svn commit: r252335 - head/sys/amd64/vmm

[Peter Grehan]

Author: grehan
Date: Fri Jun 28 06:05:33 2013
New Revision: 252335
URL: http://svnweb.freebsd.org/changeset/base/252335

Log:
  Make sure all CPUID values are handled, instead of exiting the
  bhyve process when an unhandled one is encountered.
  
  Hide some additional capabilities from the guest (e.g. debug store).
  
  This fixes the issue with FreeBSD 9.1 MP guests exiting the VM on
  AP spinup (where CPUID is used when sync'ing the TSCs) and the
  issue with the Java build where CPUIDs are issued from a guest
  userspace.
  
  Submitted by: tycho nightingale at pluribusnetworks com
  Reviewed by:  neel
  Reported by:  many

Modified:
  head/sys/amd64/vmm/x86.c

Modified: head/sys/amd64/vmm/x86.c
==
--- head/sys/amd64/vmm/x86.cFri Jun 28 05:21:59 2013(r252334)
+++ head/sys/amd64/vmm/x86.cFri Jun 28 06:05:33 2013(r252335)
@@ -45,7 +45,9 @@ __FBSDID($FreeBSD$);
 
 #defineCPUID_VM_HIGH   0x4000
 
-static const char bhyve_id[12] = BHyVE BHyVE ;
+static const char bhyve_id[12] = bhyve bhyve ;
+
+static uint64_t bhyve_xcpuids;
 
 int
 x86_emulate_cpuid(struct vm *vm, int vcpu_id,
@@ -77,15 +79,13 @@ x86_emulate_cpuid(struct vm *vm, int vcp
 * no multi-core or SMT.
 */
switch (func) {
+   /*
+* Pass these through to the guest
+*/
case CPUID__:
case CPUID__0002:
case CPUID__0003:
-   case CPUID__000A:
-   cpuid_count(*eax, *ecx, regs);
-   break;
-
case CPUID_8000_:
-   case CPUID_8000_0001:
case CPUID_8000_0002:
case CPUID_8000_0003:
case CPUID_8000_0004:
@@ -94,6 +94,15 @@ x86_emulate_cpuid(struct vm *vm, int vcp
cpuid_count(*eax, *ecx, regs);
break;
 
+   case CPUID_8000_0001:
+   /*
+* Hide rdtscp/ia32_tsc_aux until we know how
+* to deal with them.
+*/
+   cpuid_count(*eax, *ecx, regs);
+   regs[3] = ~AMDID_RDTSCP;
+   break;
+
case CPUID_8000_0007:
cpuid_count(*eax, *ecx, regs);
/*
@@ -150,6 +159,11 @@ x86_emulate_cpuid(struct vm *vm, int vcp
 */
regs[2] = ~CPUID2_MON;
 
+/*
+* Hide the performance and debug features.
+*/
+   regs[2] = ~CPUID2_PDCM;
+   
/*
 * Hide thermal monitoring
 */
@@ -161,6 +175,11 @@ x86_emulate_cpuid(struct vm *vm, int vcp
 */
regs[3] = ~(CPUID_MCA | CPUID_MCE | CPUID_MTRR);
 
+/*
+* Hide the debug store capability.
+*/
+   regs[3] = ~CPUID_DS;
+
/*
 * Disable multi-core.
 */
@@ -180,6 +199,7 @@ x86_emulate_cpuid(struct vm *vm, int vcp
 
case CPUID__0006:
case CPUID__0007:
+   case CPUID__000A:
/*
 * Handle the access, but report 0 for
 * all options
@@ -203,17 +223,25 @@ x86_emulate_cpuid(struct vm *vm, int vcp
case 0x4000:
regs[0] = CPUID_VM_HIGH;
bcopy(bhyve_id, regs[1], 4);
-   bcopy(bhyve_id, regs[2], 4);
-   bcopy(bhyve_id, regs[3], 4);
+   bcopy(bhyve_id + 4, regs[2], 4);
+   bcopy(bhyve_id + 8, regs[3], 4);
break;
+
default:
-   /* XXX: Leaf 5? */
-   return (0);
+   /*
+* The leaf value has already been clamped so
+* simply pass this through, keeping count of
+* how many unhandled leaf values have been seen.
+*/
+   atomic_add_long(bhyve_xcpuids, 1);
+   cpuid_count(*eax, *ecx, regs);
+   break;
}
 
*eax = regs[0];
*ebx = regs[1];
*ecx = regs[2];
*edx = regs[3];
+
return (1);
 }
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252101 - head/release

[Garrett Cooper]

On Jun 22, 2013, at 3:51 PM, Glen Barber wrote:

 On Sat, Jun 22, 2013 at 03:39:41PM -0700, Garrett Cooper wrote:
 On Jun 22, 2013, at 3:12 PM, Glen Barber wrote:
 
 Author: gjb
 Date: Sat Jun 22 22:12:24 2013
 New Revision: 252101
 URL: http://svnweb.freebsd.org/changeset/base/252101
 
 Log:
 If MAKE_CONF and/or SRC_CONF are set and not character devices,
 copy to chroot to use with release build.
 
 Submitted by:   Garrett Cooper (original version)
 Approved by:kib (mentor)
 MFC after:  1 week
 
 Modified:
 head/release/release.sh
 
 Technically copying /dev/null to a file creates a 0 byte file.
 That's why I used cp in the patch I submitted over email.
 Thanks!
 
 Right.  But did you read the script?
 
 Both __MAKE_CONF and SRCCONF are set to the path of the local files.

Oh, interesting… I see what you did there instead of what I proposed; I 
copied __MAKE_CONF to /etc/make.conf and SRCCONF to /etc/src.conf for symmetry 
with the base system, but your method works more than mine did (I forgot that 
you were passing those vars in the chroot environment). Should the files be 
installed to /etc/ in the system default locations ?
Thanks!
-Garrett
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252336 - head/usr.sbin/bhyve

[Peter Grehan]

Author: grehan
Date: Fri Jun 28 06:25:04 2013
New Revision: 252336
URL: http://svnweb.freebsd.org/changeset/base/252336

Log:
  Allow 8259 registers to be read. This is a transient condition
  during Linux boot.
  
  Submitted by: tycho nightingale at pluribusnetworks com
  Reviewed by:  neel

Modified:
  head/usr.sbin/bhyve/atpic.c

Modified: head/usr.sbin/bhyve/atpic.c
==
--- head/usr.sbin/bhyve/atpic.c Fri Jun 28 06:05:33 2013(r252335)
+++ head/usr.sbin/bhyve/atpic.c Fri Jun 28 06:25:04 2013(r252336)
@@ -37,13 +37,6 @@ __FBSDID($FreeBSD$);
 
 #include inout.h
 
-/*
- * FreeBSD only writes to the 8259 interrupt controllers to put them in a
- * shutdown state.
- *
- * So, we just ignore the writes.
- */
-
 #defineIO_ICU1 0x20
 #defineIO_ICU2 0xA0
 #defineICU_IMR_OFFSET  1
@@ -55,8 +48,14 @@ atpic_handler(struct vmctx *ctx, int vcp
if (bytes != 1)
return (-1);
 
-   if (in)
-   return (-1);
+   if (in) {
+   if (port  ICU_IMR_OFFSET) {
+   /* all interrupts masked */
+   *eax = 0xff;
+   } else {
+   *eax = 0x00;
+   }
+   }
 
/* Pretend all writes to the 8259 are alright */
return (0);
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252313 - head/sys/kern

[Konstantin Belousov]

On Fri, Jun 28, 2013 at 03:03:46AM +0200, Mateusz Guzik wrote:
 On Thu, Jun 27, 2013 at 07:14:04PM +, Mikolaj Golub wrote:
  Author: trociny
  Date: Thu Jun 27 19:14:03 2013
  New Revision: 252313
  URL: http://svnweb.freebsd.org/changeset/base/252313
  
  Log:
To avoid LOR, always drop the filedesc lock before exporting fd to sbuf.

Reviewed by:  kib
MFC after:3 days
  
  Modified:
head/sys/kern/kern_descrip.c
  
  Modified: head/sys/kern/kern_descrip.c
  ==
  --- head/sys/kern/kern_descrip.cThu Jun 27 18:59:07 2013
  (r252312)
  +++ head/sys/kern/kern_descrip.cThu Jun 27 19:14:03 2013
  (r252313)
  @@ -3427,12 +3427,10 @@ kern_proc_filedesc_out(struct proc *p,  
   * re-validate and re-evaluate its properties when
   * the loop continues.
   */
  -   if (type == KF_TYPE_VNODE || type == KF_TYPE_FIFO)
  -   FILEDESC_SUNLOCK(fdp);
  +   FILEDESC_SUNLOCK(fdp);
  error = export_fd_to_sb(data, type, i, fflags, refcnt,
  offset, fd_cap_rights, kif, sb, remainder);
  -   if (type == KF_TYPE_VNODE || type == KF_TYPE_FIFO)
  -   FILEDESC_SLOCK(fdp);
  +   FILEDESC_SLOCK(fdp);
  if (error)
  break;
  }
 
 Is this really ok? What prevents given fd from going away during
 export_fd_to_sb execution? Both DTYPE_VNODE and DTYPE_FIFO pass down
 a vrefed vnode so these are safe. But for example DTYPE_SOCKET goes with
 fp-f_data, which can go away in the meantime (or I'm misreading the code).
 
 I suggest obtainng ref to fp and passing it down in all cases.

Oops, I am sorry for missed this. But, I do not actually like the
idea of referencing the fd.  It de-facto means that the process calling
the sysctl duped the descriptor, potentially causing the close to be
performed in the sysctl context on dereference.

Ideal solution would be to drop the filedesc lock between processing
of the filedescriptors and draining sbuf while lock is dropped.


pgpZwWP2jtCkR.pgp
Description: PGP signature

svn commit: r252337 - head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs

[Gavin Atkinson]

Author: gavin
Date: Fri Jun 28 07:51:12 2013
New Revision: 252337
URL: http://svnweb.freebsd.org/changeset/base/252337

Log:
  Don't try to re-insert an already present but invalid page.
  
  This could happen if a thread doing a page-in loses a ZFS range lock
  race to a thread writing to the same range
  
  This fixes panic: vm_page_alloc: pindex already allocated in
  http://docs.FreeBSD.org/cgi/mid.cgi?1372165971.96049.42.camel
  
  Submitted by: avg
  MFC after:1 week

Modified:
  head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vnops.c

Modified: head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vnops.c
==
--- head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vnops.c Fri Jun 
28 06:25:04 2013(r252336)
+++ head/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vnops.c Fri Jun 
28 07:51:12 2013(r252337)
@@ -345,10 +345,13 @@ page_busy(vnode_t *vp, int64_t start, in
vm_page_sleep(pp, zfsmwb);
continue;
}
-   } else {
+   } else if (pp == NULL) {
pp = vm_page_alloc(obj, OFF_TO_IDX(start),
VM_ALLOC_SYSTEM | VM_ALLOC_IFCACHED |
VM_ALLOC_NOBUSY);
+   } else {
+   ASSERT(pp != NULL  !pp-valid);
+   pp = NULL;
}
 
if (pp != NULL) {
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252209 - in head: share/man/man9 sys/kern sys/sys

[Attilio Rao]

On Thu, Jun 27, 2013 at 4:34 PM, John Baldwin j...@freebsd.org wrote:
 On Wednesday, June 26, 2013 3:26:38 am Andre Oppermann wrote:
 On 25.06.2013 20:44, John Baldwin wrote:
  Author: jhb
  Date: Tue Jun 25 18:44:15 2013
  New Revision: 252209
  URL: http://svnweb.freebsd.org/changeset/base/252209
 
  Log:
 Several improvements to rmlock(9).  Many of these are based on patches
 provided by Isilon.
 - Add an rm_assert() supporting various lock assertions similar to other
   locking primitives.  Because rmlocks track readers the assertions are
   always fully accurate unlike rw_assert() and sx_assert().
 - Flesh out the lock class methods for rmlocks to support sleeping via
   condvars and rm_sleep() (but only while holding write locks), rmlock
   details in 'show lock' in DDB, and the lc_owner method used by
   dtrace.
 - Add an internal destroyed cookie so that API functions can assert
   that an rmlock is not destroyed.
 - Make use of rm_assert() to add various assertions to the API (e.g.
   to assert locks are held when an unlock routine is called).
 - Give RM_SLEEPABLE locks their own lock class and always use the
   rmlock's own lock_object with WITNESS.
 - Use THREAD_NO_SLEEPING() / THREAD_SLEEPING_OK() to disallow sleeping
   while holding a read lock on an rmlock.

 Thanks!

 Would it make sense to move struct rm_queue from struct pcpu itself to
 using DPCPU as a next step?

 Perhaps.  It might make pcpu.h cleaner, aside from that concern I don't think
 it really matters much.

It cannot for performance reasons. I had a comment ready for this but
I'm not sure if it was ever committed.

Attilio


--
Peace can only be achieved by understanding - A. Einstein
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252330 - in head/sys: conf geom kern sys vm

[Adrian Chadd]

Hi,

Do we really need another allocator / resource manager just for this?




Adrian

On 27 June 2013 20:51, Jeff Roberson j...@freebsd.org wrote:
 Author: jeff
 Date: Fri Jun 28 03:51:20 2013
 New Revision: 252330
 URL: http://svnweb.freebsd.org/changeset/base/252330

 Log:
- Add a general purpose resource allocator, vmem, from NetBSD.  It was
  originally inspired by the Solaris vmem detailed in the proceedings
  of usenix 2001.  The NetBSD version was heavily refactored for bugs
  and simplicity.
- Use this resource allocator to allocate the buffer and transient maps.
  Buffer cache defrags are reduced by 25% when used by filesystems with
  mixed block sizes.  Ultimately this may permit dynamic buffer cache
  sizing on low KVA machines.

   Discussed with:   alc, kib, attilio
   Tested by:pho
   Sponsored by: EMC / Isilon Storage Division

 Added:
   head/sys/kern/subr_vmem.c   (contents, props changed)
   head/sys/sys/vmem.h   (contents, props changed)
 Modified:
   head/sys/conf/files
   head/sys/geom/geom_io.c
   head/sys/kern/vfs_bio.c
   head/sys/sys/malloc.h
   head/sys/vm/vm.h
   head/sys/vm/vm_init.c
   head/sys/vm/vm_kern.c
   head/sys/vm/vm_kern.h
   head/sys/vm/vm_object.c
   head/sys/vm/vm_pager.c
   head/sys/vm/vm_pager.h

 Modified: head/sys/conf/files
 ==
 --- head/sys/conf/files Fri Jun 28 03:41:23 2013(r252329)
 +++ head/sys/conf/files Fri Jun 28 03:51:20 2013(r252330)
 @@ -2797,6 +2797,7 @@ kern/subr_trap.c  standard
  kern/subr_turnstile.c  standard
  kern/subr_uio.cstandard
  kern/subr_unit.c   standard
 +kern/subr_vmem.c   standard
  kern/subr_witness.coptional witness
  kern/sys_capability.c  standard
  kern/sys_generic.c standard

 Modified: head/sys/geom/geom_io.c
 ==
 --- head/sys/geom/geom_io.c Fri Jun 28 03:41:23 2013(r252329)
 +++ head/sys/geom/geom_io.c Fri Jun 28 03:51:20 2013(r252330)
 @@ -49,6 +49,7 @@ __FBSDID($FreeBSD$);
  #include sys/proc.h
  #include sys/stack.h
  #include sys/sysctl.h
 +#include sys/vmem.h

  #include sys/errno.h
  #include geom/geom.h
 @@ -626,7 +627,6 @@ g_io_transient_map_bio(struct bio *bp)
 vm_offset_t addr;
 long size;
 u_int retried;
 -   int rv;

 KASSERT(unmapped_buf_allowed, (unmapped disabled));

 @@ -636,10 +636,7 @@ g_io_transient_map_bio(struct bio *bp)
 retried = 0;
 atomic_add_long(transient_maps, 1);
  retry:
 -   vm_map_lock(bio_transient_map);
 -   if (vm_map_findspace(bio_transient_map, vm_map_min(bio_transient_map),
 -   size, addr)) {
 -   vm_map_unlock(bio_transient_map);
 +   if (vmem_alloc(transient_arena, size, M_BESTFIT | M_NOWAIT, addr)) {
 if (transient_map_retries != 0 
 retried = transient_map_retries) {
 g_io_deliver(bp, EDEADLK/* XXXKIB */);
 @@ -651,7 +648,7 @@ retry:
 /*
  * Naive attempt to quisce the I/O to get more
  * in-flight requests completed and defragment
 -* the bio_transient_map.
 +* the transient_arena.
  */
 CTR3(KTR_GEOM, g_down retrymap bp %p provider %s r 
 %d,
 bp, bp-bio_to-name, retried);
 @@ -661,12 +658,6 @@ retry:
 goto retry;
 }
 }
 -   rv = vm_map_insert(bio_transient_map, NULL, 0, addr, addr + size,
 -   VM_PROT_RW, VM_PROT_RW, MAP_NOFAULT);
 -   KASSERT(rv == KERN_SUCCESS,
 -   (vm_map_insert(bio_transient_map) rv %d %jx %lx,
 -   rv, (uintmax_t)addr, size));
 -   vm_map_unlock(bio_transient_map);
 atomic_add_int(inflight_transient_maps, 1);
 pmap_qenter((vm_offset_t)addr, bp-bio_ma, OFF_TO_IDX(size));
 bp-bio_data = (caddr_t)addr + bp-bio_ma_offset;

 Added: head/sys/kern/subr_vmem.c
 ==
 --- /dev/null   00:00:00 1970   (empty, because file is newly added)
 +++ head/sys/kern/subr_vmem.c   Fri Jun 28 03:51:20 2013(r252330)
 @@ -0,0 +1,1372 @@
 +/*-
 + * Copyright (c)2006,2007,2008,2009 YAMAMOTO Takashi,
 + * Copyright (c) 2013 EMC Corp.
 + * All rights reserved.
 + *
 + * Redistribution and use in source and binary forms, with or without
 + * modification, are permitted provided that the following conditions
 + * are met:
 + * 1. Redistributions of source code must retain the above copyright
 + *notice, this list of conditions and the following disclaimer.
 + * 2. Redistributions in binary form must reproduce the above copyright
 + *

Re: svn commit: r252313 - head/sys/kern

[Mikolaj Golub]

On Fri, Jun 28, 2013 at 09:44:30AM +0300, Konstantin Belousov wrote:
 On Fri, Jun 28, 2013 at 03:03:46AM +0200, Mateusz Guzik wrote:
  On Thu, Jun 27, 2013 at 07:14:04PM +, Mikolaj Golub wrote:
   Author: trociny
   Date: Thu Jun 27 19:14:03 2013
   New Revision: 252313
   URL: http://svnweb.freebsd.org/changeset/base/252313
   
   Log:
 To avoid LOR, always drop the filedesc lock before exporting fd to sbuf.
 
 Reviewed by:kib
 MFC after:  3 days
   
   Modified:
 head/sys/kern/kern_descrip.c
   
   Modified: head/sys/kern/kern_descrip.c
   ==
   --- head/sys/kern/kern_descrip.c  Thu Jun 27 18:59:07 2013
   (r252312)
   +++ head/sys/kern/kern_descrip.c  Thu Jun 27 19:14:03 2013
   (r252313)
   @@ -3427,12 +3427,10 @@ kern_proc_filedesc_out(struct proc *p,  
  * re-validate and re-evaluate its properties when
  * the loop continues.
  */
   - if (type == KF_TYPE_VNODE || type == KF_TYPE_FIFO)
   - FILEDESC_SUNLOCK(fdp);
   + FILEDESC_SUNLOCK(fdp);
 error = export_fd_to_sb(data, type, i, fflags, refcnt,
 offset, fd_cap_rights, kif, sb, remainder);
   - if (type == KF_TYPE_VNODE || type == KF_TYPE_FIFO)
   - FILEDESC_SLOCK(fdp);
   + FILEDESC_SLOCK(fdp);
 if (error)
 break;
 }
  
  Is this really ok? What prevents given fd from going away during
  export_fd_to_sb execution? Both DTYPE_VNODE and DTYPE_FIFO pass down
  a vrefed vnode so these are safe. But for example DTYPE_SOCKET goes with
  fp-f_data, which can go away in the meantime (or I'm misreading the code).
  

Thanks!

  I suggest obtainng ref to fp and passing it down in all cases.
 
 Oops, I am sorry for missed this. But, I do not actually like the
 idea of referencing the fd.  It de-facto means that the process calling
 the sysctl duped the descriptor, potentially causing the close to be
 performed in the sysctl context on dereference.
 
 Ideal solution would be to drop the filedesc lock between processing
 of the filedescriptors and draining sbuf while lock is dropped.

You mean something like below? (not well tested yet)

-- 
Mikolaj Golub
Index: sys/kern/kern_descrip.c
===
--- sys/kern/kern_descrip.c	(revision 252313)
+++ sys/kern/kern_descrip.c	(working copy)
@@ -3176,10 +3176,16 @@
 CTASSERT(sizeof(struct kinfo_file) == KINFO_FILE_SIZE);
 #endif
 
+struct export_fd_buf {
+	struct filedesc		*fdp;
+	struct sbuf 		*sb;
+	ssize_t			remainder;
+	struct kinfo_file	kif;
+};
+
 static int
 export_fd_to_sb(void *data, int type, int fd, int fflags, int refcnt,
-int64_t offset, cap_rights_t fd_cap_rights, struct kinfo_file *kif,
-struct sbuf *sb, ssize_t *remainder)
+int64_t offset, cap_rights_t fd_cap_rights, struct export_fd_buf *efbuf)
 {
 	struct {
 		int	fflag;
@@ -3202,16 +3208,20 @@
 		{ O_TRUNC, KF_FLAG_TRUNC }
 	};
 #define	NFFLAGS	(sizeof(fflags_table) / sizeof(*fflags_table))
+	struct kinfo_file *kif;
 	struct vnode *vp;
 	int error;
 	unsigned int i;
 
-	if (*remainder == 0)
+	if (efbuf-remainder == 0)
 		return (0);
+	kif = efbuf-kif;
 	bzero(kif, sizeof(*kif));
 	switch (type) {
 	case KF_TYPE_FIFO:
 	case KF_TYPE_VNODE:
+		if (efbuf-fdp != NULL)
+			FILEDESC_SUNLOCK(efbuf-fdp);
 		vp = (struct vnode *)data;
 		error = fill_vnode_info(vp, kif);
 		vrele(vp);
@@ -3255,15 +3265,19 @@
 	kif-kf_structsize = offsetof(struct kinfo_file, kf_path) +
 	strlen(kif-kf_path) + 1;
 	kif-kf_structsize = roundup(kif-kf_structsize, sizeof(uint64_t));
-	if (*remainder != -1) {
-		if (*remainder  kif-kf_structsize) {
+	if (efbuf-remainder != -1) {
+		if (efbuf-remainder  kif-kf_structsize) {
 			/* Terminate export. */
-			*remainder = 0;
+			efbuf-remainder = 0;
 			return (0);
 		}
-		*remainder -= kif-kf_structsize;
+		efbuf-remainder -= kif-kf_structsize;
 	}
-	error = sbuf_bcat(sb, kif, kif-kf_structsize);
+	if (efbuf-fdp != NULL  type != KF_TYPE_FIFO  type != KF_TYPE_VNODE)
+		FILEDESC_SUNLOCK(efbuf-fdp);
+	error = sbuf_bcat(efbuf-sb, kif, kif-kf_structsize);
+	if (efbuf-fdp != NULL)
+		FILEDESC_SLOCK(efbuf-fdp);
 	return (error);
 }
 
@@ -3277,18 +3291,16 @@
 {
 	struct file *fp;
 	struct filedesc *fdp;
-	struct kinfo_file *kif;
+	struct export_fd_buf *efbuf;
 	struct vnode *cttyvp, *textvp, *tracevp;
 	int64_t offset;
 	void *data;
-	ssize_t remainder;
 	int error, i;
 	int type, refcnt, fflags;
 	cap_rights_t fd_cap_rights;
 
 	PROC_LOCK_ASSERT(p, MA_OWNED);
 
-	remainder = maxlen;
 	/* ktrace vnode */
 	tracevp = p-p_tracevp;
 	if (tracevp != NULL)
@@ -3306,46 +3318,44 @@
 	}
 	fdp = fdhold(p);
 	PROC_UNLOCK(p);
-	kif = malloc(sizeof(*kif), M_TEMP, M_WAITOK);
+	efbuf = malloc(sizeof(*efbuf), M_TEMP, M_WAITOK);
+	efbuf-fdp = NULL;
+	efbuf-sb = sb;
+	efbuf-remainder = maxlen;
 	if 

svn commit: r252338 - head/crypto/openssh

[Dag-Erling Smørgrav]

Author: des
Date: Fri Jun 28 09:41:59 2013
New Revision: 252338
URL: http://svnweb.freebsd.org/changeset/base/252338

Log:
  r251088 reverted the default value for UsePrivilegeSeparation from
  sandbox to yes, but did not update the documentation to match.

Modified:
  head/crypto/openssh/sshd_config
  head/crypto/openssh/sshd_config.5

Modified: head/crypto/openssh/sshd_config
==
--- head/crypto/openssh/sshd_config Fri Jun 28 07:51:12 2013
(r252337)
+++ head/crypto/openssh/sshd_config Fri Jun 28 09:41:59 2013
(r252338)
@@ -105,7 +105,7 @@
 #PrintLastLog yes
 #TCPKeepAlive yes
 #UseLogin no
-#UsePrivilegeSeparation sandbox
+#UsePrivilegeSeparation yes
 #PermitUserEnvironment no
 #Compression delayed
 #ClientAliveInterval 0

Modified: head/crypto/openssh/sshd_config.5
==
--- head/crypto/openssh/sshd_config.5   Fri Jun 28 07:51:12 2013
(r252337)
+++ head/crypto/openssh/sshd_config.5   Fri Jun 28 09:41:59 2013
(r252338)
@@ -1169,7 +1169,7 @@ the privilege of the authenticated user.
 The goal of privilege separation is to prevent privilege
 escalation by containing any corruption within the unprivileged processes.
 The default is
-.Dq sandbox .
+.Dq yes .
 If
 .Cm UsePrivilegeSeparation
 is set to
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r251886 - in head: contrib/apr contrib/apr-util contrib/serf contrib/sqlite3 contrib/subversion share/mk usr.bin usr.bin/svn usr.bin/svn/lib usr.bin/svn/lib/libapr usr.bin/svn/lib/liba

[Jan Beich]

Peter Wemm pe...@freebsd.org writes:

 Author: peter
 Date: Tue Jun 18 02:53:45 2013
 New Revision: 251886
 URL: http://svnweb.freebsd.org/changeset/base/251886

 Log:
   Introduce svnlite so that we can check out our source code again.

Can you teach apr about WITH_ICONV?

  $ svnsync sync file:///freebsd/base
  svnsync: E22: Safe data 'In the Xen block front driver, take advantage of 
backends that
  support cache flush and write barrier commands.
  ...
  Submitted by:   Roger Pau Monn' was followed by non-ASCII byte 195: unable to 
convert to/from UTF-8
  Exit 1

  $ svn info file:///freebsd/base | awk '/^Rev/ { print $2 }'
  252259
  
  $ locale | awk -F. '/ctype/I { print $2; exit }'
  UTF-8

Not sure why svnsync tries to convert from UTF-8 to UTF-8 only to
fail without iconv().

Index: usr.bin/svn/lib/libapr_util/Makefile
===
--- usr.bin/svn/lib/libapr_util/Makefile(revision 252337)
+++ usr.bin/svn/lib/libapr_util/Makefile(working copy)
@@ -1,6 +1,7 @@
 # $FreeBSD$
 
 .include ${.CURDIR}/../Makefile.inc
+.include bsd.own.mk
 
 INTERNALLIB=   yes
 LIB=   apr-util
@@ -28,5 +29,11 @@ CFLAGS+= -DHAVE_CONFIG_H \
-I${APR}/include/arch/unix \
-I${APR}/include
 
+.if ${MK_ICONV} != no
+CFLAGS+=   -DHAVE_ICONV_H -DAPU_HAVE_ICONV=1
+.else
+CFLAGS+=   -DAPU_HAVE_ICONV=0
+.endif
+
 .include bsd.lib.mk
 
Index: usr.bin/svn/lib/libapr_util/apu.h
===
--- usr.bin/svn/lib/libapr_util/apu.h   (revision 252337)
+++ usr.bin/svn/lib/libapr_util/apu.h   (working copy)
@@ -85,7 +85,6 @@
 #define APU_HAVE_SQLITE2   0
 
 #define APU_HAVE_APR_ICONV 0
-#define APU_HAVE_ICONV 0
 #define APR_HAS_XLATE  (APU_HAVE_APR_ICONV || APU_HAVE_ICONV)
 
 #endif /* APU_H */
Index: usr.bin/svn/lib/libapr_util/apu_config.h
===
--- usr.bin/svn/lib/libapr_util/apu_config.h(revision 252337)
+++ usr.bin/svn/lib/libapr_util/apu_config.h(working copy)
@@ -24,9 +24,6 @@
 /* Define if expat.h is available */
 #define HAVE_EXPAT_H 1
 
-/* Define to 1 if you have the iconv.h header file. */
-/* #undef HAVE_ICONV_H */
-
 /* Define to 1 if you have the inttypes.h header file. */
 #define HAVE_INTTYPES_H 1
 
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252330 - in head/sys: conf geom kern sys vm

[Jeff Roberson]

On Fri, 28 Jun 2013, Adrian Chadd wrote:


Hi,

Do we really need another allocator / resource manager just for this?



No, however;  I have a follow-up patch to replace kmem with this.  And 
then we will use it for NUMA allocations in the kernel.  After that it is 
likely that we could replace several other less efficient allocators with 
this.  Solaris uses it for pids, tids, device unit numbers.  etc.  We 
could easily do the same.  The existing allocators have failure modes, big 
O cost, and allocation requirements that are not tolerable for use in the 
vm.  This also has a very nice feature that works with UMA to provide 
per-cpu caches of arbitrary number ranges.  So it is more scalable as 
well as providing for less fragmentation.


Thanks,
Jeff





Adrian

On 27 June 2013 20:51, Jeff Roberson j...@freebsd.org wrote:

Author: jeff
Date: Fri Jun 28 03:51:20 2013
New Revision: 252330
URL: http://svnweb.freebsd.org/changeset/base/252330

Log:
   - Add a general purpose resource allocator, vmem, from NetBSD.  It was
 originally inspired by the Solaris vmem detailed in the proceedings
 of usenix 2001.  The NetBSD version was heavily refactored for bugs
 and simplicity.
   - Use this resource allocator to allocate the buffer and transient maps.
 Buffer cache defrags are reduced by 25% when used by filesystems with
 mixed block sizes.  Ultimately this may permit dynamic buffer cache
 sizing on low KVA machines.

  Discussed with:   alc, kib, attilio
  Tested by:pho
  Sponsored by: EMC / Isilon Storage Division

Added:
  head/sys/kern/subr_vmem.c   (contents, props changed)
  head/sys/sys/vmem.h   (contents, props changed)
Modified:
  head/sys/conf/files
  head/sys/geom/geom_io.c
  head/sys/kern/vfs_bio.c
  head/sys/sys/malloc.h
  head/sys/vm/vm.h
  head/sys/vm/vm_init.c
  head/sys/vm/vm_kern.c
  head/sys/vm/vm_kern.h
  head/sys/vm/vm_object.c
  head/sys/vm/vm_pager.c
  head/sys/vm/vm_pager.h

Modified: head/sys/conf/files
==
--- head/sys/conf/files Fri Jun 28 03:41:23 2013(r252329)
+++ head/sys/conf/files Fri Jun 28 03:51:20 2013(r252330)
@@ -2797,6 +2797,7 @@ kern/subr_trap.c  standard
 kern/subr_turnstile.c  standard
 kern/subr_uio.cstandard
 kern/subr_unit.c   standard
+kern/subr_vmem.c   standard
 kern/subr_witness.coptional witness
 kern/sys_capability.c  standard
 kern/sys_generic.c standard

Modified: head/sys/geom/geom_io.c
==
--- head/sys/geom/geom_io.c Fri Jun 28 03:41:23 2013(r252329)
+++ head/sys/geom/geom_io.c Fri Jun 28 03:51:20 2013(r252330)
@@ -49,6 +49,7 @@ __FBSDID($FreeBSD$);
 #include sys/proc.h
 #include sys/stack.h
 #include sys/sysctl.h
+#include sys/vmem.h

 #include sys/errno.h
 #include geom/geom.h
@@ -626,7 +627,6 @@ g_io_transient_map_bio(struct bio *bp)
vm_offset_t addr;
long size;
u_int retried;
-   int rv;

KASSERT(unmapped_buf_allowed, (unmapped disabled));

@@ -636,10 +636,7 @@ g_io_transient_map_bio(struct bio *bp)
retried = 0;
atomic_add_long(transient_maps, 1);
 retry:
-   vm_map_lock(bio_transient_map);
-   if (vm_map_findspace(bio_transient_map, vm_map_min(bio_transient_map),
-   size, addr)) {
-   vm_map_unlock(bio_transient_map);
+   if (vmem_alloc(transient_arena, size, M_BESTFIT | M_NOWAIT, addr)) {
if (transient_map_retries != 0 
retried = transient_map_retries) {
g_io_deliver(bp, EDEADLK/* XXXKIB */);
@@ -651,7 +648,7 @@ retry:
/*
 * Naive attempt to quisce the I/O to get more
 * in-flight requests completed and defragment
-* the bio_transient_map.
+* the transient_arena.
 */
CTR3(KTR_GEOM, g_down retrymap bp %p provider %s r %d,
bp, bp-bio_to-name, retried);
@@ -661,12 +658,6 @@ retry:
goto retry;
}
}
-   rv = vm_map_insert(bio_transient_map, NULL, 0, addr, addr + size,
-   VM_PROT_RW, VM_PROT_RW, MAP_NOFAULT);
-   KASSERT(rv == KERN_SUCCESS,
-   (vm_map_insert(bio_transient_map) rv %d %jx %lx,
-   rv, (uintmax_t)addr, size));
-   vm_map_unlock(bio_transient_map);
atomic_add_int(inflight_transient_maps, 1);
pmap_qenter((vm_offset_t)addr, bp-bio_ma, OFF_TO_IDX(size));
bp-bio_data = (caddr_t)addr + bp-bio_ma_offset;

Added: head/sys/kern/subr_vmem.c
==
--- /dev/null   00:00:00 1970   (empty, because file is newly added)
+++ 

svn commit: r252341 - head/sys/dev/re

[Sergey Kandaurov]

Author: pluknet
Date: Fri Jun 28 14:41:06 2013
New Revision: 252341
URL: http://svnweb.freebsd.org/changeset/base/252341

Log:
  Correct comment typos.

Modified:
  head/sys/dev/re/if_re.c

Modified: head/sys/dev/re/if_re.c
==
--- head/sys/dev/re/if_re.c Fri Jun 28 10:21:49 2013(r252340)
+++ head/sys/dev/re/if_re.c Fri Jun 28 14:41:06 2013(r252341)
@@ -1321,7 +1321,7 @@ re_attach(device_t dev)
SYS_RES_IRQ, rid, RF_ACTIVE);
if (sc-rl_irq[i] == NULL) {
device_printf(dev,
-   couldn't llocate IRQ resources for 
+   couldn't allocate IRQ resources for 
message %d\n, rid);
error = ENXIO;
goto fail;
@@ -1632,7 +1632,7 @@ re_attach(device_t dev)
/*
 * Don't enable TSO by default.  It is known to generate
 * corrupted TCP segments(bad TCP options) under certain
-* circumtances.
+* circumstances.
 */
ifp-if_hwassist = ~CSUM_TSO;
ifp-if_capenable = ~(IFCAP_TSO4 | IFCAP_VLAN_HWTSO);
@@ -2784,7 +2784,7 @@ re_encap(struct rl_softc *sc, struct mbu
/*
 * Unconditionally enable IP checksum if TCP or UDP
 * checksum is required. Otherwise, TCP/UDP checksum
-* does't make effects.
+* doesn't make effects.
 */
if (((*m_head)-m_pkthdr.csum_flags  RE_CSUM_FEATURES) != 0) {
if ((sc-rl_flags  RL_FLAG_DESCV2) == 0) {
@@ -3247,7 +3247,7 @@ re_init_locked(struct rl_softc *sc)
if ((sc-rl_flags  RL_FLAG_JUMBOV2) != 0) {
/*
 * For controllers that use new jumbo frame scheme,
-* set maximum size of jumbo frame depedning on
+* set maximum size of jumbo frame depending on
 * controller revisions.
 */
if (ifp-if_mtu  RL_MTU)
@@ -3948,7 +3948,7 @@ re_sysctl_stats(SYSCTL_HANDLER_ARGS)
RL_UNLOCK(sc);
if (i == 0) {
device_printf(sc-rl_dev,
-   DUMP statistics request timedout\n);
+   DUMP statistics request timed out\n);
return (ETIMEDOUT);
}
 done:
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252101 - head/release

[Glen Barber]

On Thu, Jun 27, 2013 at 12:13:51PM -0700, Garrett Cooper wrote:
 On Jun 22, 2013, at 3:51 PM, Glen Barber wrote:
 
  On Sat, Jun 22, 2013 at 03:39:41PM -0700, Garrett Cooper wrote:
  On Jun 22, 2013, at 3:12 PM, Glen Barber wrote:
  
  Author: gjb
  Date: Sat Jun 22 22:12:24 2013
  New Revision: 252101
  URL: http://svnweb.freebsd.org/changeset/base/252101
  
  Log:
  If MAKE_CONF and/or SRC_CONF are set and not character devices,
  copy to chroot to use with release build.
  
  Submitted by: Garrett Cooper (original version)
  Approved by:  kib (mentor)
  MFC after:1 week
  
  Modified:
  head/release/release.sh
  
  Technically copying /dev/null to a file creates a 0 byte file.
  That's why I used cp in the patch I submitted over email.
  Thanks!
  
  Right.  But did you read the script?
  
  Both __MAKE_CONF and SRCCONF are set to the path of the local files.
 
 Oh, interesting… I see what you did there instead of what
 I proposed; I copied __MAKE_CONF to /etc/make.conf and SRCCONF to
 /etc/src.conf for symmetry with the base system, but your method
 works more than mine did (I forgot that you were passing those vars
 in the chroot environment). Should the files be installed to /etc/
 in the system default locations ?

My initial reaction is no.  Those who are using __MAKE_CONF and
SRCCONF in such an environment know why they are doing so.  So, in my
opinion, it is up to them to modify the end-release as needed.

Since __MAKE_CONF and SRCCONF are pointing to /dev/null in release
builds, I do not see this as necessary.  Moreso, I do not want to
introduce a level of confusion where /etc/make.conf and/or /etc/src.conf
are modified within the release, or worse, in an altogether different
location.  There is too much complexity here.

Glen



pgpwry168PK1B.pgp
Description: PGP signature

Re: svn commit: r252330 - in head/sys: conf geom kern sys vm

[Adrian Chadd]

On 28 June 2013 03:36, Jeff Roberson jrober...@jroberson.net wrote:

 Do we really need another allocator / resource manager just for this?


 No, however;  I have a follow-up patch to replace kmem with this.  And then
 we will use it for NUMA allocations in the kernel.  After that it is likely
 that we could replace several other less efficient allocators with this.
 Solaris uses it for pids, tids, device unit numbers.  etc.  We could easily
 do the same.  The existing allocators have failure modes, big O cost, and
 allocation requirements that are not tolerable for use in the vm.  This also
 has a very nice feature that works with UMA to provide per-cpu caches of
 arbitrary number ranges.  So it is more scalable as well as providing for
 less fragmentation.

Sweet. :)

Thanks,



-adrian
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252342 - head/sys/kern

[Ryan Stone]

Author: rstone
Date: Fri Jun 28 15:55:30 2013
New Revision: 252342
URL: http://svnweb.freebsd.org/changeset/base/252342

Log:
  Correct a bug that prevented deadlkres from (almost) ever firing.
  
  deadlkres was using a reversed test to check whether ticks had rolled over.
  This meant that deadlkres could only fire after ticks had rolled over.
  This test was actually unnecessary as deadlkres only ever took the
  difference of ticks values which is safe even in the presence of ticks
  rollover.  Remove the tests entirely.  Now deadlkres will properly fire
  after a lock has been held after the timeout period.
  
  MFC after:1 month

Modified:
  head/sys/kern/kern_clock.c

Modified: head/sys/kern/kern_clock.c
==
--- head/sys/kern/kern_clock.c  Fri Jun 28 14:41:06 2013(r252341)
+++ head/sys/kern/kern_clock.c  Fri Jun 28 15:55:30 2013(r252342)
@@ -216,13 +216,8 @@ deadlkres(void)
}
FOREACH_THREAD_IN_PROC(p, td) {
 
-   /*
-* Once a thread is found in interesting
-* state a possible ticks wrap-up needs to be
-* checked.
-*/
thread_lock(td);
-   if (TD_ON_LOCK(td)  ticks  td-td_blktick) {
+   if (TD_ON_LOCK(td)) {
 
/*
 * The thread should be blocked on a
@@ -247,8 +242,7 @@ deadlkres(void)
__func__, td, tticks);
}
} else if (TD_IS_SLEEPING(td) 
-   TD_ON_SLEEPQ(td) 
-   ticks  td-td_blktick) {
+   TD_ON_SLEEPQ(td)) {
 
/*
 * Check if the thread is sleeping on a
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252346 - head/share/man/man9

[John Baldwin]

Author: jhb
Date: Fri Jun 28 16:33:45 2013
New Revision: 252346
URL: http://svnweb.freebsd.org/changeset/base/252346

Log:
  Make a pass over this page to correct and clarify a few things as well as
  some general word-smithing.
  - Don't claim that adaptive mutexes have a timeout (they don't).
  - Don't treat pool mutexes as a separate primitive in a few places.
  - Describe sleepable read-mostly locks as a separate lock type and add
them to the various tables.
  - Don't claim that sx locks are less efficient.  That hasn't been true in
a few years now.
  - Describe lockmanager locks next to sx locks since they are very similar
in terms of rules, etc., and so that all the lock primitives are
grouped together before the non-lock primitives.
  - Similarly, move the section on Giant after the description of all the
non-lock primitives to preserve grouping.
  - Condition variables work on several types of locks, not just mutexes.
  - Add a bit of language to compare/contrast condition variables with
sleep/wakeup.
  - Add a note about why pause(9) is unique.
  - Add some language to define bounded vs unbounded sleeps and explain
why they are treated separately (bounded sleeps only need CPU time
to make forward progress).
  - Don't state that using mtx_sleep() is a bad idea.  It is in fact rather
necessary.
  - Rework the interaction table a bit.  First, it did not include really
include sleepable rmlocks and it left out lockmgr entirely.  To get
things to fit, combine similar lock types into the same column / row,
and explicitly state what sleep means.  The notes about recursion
and lock order were also a bit banal (lock order is always important,
not just in the few places annotated here), so remove them.  In
particular, the lock order note would need to be on just about every
cell.  If we want to document recursion I think a better approach
would be a separate table summarizing the recursion rules for each
lock as having too many notes clutters the table.
  - Tweak the tables to use less indentation so everything still fits with
the added columns.
  - Correct a few cells in the context mode table.
  - Use mdoc markup instead of explicit markup in a few places.
  
  Requested by: julian
  MFC after:2 weeks

Modified:
  head/share/man/man9/locking.9

Modified: head/share/man/man9/locking.9
==
--- head/share/man/man9/locking.9   Fri Jun 28 16:24:14 2013
(r252345)
+++ head/share/man/man9/locking.9   Fri Jun 28 16:33:45 2013
(r252346)
@@ -33,53 +33,52 @@
 .Sh DESCRIPTION
 The
 .Em FreeBSD
-kernel is written to run across multiple CPUs and as such requires
-several different synchronization primitives to allow the developers
-to safely access and manipulate the many data types required.
+kernel is written to run across multiple CPUs and as such provides
+several different synchronization primitives to allow developers
+to safely access and manipulate many data types.
 .Ss Mutexes
-Mutexes (also erroneously called sleep mutexes) are the most commonly used
+Mutexes (also called blocking mutexes) are the most commonly used
 synchronization primitive in the kernel.
 A thread acquires (locks) a mutex before accessing data shared with other
 threads (including interrupt threads), and releases (unlocks) it afterwards.
 If the mutex cannot be acquired, the thread requesting it will wait.
-Mutexes are by default adaptive, meaning that
+Mutexes are adaptive by default, meaning that
 if the owner of a contended mutex is currently running on another CPU,
-then a thread attempting to acquire the mutex will briefly spin
-in the hope that the owner is only briefly holding it,
-and might release it shortly.
-If the owner does not do so, the waiting thread proceeds to yield the 
processor,
-allowing other threads to run.
-If the owner is not currently actually running then the spin step is skipped.
+then a thread attempting to acquire the mutex will spin rather than yielding
+the processor.
 Mutexes fully support priority propagation.
 .Pp
 See
 .Xr mutex 9
 for details.
-.Ss Spin mutexes
-Spin mutexes are variation of basic mutexes; the main difference between
-the two is that spin mutexes never yield the processor - instead, they spin,
-waiting for the thread holding the lock,
-(which must be running on another CPU), to release it.
-Spin mutexes disable interrupts while the held so as to not get pre-empted.
-Since disabling interrupts is expensive, they are also generally slower.
-Spin mutexes should be used only when necessary, e.g. to protect data shared
+.Ss Spin Mutexes
+Spin mutexes are a variation of basic mutexes; the main difference between
+the two is that spin mutexes never block.
+Instead, they spin while waiting for the lock to be released.
+Note that a thread that holds a spin mutex must never yield its CPU to
+avoid deadlock.
+Unlike 

Re: svn commit: r252346 - head/share/man/man9

[Benjamin Kaduk]

On Fri, Jun 28, 2013 at 12:33 PM, John Baldwin j...@freebsd.org wrote:

 Author: jhb
 Date: Fri Jun 28 16:33:45 2013
 New Revision: 252346
 URL: http://svnweb.freebsd.org/changeset/base/252346

 Log:
   Make a pass over this page to correct and clarify a few things as well as
   some general word-smithing.


Many thanks for this!  A few nits...


 Modified: head/share/man/man9/locking.9

 ==
 --- head/share/man/man9/locking.9   Fri Jun 28 16:24:14 2013
  (r252345)
 +++ head/share/man/man9/locking.9   Fri Jun 28 16:33:45 2013
  (r252346)
 -.Ss Spin mutexes
 -Spin mutexes are variation of basic mutexes; the main difference between
 -the two is that spin mutexes never yield the processor - instead, they
 spin,
 -waiting for the thread holding the lock,
 -(which must be running on another CPU), to release it.
 -Spin mutexes disable interrupts while the held so as to not get
 pre-empted.
 -Since disabling interrupts is expensive, they are also generally slower.
 -Spin mutexes should be used only when necessary, e.g. to protect data
 shared
 +.Ss Spin Mutexes
 +Spin mutexes are a variation of basic mutexes; the main difference between
 +the two is that spin mutexes never block.
 +Instead, they spin while waiting for the lock to be released.
 +Note that a thread that holds a spin mutex must never yield its CPU to
 +avoid deadlock.


I think this needs a comma after CPU (or reordering to emphasize the
deadlock avoidance part).


 +Unlike ordinary mutexes, spin mutexes disable interrupts when acquired.
 +Since disabling interrupts can be expensive, they are generally slower to
 +acquire and release.
 +Spin mutexes should be used only when absolutely necessary,
 +e.g. to protect data shared
  with interrupt filter code (see
  .Xr bus_setup_intr 9
 -for details).
 -.Ss Pool mutexes
 -With most synchronization primitives, such as mutexes, programmer must
 -provide a piece of allocated memory to hold the primitive.
 +for details),
 +or for scheduler internals.

[...]

 @@ -122,21 +111,41 @@ data structure.
 +.Ss Lockmanager locks
 +Lockmanager locks are sleepable shared/exclusive locks used mostly in
 +.Xr VFS 9
 +.Po
 +as a
 +.Xr vnode 9
 +lock
 +.Pc
 +and in the buffer cache
 +.Po
 +.Xr BUF_LOCK 9
 +.Pc .
 +They have features other lock types do not have such as sleep
 +timeouts, blocking upgrades,
 +writer starvation avoidance, draining, and an interlock mutex,
 +but this makes them complicated to both use and implement;


I think complicated both to use and to implement flows better; in the
present text, both is something of an interruption.


 +for this reason, they should be avoided.
 +.Pp
 +See
 +.Xr lock 9
 +for details.
 @@ -194,7 +181,12 @@ The functions
  .Fn wakeup ,
  and
  .Fn wakeup_one
 -handle event-based thread blocking.
 +also handle event-based thread blocking.
 +Unlike condition variables,
 +arbitrary addresses may be used as wait channels and an dedicated


s/an /a /


 +structure does not need to be allocated.
 +However, care must be taken to ensure that wait channel addresses are
 +unique to an event.
  If a thread must wait for an external event, it is put to sleep by
  .Fn tsleep ,
  .Fn msleep ,
 @@ -232,122 +225,168 @@ includes the
  .Sh INTERACTIONS
 -The primitives interact and have a number of rules regarding how
 +The primitives can interact and have a number of rules regarding how
  they can and can not be combined.
 -Many of these rules are checked using the
 -.Xr witness 4
 -code.
 -.Ss Bounded vs. unbounded sleep
 -The following primitives perform bounded sleep:
 - mutexes, pool mutexes, reader/writer locks and read-mostly locks.
 -.Pp
 -The following primitives may perform an unbounded sleep:
 -shared/exclusive locks, counting semaphores, condition variables,
 sleep/wakeup and lockmanager locks.
 -.Pp
 +Many of these rules are checked by
 +.Xr witness 4 .
 +.Ss Bounded vs. Unbounded Sleep
 +A bounded sleep
 +.Pq or blocking


The Pq should probably go before sleep?


 +is a sleep where the only resource needed to resume execution of a thread
 +is CPU time for the owner of a lock that the thread is waiting to acquire.
 +An unbounded sleep
 +.Po
 +often referred to as simply
 +.Dq sleeping
 +.Pc
 +is a sleep where a thread is waiting for an external event or for a
 condition
 +to become true.
 +In particular,
 +since there is always CPU time available,
 +a dependency chain of threads in bounded sleeps should always make forward
 +progress.


This sentence is a little bit misleading, as the key fact that the threads
in question are in bounded sleeps is buried in the last clause.  On first
reading, one might think that we are still talking about unbounded sleeps,
so having CPU time always available is quite the non sequitur.  I would
reorder things so that since there is always CPU time available is last.


 +This requires that no thread in a bounded sleep is waiting for a lock held
 +by a thread in an 

Re: svn commit: r252346 - head/share/man/man9

[John Baldwin]

On Friday, June 28, 2013 1:08:36 pm Benjamin Kaduk wrote:
  +.Ss Spin Mutexes
  +Spin mutexes are a variation of basic mutexes; the main difference between
  +the two is that spin mutexes never block.
  +Instead, they spin while waiting for the lock to be released.
  +Note that a thread that holds a spin mutex must never yield its CPU to
  +avoid deadlock.
 
 
 I think this needs a comma after CPU (or reordering to emphasize the
 deadlock avoidance part).

Hmm, perhaps:

To avoid deadlock, a thread that holds a spin mutex must never yield its CPU.

?
  
 I think complicated both to use and to implement flows better; in the
 present text, both is something of an interruption.

Yes, both to does flow better.

  +Unlike condition variables,
  +arbitrary addresses may be used as wait channels and an dedicated
 
 
 s/an /a /

Agreed.
  
  +.Ss Bounded vs. Unbounded Sleep
  +A bounded sleep
  +.Pq or blocking
 
 
 The Pq should probably go before sleep?

I think not.  The root sentence is A bounded sleep is a sleep where...

If you move the parent you get:

A bounded is a sleep where...

However, the sleep is a sleep bit is awkward.  Perhaps:

In a bounded sleep (or blocking), the only resource needed...

and then

In an unbounded sleep (often referred to as simply sleeping), a thread
waits for an external event...

 
  +is a sleep where the only resource needed to resume execution of a thread
  +is CPU time for the owner of a lock that the thread is waiting to acquire.
  +An unbounded sleep
  +.Po
  +often referred to as simply
  +.Dq sleeping
  +.Pc
  +is a sleep where a thread is waiting for an external event or for a
  condition
  +to become true.
  +In particular,
  +since there is always CPU time available,
  +a dependency chain of threads in bounded sleeps should always make forward
  +progress.
 
 
 This sentence is a little bit misleading, as the key fact that the threads
 in question are in bounded sleeps is buried in the last clause.  On first
 reading, one might think that we are still talking about unbounded sleeps,
 so having CPU time always available is quite the non sequitur.  I would
 reorder things so that since there is always CPU time available is last.

Agreed.

 Let me know if I should just send a patch off to my mentor for approval.

I can review it.

-- 
John Baldwin
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252349 - head/sys/kern

[Mikolaj Golub]

Author: trociny
Date: Fri Jun 28 18:07:41 2013
New Revision: 252349
URL: http://svnweb.freebsd.org/changeset/base/252349

Log:
  Rework r252313:
  
  The filedesc lock may not be dropped unconditionally before exporting
  fd to sbuf: fd might go away during execution.  While it is ok for
  DTYPE_VNODE and DTYPE_FIFO because the export is from a vrefed vnode
  here, for other types it is unsafe.
  
  Instead, drop the lock in export_fd_to_sb(), after preparing data in
  memory and before writing to sbuf.
  
  Spotted by:   mjg
  Suggested by: kib
  Review by:kib
  MFC after:1 week

Modified:
  head/sys/kern/kern_descrip.c

Modified: head/sys/kern/kern_descrip.c
==
--- head/sys/kern/kern_descrip.cFri Jun 28 17:18:28 2013
(r252348)
+++ head/sys/kern/kern_descrip.cFri Jun 28 18:07:41 2013
(r252349)
@@ -3176,10 +3176,16 @@ static SYSCTL_NODE(_kern_proc, KERN_PROC
 CTASSERT(sizeof(struct kinfo_file) == KINFO_FILE_SIZE);
 #endif
 
+struct export_fd_buf {
+   struct filedesc *fdp;
+   struct sbuf *sb;
+   ssize_t remainder;
+   struct kinfo_file   kif;
+};
+
 static int
 export_fd_to_sb(void *data, int type, int fd, int fflags, int refcnt,
-int64_t offset, cap_rights_t fd_cap_rights, struct kinfo_file *kif,
-struct sbuf *sb, ssize_t *remainder)
+int64_t offset, cap_rights_t fd_cap_rights, struct export_fd_buf *efbuf)
 {
struct {
int fflag;
@@ -3202,16 +3208,23 @@ export_fd_to_sb(void *data, int type, in
{ O_TRUNC, KF_FLAG_TRUNC }
};
 #defineNFFLAGS (sizeof(fflags_table) / sizeof(*fflags_table))
+   struct kinfo_file *kif;
struct vnode *vp;
-   int error;
+   int error, locked;
unsigned int i;
 
-   if (*remainder == 0)
+   if (efbuf-remainder == 0)
return (0);
+   kif = efbuf-kif;
bzero(kif, sizeof(*kif));
+   locked = efbuf-fdp != NULL;
switch (type) {
case KF_TYPE_FIFO:
case KF_TYPE_VNODE:
+   if (locked) {
+   FILEDESC_SUNLOCK(efbuf-fdp);
+   locked = 0;
+   }
vp = (struct vnode *)data;
error = fill_vnode_info(vp, kif);
vrele(vp);
@@ -3255,15 +3268,19 @@ export_fd_to_sb(void *data, int type, in
kif-kf_structsize = offsetof(struct kinfo_file, kf_path) +
strlen(kif-kf_path) + 1;
kif-kf_structsize = roundup(kif-kf_structsize, sizeof(uint64_t));
-   if (*remainder != -1) {
-   if (*remainder  kif-kf_structsize) {
+   if (efbuf-remainder != -1) {
+   if (efbuf-remainder  kif-kf_structsize) {
/* Terminate export. */
-   *remainder = 0;
+   efbuf-remainder = 0;
return (0);
}
-   *remainder -= kif-kf_structsize;
+   efbuf-remainder -= kif-kf_structsize;
}
-   error = sbuf_bcat(sb, kif, kif-kf_structsize);
+   if (locked)
+   FILEDESC_SUNLOCK(efbuf-fdp);
+   error = sbuf_bcat(efbuf-sb, kif, kif-kf_structsize);
+   if (efbuf-fdp != NULL)
+   FILEDESC_SLOCK(efbuf-fdp);
return (error);
 }
 
@@ -3277,18 +3294,16 @@ kern_proc_filedesc_out(struct proc *p,  
 {
struct file *fp;
struct filedesc *fdp;
-   struct kinfo_file *kif;
+   struct export_fd_buf *efbuf;
struct vnode *cttyvp, *textvp, *tracevp;
int64_t offset;
void *data;
-   ssize_t remainder;
int error, i;
int type, refcnt, fflags;
cap_rights_t fd_cap_rights;
 
PROC_LOCK_ASSERT(p, MA_OWNED);
 
-   remainder = maxlen;
/* ktrace vnode */
tracevp = p-p_tracevp;
if (tracevp != NULL)
@@ -3306,46 +3321,44 @@ kern_proc_filedesc_out(struct proc *p,  
}
fdp = fdhold(p);
PROC_UNLOCK(p);
-   kif = malloc(sizeof(*kif), M_TEMP, M_WAITOK);
+   efbuf = malloc(sizeof(*efbuf), M_TEMP, M_WAITOK);
+   efbuf-fdp = NULL;
+   efbuf-sb = sb;
+   efbuf-remainder = maxlen;
if (tracevp != NULL)
export_fd_to_sb(tracevp, KF_TYPE_VNODE, KF_FD_TYPE_TRACE,
-   FREAD | FWRITE, -1, -1, 0, kif, sb, remainder);
+   FREAD | FWRITE, -1, -1, 0, efbuf);
if (textvp != NULL)
export_fd_to_sb(textvp, KF_TYPE_VNODE, KF_FD_TYPE_TEXT,
-   FREAD, -1, -1, 0, kif, sb, remainder);
+   FREAD, -1, -1, 0, efbuf);
if (cttyvp != NULL)
export_fd_to_sb(cttyvp, KF_TYPE_VNODE, KF_FD_TYPE_CTTY,
-   FREAD | FWRITE, -1, -1, 0, kif, sb, remainder);
+   FREAD | FWRITE, -1, -1, 0, efbuf);
error = 0;
if (fdp == NULL)
 

svn commit: r252351 - head/sys/kern

[Mateusz Guzik]

Author: mjg
Date: Fri Jun 28 18:32:46 2013
New Revision: 252351
URL: http://svnweb.freebsd.org/changeset/base/252351

Log:
  Remove duplicate NULL check in kern_proc_filedesc_out.
  No functional changes.
  
  MFC after:1 week

Modified:
  head/sys/kern/kern_descrip.c

Modified: head/sys/kern/kern_descrip.c
==
--- head/sys/kern/kern_descrip.cFri Jun 28 18:25:04 2013
(r252350)
+++ head/sys/kern/kern_descrip.cFri Jun 28 18:32:46 2013
(r252351)
@@ -3446,9 +3446,8 @@ kern_proc_filedesc_out(struct proc *p,  
break;
}
FILEDESC_SUNLOCK(fdp);
+   fddrop(fdp);
 fail:
-   if (fdp != NULL)
-   fddrop(fdp);
free(efbuf, M_TEMP);
return (error);
 }
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252352 - head/sys/fs/smbfs

[Davide Italiano]

Author: davide
Date: Fri Jun 28 20:07:24 2013
New Revision: 252352
URL: http://svnweb.freebsd.org/changeset/base/252352

Log:
  Plug a couple of leakages in smbfs_lookup().

Modified:
  head/sys/fs/smbfs/smbfs_vnops.c

Modified: head/sys/fs/smbfs/smbfs_vnops.c
==
--- head/sys/fs/smbfs/smbfs_vnops.c Fri Jun 28 18:32:46 2013
(r252351)
+++ head/sys/fs/smbfs/smbfs_vnops.c Fri Jun 28 20:07:24 2013
(r252352)
@@ -1282,11 +1282,14 @@ smbfs_lookup(ap)
error = vfs_busy(mp, 0);
vn_lock(dvp, LK_EXCLUSIVE | LK_RETRY);
vfs_rel(mp);
-   if (error)
-   return (ENOENT);
+   if (error) {
+   error = ENOENT;
+   goto out;
+   }
if ((dvp-v_iflag  VI_DOOMED) != 0) {
vfs_unbusy(mp);
-   return (ENOENT);
+   error = ENOENT;
+   goto out;
}
}   
VOP_UNLOCK(dvp, 0);
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252353 - head/sys/fs/smbfs

[Davide Italiano]

Author: davide
Date: Fri Jun 28 20:14:30 2013
New Revision: 252353
URL: http://svnweb.freebsd.org/changeset/base/252353

Log:
  Garbage collect an useless check. smp should be never NULL.

Modified:
  head/sys/fs/smbfs/smbfs_vfsops.c

Modified: head/sys/fs/smbfs/smbfs_vfsops.c
==
--- head/sys/fs/smbfs/smbfs_vfsops.cFri Jun 28 20:07:24 2013
(r252352)
+++ head/sys/fs/smbfs/smbfs_vfsops.cFri Jun 28 20:14:30 2013
(r252353)
@@ -324,11 +324,6 @@ smbfs_root(struct mount *mp, int flags, 
td = curthread;
cred = td-td_ucred;
 
-   if (smp == NULL) {
-   SMBERROR(smp == NULL (bug in umount)\n);
-   vfs_mount_error(mp, smp == NULL (bug in umount));
-   return EINVAL;
-   }
if (smp-sm_root) {
*vpp = SMBTOV(smp-sm_root);
return vget(*vpp, LK_EXCLUSIVE | LK_RETRY, td);
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252354 - head/sys/netsmb

[Davide Italiano]

Author: davide
Date: Fri Jun 28 20:21:13 2013
New Revision: 252354
URL: http://svnweb.freebsd.org/changeset/base/252354

Log:
  Remove a reference to LK_DRAIN now that lockmgr(9) is gone from this
  piece of code.
  
  Reported by:  attilio

Modified:
  head/sys/netsmb/smb_conn.c

Modified: head/sys/netsmb/smb_conn.c
==
--- head/sys/netsmb/smb_conn.c  Fri Jun 28 20:14:30 2013(r252353)
+++ head/sys/netsmb/smb_conn.c  Fri Jun 28 20:21:13 2013(r252354)
@@ -530,7 +530,6 @@ smb_vc_free(struct smb_connobj *cp)
 
 /*
  * Called when use count of VC dropped to zero.
- * VC should be locked on enter with LK_DRAIN.
  */
 static void
 smb_vc_gone(struct smb_connobj *cp, struct smb_cred *scred)
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252355 - head/sys/fs/smbfs

[Davide Italiano]

Author: davide
Date: Fri Jun 28 20:32:48 2013
New Revision: 252355
URL: http://svnweb.freebsd.org/changeset/base/252355

Log:
  Properly use v_data field. This magically worked (even if wrong) until
  now because v_data is the first field of the structure, but it's not
  something we should rely on.

Modified:
  head/sys/fs/smbfs/smbfs_node.c

Modified: head/sys/fs/smbfs/smbfs_node.c
==
--- head/sys/fs/smbfs/smbfs_node.c  Fri Jun 28 20:21:13 2013
(r252354)
+++ head/sys/fs/smbfs/smbfs_node.c  Fri Jun 28 20:32:48 2013
(r252355)
@@ -89,7 +89,7 @@ smbfs_vnode_cmp(struct vnode *vp, void *
struct smbnode *np;
struct smbcmp *sc;
 
-   np = (struct smbnode *) vp;
+   np = (struct smbnode *) vp-v_data;
sc = (struct smbcmp *) _sc;
if (np-n_parent != sc-n_parent || np-n_nmlen != sc-n_nmlen ||
bcmp(sc-n_name, np-n_name, sc-n_nmlen) != 0)
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252356 - in head: contrib/smbfs/mount_smbfs etc/defaults etc/mtree include lib lib/libprocstat rescue/rescue sbin/mount share/examples share/examples/etc share/mk sys/conf sys/kern sys...

[Davide Italiano]

Author: davide
Date: Fri Jun 28 21:00:08 2013
New Revision: 252356
URL: http://svnweb.freebsd.org/changeset/base/252356

Log:
  - Trim an unused and bogus Makefile for mount_smbfs.
  - Reconnect with some minor modifications, in particular now selsocket()
  internals are adapted to use sbintime units after recent'ish calloutng
  switch.

Deleted:
  head/contrib/smbfs/mount_smbfs/Makefile
Modified:
  head/etc/defaults/rc.conf
  head/etc/mtree/BSD.include.dist
  head/etc/mtree/BSD.usr.dist
  head/include/Makefile
  head/lib/Makefile
  head/lib/libprocstat/Makefile
  head/lib/libprocstat/common_kvm.h
  head/lib/libprocstat/libprocstat.c
  head/rescue/rescue/Makefile
  head/sbin/mount/mount.8
  head/sbin/mount/mount.c
  head/share/examples/Makefile
  head/share/examples/etc/README.examples
  head/share/mk/bsd.libnames.mk
  head/sys/conf/NOTES
  head/sys/conf/files
  head/sys/conf/files.amd64
  head/sys/conf/files.arm
  head/sys/conf/files.i386
  head/sys/conf/files.ia64
  head/sys/conf/files.mips
  head/sys/conf/files.pc98
  head/sys/conf/files.powerpc
  head/sys/conf/files.sparc64
  head/sys/conf/options
  head/sys/kern/Make.tags.inc
  head/sys/kern/sys_generic.c
  head/sys/modules/Makefile
  head/sys/sys/socketvar.h
  head/usr.bin/Makefile.amd64
  head/usr.bin/Makefile.i386
  head/usr.bin/Makefile.ia64
  head/usr.bin/Makefile.powerpc
  head/usr.bin/Makefile.sparc64
  head/usr.bin/kdump/kdump.c
  head/usr.sbin/Makefile.amd64
  head/usr.sbin/Makefile.i386
  head/usr.sbin/Makefile.ia64
  head/usr.sbin/Makefile.powerpc
  head/usr.sbin/Makefile.sparc64

Modified: head/etc/defaults/rc.conf
==
--- head/etc/defaults/rc.conf   Fri Jun 28 20:32:48 2013(r252355)
+++ head/etc/defaults/rc.conf   Fri Jun 28 21:00:08 2013(r252356)
@@ -88,7 +88,7 @@ fsck_y_enable=NO# Set to YES to do fs
 fsck_y_flags=# Additional flags for fsck -y
 background_fsck=YES  # Attempt to run fsck in the background where possible.
 background_fsck_delay=60 # Time to wait (seconds) before starting the fsck.
-netfs_types=nfs:NFS oldnfs:OLDNFS # Net filesystems.
+netfs_types=nfs:NFS oldnfs:OLDNFS smbfs:SMB # Net filesystems.
 extra_netfs_types=NO # List of network extra filesystem types for delayed
# mount at startup (or NO).
 

Modified: head/etc/mtree/BSD.include.dist
==
--- head/etc/mtree/BSD.include.dist Fri Jun 28 20:32:48 2013
(r252355)
+++ head/etc/mtree/BSD.include.dist Fri Jun 28 21:00:08 2013
(r252356)
@@ -176,6 +176,8 @@
 ..
 procfs
 ..
+smbfs
+..
 udf
 ..
 unionfs

Modified: head/etc/mtree/BSD.usr.dist
==
--- head/etc/mtree/BSD.usr.dist Fri Jun 28 20:32:48 2013(r252355)
+++ head/etc/mtree/BSD.usr.dist Fri Jun 28 21:00:08 2013(r252356)
@@ -309,6 +309,10 @@
 srcs
 ..
 ..
+smbfs
+print
+..
+..
 sunrpc
 dir
 ..

Modified: head/include/Makefile
==
--- head/include/Makefile   Fri Jun 28 20:32:48 2013(r252355)
+++ head/include/Makefile   Fri Jun 28 21:00:08 2013(r252356)
@@ -37,7 +37,7 @@ LHDRS=aio.h errno.h fcntl.h linker_set.
syslog.h ucontext.h
 
 LDIRS= bsm cam geom net net80211 netatalk netgraph netinet netinet6 \
-   netipsec ${_netipx} netnatm \
+   netipsec ${_netipx} netnatm netsmb \
nfs nfsclient nfsserver \
sys vm
 
@@ -48,7 +48,7 @@ LSUBDIRS= cam/ata cam/scsi \
dev/ofw dev/pbio dev/pci ${_dev_powermac_nvram} dev/ppbus dev/smbus \
dev/speaker dev/usb dev/utopia dev/vkbd dev/wi \
fs/devfs fs/fdescfs fs/msdosfs fs/nandfs fs/nfs fs/nullfs \
-   fs/procfs fs/udf fs/unionfs \
+   fs/procfs fs/smbfs fs/udf fs/unionfs \
geom/cache geom/concat geom/eli geom/gate geom/journal geom/label \
geom/mirror geom/mountver geom/multipath geom/nop \
geom/raid geom/raid3 geom/shsec geom/stripe geom/virstor \
@@ -80,6 +80,7 @@ INCS+=iconv.h
 LSUBSUBDIRS+=  netgraph/bluetooth/include
 .endif
 
+# XXX unconditionally needed by netsmb/netbios.h
 #.if ${MK_IPX} != no
 _netipx=   netipx
 #.endif

Modified: head/lib/Makefile
==
--- head/lib/Makefile   Fri Jun 28 20:32:48 2013(r252355)
+++ head/lib/Makefile   Fri Jun 28 21:00:08 2013(r252356)
@@ -102,6 +102,7 @@ SUBDIR= ${SUBDIR_ORDERED} \
${_librtld_db} \
${_libsdp} \
${_libsm} \
+   ${_libsmb} \
${_libsmdb} \
${_libsmutil} \

svn commit: r252357 - head/sys/kern

[Davide Italiano]

Author: davide
Date: Fri Jun 28 21:04:15 2013
New Revision: 252357
URL: http://svnweb.freebsd.org/changeset/base/252357

Log:
  Correct the comment above _sleep() function which still mentions 'timo'
  instead of 'sbintime_t'.
  
  Reported by:  kan

Modified:
  head/sys/kern/kern_synch.c

Modified: head/sys/kern/kern_synch.c
==
--- head/sys/kern/kern_synch.c  Fri Jun 28 21:00:08 2013(r252356)
+++ head/sys/kern/kern_synch.c  Fri Jun 28 21:04:15 2013(r252357)
@@ -132,7 +132,7 @@ sleepinit(void)
 /*
  * General sleep call.  Suspends the current thread until a wakeup is
  * performed on the specified identifier.  The thread will then be made
- * runnable with the specified priority.  Sleeps at most timo/hz seconds
+ * runnable with the specified priority.  Sleeps at most sbt units of time
  * (0 means no timeout).  If pri includes the PCATCH flag, let signals
  * interrupt the sleep, otherwise ignore them while sleeping.  Returns 0 if
  * awakened, EWOULDBLOCK if the timeout expires.  If PCATCH is set and a
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252356 - in head: contrib/smbfs/mount_smbfs etc/defaults etc/mtree include lib lib/libprocstat rescue/rescue sbin/mount share/examples share/examples/etc share/mk sys/conf sys/kern sy

[Attilio Rao]

On Fri, Jun 28, 2013 at 11:00 PM, Davide Italiano dav...@freebsd.org wrote:
 Author: davide
 Date: Fri Jun 28 21:00:08 2013
 New Revision: 252356
 URL: http://svnweb.freebsd.org/changeset/base/252356

 Log:
   - Trim an unused and bogus Makefile for mount_smbfs.
   - Reconnect with some minor modifications, in particular now selsocket()
   internals are adapted to use sbintime units after recent'ish calloutng
   switch.

I've updated the page at:
https://wiki.freebsd.org/NONMPSAFE_DEORBIT_VFS

which reflects completition of the project and the re-insert of smbfs.
During VFS Non-MPSAFE deorbit we lost the following in-kernel support:
CodaFS, HPFS, NTFS, NWFS, PortalFS, XFS

even if for some of those (most notably NTFS) there is an userland
module for FUSE available.

Thanks,
Attilio


--
Peace can only be achieved by understanding - A. Einstein
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252358 - head/sys/vm

[Davide Italiano]

Author: davide
Date: Fri Jun 28 21:13:19 2013
New Revision: 252358
URL: http://svnweb.freebsd.org/changeset/base/252358

Log:
  Remove a spurious keg lock acquisition.

Modified:
  head/sys/vm/uma_core.c

Modified: head/sys/vm/uma_core.c
==
--- head/sys/vm/uma_core.c  Fri Jun 28 21:04:15 2013(r252357)
+++ head/sys/vm/uma_core.c  Fri Jun 28 21:13:19 2013(r252358)
@@ -493,7 +493,7 @@ keg_timeout(uma_keg_t keg)
 
KEG_UNLOCK(keg);
hash_free(oldhash);
-   KEG_LOCK(keg);
+   return;
}
}
KEG_UNLOCK(keg);
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252359 - head/bin/sh

[Jilles Tjoelker]

Author: jilles
Date: Fri Jun 28 21:47:08 2013
New Revision: 252359
URL: http://svnweb.freebsd.org/changeset/base/252359

Log:
  sh: Do not close(-1) if pipe() fails.

Modified:
  head/bin/sh/eval.c

Modified: head/bin/sh/eval.c
==
--- head/bin/sh/eval.c  Fri Jun 28 21:13:19 2013(r252358)
+++ head/bin/sh/eval.c  Fri Jun 28 21:47:08 2013(r252359)
@@ -589,7 +589,8 @@ evalpipe(union node *n)
pip[1] = -1;
if (lp-next) {
if (pipe(pip)  0) {
-   close(prevfd);
+   if (prevfd = 0)
+   close(prevfd);
error(Pipe call failed: %s, strerror(errno));
}
}
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252360 - head/etc

[Xin LI]

Author: delphij
Date: Fri Jun 28 22:25:37 2013
New Revision: 252360
URL: http://svnweb.freebsd.org/changeset/base/252360

Log:
  Don't attempt to do DHCP on certain interfaces, similar to what's done for
  ipv6_autoconfif() in r212577.
  
  MFC after:1 week

Modified:
  head/etc/network.subr

Modified: head/etc/network.subr
==
--- head/etc/network.subr   Fri Jun 28 21:47:08 2013(r252359)
+++ head/etc/network.subr   Fri Jun 28 22:25:37 2013(r252360)
@@ -318,6 +318,15 @@ dhcpif()
local _tmpargs _arg
_tmpargs=`_ifconfig_getargs $1`
 
+   case $1 in
+   lo[0-9]*|\
+   stf[0-9]*|\
+   faith[0-9]*|\
+   lp[0-9]*|\
+   sl[0-9]*)
+   return 1
+   ;;
+   esac
if noafif $1; then
return 1
fi
@@ -509,7 +518,7 @@ ipv6_autoconfif()
_if=$1
 
case $_if in
-   lo0|\
+   lo[0-9]*|\
stf[0-9]*|\
faith[0-9]*|\
lp[0-9]*|\
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252361 - in head/sys/arm: arm include

[Aleksandr Rybalko]

Author: ray
Date: Fri Jun 28 22:31:17 2013
New Revision: 252361
URL: http://svnweb.freebsd.org/changeset/base/252361

Log:
  Add identification for Cortex-A15 (R0) cores.
  
  Submitted by: Ruslan Bukin b...@bsdpad.com

Modified:
  head/sys/arm/arm/cpufunc.c
  head/sys/arm/arm/identcpu.c
  head/sys/arm/include/armreg.h

Modified: head/sys/arm/arm/cpufunc.c
==
--- head/sys/arm/arm/cpufunc.c  Fri Jun 28 22:25:37 2013(r252360)
+++ head/sys/arm/arm/cpufunc.c  Fri Jun 28 22:31:17 2013(r252361)
@@ -1481,7 +1481,8 @@ set_cpufuncs()
cputype == CPU_ID_CORTEXA8R3 ||
cputype == CPU_ID_CORTEXA9R1 ||
cputype == CPU_ID_CORTEXA9R2 ||
-   cputype == CPU_ID_CORTEXA9R3) {
+   cputype == CPU_ID_CORTEXA9R3 ||
+   cputype == CPU_ID_CORTEXA15 ) {
cpufuncs = cortexa_cpufuncs;
cpu_reset_needs_v4_MMU_disable = 1; /* V4 or higher */
get_cachetype_cp15();

Modified: head/sys/arm/arm/identcpu.c
==
--- head/sys/arm/arm/identcpu.c Fri Jun 28 22:25:37 2013(r252360)
+++ head/sys/arm/arm/identcpu.c Fri Jun 28 22:31:17 2013(r252361)
@@ -248,6 +248,8 @@ const struct cpuidtab cpuids[] = {
  generic_steppings },
{ CPU_ID_CORTEXA9R3,CPU_CLASS_CORTEXA,  Cortex A9-r3,
  generic_steppings },
+   { CPU_ID_CORTEXA15, CPU_CLASS_CORTEXA,  Cortex A15,
+ generic_steppings },
 
{ CPU_ID_SA110, CPU_CLASS_SA1,  SA-110,
  sa110_steppings },

Modified: head/sys/arm/include/armreg.h
==
--- head/sys/arm/include/armreg.h   Fri Jun 28 22:25:37 2013
(r252360)
+++ head/sys/arm/include/armreg.h   Fri Jun 28 22:31:17 2013
(r252361)
@@ -153,6 +153,7 @@
 #define CPU_ID_CORTEXA9R1  0x411fc090
 #define CPU_ID_CORTEXA9R2  0x412fc090
 #define CPU_ID_CORTEXA9R3  0x413fc090
+#define CPU_ID_CORTEXA15   0x410fc0f0
 #define CPU_ID_SA110   0x4401a100
 #define CPU_ID_SA1100  0x4401a110
 #defineCPU_ID_TI925T   0x54029250
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252362 - head/sys/arm/include

[Aleksandr Rybalko]

Author: ray
Date: Fri Jun 28 22:47:33 2013
New Revision: 252362
URL: http://svnweb.freebsd.org/changeset/base/252362

Log:
  Bump max number of IRQs for Cortex-Ax family to cover Exynos5 requirement.
  
  Submitted by: Ruslan Bukin b...@bsdpad.com

Modified:
  head/sys/arm/include/intr.h

Modified: head/sys/arm/include/intr.h
==
--- head/sys/arm/include/intr.h Fri Jun 28 22:31:17 2013(r252361)
+++ head/sys/arm/include/intr.h Fri Jun 28 22:47:33 2013(r252362)
@@ -51,7 +51,7 @@
 defined(CPU_XSCALE_IXP435)
 #define NIRQ   64
 #elif defined(CPU_CORTEXA)
-#define NIRQ   128
+#define NIRQ   160
 #elif defined(CPU_ARM1136) || defined(CPU_ARM1176)
 #define NIRQ   128
 #elif defined(SOC_MV_ARMADAXP)
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252356 - in head: contrib/smbfs/mount_smbfs etc/defaults etc/mtree include lib lib/libprocstat rescue/rescue sbin/mount share/examples share/examples/etc share/mk sys/conf sys/kern sy

[Jilles Tjoelker]

On Fri, Jun 28, 2013 at 09:00:08PM +, Davide Italiano wrote:
 Author: davide
 Date: Fri Jun 28 21:00:08 2013
 New Revision: 252356
 URL: http://svnweb.freebsd.org/changeset/base/252356

 Log:
   - Trim an unused and bogus Makefile for mount_smbfs.
   - Reconnect with some minor modifications, in particular now selsocket()
   internals are adapted to use sbintime units after recent'ish calloutng
   switch.

 Modified: head/sys/kern/sys_generic.c
 ==
 --- head/sys/kern/sys_generic.c   Fri Jun 28 20:32:48 2013
 (r252355)
 +++ head/sys/kern/sys_generic.c   Fri Jun 28 21:00:08 2013
 (r252356)
 @@ -1498,6 +1498,61 @@ sys_openbsd_poll(td, uap)
  }
  
  /*
 + * XXX This was created specifically to support netncp and netsmb.  This
 + * allows the caller to specify a socket to wait for events on.  It returns
 + * 0 if any events matched and an error otherwise.  There is no way to
 + * determine which events fired.
 + */
 +int
 +selsocket(struct socket *so, int events, struct timeval *tvp, struct thread 
 *td)
 +{
 + struct timeval rtv;
 + sbintime_t asbt, precision, rsbt;
 + int error;
 +
 + if (tvp != NULL) {
 + rtv = *tvp;
 + if (rtv.tv_sec  0 || rtv.tv_usec  0 || 
 + rtv.tv_usec = 100)
 + return (EINVAL);
 + if (!timevalisset(rtv))
 + asbt = 0;
 + else if (rtv.tv_sec = INT32_MAX) {
 + rsbt = tvtosbt(rtv);
 + precision = rsbt;
 + precision = tc_precexp;
 + if (TIMESEL(asbt, rsbt))
 + asbt += tc_tick_sbt;
 + if (asbt = INT64_MAX - rsbt)
 + asbt += rsbt;
 + else
 + asbt = -1;
 + } else
 + asbt = -1;
 + } else
 + asbt = -1;
 + seltdinit(td);
 + /*
 +  * Iterate until the timeout expires or the socket becomes ready.
 +  */
 + for (;;) {
 + selfdalloc(td, NULL);
 + error = sopoll(so, events, NULL, td);
 + /* error here is actually the ready events. */
 + if (error)
 + return (0);
 + error = seltdwait(td, asbt, precision);
 + if (error)
 + break;
 + }
 + seltdclear(td);
 + /* XXX Duplicates ncp/smb behavior. */
 + if (error == ERESTART)
 + error = 0;

This if looks like it may introduce unexpected differences based on
whether SA_RESTART is set for an interrupting signal. When ignoring
ERESTART or EINTR like this, there is a danger of busy-waiting because
once a sleep with PCATCH has returned ERESTART or EINTR, all further
sleeps with PCATCH will immediately return the same until you return to
the user boundary so the signal can be handled.

In quick inspection, the (single) caller looks like it can handle
ERESTART like EINTR.

The file sys/netsmb/smb_trantcp.c does appear to contain incorrect
ERESTART/EINTR handling in nbssn_recv(), which can busy-wait for the
rest of the data if a signal occurs. You could mask some signals
temporarily if proper ERESTART/EINTR handling would be too hard or would
confuse applications that do not expect partial success or [EINTR] on
regular files. The NFS client does some of these things although it
contains bugs as well (such as losing the distinction between ERESTART
and EINTR).

 + return (error);
 +}
 +
 +/*
   * Preallocate two selfds associated with 'cookie'.  Some fo_poll routines
   * have two select sets, one for read and another for write.
   */

-- 
Jilles Tjoelker
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252363 - in head/usr.sbin/pkg_install: add create lib

[David E. O'Brien]

Author: obrien
Date: Sat Jun 29 00:37:49 2013
New Revision: 252363
URL: http://svnweb.freebsd.org/changeset/base/252363

Log:
  sysconf(3) returns a long, not an int.

Modified:
  head/usr.sbin/pkg_install/add/extract.c
  head/usr.sbin/pkg_install/create/pl.c
  head/usr.sbin/pkg_install/lib/exec.c

Modified: head/usr.sbin/pkg_install/add/extract.c
==
--- head/usr.sbin/pkg_install/add/extract.c Fri Jun 28 22:47:33 2013
(r252362)
+++ head/usr.sbin/pkg_install/add/extract.c Sat Jun 29 00:37:49 2013
(r252363)
@@ -110,7 +110,8 @@ extract_plist(const char *home, Package 
 PackingList p = pkg-head;
 char *last_file, *prefix = NULL;
 char *where_args, *perm_args, *last_chdir;
-int maxargs, where_count = 0, perm_count = 0, add_count;
+long maxargs;
+int where_count = 0, perm_count = 0, add_count;
 Boolean preserve;
 
 maxargs = sysconf(_SC_ARG_MAX) / 2;/* Just use half the argument 
space */

Modified: head/usr.sbin/pkg_install/create/pl.c
==
--- head/usr.sbin/pkg_install/create/pl.c   Fri Jun 28 22:47:33 2013
(r252362)
+++ head/usr.sbin/pkg_install/create/pl.c   Sat Jun 29 00:37:49 2013
(r252363)
@@ -140,7 +140,8 @@ copy_plist(const char *home, Package *pl
 const char *there = NULL, *mythere;
 char *where_args, *prefix = NULL;
 const char *last_chdir, *root = /;
-int maxargs, where_count = 0, add_count;
+long maxargs;
+int where_count = 0, add_count;
 struct stat stb;
 dev_t curdir;
 

Modified: head/usr.sbin/pkg_install/lib/exec.c
==
--- head/usr.sbin/pkg_install/lib/exec.cFri Jun 28 22:47:33 2013
(r252362)
+++ head/usr.sbin/pkg_install/lib/exec.cSat Jun 29 00:37:49 2013
(r252363)
@@ -65,7 +65,7 @@ vpipe(const char *fmt, ...)
 {
FILE *fp;
char *cmd, *rp;
-   int maxargs;
+   long maxargs;
va_list args;
 
 rp = malloc(MAXPATHLEN);
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252364 - head/sys/fs/ext2fs

[Pedro F. Giffuni]

Author: pfg
Date: Sat Jun 29 01:35:28 2013
New Revision: 252364
URL: http://svnweb.freebsd.org/changeset/base/252364

Log:
  Bring some updates from ufs_lookup to ext2fs.
  
  r156418:
  
  Don't set IN_CHANGE and IN_UPDATE on inodes for potentially suspended
  file systems.  This could cause deadlocks when creating snapshots.
  (We can't do snapshots on ext2fs but it is useful to keep things in sync).
  
  r183079:
  
  - Only set i_offset in the parent directory's i-node during a lookup for
non-LOOKUP operations.
  - Relax a VOP assertion for a DELETE lookup.
  
  r187528:
  
  Move the code from ufs_lookup.c used to do dotdot lookup, into
  the helper function. It is supposed to be useful for any filesystem
  that has to unlock dvp to walk to the .. entry in lookup routine.
  
  MFC after:5 days

Modified:
  head/sys/fs/ext2fs/ext2_lookup.c

Modified: head/sys/fs/ext2fs/ext2_lookup.c
==
--- head/sys/fs/ext2fs/ext2_lookup.cSat Jun 29 00:37:49 2013
(r252363)
+++ head/sys/fs/ext2fs/ext2_lookup.cSat Jun 29 01:35:28 2013
(r252364)
@@ -289,9 +289,9 @@ ext2_lookup_ino(struct vnode *vdp, struc
int entryoffsetinblock; /* offset of ep in bp's buffer */
enum {NONE, COMPACT, FOUND} slotstatus;
doff_t slotoffset;  /* offset of area with free space */
-   int slotsize;   /* size of area at slotoffset */
doff_t i_diroff;/* cached i_diroff value */
doff_t i_offset;/* cached i_offset value */
+   int slotsize;   /* size of area at slotoffset */
int slotfreespace;  /* amount of space free in slot */
int slotneeded; /* size of the entry we're seeking */
int numdirpasses;   /* strategy for directory search */
@@ -476,7 +476,6 @@ searchloop:
endsearch = i_diroff;
goto searchloop;
}
-   dp-i_offset = i_offset;
if (bp != NULL)
brelse(bp);
/*
@@ -512,7 +511,6 @@ searchloop:
enduseful = slotoffset + slotsize;
}
dp-i_endoff = roundup2(enduseful, DIRBLKSIZ);
-   dp-i_flag |= IN_CHANGE | IN_UPDATE;
/*
 * We return with the directory locked, so that
 * the parameters we set up above will still be
@@ -560,12 +558,13 @@ found:
 */
if ((flags  ISLASTCN)  nameiop == LOOKUP)
dp-i_diroff = i_offset ~ (DIRBLKSIZ - 1);
-   dp-i_offset = i_offset;
/*
 * If deleting, and at end of pathname, return
 * parameters which can be used to remove file.
 */
if (nameiop == DELETE  (flags  ISLASTCN)) {
+   if (flags  LOCKPARENT)
+   ASSERT_VOP_ELOCKED(vdp, __FUNCTION__);
/*
 * Write access to directory required to delete files.
 */
@@ -576,7 +575,13 @@ found:
 * and distance past previous entry (if there
 * is a previous entry in this block) in dp-i_count.
 * Save directory inode pointer in ndp-ni_dvp for dirremove().
+*
+* Technically we shouldn't be setting these in the
+* WANTPARENT case (first lookup in rename()), but any
+* lookups that will result in directory changes will
+* overwrite these.
 */
+   dp-i_offset = i_offset;
if ((dp-i_offset  (DIRBLKSIZ - 1)) == 0)
dp-i_count = 0;
else
@@ -621,6 +626,7 @@ found:
 * Careful about locking second inode.
 * This can only occur if the target is ..
 */
+   dp-i_offset = i_offset;
if (dp-i_number == ino)
return (EISDIR);
if (dd_ino != NULL)
@@ -656,10 +662,7 @@ found:
 */
pdp = vdp;
if (flags  ISDOTDOT) {
-   ltype = VOP_ISLOCKED(pdp);
-   VOP_UNLOCK(pdp, 0); /* race to get the inode */
-   error = VFS_VGET(vdp-v_mount, ino, cnp-cn_lkflags, tdp);
-   vn_lock(pdp, ltype | LK_RETRY);
+   error = vn_vget_ino(pdp, ino, cnp-cn_lkflags, tdp);
if (pdp-v_iflag  VI_DOOMED) {
if (error == 0)
vput(tdp);
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252367 - head/sys/kern

[Peter Wemm]

Author: peter
Date: Sat Jun 29 04:35:04 2013
New Revision: 252367
URL: http://svnweb.freebsd.org/changeset/base/252367

Log:
  Help out gcc.  clang understands.
  
  sys_generic.c:1510: warning: 'precision' may be used uninitialized
  *** [sys_generic.o] Error code 1

Modified:
  head/sys/kern/sys_generic.c
  head/sys/kern/vfs_init.c

Modified: head/sys/kern/sys_generic.c
==
--- head/sys/kern/sys_generic.c Sat Jun 29 04:27:04 2013(r252366)
+++ head/sys/kern/sys_generic.c Sat Jun 29 04:35:04 2013(r252367)
@@ -1510,6 +1510,7 @@ selsocket(struct socket *so, int events,
sbintime_t asbt, precision, rsbt;
int error;
 
+   precision = 0;  /* stupid gcc! */
if (tvp != NULL) {
rtv = *tvp;
if (rtv.tv_sec  0 || rtv.tv_usec  0 || 

Modified: head/sys/kern/vfs_init.c
==
--- head/sys/kern/vfs_init.cSat Jun 29 04:27:04 2013(r252366)
+++ head/sys/kern/vfs_init.cSat Jun 29 04:35:04 2013(r252367)
@@ -201,6 +201,7 @@ vfs_register(struct vfsconf *vfc)
} else
vfc-vfc_typenum = maxvfsconf++;
TAILQ_INSERT_TAIL(vfsconf, vfc, vfc_list);
+   vfc-vfc_flags |= VFCF_JAIL;
 
/*
 * If this filesystem has a sysctl node under vfs
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

Re: svn commit: r252367 - head/sys/kern

[Garrett Cooper]

On Jun 28, 2013, at 9:35 PM, Peter Wemm wrote:

 Author: peter
 Date: Sat Jun 29 04:35:04 2013
 New Revision: 252367
 URL: http://svnweb.freebsd.org/changeset/base/252367
 
 Log:
  Help out gcc.  clang understands.
 
  sys_generic.c:1510: warning: 'precision' may be used uninitialized
  *** [sys_generic.o] Error code 1

...

 Modified: head/sys/kern/vfs_init.c
 ==
 --- head/sys/kern/vfs_init.c  Sat Jun 29 04:27:04 2013(r252366)
 +++ head/sys/kern/vfs_init.c  Sat Jun 29 04:35:04 2013(r252367)
 @@ -201,6 +201,7 @@ vfs_register(struct vfsconf *vfc)
   } else
   vfc-vfc_typenum = maxvfsconf++;
   TAILQ_INSERT_TAIL(vfsconf, vfc, vfc_list);
 + vfc-vfc_flags |= VFCF_JAIL;

Was this portion of the diff intentional?
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org

svn commit: r252368 - head/sys/kern

[Peter Wemm]

Author: peter
Date: Sat Jun 29 05:05:57 2013
New Revision: 252368
URL: http://svnweb.freebsd.org/changeset/base/252368

Log:
  Revert accidental commit.
  
  Pointy hat to: peter

Modified:
  head/sys/kern/vfs_init.c

Modified: head/sys/kern/vfs_init.c
==
--- head/sys/kern/vfs_init.cSat Jun 29 04:35:04 2013(r252367)
+++ head/sys/kern/vfs_init.cSat Jun 29 05:05:57 2013(r252368)
@@ -201,7 +201,6 @@ vfs_register(struct vfsconf *vfc)
} else
vfc-vfc_typenum = maxvfsconf++;
TAILQ_INSERT_TAIL(vfsconf, vfc, vfc_list);
-   vfc-vfc_flags |= VFCF_JAIL;
 
/*
 * If this filesystem has a sysctl node under vfs
___
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to svn-src-head-unsubscr...@freebsd.org