[odc] Daily src changes for 2006-05-07
ODC
auto at squish.net
Mon May 8 08:18:20 BST 2006
OpenBSD src changes summary for 2006-05-07
==========================================
distrib/sets lib/libkvm
libexec/ld.so sbin/bioctl
share/man sys/arch/i386/i386
sys/arch/m88k/include sys/arch/m88k/m88k
sys/arch/mvme88k/mvme88k sys/arch/sgi/sgi
sys/arch/sparc/include sys/arch/sparc64/include
sys/dev/ic sys/dev/pci
sys/kern sys/nfs
sys/scsi sys/sys
sys/uvm usr.bin/getent
== distrib =========================================================== 01/07 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/distrib
sets
~ lists/base/md.alpha ~ lists/base/md.amd64
~ lists/base/md.cats ~ lists/base/md.hp300
~ lists/base/md.hppa ~ lists/base/md.hppa64
~ lists/base/md.i386 ~ lists/base/md.mac68k
~ lists/base/md.macppc ~ lists/base/md.mvme68k
~ lists/base/md.mvmeppc ~ lists/base/md.sgi
~ lists/base/md.sparc ~ lists/base/md.sparc64
~ lists/base/md.zaurus
> sync (deraadt@)
== lib =============================================================== 02/07 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/lib
libkvm
~ kvm_i386.c
> Unbreak. Fix some of the lint warnings that mickey reintroduced.
> ok miod@ (kettenis@)
== libexec =========================================================== 03/07 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/libexec
ld.so
~ prebind/objarray.c
> Initialize oprebind_data so that unintialized memory is not referenced.
> caught by niallo (drahn@)
== sbin ============================================================== 04/07 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sbin
bioctl
~ bioctl.c
> Add a little debug to test BIOCINQ primitive. (marco@)
== share ============================================================= 05/07 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/share
man
~ man9/rwlock.9
> name arg for rw_init (tedu@)
~ man9/Makefile ~ man9/pool.9
> Pool drains are gone, okay tedu@ (pedro@)
== sys =============================================================== 06/07 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/sys
arch/i386/i386
~ locore.s
> .globl PTDsize; mickey obviously doesn't test before he commits (kettenis@)
arch/m88k/include
~ vmparam.h
> Move the userland stack at the top of the userland address space. (miod@)
arch/m88k/m88k
~ db_trace.c
> Nuke TRACE_DEBUG code. Half of it was useless, and the other half was
> either
> not smart enough to cope with gcc -O2 or with the no longer fixed address
> for the kernels stack. (miod@)
arch/mvme88k/mvme88k
~ m188_machdep.c
> Do not panic on MVME188 if the abort switch is pressed while we are at ipl
> 7
> (fallout from always enabling the abort switch). (miod@)
~ m88410.c
> No need to mask misaligned address exceptions for global flush or
> invalidate
> operations (those do not take a page number in the low address bits of the
> 88410 control register). (miod@)
arch/sgi/sgi
~ machdep.c
> fix pmap debug code and only compile it when PMAPDEBUG is defined.
> this change also unbreaks builds with DEBUG defined.
> ok miod@ (robert@)
arch/sparc/include
~ reloc.h
> Long due update from a.out to ELF (slightly different) relocation defines,
> from sparc64 with love. Spotted by drahn@ (miod@)
arch/sparc64/include
~ reloc.h
> Missing license block; from art@ (miod@)
dev/ic
~ wdc.c
> Default to Ultra DMA Mode 6 not 5 when we know we have a native
> SATA drive.
> ok grange@ (jsg@)
~ dc.c
> - Return if there is nothing to do in the interrupt handler.
> - Check for IFF_RUNNING in the interrupt loop. (brad@)
~ lm78.c ~ lm78var.h
> Add abstraction for resistor factor; makes it easier to compare them to
> the datasheets. Fix a few typos too.
> From Constantine Murenin <mureninc at gmail.org(kettenis@)
~ aacvar.h
> add a name to rwlock so that we can tell where procs are getting stuck
> without breaking into ddb. doubles the size of rwlock [1], but moving
> forward this really helps. ok/tested pedro fgsch millert krw
> [1 - next person to add a field to this struct gets whipped with a wet
> noodle] (tedu@)
~ gdt_common.c ~ gdtvar.h
> Add bio framework (not functional yet).
> Replace index into struct for device name with DEVNAME macro.
> Started translating linux driver back into swedish for gdt->sc_more_proc
> part.
> Added initial scsi passthrough magic.
> Added 2 new debug bits.
> ok krw (marco@)
~ gdt_common.c
> oops, use correct size for GET_IOC_CHAN_DESC (marco@)
~ wdcreg.h
> Fix cut'n'paste abuse in previous typo fix; spotted by Constantine A.
> Murenin,
> thanks! (miod@)
~ gdt_common.c ~ gdtreg.h
~ gdtvar.h
> Add remaining glue for physical disk count.
> Fix a bugglet in the size of GDT_IOCHAN_DESC structure. (marco@)
dev/pci
~ if_em.c
> - Remove unreachable bus_dmamap_unload() in em_dma_malloc().
> - Set the dma_tag to NULL upon failure in em_dma_malloc().
> - In em_dma_free(), return if dma_tag is NULL. (brad@)
~ if_em.c
> fix a typo and some KNF. (brad@)
kern
~ vfs_subr.c
> forgot to remove this sentence from the comment
> ok pedro (sturm@)
~ subr_pool.c ~ uipc_mbuf.c
> remove drain hooks from pool.
> 1. drain hooks and lists of allocators make the code complicated
> 2. the only hooks in the system are the mbuf reclaim routines
> 3. if reclaim is actually able to put a meaningful amount of memory back
> in the system, i think something else is dicked up. ie, if reclaiming
> your ip fragment buffers makes the difference thrashing swap and not,
> your system is in a load of trouble.
> 4. it's a scary amount of code running with very weird spl requirements
> and i'd say it's pretty much totally untested. raise your hand if your
> router is running at the edge of swap.
> 5. the reclaim stuff goes back to when mbufs lived in a tiny vm_map and
> you could run out of va. that's very unlikely (like impossible) now.
> ok/tested pedro krw sturm (tedu@)
~ kern_descrip.c ~ kern_rwlock.c
> add a name to rwlock so that we can tell where procs are getting stuck
> without breaking into ddb. doubles the size of rwlock [1], but moving
> forward this really helps. ok/tested pedro fgsch millert krw
> [1 - next person to add a field to this struct gets whipped with a wet
> noodle] (tedu@)
nfs
~ nfs_node.c
> add a name to rwlock so that we can tell where procs are getting stuck
> without breaking into ddb. doubles the size of rwlock [1], but moving
> forward this really helps. ok/tested pedro fgsch millert krw
> [1 - next person to add a field to this struct gets whipped with a wet
> noodle] (tedu@)
scsi
~ scsi_base.c
> Zap a bunch of trailing whitespace. (krw@)
~ scsi_base.c
> If a device says a MODE SENSE command worked fine, do *not* try a MODE
> SENSE BIG command even if no page data was returned. Just accept the
> fact that the device does not have any such page data.
> This is more in line with the spec, and unbreaks devices (like the
> Neodio 8-in-1 USB Card Reader donated by Manuel Pata) which freak out
> when getting a MODE SENSE BIG.
> Should not break any 'working' device.
> ok pedro@ dlg@ marco@ 'Looks correct.' miod@ (krw@)
sys
~ pool.h
> remove drain hooks from pool.
> 1. drain hooks and lists of allocators make the code complicated
> 2. the only hooks in the system are the mbuf reclaim routines
> 3. if reclaim is actually able to put a meaningful amount of memory back
> in the system, i think something else is dicked up. ie, if reclaiming
> your ip fragment buffers makes the difference thrashing swap and not,
> your system is in a load of trouble.
> 4. it's a scary amount of code running with very weird spl requirements
> and i'd say it's pretty much totally untested. raise your hand if your
> router is running at the edge of swap.
> 5. the reclaim stuff goes back to when mbufs lived in a tiny vm_map and
> you could run out of va. that's very unlikely (like impossible) now.
> ok/tested pedro krw sturm (tedu@)
~ rwlock.h
> add a name to rwlock so that we can tell where procs are getting stuck
> without breaking into ddb. doubles the size of rwlock [1], but moving
> forward this really helps. ok/tested pedro fgsch millert krw
> [1 - next person to add a field to this struct gets whipped with a wet
> noodle] (tedu@)
uvm
~ uvm_pdaemon.c
> remove drain hooks from pool.
> 1. drain hooks and lists of allocators make the code complicated
> 2. the only hooks in the system are the mbuf reclaim routines
> 3. if reclaim is actually able to put a meaningful amount of memory back
> in the system, i think something else is dicked up. ie, if reclaiming
> your ip fragment buffers makes the difference thrashing swap and not,
> your system is in a load of trouble.
> 4. it's a scary amount of code running with very weird spl requirements
> and i'd say it's pretty much totally untested. raise your hand if your
> router is running at the edge of swap.
> 5. the reclaim stuff goes back to when mbufs lived in a tiny vm_map and
> you could run out of va. that's very unlikely (like impossible) now.
> ok/tested pedro krw sturm (tedu@)
== usr.bin =========================================================== 07/07 ==
http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin
getent
~ getent.c
> getservbyport() wants network byte order; from NetBSD (otto@)
===============================================================================
More information about the odc
mailing list