| 183029184352_586360007172 | Tejun Heo |
|
There are a number of make_request based drivers which dont support
cache flushes. Filter out flush bios in __generic_make_request() so
that they dont have to worry about them. All FLUSH/FUA reques |
|
| |
| 103425954098_544260007481 | Nicholas A Bellinger |
|
Hello, Jens.
I added Mikes cleanup patch and Kiyoshis Tested-by and updated
flush-fua branch accordingly.
git://git.kernel.org/pub/scm/linux/kernel/git/tj/misc.git flush-fua
I guess its about re |
|
| |
| 121720194931_503360007073 | Tejun Heo |
|
Currently __blk_rq_prep_clone() copies only REQ_WRITE and REQ_DISCARD.
Theres no reason to omit other command flags and REQ_FUA needs to be
copied to implement FUA support in request-based dm.
REQ_C |
|
| |
| 144727304533_566160007542 | Joel Becker |
|
I NAKd this a while back. Im willing to be convinced, but so
far it remains that way.
Hi Joel,
Thanks for bringing this point up again. So a brief refresh on why this
is currentl |
|
| |
| 153720914395_517060007351 | Tejun Heo |
|
Remove now unused REQ_HARDBARRIER support. virtio_blk already
supports REQ_FLUSH and the usefulness of REQ_FUA for virtio_blk is
questionable at this point, so theres nothing else to do to support
n |
|
| |
| 184222344268_519460007389 | Joel Becker |
|
On Tue, 2010-09-07 at 15:44 -0700, Joel Becker wrote:
On Tue, Sep 07, 2010 at 05:01:01PM -0400, Konrad Rzeszutek Wilk wrote:
I NAKd this a while back. Im willing to be convinced, but so
|
|
| |
| 136621314984_546560007842 | Tejun Heo |
|
Unplugging from a request function doesnt really help much (its
already in the request_fn) and soon block layer will be updated to mix
barrier sequence with other commands, so theres no need to treat |
|
| |
| 194627924861_587760007352 | Joel Becker |
|
On Wed, 2010-09-08 at 12:26 -0700, Joel Becker wrote:
On Tue, Sep 07, 2010 at 07:08:59PM -0700, Nicholas A. Bellinger wrote:
On Tue, 2010-09-07 at 15:44 -0700, Joel Becker wrote:
Nick, c |
|
| |
| 195122724082_548360007856 | Joel Becker |
|
On Fri, 2010-09-10 at 08:28 -0700, Joel Becker wrote:
On Wed, Sep 08, 2010 at 01:53:27PM -0700, Nicholas A. Bellinger wrote:
On Wed, 2010-09-08 at 12:26 -0700, Joel Becker wrote:
So after r |
|
| |
| 102524704948_542060007352 | Jeff Layton |
|
On Fri, 2010-09-10 at 12:44 -0700, Joel Becker wrote:
On Fri, Sep 10, 2010 at 12:06:46PM -0700, Nicholas A. Bellinger wrote:
On Fri, 2010-09-10 at 08:28 -0700, Joel Becker wrote:
The tri |
|
| |
| 177124834587_503960007684 | Dave Chinner |
|
I switched from 2.6.35 to 2.6.36-rc3 a few days ago and while doing so
enabled lockdep. Just got the following report during normal desktop
usage:
=================================
[ INFO: inconsis |
|
| |
| 142820444816_541260007390 | Guenter Roeck |
|
If a CD has both Rock Ridge and Joliet extensions and the ISO root directory
is empty, no files are visible. Disable Rock Ridge extensions in this case and
use Joliet root directory instead.
Signed- |
|
| |
| 125324544181_511460007006 | Ondrej Zary |
|
On Tuesday 31 August 2010 22:44:55 Guenter Roeck wrote:
On Tue, 2010-08-31 at 14:10 -0400, Ondrej Zary wrote:
If a CD has both Rock Ridge and Joliet extensions and the ISO root
directory is |
|
| |
| 146622814279_554860007212 | Andreas Dilger |
|
Im investigating a performance degradation in yaffs when xattribs are
enabled. Yes, I know yaffs is not part of mainline Linux yet, but this is
really a vfs question.
What I notice is that on eve |
|
| |
| 140427904683_535260007324 | Dan Magenheimer |
|
[PATCH V4 8/8] Cleancache: ocfs2 hook for cleancache
Filesystems must explicitly enable cleancache by calling
cleancache_init_fs anytime a instance of the filesystem
is mounted and must save the ret |
|
| |
| 170429814808_529060007485 | Jeremy Fitzhardinge |
|
[PATCH V4 5/8] Cleancache: ext3 hook for cleancache
Filesystems must explicitly enable cleancache by calling
cleancache_init_fs anytime a instance of the filesystem
is mounted and must save the retu |
|
| |
| 197524494529_579060007037 | Dan Magenheimer |
|
@@ -1349,6 +1350,7 @@ static int ext3_setup_super(struct super
} else {
ext3_msg(sb, KERN_INFO, "using internal journal");
}
+ sb- cleancache_poolid = cleancache_ |
|
| |
| 118021144492_536160007823 | Jeremy Fitzhardinge |
|
[PATCH V4 3/8] Cleancache: core ops functions and configuration
Cleancache core ops functions and configuration
Credits: Cleancache_ops design derived from Jeremy Fitzhardinge
design for tmem; sysf |
|
| |
| 136726954103_564660007868 | Dan Magenheimer |
|
+#ifdef CONFIG_CLEANCACHE
+#define cleancache_enabled (cleancache_ops.init_fs)
Pointers can be used in a boolean context, but it would probably be
cleaner to have this evaluate to a pr |
|
| |
| 185926474657_554160007867 | Dan Magenheimer |
|
On 08/30/2010 03:32 PM, Dan Magenheimer wrote:
[PATCH V4 4/8] Cleancache: VFS hooks for cleancache
Implement core hooks in VFS for: initializing cleancache
per filesystem; capturing clean pa |
|
| |
| 186627034848_567360007868 | Dan Magenheimer |
|
[PATCH V4 1/8] Cleancache: Documentation
Add cleancache documentation to Documentation/vm and
sysfs ABI documentation to Documentation/ABI
Signed-off-by: Dan Magenheimer <dan.magenheimer@xxxxxxx |
|
| |
| 115729684316_568760007177 | Valerie Aurora |
|
[PATCH V4 0/8] Cleancache: overview
Changes from V3 to V4:
- Rebased to 2.6.36-rc3
- Use exportfs/filehandle for unique file identification on next-gen FSs
(Many thanks to Andreas Dilger for guida |
|
| |
| 172429384291_521660007520 | Aneesh Kumar KV |
|
On Mon, Aug 30, 2010 at 05:23:12PM -0400, Valerie Aurora wrote:
Documentation: Fix trivial typo in filesystems/sharedsubtree.txt
This typo is easy to ignore unless you have spent a great d |
|
| |
| 103826664214_514260007457 | Tejun Heo |
|
We should use the cached dentry operation only if caching mode is enabled
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@xxxxxxxxxxxxxxxxxx
---
fs/9p/vfs_inode.c | 5 ++++-
1 files changed, 4 |
|
| |
| 100921704104_566760007103 | Mike Snitzer |
|
This patch converts request-based dm to support the new REQ_FLUSH/FUA.
The original request-based flush implementation depended on
request_queue blocking other requests while a barrier sequence is i |
|
| |
| 153325784806_538060007799 | Mike Snitzer |
|
This patch converts request-based dm to support the new REQ_FLUSH/FUA.
The original request-based flush implementation depended on
request_queue blocking other requests while a barrier sequence is i |
|
| |
| 124824084767_556560007524 | Tejun Heo |
|
Hello,
On 08/30/2010 03:28 PM, Mike Snitzer wrote:
+ clone- cmd = rq- cmd;
+ clone- cmd_len = rq- cmd_len;
+ clone- sense = rq- sense;
+ clone- buffer = rq- buffer;
clone- end_io = |
|
| |
| 174228194462_503360007761 | Tejun Heo |
|
Hello,
On 08/30/2010 03:28 PM, Mike Snitzer wrote:
+ clone- cmd = rq- cmd;
+ clone- cmd_len = rq- cmd_len;
+ clone- sense = rq- sense;
+ clone- buffer = rq- buffer;
clone- end_io = |
|
| |
| 155320734427_576860007739 | Mike Snitzer |
|
On Mon, Aug 30 2010 at 11:07am -0400,
Tejun Heo <tj@xxxxxxxxxx wrote:
On 08/30/2010 03:59 PM, Tejun Heo wrote:
Ah... thats probably from "if (!elv_queue_empty(q))" check below,
|
|
| |
| 122522254910_532960007622 | Mike Snitzer |
|
On Mon, Aug 30 2010 at 11:07am -0400,
Tejun Heo <tj@xxxxxxxxxx wrote:
On 08/30/2010 03:59 PM, Tejun Heo wrote:
Ah... thats probably from "if (!elv_queue_empty(q))" check below,
|
|
| |
| 143224264199_523760007450 | Mike Snitzer |
|
On 08/30/2010 11:28 PM, Mike Snitzer wrote:
On Mon, Aug 30 2010 at 3:08pm -0400,
Mike Snitzer <snitzer@xxxxxxxxxx wrote:
On Mon, Aug 30 2010 at 11:07am -0400,
Tejun Heo <tj@xxxxx |
|
| |
| 103620544858_580360007952 | Mike Snitzer |
|
On 08/30/2010 11:28 PM, Mike Snitzer wrote:
On Mon, Aug 30 2010 at 3:08pm -0400,
Mike Snitzer <snitzer@xxxxxxxxxx wrote:
On Mon, Aug 30 2010 at 11:07am -0400,
Tejun Heo <tj@xxxxx |
|
| |
| 132723614770_579060007805 | Tejun Heo |
|
Hello,
On 08/31/2010 03:02 PM, Mike Snitzer wrote:
On Tue, Aug 31 2010 at 6:29am -0400,
Tejun Heo <tj@xxxxxxxxxx wrote:
On 08/30/2010 11:28 PM, Mike Snitzer wrote:
Hmm, but unfort |
|
| |
| 155029494878_597860007093 | Tejun Heo |
|
Hello,
On 08/31/2010 03:02 PM, Mike Snitzer wrote:
On Tue, Aug 31 2010 at 6:29am -0400,
Tejun Heo <tj@xxxxxxxxxx wrote:
On 08/30/2010 11:28 PM, Mike Snitzer wrote:
Hmm, but unfort |
|
| |
| 171729424319_562760007211 | Mike Snitzer |
|
This patch converts request-based dm to support the new REQ_FLUSH/FUA.
The original request-based flush implementation depended on
request_queue blocking other requests while a barrier sequence is i |
|
| |
| 124827034637_544860007443 | Mike Snitzer |
|
Hi Tejun,
On 08/31/2010 12:45 AM +0900, Tejun Heo wrote:
This patch converts request-based dm to support the new REQ_FLUSH/FUA.
The original request-based flush implementation depended on
r |
|
| |
| 189829774498_575660007117 | Tejun Heo |
|
Hello,
On 09/01/2010 09:15 AM, Kiyoshi Ueda wrote:
I dont see any obvious problem on this patch.
However, I hit a NULL pointer dereference below when I use a mpath
device with barrier option o |
|
| |
| 170721074478_547760007874 | Tejun Heo |
|
Hi Tejun,
On 09/02/2010 10:22 PM +0900, Tejun Heo wrote:
On 09/01/2010 09:15 AM, Kiyoshi Ueda wrote:
@@ -2619,9 +2458,8 @@ int dm_suspend(struct mapped_device *md,
up_write(&md- io_l |
|
| |
| 128124464822_515960007537 | Tejun Heo |
|
Hi Tejun,
On 09/03/2010 02:43 AM +0900, Tejun Heo wrote:
rq- rq_disk and bio- bi_bdev- bd_disk may differ if a request has
passed through remapping drivers. FSEQ_DATA request incorrectly
foll |
|
| |
| 125221404478_557260007749 | Tejun Heo |
|
Hi Tejun,
On 09/03/2010 06:33 PM +0900, Tejun Heo wrote:
On 09/03/2010 07:47 AM, Kiyoshi Ueda wrote:
By the way, I had been considering a block-layer interface which remaps
struct request an |
|
| |
| 178420154452_542160007392 | Tejun Heo |
|
Hi Tejun,
On 09/03/2010 08:42 PM +0900, Tejun Heo wrote:
Hello,
On 09/03/2010 12:28 PM, Kiyoshi Ueda wrote:
Thank you for pointing it.
Yes, the interface should also set rq- q.
Ab |
|
| |
| 186423614369_536260007149 | Tejun Heo |
|
On Mon, Aug 30 2010 at 5:58am -0400,
Tejun Heo <tj@xxxxxxxxxx wrote:
Unlike REQ_HARDBARRIER, REQ_FLUSH/FUA doesnt mandate any ordering
against other bios. This patch relaxes ordering aroun |
|
| |
| 147928244256_595860007152 | Tejun Heo |
|
Hi Tejun,
On 08/30/2010 06:58 PM +0900, Tejun Heo wrote:
Unlike REQ_HARDBARRIER, REQ_FLUSH/FUA doesnt mandate any ordering
against other bios. This patch relaxes ordering around flushes.
...
|
|
| |
| 183125814599_565260007022 | Tejun Heo |
|
From: Christoph Hellwig <hch@xxxxxxxxxxxxx
Its unused now.
Signed-off-by: Christoph Hellwig <hch@xxxxxx
Signed-off-by: Tejun Heo <tj@xxxxxxxxxx
---
include/linux/fs.h | 3 ---
1 fil |
|
| |
| 118929934162_566860007930 | Tejun Heo |
|
On Mon, Aug 30, 2010 at 11:58:12AM +0200, Tejun Heo wrote:
Currently __blk_rq_prep_clone() copies only REQ_WRITE and REQ_DISCARD.
Theres no reason to omit other command flags and REQ_FUA needs to |
|
| |
| 141723484945_586960007407 | Tejun Heo |
|
On Mon, Aug 30 2010 at 5:58am -0400,
Tejun Heo <tj@xxxxxxxxxx wrote:
This patch converts bio-based dm to support REQ_FLUSH/FUA instead of
now deprecated REQ_HARDBARRIER.
* -EOPNOTSUPP |
|
| |
| 126126334768_563160007885 | Tejun Heo |
|
On Wed, Sep 01 2010 at 9:50am -0400,
Tejun Heo <tj@xxxxxxxxxx wrote:
Hello,
On 09/01/2010 03:43 PM, Mike Snitzer wrote:
On Mon, Aug 30 2010 at 5:58am -0400,
Tejun Heo <tj@xxx |
|
| |
| 178024744708_503560007007 | KOSAKI Motohiro |
|
Brad Spengler published a local memory-allocation DoS that
evades the OOM-killer (though not the virtual memory RLIMIT):
rel="nofollow" www.grsecurity.net/~spender/64bit_dos.c www.grsecurity.ne |
|
| |
| 118822624589_578760007329 | Solar Designer |
|
IMHO unlimited should mean unlimited. So, on that score, Id leave this
constraint out and just say whatever deficiencies in the OOM killer (or in
whatever should make a manifestly too-large allocati |
|
| |
| 180921044322_597260007887 | Solar Designer |
|
IIRC, prior to that fix, I was able to cause the kernel to loop for tens
of minutes in a single execve() call on an Alpha with 128 MB RAM, by
using repeated mappings of the same pages (almost 2 |
|
| |