svn commit: r194569 - head/sys/dev/fxp

Author: yongari Date: Sun Jun 21 06:06:43 2009 New Revision: 194569 URL: http://svn.freebsd.org/changeset/base/194569 Log: Introduce Rx mbuf dma tag and use it in Rx path. Previously it used common mbuf dma tag for both Tx and Rx path but Rx buffer should have single DMA segment and maximum

svn commit: r194570 - head/sys/dev/fxp

Author: yongari Date: Sun Jun 21 06:18:19 2009 New Revision: 194570 URL: http://svn.freebsd.org/changeset/base/194570 Log: Due to possible PCI bus lock-up issues fxp(4) didn't perform full hardware reset in attach phase. Selective reset does not clear configured parameters so I think full

svn commit: r194571 - head/sys/dev/fxp

Author: yongari Date: Sun Jun 21 06:27:35 2009 New Revision: 194571 URL: http://svn.freebsd.org/changeset/base/194571 Log: Don't blindly enable Rx lock-up workaround. Newer chips do not need the Rx lock-up workaround. Obtained from:NetBSD Modified: head/sys/dev/fxp/if_fxp.c

svn commit: r194572 - head/sys/dev/fxp

Author: yongari Date: Sun Jun 21 06:46:32 2009 New Revision: 194572 URL: http://svn.freebsd.org/changeset/base/194572 Log: Always check fxp(4) is running, see if it can accept frames from upper stack in fxp_start_body(). fxp(4) drops driver lock in Rx path so check the fxp(4) is still

svn commit: r194573 - head/sys/dev/fxp

Author: yongari Date: Sun Jun 21 07:17:49 2009 New Revision: 194573 URL: http://svn.freebsd.org/changeset/base/194573 Log: Overhaul fxp(4) multicast filter programming. fxp(4) hardwares do not allow multicast filter programming when controller is busy to send/receive frames. So it used to

svn commit: r194574 - head/sys/dev/fxp

Author: yongari Date: Sun Jun 21 07:34:12 2009 New Revision: 194574 URL: http://svn.freebsd.org/changeset/base/194574 Log: For ICH based fxp(4) controllers treat them as 82559 compatibles. To detect which controller is ICH based one, add a new member variable ich to struct fxp_ident and

svn commit: r194575 - head/sys/kern

Author: rdivacky Date: Sun Jun 21 07:54:47 2009 New Revision: 194575 URL: http://svn.freebsd.org/changeset/base/194575 Log: In non-debugging mode make this define (void)0 instead of nothing. This helps to catch bugs like the below with clang. if (cond); --- note the

svn commit: r194576 - head/sys/fs/ntfs

Author: rdivacky Date: Sun Jun 21 08:36:30 2009 New Revision: 194576 URL: http://svn.freebsd.org/changeset/base/194576 Log: In non-debugging mode make this define (void)0 instead of nothing. This helps to catch bugs like the below with clang. if (cond); --- note the

svn commit: r194577 - head/sys/net

Author: rdivacky Date: Sun Jun 21 08:49:06 2009 New Revision: 194577 URL: http://svn.freebsd.org/changeset/base/194577 Log: In non-debugging mode make this define (void)0 instead of nothing. This helps to catch bugs like the below with clang. if (cond); --- note the

svn commit: r194578 - head/sys/sys

Author: rdivacky Date: Sun Jun 21 09:01:12 2009 New Revision: 194578 URL: http://svn.freebsd.org/changeset/base/194578 Log: In non-debugging mode make this define (void)0 instead of nothing. This helps to catch bugs like the below with clang. if (cond); --- note the

svn commit: r194580 - head/sys/netipx

Author: rwatson Date: Sun Jun 21 10:10:44 2009 New Revision: 194580 URL: http://svn.freebsd.org/changeset/base/194580 Log: Remove historical support for capturing IPX packets in the output path using raw IPX sockets. While functional, this support is disabled using a flag that can't be

svn commit: r194581 - in head/sys: net netinet netinet6

Author: rdivacky Date: Sun Jun 21 10:29:31 2009 New Revision: 194581 URL: http://svn.freebsd.org/changeset/base/194581 Log: Switch cmd argument to u_long. This matches what if_ethersubr.c does and allows the code to compile cleanly on amd64 with clang. Reviewed by: rwatson Approved

svn commit: r194582 - in head/sys/dev/usb: . storage

Author: remko Date: Sun Jun 21 11:21:16 2009 New Revision: 194582 URL: http://svn.freebsd.org/changeset/base/194582 Log: Add support for the Myson Heden 8813. Note that I also added the usbdev to the list, because the 8813 version is not yet known there. I might have twisted the sorting

svn commit: r194583 - head/lib/libc/arm/string

Author: stas Date: Sun Jun 21 12:58:56 2009 New Revision: 194583 URL: http://svn.freebsd.org/changeset/base/194583 Log: - Fix strncmp on arm. Return 0 as result without performing the main cycle only if the len passed is equal to 0. If end address overflows use last possible address as

Re: svn commit: r194578 - head/sys/sys

On Sun, 21 Jun 2009 09:01:12 + (UTC) Roman Divacky rdiva...@freebsd.org mentioned: Modified: head/sys/sys/lock_profile.h == --- head/sys/sys/lock_profile.h Sun Jun 21 08:49:06 2009 (r194577) +++

svn commit: r194584 - head/sys/dev/usb/storage

Author: remko Date: Sun Jun 21 13:13:13 2009 New Revision: 194584 URL: http://svn.freebsd.org/changeset/base/194584 Log: use PROTO_DEFAULT. Requested by: hps Modified: head/sys/dev/usb/storage/umass.c Modified: head/sys/dev/usb/storage/umass.c

svn commit: r194586 - in head/sys: cddl/compat/opensolaris/kern cddl/compat/opensolaris/sys cddl/contrib/opensolaris/uts/common/fs/zfs kern sys

Author: kib Date: Sun Jun 21 13:41:32 2009 New Revision: 194586 URL: http://svn.freebsd.org/changeset/base/194586 Log: Add another flags argument to vn_open_cred. Use it to specify that some vn_open_cred invocations shall not audit namei path. In particular, specify VN_OPEN_NOAUDIT for

Re: svn commit: r194493 - head/usr.bin/catman

Bruce Evans wrote: I've never seen a machine that actually runs catman. I remove the cat directories on all my machines to prevent any saving of cat pages. All FreeBSD cluster machines that I checked (just 3) have 0, 1 and 2 saved cat pages. This shows shows that catman isn't run on them

Re: svn commit: r194493 - head/usr.bin/catman

Alexey Dokuchaev da...@freebsd.org writes: This kinda brings up the question, do we actually need catman(1) at all these days of fast CPU and disks and plenty of RAM? Show me the fast CPU and plenty of RAM on this board: http://www.soekris.com/net4801.htm DES -- Dag-Erling Smørgrav -

Re: svn commit: r194493 - head/usr.bin/catman

Bruce Evans b...@optusnet.com.au writes: I've never seen a machine that actually runs catman. I remove the cat directories on all my machines to prevent any saving of cat pages. All FreeBSD cluster machines that I checked (just 3) have 0, 1 and 2 saved cat pages. This shows shows that

svn commit: r194587 - stable/7/bin/cp

Author: jilles Date: Sun Jun 21 15:36:10 2009 New Revision: 194587 URL: http://svn.freebsd.org/changeset/base/194587 Log: MFC r193086: Preserve file flags on symlinks in cp -Rp. This reported ENOSYS before. PR: bin/111226 (part of) Submitted by: Martin Kammerhofer Approved

svn commit: r194590 - head/sys/netipx

Author: rwatson Date: Sun Jun 21 16:11:26 2009 New Revision: 194590 URL: http://svn.freebsd.org/changeset/base/194590 Log: Update copyright on netipx. Modified: head/sys/netipx/README Modified: head/sys/netipx/README

svn commit: r194591 - head/sys/netipx

Author: rwatson Date: Sun Jun 21 16:11:40 2009 New Revision: 194591 URL: http://svn.freebsd.org/changeset/base/194591 Log: Remove unuxed ipx_zerohost. MFC after:3 days Modified: head/sys/netipx/ipx_input.c head/sys/netipx/ipx_var.h Modified: head/sys/netipx/ipx_input.c

Re: svn commit: r194493 - head/usr.bin/catman

On Sun, Jun 21, 2009 at 04:48:09PM +0200, Dag-Erling Sm??rgrav wrote: Alexey Dokuchaev da...@freebsd.org writes: This kinda brings up the question, do we actually need catman(1) at all these days of fast CPU and disks and plenty of RAM? Show me the fast CPU and plenty of RAM on this board:

svn commit: r194599 - in stable/6/sys: . compat/linux contrib/pf dev/cxgb

Author: dchagin Date: Sun Jun 21 19:02:32 2009 New Revision: 194599 URL: http://svn.freebsd.org/changeset/base/194599 Log: MFC r164890 (by jkim): Fixes for 'blocking in fifoor state' problem of LTP tests. linux_*stat*() functions were opening files with O_RDONLY to get major/minor pair

svn commit: r194600 - head/sys/dev/ic

Author: sam Date: Sun Jun 21 19:17:22 2009 New Revision: 194600 URL: http://svn.freebsd.org/changeset/base/194600 Log: add %b formats for various registers Modified: head/sys/dev/ic/ns16550.h Modified: head/sys/dev/ic/ns16550.h

svn commit: r194602 - in head/sys: net netatalk netinet netinet6 netipx

Author: rwatson Date: Sun Jun 21 19:30:33 2009 New Revision: 194602 URL: http://svn.freebsd.org/changeset/base/194602 Log: Clean up common ifaddr management: - Unify reference count and lock initialization in a single function, ifa_init(). - Move tear-down from a macro (IFAFREE) to a

svn commit: r194603 - in stable/7/sys: . contrib/pf

Author: des Date: Sun Jun 21 19:31:41 2009 New Revision: 194603 URL: http://svn.freebsd.org/changeset/base/194603 Log: Record r192973 as merged so it won't show up as eligible. Modified: stable/7/sys/ (props changed) stable/7/sys/contrib/pf/ (props changed)

svn commit: r194604 - in stable/7/sys: . contrib/pf kern sys

Author: des Date: Sun Jun 21 19:39:34 2009 New Revision: 194604 URL: http://svn.freebsd.org/changeset/base/194604 Log: merge r193027, r193557, r192900, r193028: fix trailing-slash symlink bug PR: kern/21768 Modified: stable/7/sys/ (props changed) stable/7/sys/contrib/pf/

svn commit: r194607 - head/sys/vm

Author: alc Date: Sun Jun 21 20:29:14 2009 New Revision: 194607 URL: http://svn.freebsd.org/changeset/base/194607 Log: Implement a mechanism within vm_phys_alloc_contig() to defer all necessary calls to vdrop() until after the free page queues lock is released. This eliminates repeatedly

svn commit: r194608 - head/sys/netipx

Author: rwatson Date: Sun Jun 21 21:04:12 2009 New Revision: 194608 URL: http://svn.freebsd.org/changeset/base/194608 Log: Introduce basic locking of global IPX address list 'ipx_ifaddr' using a new rwlock, ipx_ifaddr_rw, wrapped with macros. This locking is necessary but not sufficient,

svn commit: r194609 - head/sys/arm/arm

Author: cognet Date: Sun Jun 21 21:38:12 2009 New Revision: 194609 URL: http://svn.freebsd.org/changeset/base/194609 Log: Disable write-back until I figure out what's wrong with it on the i81342. There's no need to disable the MMU once we're done inflating the kernel. Modified:

svn commit: r194610 - head/sys/netipx

Author: rwatson Date: Sun Jun 21 21:42:29 2009 New Revision: 194610 URL: http://svn.freebsd.org/changeset/base/194610 Log: Add ipx_ifaddr locking to ipx_control(), which should close most remaining potential races in ifconfig's management of IPX addresses. This is largely accomplished by

Re: svn commit: r194493 - head/usr.bin/catman

Brooks Davis wrote: On Sun, Jun 21, 2009 at 04:48:09PM +0200, Dag-Erling Sm??rgrav wrote: Alexey Dokuchaev da...@freebsd.org writes: This kinda brings up the question, do we actually need catman(1) at all these days of fast CPU and disks and plenty of RAM? Show me the fast CPU

Re: svn commit: r194493 - head/usr.bin/catman

On Sun, Jun 21, 2009 at 04:48:09PM +0200 I heard the voice of Dag-Erling Smørgrav, and lo! it spake thus: Alexey Dokuchaev da...@freebsd.org writes: This kinda brings up the question, do we actually need catman(1) at all these days of fast CPU and disks and plenty of RAM? Show me the fast

Re: svn commit: r194493 - head/usr.bin/catman

Bruce Simpson b...@incunabulum.net writes: Also, why are people expecting to render man pages on an embedded platform anyway? You'd be surprised how often I type 'man pf.conf' on that box :) DES -- Dag-Erling Smørgrav - d...@des.no ___

Re: svn commit: r194586 - in head/sys: cddl/compat/opensolaris/kern cddl/compat/opensolaris/sys cddl/contrib/opensolaris/uts/common/fs/zfs kern sys

On Sun, 21 Jun 2009, Konstantin Belousov wrote: Log: Add another flags argument to vn_open_cred. Use it to specify that some vn_open_cred invocations shall not audit namei path. Modified: head/sys/cddl/compat/opensolaris/kern/opensolaris_kobj.c

svn commit: r194611 - in head/sys: amd64/include i386/include

Author: alc Date: Mon Jun 22 04:21:02 2009 New Revision: 194611 URL: http://svn.freebsd.org/changeset/base/194611 Log: Eliminate dead code. These definitions should have been deleted with the introduction of i686_mem.c in r45405. Merge adjacent #ifdef _KERNEL/#endif blocks. Modified: