ~ usb.c
~ usbdi.c ~ usbdivar.h
> instead of running usb_explore() from individual kthreads for each
> USB bus, make usb_explore() a usb_task. reduces races during normal
> USB device detach, since now usb_tasks and detach happen in the same
> process.
> ok yuo@, matthew@ helped with the task thread loop (jakemsr@)
~ usb.c
> instead of waiting the full power-up-to-power-steady time for root
> hubs before their first port explore, check how long it's actually
> been since power up and only wait as necessary. saves a little time
> on boot, especially now that explore tasks are serialized.
> ok yuo@ (jakemsr@)
~ uhub.c ~ usb.c
~ usbdi.c ~ usbdivar.h
> add a 'dying' flag to struct usbd_bus. use this to signify the bus
> is dying, instead of setting a flag in struct usb_softc. as
> usbd_device_handle has a pointer to the usbd_bus it's attached to,
> usb devices, and functions they run or functions run on their behalf,
> can now easily check if their bus is dying. use this to stop
> usbd_do_request* from running and the usb task thread from adding
> new tasks when a device's bus is dying. (jakemsr@)
~ usb.c ~ usbdi.h
> add two members to struct usb_task
> * usbd_device_handle dev - the device responsible for the task. use
> this to not run the task if the device's hub is dying.
> * int running - a flag to be set when the task is running.
> add usb_rem_wait_task(), a wrapper for usb_rem_task() that waits for
> the task to complete if the task is already running.
> s/usb_rem_task/usb_rem_wait_task/ in usb_detach(). probably most
> drivers using usb_tasks should do this as well. although device
> attach/detach is serialized in normal cases, in the special case
> where the usb bus is hotpluggable (like cardbus/pcmcia), devices
> are not detached in the task thread. (jakemsr@)
~ ehci.c
> returning here can cause the usb task thread to go to sleep forever,
> becaue it skips a wakeup() later on. this is the only error condition
> that returns before the wakeup. not sure why this particular case is
> handled differently, and not certain what the error code should be, but
> this is better than breaking all of usb because of a problem on one port.
> dicussed with deraadt (jakemsr@)
- uyurex.c ~ files.usb
+ utwitch.c
> rename uyurex(4) to utwitch(4)
> ok deraadt@ (yuo@)
~ if_axe.c
> remove dying flag in the detach() function.
> the device driver which has DEACTIVATE processing seems to no need to
> set dying flag in the detach() function because the flag will be set
> in the DEACTIVATE process before detach().
> ok jsg@, ok for experiment deraadt@ (yuo@)
kern
~ uipc_mbuf.c
> tweak the mclgeti algorithm to behave better under load.
> instead of letting hardware rings grow on every interrupt, restrict
> it so it can only grow once per softclock tick. we can only punish
> the rings on softclock ticks, so it make sense to only grow on
> softclock tick boundaries too.
> the rings are now punished after >1 lost softclock tick rather than
> >2. mclgeti is now more aggressive at detecting livelock.
> the rings get punished by an 8th, rather than by half.
> we now allow the rings to be punished again even if the system is
> already considered in livelock.
> without this diff a livelocked system will have its rx ring sizes
> scale up and down very rapidly, while holding the rings low for too
> long. this affected throughput significantly.
> discussed and tested heavily at j2k10. there are still some games
> with softnet we can play, but this is a good first step.
> "put it in" and ok deraadt@
> ok claudio@ krw@ henning@ mcbride@
> if we find out that it sucks we can pull it out again later. till then
> we'll run with it and see how it goes. (dlg@)
~ kern_sysctl.c
> Include the disklabel UID in hw.disknames.
> ok deraadt@ krw@ (jsing@)
~ subr_disk.c
> When a disk is attached create a workq task to read the disklabel,
> providing the DKF_NOLABELREAD flag is not set. This provides the kernel
> with the actual disklabel which includes the disklabel UID.
> ok deraadt@ miod@ krw@ (jsing@)
~ subr_disk.c
> The only sensible argument for VOP_* calls that take a struct proc pointer
> is
> curproc. A bunch of callers were passing in 0 (not even NULL, 0) as this
> pointer, which was fine until the called vnode function tried to do
> something with it. Typically, this code was then copy/pasted to various
> parts of the tree.
> Accept the facts of life and switch all of these over to passing curproc
> for now until the argument can be removed.
> Discovered by stsp trying to create a softraid on top of a vnd, which
> crashed with a NULL deref in vndioctl.
> softraid bits tested by mikeb and jsing. raidframe bits tested by pea,
> matthieu and naddy. The rest tested by at least thib, jsing and myself.
> ok thib@, jsing at . (oga@)
miscfs/procfs
~ procfs_vfsops.c
> In procfs_root(), while the vn_lock on a newly-allocated vnode can not
> fail it is still good practise to either check the return value or to
> provide a flag that won't (to avoid copy pasted code that gets it
> wrong).
> So change LK_EXCLUSIVE to LK_EXCLUSIVE | LK_RETRY, removing the one
> caller of vn_lock in the tree that doesn't either pass LK_RETRY or check
> the return value.
> ok thib@, jsing@ as part of a larger diff. (oga@)
msdosfs
~ msdosfs_vfsops.c
> Change:
> /* something */
> if (error) {
> VOP_UNLOCK();
> return;
> }
> VOP_UNLOCK();
> to the clearer and shorter:
> VOP_UNLOCK();
> if (error)
> return;
> ok thib@, jsing@ as part of a larger diff. (oga@)
net
~ if_tun.c
> If tunread() fails, we should increment if_oerrors, not if_ierrors.
> "yup" deraadt@ (matthew@)
~ if.h
> tweak the mclgeti algorithm to behave better under load.
> instead of letting hardware rings grow on every interrupt, restrict
> it so it can only grow once per softclock tick. we can only punish
> the rings on softclock ticks, so it make sense to only grow on
> softclock tick boundaries too.
> the rings are now punished after >1 lost softclock tick rather than
> >2. mclgeti is now more aggressive at detecting livelock.
> the rings get punished by an 8th, rather than by half.
> we now allow the rings to be punished again even if the system is
> already considered in livelock.
> without this diff a livelocked system will have its rx ring sizes
> scale up and down very rapidly, while holding the rings low for too
> long. this affected throughput significantly.
> discussed and tested heavily at j2k10. there are still some games
> with softnet we can play, but this is a good first step.
> "put it in" and ok deraadt@
> ok claudio@ krw@ henning@ mcbride@
> if we find out that it sucks we can pull it out again later. till then
> we'll run with it and see how it goes. (dlg@)
~ if_gre.c
> fix indentation to not confusingly align with an unassociated while()
> statement
> ok claudio@ (blambert@)
~ pf.c
> Break out rule counter update code into a separate function, makes the
> behaviour consistent between IPv4 and IPv6.
> From martin.pelikan at gmail.com (mcbride@)
netatalk
~ at.h ~ at_proto.c
> Do not declare the domain and protosw structures in at.h fixes build with
> netatalk enabled. (claudio@)
netinet
~ in.h ~ in_pcb.h
~ ip_output.c
> add a new IP level socket option IP_PIPEX. This option is used for L2TP
> support by pipex.
> OK henning@, "Carry on" blambert@ (yasuoka@)
~ ip_esp.c
> remove m_pad in favor of m_inject as it's equivalent to m_inject
> with an offset equal to the actual data length.
> ok henning blambert (mikeb@)
~ ip_ipsp.h
> remove m_pad in favor of m_inject as it's equivalent to m_inject
> with an offset equal to the actual data length.
> ok henning blambert (mikeb@)
netinet6
~ in6.h ~ ip6_output.c
> add a new IP level socket option IP_PIPEX. This option is used for L2TP
> support by pipex.
> OK henning@, "Carry on" blambert@ (yasuoka@)
nnpfs
~ nnpfs_vnodeops.h
> The only sensible argument for VOP_* calls that take a struct proc pointer
> is
> curproc. A bunch of callers were passing in 0 (not even NULL, 0) as this
> pointer, which was fine until the called vnode function tried to do
> something with it. Typically, this code was then copy/pasted to various
> parts of the tree.
> Accept the facts of life and switch all of these over to passing curproc
> for now until the argument can be removed.
> Discovered by stsp trying to create a softraid on top of a vnd, which
> crashed with a NULL deref in vndioctl.
> softraid bits tested by mikeb and jsing. raidframe bits tested by pea,
> matthieu and naddy. The rest tested by at least thib, jsing and myself.
> ok thib@, jsing at . (oga@)
scsi
~ scsi_disk.h
> descripe a few bits of the write same commands. (dlg@)
~ cd.c
> Introduce a DKF_NOLABELREAD flag, which is used by a device to prevent
> automatic reading of disklabel on attach.
> ok deraadt@ miod@ krw@ (jsing@)
sys
~ disk.h
> Introduce a DKF_NOLABELREAD flag, which is used by a device to prevent
> automatic reading of disklabel on attach.
> ok deraadt@ miod@ krw@ (jsing@)
ufs/ext2fs
~ ext2fs_vfsops.c
> Change:
> /* something */
> if (error) {
> VOP_UNLOCK();
> return;
> }
> VOP_UNLOCK();
> to the clearer and shorter:
> VOP_UNLOCK();
> if (error)
> return;
> ok thib@, jsing@ as part of a larger diff. (oga@)
~ ext2fs_lookup.c ~ ext2fs_vnops.c
> The only sensible argument for VOP_* calls that take a struct proc pointer
> is
> curproc. A bunch of callers were passing in 0 (not even NULL, 0) as this
> pointer, which was fine until the called vnode function tried to do
> something with it. Typically, this code was then copy/pasted to various
> parts of the tree.
> Accept the facts of life and switch all of these over to passing curproc
> for now until the argument can be removed.
> Discovered by stsp trying to create a softraid on top of a vnd, which
> crashed with a NULL deref in vndioctl.
> softraid bits tested by mikeb and jsing. raidframe bits tested by pea,
> matthieu and naddy. The rest tested by at least thib, jsing and myself.
> ok thib@, jsing at . (oga@)
ufs/ufs
~ ufs_lookup.c ~ ufs_vnops.c
> The only sensible argument for VOP_* calls that take a struct proc pointer
> is
> curproc. A bunch of callers were passing in 0 (not even NULL, 0) as this
> pointer, which was fine until the called vnode function tried to do
> something with it. Typically, this code was then copy/pasted to various
> parts of the tree.
> Accept the facts of life and switch all of these over to passing curproc
> for now until the argument can be removed.
> Discovered by stsp trying to create a softraid on top of a vnd, which
> crashed with a NULL deref in vndioctl.
> softraid bits tested by mikeb and jsing. raidframe bits tested by pea,
> matthieu and naddy. The rest tested by at least thib, jsing and myself.
> ok thib@, jsing at . (oga@)
== usr.bin =========================================================== 08/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
usr.bin
~ Makefile
> Unlink tn3270 from the build.
> Okay deraadt@, millert@ and all around general approval. (pirofti@)
calendar
~ calendars/calendar.history
> "British defeat the French at the Plains of Abraham" was 1759, not 1788;
> from Douglas Berry, freebsd pr misc/150504 (jmc@)
cvs
~ import.c ~ util.c
~ util.h
> Add cvs_mkdir() for recursive directory creation on import, based on a
> diff from Michael W Bombardieri in PR 6398.
> tested by & ok jasper (nicm@)
mandoc
~ term.c
> Count trailing escaped blanks correctly;
> those ruined the alignment of columns.
> Tested by jmc@, and kristaps@ agrees with the direction. (schwarze@)
~ mdoc_term.c
> When the HEAD of an .Nm block in the SYNOPSIS might be wider
> than the column containing it, the TERMP_HANG flag is required,
> but avoid the flag when we know that the HEAD is shorter,
> because in that case, the flag might ruin the alignment.
> Problem originally reported by jmc@, who also spotted a regression
> in an earlier version of this patch.
> "feel free to commit" kristaps@ (schwarze@)
rcs
~ rcs.c
> Plug memory leak in rcs_getrev, from zinovik.
> ok me stsp (nicm@)
ssh
~ sftp.c
> add [-l limit] to usage(); (jmc@)
~ scp.1 ~ sftp.1
> add KexAlgorithms to the -o list; (jmc@)
systat
~ mbufs.c
> tweak the mclgeti algorithm to behave better under load.
> instead of letting hardware rings grow on every interrupt, restrict
> it so it can only grow once per softclock tick. we can only punish
> the rings on softclock ticks, so it make sense to only grow on
> softclock tick boundaries too.
> the rings are now punished after >1 lost softclock tick rather than
> >2. mclgeti is now more aggressive at detecting livelock.
> the rings get punished by an 8th, rather than by half.
> we now allow the rings to be punished again even if the system is
> already considered in livelock.
> without this diff a livelocked system will have its rx ring sizes
> scale up and down very rapidly, while holding the rings low for too
> long. this affected throughput significantly.
> discussed and tested heavily at j2k10. there are still some games
> with softnet we can play, but this is a good first step.
> "put it in" and ok deraadt@
> ok claudio@ krw@ henning@ mcbride@
> if we find out that it sucks we can pull it out again later. till then
> we'll run with it and see how it goes. (dlg@)
tn3270
- Makefile - Makefile.inc
- api/api.order - api/api_bsd.c
- api/api_exch.c - api/api_exch.h
- api/apilib.c - api/apilib.h
- api/asc_ebc.c - api/asc_ebc.h
- api/astosc.c - api/astosc.h
- api/dctype.c - api/dctype.h
- api/disp_asc.c - api/disp_asc.h
- api/ebc_disp.c - api/ebc_disp.h
- ascii/ascii.order - ascii/default.map
- ascii/map3270.c - ascii/map3270.h
- ascii/mset.c - ascii/state.h
- ascii/termin.c - ctlr/3180.kbd
- ctlr/3270pc.kbd - ctlr/api.c
- ctlr/api.h - ctlr/ctlr.order
- ctlr/declare.h - ctlr/externs.h
- ctlr/function.c - ctlr/function.h
- ctlr/hostctlr.h - ctlr/inbound.c
- ctlr/oia.c - ctlr/oia.h
- ctlr/options.c - ctlr/options.h
- ctlr/outbound.c - ctlr/screen.h
- ctlr/scrnctlr.h - ctlr/unix.kbd
- general/genbsubs.c - general/general.h
- general/general.order - general/globals.c
- general/globals.h - general/vaxbsubs.s
- mset/Makefile - mset/map3270.5
- mset/mset.1 - sys_curses/system.c
- sys_curses/telextrn.h - sys_curses/terminal.h
- sys_curses/termout.c - tn3270/Makefile
- tn3270/tn3270.1 - tools/Makefile
- tools/Makefile.inc - tools/mkmake.y
- tools/prt3270.c - tools/mkastods/Makefile
- tools/mkastods/mkastods.c - tools/mkastosc/Makefile
- tools/mkastosc/mkastosc.c - tools/mkdctype/Makefile
- tools/mkdctype/ectype.c - tools/mkdctype/ectype.h
- tools/mkdctype/mkdctype.c - tools/mkdstoas/Makefile
- tools/mkdstoas/mkdstoas.c - tools/mkhits/Makefile
- tools/mkhits/dohits.c - tools/mkhits/dohits.h
- tools/mkhits/mkhits.c
> Kill tn3270.
> Okay deraadt@, millert@ and all around general approval. (pirofti@)
== usr.sbin ========================================================== 09/09 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
npppd
- common/rt_zebra.c - common/rt_zebra.h
- common/rt_zebra_local.h ~ npppd/Makefile
~ npppd/npppd.c ~ npppd/npppd_subr.c
> remove support for talking to the zebra routing socket which
> was not enabled, we use a kernel routing socket for such things.
> ok yasuoka@ claudio@ (jsg@)
~ common/ipsec_util.c ~ common/ipsec_util_local.h
~ common/time_utils.c ~ common/time_utils.h
~ npppdctl/npppdctl.8
> Add the same license text to these files as the rest of npppd.
> ok yasuoka@ (jsg@)
snmpd
~ mib.c
> tweak the mclgeti algorithm to behave better under load.
> instead of letting hardware rings grow on every interrupt, restrict
> it so it can only grow once per softclock tick. we can only punish
> the rings on softclock ticks, so it make sense to only grow on
> softclock tick boundaries too.
> the rings are now punished after >1 lost softclock tick rather than
> >2. mclgeti is now more aggressive at detecting livelock.
> the rings get punished by an 8th, rather than by half.
> we now allow the rings to be punished again even if the system is
> already considered in livelock.
> without this diff a livelocked system will have its rx ring sizes
> scale up and down very rapidly, while holding the rings low for too
> long. this affected throughput significantly.
> discussed and tested heavily at j2k10. there are still some games
> with softnet we can play, but this is a good first step.
> "put it in" and ok deraadt@
> ok claudio@ krw@ henning@ mcbride@
> if we find out that it sucks we can pull it out again later. till then
> we'll run with it and see how it goes. (dlg@)
===============================================================================
More information about the odc
mailing list