On Sun, 20 Mar 2011, Kirk McKusick wrote:
Date: Sun, 20 Mar 2011 13:25:20 -0700
From: Doug Barton
To: Marius Strobl
CC: Kirk McKusick ,
Nathan Whitehorn , svn-src-head@FreeBSD.org,
Jeff Roberson , Gavin Atkinson ,
svn-src-...@freebsd.org, src-committ...@freebsd.org
On Sun, 20 Mar 2011, Doug Barton wrote:
On 03/20/2011 09:22, Marius Strobl wrote:
I fear it's still a bit premature for enable SU+J by default. Rather
recently I was told about a SU+J filesystems lost after a panic
that happend after snapshotting it (report CC'ed, maybe he can
provide some mor
On Mon, 21 Mar 2011, Michael Moll wrote:
Hi All,
On Sun, Mar 20, 2011 at 05:22:12PM +0100, Marius Strobl wrote:
I fear it's still a bit premature for enable SU+J by default. Rather
recently I was told about a SU+J filesystems lost after a panic
that happend after snapshotting it (report CC'ed,
Author: jeff
Date: Mon Mar 21 21:35:19 2011
New Revision: 219849
URL: http://svn.freebsd.org/changeset/base/219849
Log:
- For now, disable ofed module build unless MK_OFED is set.
Modified:
head/sys/modules/Makefile
Modified: head/sys/modules/Makefile
===
Author: jeff
Date: Tue Mar 22 04:50:47 2011
New Revision: 219859
URL: http://svn.freebsd.org/changeset/base/219859
Log:
- Don't use a separate set of rx queues for UDP, hash them into the same
set as TCP.
- Eliminate the fully linear non-scatter/gather rx path, there is no
harm in
Author: jeff
Date: Wed Mar 23 02:47:04 2011
New Revision: 219893
URL: http://svn.freebsd.org/changeset/base/219893
Log:
- Correct the vlan filter programming. The device filter is built in
reverse order.
- Name the cq taskqueues according to whether they handle rx or tx.
- Default L
Author: jeff
Date: Wed Mar 23 08:27:57 2011
New Revision: 219898
URL: http://svn.freebsd.org/changeset/base/219898
Log:
- Move ofed modules into the i386 and amd64 specific sections to fix
universe on other architectures.
Modified:
head/sys/modules/Makefile
Modified: head/sys/modules/M
Author: jeff
Date: Sat Mar 26 00:54:01 2011
New Revision: 220016
URL: http://svn.freebsd.org/changeset/base/220016
Log:
- Implement wake-on-lan support in mlxen.
Modified:
head/sys/ofed/drivers/net/mlx4/en_ethtool.c
head/sys/ofed/drivers/net/mlx4/en_netdev.c
head/sys/ofed/drivers/net/mlx
Author: jeff
Date: Sat Apr 2 21:52:58 2011
New Revision: 220282
URL: http://svn.freebsd.org/changeset/base/220282
Log:
Fix problems that manifested from filesystem full conditions:
- In softdep_revert_mkdir() find the dotaddref before we attempt to cancel
the jaddref so we can make a
Author: jeff
Date: Thu Apr 7 03:19:10 2011
New Revision: 220406
URL: http://svn.freebsd.org/changeset/base/220406
Log:
- Don't invalidate jnewblks immediately upon discovering that the block
will be removed. Permit the journal to proceed so that we don't leave
a rollback in a cg for
Author: jeff
Date: Sun Apr 10 03:49:53 2011
New Revision: 220511
URL: http://svn.freebsd.org/changeset/base/220511
Log:
Fix a long standing SUJ performance problem:
- Keep a hash of indirect blocks that have recently been freed and are
still referenced in the journal.
- Lookup bloc
On Sun, 10 Apr 2011, Jeff Roberson wrote:
Author: jeff
Date: Sun Apr 10 03:49:53 2011
New Revision: 220511
URL: http://svn.freebsd.org/changeset/base/220511
Log:
Fix a long standing SUJ performance problem:
This brought my dbench performance to within 10-15% of softupdates on
a real disk
Author: jeff
Date: Mon Apr 11 01:43:59 2011
New Revision: 220532
URL: http://svn.freebsd.org/changeset/base/220532
Log:
- Refactor softdep_setup_freeblocks() into a set of functions to prepare
for a new journal specific partial truncate routine.
- Use dep_current[] in place of specific
Author: jeff
Date: Tue Apr 26 07:30:52 2011
New Revision: 221055
URL: http://svn.freebsd.org/changeset/base/221055
Log:
- Catch up to falloc() changes.
- PHOLD() before using a task structure on the stack.
- Fix a LOR between the sleepq lock and thread lock in _intr_drain().
Modified:
Author: jeff
Date: Sat Apr 24 07:05:35 2010
New Revision: 207141
URL: http://svn.freebsd.org/changeset/base/207141
Log:
- Merge soft-updates journaling from projects/suj/head into head. This
brings in support for an optional intent log which eliminates the need
for background fsck on
Author: jeff
Date: Sat Apr 24 07:59:45 2010
New Revision: 207145
URL: http://svn.freebsd.org/changeset/base/207145
Log:
- Temporarily lower WARNS until I fix alignment warnings on sparc64.
Reported by: Florian Smeets
Modified:
head/sbin/tunefs/Makefile
Modified: head/sbin/tunefs/Makef
On Sat, 24 Apr 2010, Pawel Jakub Dawidek wrote:
Author: pjd
Date: Sat Apr 24 07:58:59 2010
New Revision: 207144
URL: http://svn.freebsd.org/changeset/base/207144
Log:
suj.c seems to contain two versions of the code.
Remove the one that doesn't compile.
Thanks pjd. I must've patched twice a
Author: jeff
Date: Wed Apr 28 07:26:41 2010
New Revision: 207309
URL: http://svn.freebsd.org/changeset/base/207309
Log:
- Fix builds without SOFTUPDATES defined in the kernel config.
Modified:
head/sys/ufs/ffs/ffs_softdep.c
Modified: head/sys/ufs/ffs/ffs_softdep.c
==
Author: jeff
Date: Wed Apr 28 07:57:37 2010
New Revision: 207310
URL: http://svn.freebsd.org/changeset/base/207310
Log:
- When canceling jaddrefs they may not yet be in the journal if this is via
a revert call. In this case don't attempt to remove something that
has not yet been adde
Author: jeff
Date: Fri Apr 30 04:21:22 2010
New Revision: 207421
URL: http://svn.freebsd.org/changeset/base/207421
Log:
- Use the path to the filesystem mountpoint to look up the statfs
structure so that we correctly reload. Note that tunefs doesn't
properly detect the need to reload
Author: jeff
Date: Fri May 7 08:20:56 2010
New Revision: 207741
URL: http://svn.freebsd.org/changeset/base/207741
Log:
- Use the correct flag mask when determining whether an inode has
successfully made it to the free list yet or not. This fixes
a deadlock that can occur with unlink
Author: jeff
Date: Fri May 7 08:45:21 2010
New Revision: 207742
URL: http://svn.freebsd.org/changeset/base/207742
Log:
- Call softdep_prealloc() before any of the balloc routines in the
snapshot code.
- Don't fsync() vnodes in prealloc if copy on write is in progress. It
is not s
Author: jeff
Date: Tue May 18 01:45:28 2010
New Revision: 208241
URL: http://svn.freebsd.org/changeset/base/208241
Log:
- Round up the journal size to the block size so we don't confuse fsck.
Reported by: Mikolaj Golub
- Only require 256k of blocks per-cg when trying to allocate co
Author: jeff
Date: Wed May 19 06:18:01 2010
New Revision: 208287
URL: http://svn.freebsd.org/changeset/base/208287
Log:
- Don't immediately re-run softdepflush if we didn't make any progress
on the last iteration. This can lead to a deadlock when we have
worklist items that cannot be
Author: jeff
Date: Tue Jul 6 07:07:29 2010
New Revision: 209716
URL: http://svn.freebsd.org/changeset/base/209716
Log:
- Permit zero length directories as a handled inconsistency. This allows
directory truncation to proceed before the link has been cleared. This
is accomplished by
Author: jeff
Date: Tue Jul 6 07:11:04 2010
New Revision: 209717
URL: http://svn.freebsd.org/changeset/base/209717
Log:
- Handle the truncation of an inode with an effective link count of 0 in
the context of the process that reduced the effective count. Previously
all truncation as a
Author: jeff
Date: Fri Jun 10 22:15:36 2011
New Revision: 222953
URL: http://svn.freebsd.org/changeset/base/222953
Log:
- When printing bufs with show buf the lblkno is often more useful than
the blkno. Print them both.
Modified:
head/sys/kern/vfs_bio.c
Modified: head/sys/kern/vfs_bio
Author: jeff
Date: Fri Jun 10 22:18:25 2011
New Revision: 222954
URL: http://svn.freebsd.org/changeset/base/222954
Log:
- If the fsync in ufs_direnter fails SUJ can later panic because we have
partially added a name. Allow ufs_direnter() to continue in the
hopes that it is a transien
Author: jeff
Date: Fri Jun 10 22:19:44 2011
New Revision: 222955
URL: http://svn.freebsd.org/changeset/base/222955
Log:
- Add support for referencing quota structures without needing the inode
pointer for softupdates.
Submitted by: mckusick
Modified:
head/sys/ufs/ufs/quota.h
head
Author: jeff
Date: Fri Jun 10 22:38:31 2011
New Revision: 222956
URL: http://svn.freebsd.org/changeset/base/222956
Log:
- Eliminate an incorrect include path from the mthca build.
Modified:
head/sys/conf/files
Modified: head/sys/conf/files
===
Author: jeff
Date: Fri Jun 10 22:48:35 2011
New Revision: 222958
URL: http://svn.freebsd.org/changeset/base/222958
Log:
Implement fully asynchronous partial truncation with softupdates journaling
to resolve errors which can cause corruption on recovery with the old
synchronous mechanism.
On Fri, 10 Jun 2011, Jeff Roberson wrote:
Author: jeff
Date: Fri Jun 10 22:48:35 2011
New Revision: 222958
URL: http://svn.freebsd.org/changeset/base/222958
Log:
Implement fully asynchronous partial truncation with softupdates journaling
to resolve errors which can cause corruption on
Author: jeff
Date: Mon Jun 20 03:25:09 2011
New Revision: 223325
URL: http://svn.freebsd.org/changeset/base/223325
Log:
- Fix directory count rollbacks by passing the mode to the journal dep
earlier.
- Add rollback/forward code for frag and cluster accounting.
- Handle the FREEDEP ca
Author: jeff
Date: Thu Jun 30 05:28:10 2011
New Revision: 223689
URL: http://svn.freebsd.org/changeset/base/223689
Log:
- Handle the JOP_SYNC case as appropriate.
Reported by: pho
Modified:
head/sbin/fsck_ffs/suj.c
Modified: head/sbin/fsck_ffs/suj.c
===
Author: jeff
Date: Mon Jul 4 20:52:23 2011
New Revision: 223769
URL: http://svn.freebsd.org/changeset/base/223769
Log:
- Fix an inode quota leak. We need to decrement the quota once and only
once.
Tested by:pho
Reviewed by: mckusick
Modified:
head/sys/ufs/ufs/ufs_inode.c
Author: jeff
Date: Mon Jul 4 20:53:55 2011
New Revision: 223770
URL: http://svn.freebsd.org/changeset/base/223770
Log:
- It is impossible to run request_cleanup() while doing a copyonwrite.
This will most likely cause new block allocations which can recurse
into request cleanup.
-
Author: jeff
Date: Mon Jul 4 21:04:25 2011
New Revision: 223771
URL: http://svn.freebsd.org/changeset/base/223771
Log:
- Handle D_JSEGDEP in the softdep_sync_buf() switch. These can now
find themselves on snapshot vnodes.
Reported by: pho
Modified:
head/sys/ufs/ffs/ffs_softdep.c
Author: jeff
Date: Mon Jul 4 22:08:04 2011
New Revision: 223772
URL: http://svn.freebsd.org/changeset/base/223772
Log:
- Speed up pendingblock processing again. Having too much delay between
ffs_blkfree() and the pending adjustment causes all kinds of
space related problems.
Modifi
On Fri, 6 Jul 2018, Rodney W. Grimes wrote:
On Friday, July 6, 2018, Rodney W. Grimes
wrote:
Author: hselasky
Date: Fri Jul 6 10:13:42 2018
New Revision: 336025
URL: https://svnweb.freebsd.org/changeset/base/336025
Log:
Make sure kernel modules built by default are portable between UP and
On Fri, 6 Jul 2018, Rodney W. Grimes wrote:
On Fri, Jul 6, 2018, 12:27 PM Rodney W. Grimes <
free...@pdx.rh.cn85.dnsmgr.net> wrote:
On Fri, Jul 6, 2018 at 9:52 AM, Rodney W. Grimes <
free...@pdx.rh.cn85.dnsmgr.net> wrote:
On Fri, Jul 6, 2018 at 9:32 AM, Rodney W. Grimes <
free...@pdx.rh.cn85
Author: jeff
Date: Sat Jul 7 01:54:45 2018
New Revision: 336055
URL: https://svnweb.freebsd.org/changeset/base/336055
Log:
Use the ticks since the last update to reduce hysteresis in the partpopq and
contention on the vm_reserv_domain lock.
This gives a roughly 8x speedup on will-it-scal
Author: jeff
Date: Tue Feb 6 22:10:07 2018
New Revision: 328954
URL: https://svnweb.freebsd.org/changeset/base/328954
Log:
Use per-domain locks for vm page queue free. Move paging control from
global to per-domain state. Protect reservations with the free lock
from the domain that they be
Author: jeff
Date: Mon Feb 12 22:53:00 2018
New Revision: 329187
URL: https://svnweb.freebsd.org/changeset/base/329187
Log:
Make v_wire_count a per-cpu counter(9) counter. This eliminates a
significant source of cache line contention from vm_page_alloc(). Use
accessors and vm_page_unwire_n
Author: jeff
Date: Tue Feb 20 00:06:07 2018
New Revision: 329612
URL: https://svnweb.freebsd.org/changeset/base/329612
Log:
Further parallelize the buffer cache.
Provide multiple clean queues partitioned into 'domains'. Each domain manages
its own bufspace and has its own bufspace daemon
Author: jeff
Date: Tue Feb 20 21:27:17 2018
New Revision: 329672
URL: https://svnweb.freebsd.org/changeset/base/329672
Log:
Fix the broken subqueue assignment for the cleanq.
Reported by: pho
Tested by:pho
Sponsored by: Netflix, Dell/EMC Isilon
Modified:
head/sys/kern/vfs_bio.c
Author: jeff
Date: Fri Feb 23 22:51:51 2018
New Revision: 329882
URL: https://svnweb.freebsd.org/changeset/base/329882
Log:
Add a generic Proportional Integral Derivative (PID) controller algorithm and
use it to regulate page daemon output.
This provides much smoother and more responsive
Author: jeff
Date: Sun Feb 25 00:35:21 2018
New Revision: 329943
URL: https://svnweb.freebsd.org/changeset/base/329943
Log:
Fix issues with sparse cpu allocation. Consistently use mp_maxid + 1.
Reported by: pho
Reviewed by: markj
Sponsored by: Netflix, Dell/EMC Isilon
Modified:
he
Author: jeff
Date: Wed Mar 7 22:04:27 2018
New Revision: 330624
URL: https://svnweb.freebsd.org/changeset/base/330624
Log:
Don't assert that the domain free lock is held until we're certain that
there is a valid reservation. This can trip erroneously when memory
falls within a domain but d
Author: jeff
Date: Thu Mar 15 19:23:07 2018
New Revision: 331018
URL: https://svnweb.freebsd.org/changeset/base/331018
Log:
Eliminate pageout wakeup races. Take another step towards lockless
vmd_free_count manipulation. Reduce the scope of the free lock by
using a pageout lock to synchroni
Author: jeff
Date: Sat Mar 17 18:14:49 2018
New Revision: 331106
URL: https://svnweb.freebsd.org/changeset/base/331106
Log:
Move the dirty queues inside the per-domain structure. This resolves a bug
where we had not hit global dirty limits but a single queue was starved
for space by dirty b
Thanks for fixing my bug. Sorry I didn't build universe for this one.
Thanks,
Jeff
On Tue, 20 Mar 2018, Justin Hibbits wrote:
Author: jhibbits
Date: Tue Mar 20 02:01:30 2018
New Revision: 331233
URL: https://svnweb.freebsd.org/changeset/base/331233
Log:
Cast through uintptr_t to narrow the
On Thu, 9 May 2013, Marcel Moolenaar wrote:
Author: marcel
Date: Thu May 9 16:28:18 2013
New Revision: 250411
URL: http://svnweb.freebsd.org/changeset/base/250411
Log:
Add option WITNESS_NO_VNODE to suppress printing LORs between VNODE
locks. To support this, VNODE locks are created with the
Author: jeff
Date: Sun May 12 04:05:01 2013
New Revision: 250551
URL: http://svnweb.freebsd.org/changeset/base/250551
Log:
- Add a new general purpose path-compressed radix trie which can be used
with any structure containing a uint64_t index. The tree code
auto-generates type safe w
Author: jeff
Date: Sun May 12 20:44:28 2013
New Revision: 250578
URL: http://svnweb.freebsd.org/changeset/base/250578
Log:
- pctrie really only requires two byte alignment so that there is a single
bit available for a flag in the pointer. However, it felt more correct
to enforce natu
Author: jeff
Date: Fri May 31 00:43:41 2013
New Revision: 251171
URL: http://svnweb.freebsd.org/changeset/base/251171
Log:
- Convert the bufobj lock to rwlock.
- Use a shared bufobj lock in getblk() and inmem().
- Convert softdep's lk to rwlock to match the bufobj lock.
- Move INFREECN
Author: jeff
Date: Wed Jun 5 23:53:00 2013
New Revision: 251446
URL: http://svnweb.freebsd.org/changeset/base/251446
Log:
- Consolidate duplicate code into support functions.
- Split the bqlock into bqclean and bqdirty locks.
- Only acquire the wakeup synchronization locks when we cross
Author: jeff
Date: Thu Jun 13 20:46:03 2013
New Revision: 251703
URL: http://svnweb.freebsd.org/changeset/base/251703
Log:
- Add a BIT_FFS() macro and use it to replace cpusetffs_obj()
Discussed with: attilio
Sponsored by: EMC / Isilon Storage Division
Modified:
head/sys/amd64/a
Author: jeff
Date: Thu Jun 13 21:05:38 2013
New Revision: 251709
URL: http://svnweb.freebsd.org/changeset/base/251709
Log:
- Convert the slab free item list from a linked array of indices to a
bitmap using sys/bitset. This is much simpler, has lower space
overhead and is cheaper in m
Author: jeff
Date: Mon Jun 17 03:43:47 2013
New Revision: 251826
URL: http://svnweb.freebsd.org/changeset/base/251826
Log:
- Add a new UMA API: uma_zcache_create(). This makes a zone without any
backing memory that is only a container for per-cpu caches of arbitrary
pointer items. T
Author: jeff
Date: Tue Jun 18 04:50:20 2013
New Revision: 251894
URL: http://svnweb.freebsd.org/changeset/base/251894
Log:
Refine UMA bucket allocation to reduce space consumption and improve
performance.
- Always free to the alloc bucket if there is space. This gives LIFO
allocati
On Tue, 18 Jun 2013, Alfred Perlstein wrote:
On 6/18/13 4:37 AM, Gleb Smirnoff wrote:
On Tue, Jun 18, 2013 at 10:25:08AM +0200, Andre Oppermann wrote:
A> There used to be a problem with per CPU caches accumulating large
amounts
A> of items without freeing back to the global (or socket) pool.
Author: jeff
Date: Wed Jun 19 02:30:32 2013
New Revision: 251983
URL: http://svnweb.freebsd.org/changeset/base/251983
Log:
- Persist the caller's flags in the bucket allocation flags so we don't
lose a M_NOVM when we recurse into a bucket allocation.
Sponsored by: EMC / Isilon Storage
Author: jeff
Date: Thu Jun 20 19:08:12 2013
New Revision: 252040
URL: http://svnweb.freebsd.org/changeset/base/252040
Log:
- Add a per-zone lock for zones without kegs.
- Be more explicit about zone vs keg locking. This functionally changes
almost nothing.
- Add a size parameter to
Author: jeff
Date: Wed Jun 26 00:57:38 2013
New Revision: 252226
URL: http://svnweb.freebsd.org/changeset/base/252226
Log:
- Resolve bucket recursion issues by passing a cookie with zone flags
through bucket_alloc() to uma_zalloc_arg() and uma_zfree_arg().
- Make some smaller buckets fo
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 Ne
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 wrote:
Author: jeff
Date: Fri Jun 28 03:51:20 2013
New Revision: 252330
URL: http
Author: jeff
Date: Thu Mar 6 00:10:07 2014
New Revision: 262812
URL: http://svnweb.freebsd.org/changeset/base/262812
Log:
- Gracefully handle truncation failures when trying to shrink directories.
This could cause dirhash panics since the dirhash state would be
successfully truncated
Author: jeff
Date: Thu Mar 6 00:13:21 2014
New Revision: 262814
URL: http://svnweb.freebsd.org/changeset/base/262814
Log:
- If we fail to do a non-blocking acquire of a buf lock while doing a
waiting sync pass we need to do a blocking acquire and restart.
Another thread, typically th
Author: jeff
Date: Sat Mar 8 00:35:06 2014
New Revision: 262917
URL: http://svnweb.freebsd.org/changeset/base/262917
Log:
- Make runq_steal_from more aggressive. Previously it would examine only
a single priority queue. If that queue had a thread or threads which
could not be migra
Author: jeff
Date: Tue Apr 7 02:18:52 2015
New Revision: 281184
URL: https://svnweb.freebsd.org/changeset/base/281184
Log:
- Simplify vm_pageout_scan() by introducing a new vm_pageout_clean()
function that does the locking and validation associated with cleaning
a page. This moves 1
Author: jeff
Date: Sat Mar 28 02:36:49 2015
New Revision: 280774
URL: https://svnweb.freebsd.org/changeset/base/280774
Log:
- Eliminate pagequeue locking in the dirty code in vm_pageout_scan().
- Use a more precise series of tests to see if the page changed while we
were locking the vno
On Tue, 30 Jun 2015, Mark Murray wrote:
Author: markm
Date: Tue Jun 30 17:00:45 2015
New Revision: 284959
URL: https://svnweb.freebsd.org/changeset/base/284959
Log:
Huge cleanup of random(4) code.
* GENERAL
- Update copyright.
- Make kernel options for RANDOM_YARROW and RANDOM_DUMMY. Set
On Wed, 22 Jul 2015, Mark R V Murray wrote:
On 22 Jul 2015, at 22:42, Jeff Roberson wrote:
On Tue, 30 Jun 2015, Mark Murray wrote:
- Add harvesting of slab allocator events. This needs to be checked for
weighing down the allocator code.
Neither filesystem operations nor allocations
Author: jeff
Date: Thu Jul 23 02:20:41 2015
New Revision: 285808
URL: https://svnweb.freebsd.org/changeset/base/285808
Log:
- Don't defeat the FIFO nature of the buffer cache by eliminating the
most recently used buffer when we are under paging pressure. This is
a perversion of the b
On Thu, 23 Jul 2015, Mark R V Murray wrote:
On 23 Jul 2015, at 00:53, Warner Losh wrote:
Neither filesystem operations nor allocations are random events. They are
trivially influenced by user code. A malicious attacker could create repeated
patterns of allocations or filesystem activity
Author: jeff
Date: Thu Jul 23 19:13:41 2015
New Revision: 285819
URL: https://svnweb.freebsd.org/changeset/base/285819
Log:
Refactor unmapped buffer address handling.
- Use pointer assignment rather than a combination of pointers and
flags to switch buffers between unmapped and mapped.
Author: jeff
Date: Tue Jul 28 20:24:09 2015
New Revision: 285981
URL: https://svnweb.freebsd.org/changeset/base/285981
Log:
- Eliminate the EMPTYKVA queue. It served as a cache of KVA allocations
attached to bufs to avoid the overhead of the vm. This purposes is now
better served by
Author: jeff
Date: Wed Jul 29 02:26:57 2015
New Revision: 285993
URL: https://svnweb.freebsd.org/changeset/base/285993
Log:
- Make 'struct buf *buf' private to vfs_bio.c. Having a global variable
'buf' is inconvenient and has lead me to some irritating to discover
bugs over the years
Author: jeff
Date: Wed Jul 29 03:06:08 2015
New Revision: 285995
URL: https://svnweb.freebsd.org/changeset/base/285995
Log:
- Remove some dead code copied from ffs.
Modified:
head/sys/fs/ext2fs/ext2_subr.c
Modified: head/sys/fs/ext2fs/ext2_subr.c
Author: jeff
Date: Tue Sep 22 23:57:52 2015
New Revision: 288137
URL: https://svnweb.freebsd.org/changeset/base/288137
Log:
Some refactoring of the buf/vm interface.
- Eliminate bogus page replacement that is inconsistently applied in the
invalidation loop in brelse. This has been a no-
Author: jeff
Date: Wed Sep 23 07:44:07 2015
New Revision: 288144
URL: https://svnweb.freebsd.org/changeset/base/288144
Log:
- Fix a nonsense reordering that somehow slipped into my last diff.
Reported by: pho
Modified:
head/sys/kern/vfs_bio.c
Modified: head/sys/kern/vfs_bio.c
Author: jeff
Date: Sun Sep 27 05:16:06 2015
New Revision: 288299
URL: https://svnweb.freebsd.org/changeset/base/288299
Log:
- Collapse vfs_vmio_truncate & vfs_vmio_release into a single function.
- Allow vfs_vmio_invalidate() to free the pages, leaving us with a
single loop and bufobj l
Author: jeff
Date: Wed Oct 14 02:10:07 2015
New Revision: 289279
URL: https://svnweb.freebsd.org/changeset/base/289279
Log:
Parallelize the buffer cache and rewrite getnewbuf(). This results in a
8x performance improvement in a micro benchmark on a 4 socket machine.
- Get buffer headers
On Sun, 1 Nov 2015, NGie Cooper wrote:
On Nov 1, 2015, at 19:20, Adrian Chadd wrote:
hiya jeff,
this broke low-memory, no-swap boards (eg MIPS.)
On a MIPS board (carambola2) with 32MB of RAM, just scp'ing a kernel
into the rootfs on USB hangs the system. After doing some digging, I
found t
On Wed, 24 Oct 2012, Attilio Rao wrote:
On Wed, Oct 24, 2012 at 8:16 PM, Andre Oppermann wrote:
On 24.10.2012 20:56, Jim Harris wrote:
On Wed, Oct 24, 2012 at 11:41 AM, Adrian Chadd wrote:
On 24 October 2012 11:36, Jim Harris wrote:
Pad tdq_lock to avoid false sharing with tdq_load
Author: jeff
Date: Fri Nov 2 21:04:06 2012
New Revision: 242492
URL: http://svn.freebsd.org/changeset/base/242492
Log:
- In cancel_mkdir_dotdot don't panic if the inodedep is not available. If
the previous diradd had already finished it could have been reclaimed
already. This would
Author: jeff
Date: Thu Nov 8 01:41:04 2012
New Revision: 242734
URL: http://svnweb.freebsd.org/changeset/base/242734
Log:
- Implement BIO_FLUSH support around journal entries. This will not 100%
solve power loss problems with dishonest write caches. However, it
should improve the s
Author: jeff
Date: Thu Nov 8 01:46:47 2012
New Revision: 242736
URL: http://svnweb.freebsd.org/changeset/base/242736
Log:
- Change ULE to use dynamic slice sizes for the timeshare queue in order
to further reduce latency for threads in this queue. This should help
as threads transit
Author: jeff
Date: Fri Nov 9 04:04:25 2012
New Revision: 242815
URL: http://svnweb.freebsd.org/changeset/base/242815
Log:
- Correct rev 242734, segments can sometimes get stuck. Be a bit more
defensive with segment state.
Reported by: b. f.
Modified:
head/sys/ufs/ffs/ffs_softd
Author: jeff
Date: Mon Nov 12 19:53:55 2012
New Revision: 242924
URL: http://svnweb.freebsd.org/changeset/base/242924
Log:
- Fix a bug that has existed since the original softdep implementation.
When a background copy of a cg is written we complete any work associated
with that bmsafe
Author: jeff
Date: Wed Nov 14 06:31:47 2012
New Revision: 243017
URL: http://svnweb.freebsd.org/changeset/base/243017
Log:
- blk_equals() is too strict. If the journal entry defines more frags
than we're claiming it should still be considered an exact match. This
would previously le
Author: jeff
Date: Wed Nov 14 06:37:43 2012
New Revision: 243018
URL: http://svnweb.freebsd.org/changeset/base/243018
Log:
- Fix a truncation bug with softdep journaling that could leak blocks on
crash. When truncating a file that never made it to disk we use the
canceled allocation
Author: jeff
Date: Thu Nov 15 00:51:57 2012
New Revision: 243046
URL: http://svnweb.freebsd.org/changeset/base/243046
Log:
- Implement run-time expansion of the KTR buffer via sysctl.
- Implement a function to ensure that all preempted threads have switched
back out at least once. Use
Author: jeff
Date: Wed Dec 19 20:08:06 2012
New Revision: 24
URL: http://svnweb.freebsd.org/changeset/base/24
Log:
- Correctly handle EWOULDBLOCK in quiesce_cpus
Discussed with: mav
Modified:
head/sys/kern/subr_smp.c
Modified: head/sys/kern/subr_smp.c
Author: jeff
Date: Wed Dec 19 20:10:00 2012
New Revision: 25
URL: http://svnweb.freebsd.org/changeset/base/25
Log:
- Add new machine parsable KTR macros for timing events.
- Use this new format to automatically handle syscalls and VOPs. This
changes the earlier format but is st
Author: jeff
Date: Tue Jul 23 22:52:38 2013
New Revision: 253583
URL: http://svnweb.freebsd.org/changeset/base/253583
Log:
- Correct a stale comment. We don't have vclean() anymore. The work is
done by vgonel() and destroy_vobject() should only be called once from
VOP_INACTIVE().
Author: jeff
Date: Wed Jul 24 01:25:56 2013
New Revision: 253587
URL: http://svnweb.freebsd.org/changeset/base/253587
Log:
- Remove the long obsolete 'vm_pageout_algorithm' experiment.
Discussed with: alc
Sponsored by: EMC / Isilon Storage Division
Modified:
head/sys/vm/vm_pageo
Author: jeff
Date: Fri Jul 26 19:06:14 2013
New Revision: 253685
URL: http://svnweb.freebsd.org/changeset/base/253685
Log:
- Use kmem_malloc rather than kmem_alloc() for GDT/LDT/tss allocations etc.
This eliminates some unusual uses of that API in favor of more typical
uses of kmem_ma
Author: jeff
Date: Fri Jul 26 23:22:05 2013
New Revision: 253697
URL: http://svnweb.freebsd.org/changeset/base/253697
Log:
Improve page LRU quality and simplify the logic.
- Don't short-circuit aging tests for unmapped objects. This biases
against unmapped file pages and transient ma
Author: jeff
Date: Mon Aug 5 00:28:03 2013
New Revision: 253949
URL: http://svnweb.freebsd.org/changeset/base/253949
Log:
- Introduce a specific function, pmap_remove_kernel_pde, for removing
huge pages in the kernel's address space. This works around several
asserts from pmap_demot
201 - 300 of 314 matches
Mail list logo
