[odc] Daily src changes for 2007-04-12
ODC
auto at squish.net
Fri Apr 13 08:00:22 BST 2007
OpenBSD src changes summary for 2007-04-12
==========================================
bin/date lib/libkvm
regress/sys share/man
sys/arch/alpha/alpha sys/arch/alpha/include
sys/arch/i386/i386 sys/arch/i386/include
sys/arch/i386/isa sys/arch/landisk/include
sys/arch/solbourne/conf sys/arch/sparc64/sparc64
sys/arch/vax/include sys/dev
sys/dev/ata sys/dev/ic
sys/dev/tc sys/kern
sys/nfs sys/ntfs
sys/scsi sys/sys
sys/uvm usr.sbin/hoststated
usr.sbin/ospfd
== bin =============================================================== 01/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/bin
date
~ date.c
> use UTC instead of GMT0 for -u; from Jared R.R.Spiegel <jrrs at ice-nine.org > ok millert@, otto@ (naddy@)
== lib =============================================================== 02/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libkvm
~ kvm_proc.c
> move p_limit and p_cred into struct process
> leave macros behind for now to keep the commit small
> ok art beck miod pedro (tedu@)
== regress =========================================================== 03/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/regress
sys
~ ffs/tests/chflags/00.t ~ ffs/tests/chflags/08.t
~ ffs/tests/chflags/09.t ~ ffs/tests/chflags/10.t
~ ffs/tests/chmod/08.t ~ ffs/tests/chown/08.t
~ ffs/tests/link/12.t ~ ffs/tests/link/13.t
~ ffs/tests/mkdir/08.t ~ ffs/tests/mkfifo/10.t
~ ffs/tests/open/09.t ~ ffs/tests/open/10.t
~ ffs/tests/rename/06.t ~ ffs/tests/rename/07.t
~ ffs/tests/rename/08.t ~ ffs/tests/rmdir/09.t
~ ffs/tests/rmdir/10.t ~ ffs/tests/symlink/09.t
~ ffs/tests/truncate/08.t ~ ffs/tests/unlink/09.t
~ ffs/tests/unlink/10.t
> Remove the SF_NOUNLINK and UF_NOUNLINK from the test files, we dont
> support them; Cuts down false-postives.
> ok pedro@ (thib@)
== share ============================================================= 04/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man4/ifmedia.4
> add in channel frequency info from old wicontrol(8) page;
> diff from Thoren McDole, cleaned up by myself; (jmc@)
~ man9/mbuf.9
> m_prepend() should never be called directly. So tell people to use
> M_PREPEND() instead. Whit help by jmc@ (claudio@)
~ man9/lock.9
> Remove traces of LK_UPGRADE and LK_FORCEUPGRADE in the error descriptions.
> Spotted by jmc@ (miod@)
== sys =============================================================== 05/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/alpha/alpha
~ trap.c
> __HAVE_CPUINFO for alpha, with bits from netbsd
> ok miod@ (martin@)
arch/alpha/include
~ _types.h ~ cpu.h
~ intr.h
> __HAVE_CPUINFO for alpha, with bits from netbsd
> ok miod@ (martin@)
arch/i386/i386
~ apicvec.s ~ lapic.c
~ machdep.c ~ vector.s
> Faster signal delivery on i386/MP.
> We need to poke the other CPU so that it processes the AST immediately
> and doesn't wait for the next interrupt or syscall.
> Since IPIs really shouldn't process ASTs, we need to trigger a soft
> interrupt on the destination CPU to process the AST. But since we can't
> send soft interrupts to other CPUs, we send an IPI, that triggers a soft
> interrupt that in turn processes the AST.
> Also, this marks the beginning of moving to slightly better IPI mechanism
> of short and optimized IPIs instead of the large and complicated IPI
> infrastructure we're using now.
> tested by many, ok tholo@ (art@)
arch/i386/include
~ pmap.h
> untypo (art@)
~ cpu.h ~ i82489var.h
~ intrdefs.h
> Faster signal delivery on i386/MP.
> We need to poke the other CPU so that it processes the AST immediately
> and doesn't wait for the next interrupt or syscall.
> Since IPIs really shouldn't process ASTs, we need to trigger a soft
> interrupt on the destination CPU to process the AST. But since we can't
> send soft interrupts to other CPUs, we send an IPI, that triggers a soft
> interrupt that in turn processes the AST.
> Also, this marks the beginning of moving to slightly better IPI mechanism
> of short and optimized IPIs instead of the large and complicated IPI
> infrastructure we're using now.
> tested by many, ok tholo@ (art@)
arch/i386/isa
~ icu.s ~ isa_machdep.c
> Faster signal delivery on i386/MP.
> We need to poke the other CPU so that it processes the AST immediately
> and doesn't wait for the next interrupt or syscall.
> Since IPIs really shouldn't process ASTs, we need to trigger a soft
> interrupt on the destination CPU to process the AST. But since we can't
> send soft interrupts to other CPUs, we send an IPI, that triggers a soft
> interrupt that in turn processes the AST.
> Also, this marks the beginning of moving to slightly better IPI mechanism
> of short and optimized IPIs instead of the large and complicated IPI
> infrastructure we're using now.
> tested by many, ok tholo@ (art@)
arch/landisk/include
~ bus.h
> Remove __BUS_SPACE_ADDRESS_SANITY correctly, no cookie for me. Spotted by
> djm@ (miod@)
arch/solbourne/conf
~ ld.script
> Discard .comment sections in the final a.out file. (miod@)
arch/sparc64/sparc64
~ db_interface.c
> Remove the lk_interlock from struct lock; Also remove the LK_INTERLOCK
> flag. This effectively makes the simplelock argument to lockmgr() fluff.
> ok miod@ (thib@)
arch/vax/include
~ bus.h
> Remove __BUS_SPACE_ADDRESS_SANITY correctly, no cookie for me. Spotted by
> djm@ (miod@)
dev
~ softraid.c ~ softraidvar.h
> Add failure statistics per workunit.
> Remove BIOC_SFAILED; we'll only use BIOC_SOFFLINE.
> Call state change in interrupt handler when a chunk fails. (marco@)
dev/ata
~ atascsi.c ~ atascsi.h
> Do a security freeze lock like we do in wd(4) to deal with systems
> that leave the security set on by default.
> "looks sane" grange@, "looks ok" pascoe@, "put it in" dlg@ (jsg@)
dev/ic
~ sili.c
> Oops, a code rearrangement meant that I didn't initialise the failed ccb
> pointer for problematic non-NCQ commands.
> spotted by and ok dlg@ (pascoe@)
dev/tc
~ if_fta.c
> Fix newlines in dmesg. (miod@)
kern
~ kern_malloc.c ~ subr_pool.c
> Allow machine-dependant overrides for the ``deadbeef'' sentinel values,
> and make sure that nothing can ever be mapped at theses addresses.
> Only i386 overrides the default for now.
> From mickey@, ok art@ miod@ (miod@)
~ init_main.c ~ kern_acct.c
~ kern_exit.c ~ kern_fork.c
~ kern_resource.c ~ kern_sysctl.c
> move p_limit and p_cred into struct process
> leave macros behind for now to keep the commit small
> ok art beck miod pedro (tedu@)
~ kern_lock.c
> Remove the lk_interlock from struct lock; Also remove the LK_INTERLOCK
> flag. This effectively makes the simplelock argument to lockmgr() fluff.
> ok miod@ (thib@)
nfs
~ nfs_vfsops.c ~ nfsmount.h
~ nfsnode.h
> remove unused members from struct nfsmount and nfsnode;
> ok pedro@ (thib@)
~ nfs_kq.c
> If VOP_GETATTR() in the nfs_kqpoll() loop returns ESTALE, catch it.
> Not doing so can lead to clients missing out if the file is for
> example removed on the server and the client is doing a 'tail -f' on it.
> If it returns ESTALE, mark the file deleted and proceed to handling the
> next entry.
> ok tedu@,art@ (thib@)
ntfs
~ ntfs_inode.h ~ ntfs_subr.c
> Remove the i_interlock simplelock from struct ntnode (thib@)
scsi
~ cd.c
> Add cd_kill_buffers() as a separate function, like sd_kill_buffers()
> and for the same reason of preparing for safer handling of users
> yanking usb devices out. No functional change.
> Copied from bluhm@'s change to sd.c. (krw@)
~ scsi_all.h ~ scsi_disk.h
> The beginnings of support for 16-byte cdb's in the scsi code. This is
> so we will be able to address devices larger than the 10-byte cdb's allow
> us to do.
> ok krw@, dlg@ (weingart@)
sys
~ proc.h ~ sysctl.h
> move p_limit and p_cred into struct process
> leave macros behind for now to keep the commit small
> ok art beck miod pedro (tedu@)
~ lock.h
> Remove the lk_interlock from struct lock; Also remove the LK_INTERLOCK
> flag. This effectively makes the simplelock argument to lockmgr() fluff.
> ok miod@ (thib@)
uvm
~ uvm_aobj.c ~ uvm_map.c
> Unbreak compile with option UVMHIST after pg_flags changes.
> from mickey (art@)
~ uvm_init.c
> Allow machine-dependant overrides for the ``deadbeef'' sentinel values,
> and make sure that nothing can ever be mapped at theses addresses.
> Only i386 overrides the default for now.
> From mickey@, ok art@ miod@ (miod@)
== usr.sbin ========================================================== 06/06 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.sbin
hoststated
~ hoststated.conf.5 ~ hoststated.h
~ parse.y ~ relay.c
> add a new relay 'path' action to filter the URL path and arguments.
> ok pyr@ (reyk@)
ospfd
~ kroute.c
> Before starting to evaluate the redistributon of a route it helps to set
> the metric to the highest and least preferred possible value else no routes
> are redistributed. OK norby@ (claudio@)
===============================================================================
More information about the odc
mailing list