Commit(afcc8a40a090f): Introduce x86_io_apic_ops.print_entries for
debugging. Using kernel parameter apci=debug/verbose, kernel print
the incomplete infomation if we don't enable CONFIG_DYNAMIC_DEBUG.
So using printk to replace.
Signed-off-by: Jianpeng Ma
---
arch/x86/kernel/apic/io_apic.c |
Hi all:
I met a problem. When i plug vga cable(only need once), the kernel
print those messages and disable IRQ 16.
[ 16.907826] irq 16: nobody cared (try booting with the "irqpoll" option)
[ 16.907834] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.14.0-rc3+ #74
[ 16.907836] Hardware
For bio with REQ_NOMERGE,it mean this bio can't merge with other bios.
And the request with this bio has the same meaning.
In blk_queue_bio, bio with REQ_NOMERGE can't be merged and get a new
request. But in init_request_from_bio, request can't test REQ_NOMERGE.
So the request can merge other bio o
Signed-off-by: Jianpeng Ma
---
block/blk-core.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/block/blk-core.c b/block/blk-core.c
index 0a00e4e..fb970fd 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1492,7 +1492,7 @@ void blk_queue_bio(struct request_queue *q, str
This patch set mainly handle some bugs for REQ_NOMERGE.
Jianpeng Ma (3):
block: Add REQ_NOMERGE into REQ_COMMON_MASK
block: Check bio can merge before call attempt_plug_merge.
md/raid5: For stripe with R5_ReadNoMerge, we replace REQ_FLUSH with
REQ_NOMERGE.
block/blk-core.c
For R5_ReadNoMerge,it mean this bio can't merge with other bios or
request.It used REQ_FLUSH to achieve this. But REQ_NOMERGE can do the
same work.
Signed-off-by: Jianpeng Ma
---
drivers/md/raid5.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/md/raid5.c b/drivers/m
For ext2 mounted with errors=remount-ro, if write a file and because the
harddisk error, the fs became ro.But the following test can't return.
int main()
{
int ret;
int i = 0;
char buff[SIZE];
int fd = open("/opt/test", O_WRONLY|O_TRUNC|O_CREAT);
if (fd < 0
>majianpeng writes:
>
>>>majianpeng writes:
>>>
>>>> For async-write on block device,if device removed,but the vfs don't know
>>>> it.
>>>> It will continue to do.
>>>> Patch1 set size of inode of block device to zer
If reread partitions of disk, the paths are:
blkdev_ioctl->blkdev_reread_part->rescan_partitions->drop_partitions
->invalidate_partition->__invalidate_device
In func __invalidate_device, if disk already mounted, it will do
shrink_dcache_sb and invalidate_inodes.
This operation will effect the fs.
S
Sorry for that, i notice the commit 73a7075e3f6e.But i misread it.
I'll modify this as soon as possible.
Thanks!
Jianpeng Ma
>Hi Sage,
>
>After merging the ceph tree, today's linux-next build (x86_64
>allmodconfig) failed like this:
>
>In file included from include/linux/kernel.h:14:0,
>
>majianpeng writes:
>
>>>majianpeng writes:
>>>
>>>> For async-write on block device,if device removed,but the vfs don't know
>>>> it.
>>>> It will continue to do.
>>>> Patch1 set size of inode of block device to zer
>majianpeng writes:
>
>> For async-write on block device,if device removed,but the vfs don't know it.
>> It will continue to do.
>> Patch1 set size of inode of block device to zero when removed disk.By
>> this,vfs know
>> disk changed.
>> Path2
Hi all,
This patch has a long long time.How about this patch?
Thanks!
Jianpeng Ma
>Hi all,
>Some time ago, I mentioned there are some problems on x86-32 system about
> handling md-block-device which size is larger than 16TB.
>And i send a patch.But there are no concern with it.
>The w
For async-write on block device,when disk removed,the vfs don't know.
It will continue do async-write.Add this check it will stop async-write
when disk removed.
Signed-off-by: Jianpeng Ma
---
fs/block_dev.c | 4
1 file changed, 4 insertions(+)
diff --git a/fs/block_dev.c b/fs/block_dev.c
i
When remove gendisk, set the size of inode of block_device to zero.
This is mainly let the vfs know disk removed.
Signed-off-by: Jianpeng Ma
---
block/genhd.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/block/genhd.c b/block/genhd.c
index dadf42b..9a4c245 100644
--- a/block/genhd.c
For async-write on block device,if device removed,but the vfs don't know it.
It will continue to do.
Patch1 set size of inode of block device to zero when removed disk.By this,vfs
know
disk changed.
Path2 add size-check on blk_aio_write.If pos of write larger than size of
inode,it will
return ze
For async-write on block device,when disk removed,the vfs don't know.
It will continue do async-write.Add this check it will stop async-write
when disk removed.
Signed-off-by: Jianpeng Ma
---
fs/block_dev.c | 4
1 file changed, 4 insertions(+)
diff --git a/fs/block_dev.c b/fs/block_dev.c
i
For async-write on block device,if device removed,but the vfs don't know it.
It will continu do.
Patch1 set size of inode of block device to zero when removed disk.By this,vfs
know
disk changed.
Path2 add size-check on blk_aio_write.If pos of write larger than size of
inode,it will
return zero.S
When remove gendisk, set the size of inode of block_device to zero.
This is mainly let the vfs know disk removed.
Signed-off-by: Jianpeng Ma
---
block/genhd.c | 14 ++
1 file changed, 14 insertions(+)
diff --git a/block/genhd.c b/block/genhd.c
index dadf42b..c3a54b4 100644
--- a/blo
Hi axboe:
How about this patch?
Thanks!
Jianpeng Ma
>In func blk_queue_bio, if list of plug is empty,it will call
>blk_trace_plug.
>If process deal with a single device,it't ok.But if process deal with
>multi devices,it only trace the first device.
>Using request_count to judge, it can sol
Hi all,
The definition of func generic_file_buffered_write is:
ssize_t generic_file_buffered_write(struct kiocb *, const
struct iovec *,
unsigned long, loff_t, loff_t *, size_t, ssize_t);
The fourth an
In func blk_queue_bio, if list of plug is empty,it will call
blk_trace_plug.
If process deal with a single device,it't ok.But if process deal with
multi devices,it only trace the first device.
Using request_count to judge, it can soleve this problem.
In addition, i modify the comment.
Signed-off-
>Hi Jianpeng Ma,
>
>On 8/1/2013 10:18 AM, majianpeng wrote:
>> We found a problem when we removed a working sd card that the irqaction
>> of omap_hsmmc can sleep to 3.6s. This cause our watchdog to work.
>> In func omap_hsmmc_reset_controller_fsm, it should watch a
We found a problem when we removed a working sd card that the irqaction
of omap_hsmmc can sleep to 3.6s. This cause our watchdog to work.
In func omap_hsmmc_reset_controller_fsm, it should watch a 0->1
transition.It used loops_per_jiffy as the timer.
The code is:
> while ((!(OMAP_HSMMC_READ(host->b
>Hi Jianpeng,
>
>On 07/26/2013 03:08 PM, majianpeng wrote:
>
>> Hi all,
>> I used splice and found a prolem(at least i call).
>> The demo is:
>> A:splice(regularfileA--->pipe);
>> B:splice(pipe--->regularfileB)
>> Before do B, we
Hi all,
I used splice and found a prolem(at least i call).
The demo is:
A:splice(regularfileA--->pipe);
B:splice(pipe--->regularfileB)
Before do B, we modify the data of regA which now in pipe. The data to
regularfileB willbe change.
If we used the buff
A:read(regA, buff);
B: write(buff, r
>?
>
>Thanks!
>sage
>
>
>On Wed, 24 Jul 2013, majianpeng wrote:
>
>> Ping
>> >Hi sage,
>> >How about this patch?Can you give some advisement?
>> >Thanks!
>> >Jianpeng Ma
>> >>At now for synchronous-fs, all write-operati
Hi all,
Now i worked on omp2 and met a probelm which someplace close_irq for
3.6second.
The kernel version is 2.6.37. I used trace to find in irq_action:omap_hsmmc_irq.
This problem occured by removed the sdcard when there are io operations.
I found the read problem is in omap_hsmmc_rese
Fix those warning message when compile:
>kernel/time/tick-sched.c: In function ‘tick_nohz_init’:
>kernel/time/tick-sched.c:346:6: warning: unused variable ‘cpu’
>[-Wunused-variable]
Signed-off-by: Jianpeng Ma
---
kernel/time/tick-sched.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/kern
In func ceph_osdc_start_request, if __mape_request failed, it should
call __unregister_request avoid cause bugs, etc memleak, umount fs
failed.
Signed-off-by: Jianpeng Ma
---
net/ceph/osd_client.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
i
Hi all,
Compile x86_64 meet those messages:
WARNING: arch/x86/mm/built-in.o(.text.unlikely+0xbf8): Section mismatch in
reference from the function __node_set.constprop.0() to the variable
.init.data:numa_nodes_parsed
The function __node_set.constprop.0() references
the variable __initdata
For ceph_sync_read,it don't call file_accessed to update atime.But the
buffer_read do it .So add it.
Signed-off-by: Jianpeng Ma
---
fs/ceph/file.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index fa74e6f..b0e6f0b 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/
Now ceph don't update atime after read.So add this function.
Signed-off-by: Jianpeng Ma
---
fs/ceph/file.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 656e169..fa74e6f 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -669,6 +669,15 @@ aga
>On Tue, 2 Jul 2013, majianpeng wrote:
>> Now, update atime only for CEPH_CAP_FILE_EXCL.Change this if
>> CEPH_CAP_FILE_RD.
>
>Can we introduce a global config optoin (bool mds_atime in
>common/config_opts.h, maybe) so that users can turn this off? And/or add
>a &#x
Now,when sent a cap msg to inode/snap, it don't include the ctime of
inode/snap.So make the mtime ahead of ctime.
BTY,for snap, i'm not exactly send the ctime of snap or null.
Signed-off-by: Jianpeng Ma
---
fs/ceph/caps.c | 11 +++
1 file changed, 7 insertions(+), 4 deletions(-)
diff -
Now, update atime only for CEPH_CAP_FILE_EXCL.Change this if
CEPH_CAP_FILE_RD.
Signed-off-by: Jianpeng Ma
---
src/mds/Locker.cc | 15 +--
1 file changed, 9 insertions(+), 6 deletions(-)
diff --git a/src/mds/Locker.cc b/src/mds/Locker.cc
index 30e014a..58f953f 100644
--- a/src/mds/Lo
Now ceph don't support updating atime after read-operation if the open
mode is CEPH_CAP_FILE_RD.There are two reasons:
1:in client of fs,it don't set dirty cap of CEPH_CAP_FILE_RD.
2:in mds,it only update the atime if the condition
"dirty & (CEPH_CAP_FILE_EXCL|CEPH_CAP_FILE_WR) is true.
But if we c
For buffer read, the func generic_file_aio_read will update atime of
file.But the ceph_sync_read don't do it.So add this.
Signed-off-by: Jianpeng Ma
---
fs/ceph/file.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 656e169..87df15a 100644
--- a/fs/ce
Now update atime of file, only for client had CEPH_CAP_FLE_EXCL.But for atime,
if one can read he can update this attribte.
For this feature,it need both client and mds modify.
PATCH1,it modify the mds to support.
PATCH2,it modify the client to suport.
PATCH3, it support this feature for sync_read
Thare's a race between elevator switching and normal io operation.
Because the allocation of struct elevator_queue and struct elevator_data
don't in a atomic operation.So there are have chance to use NULL
->elevator_data.
For example:
Thread A:
At now for synchronous-fs, all write-operations use ceph_sync_mode.
But for the file which opened with O_SYNC, it don't use sync_mode.
The behaviour of them should be the same.
For fs which mounted using '-o sync', it want all I/O to the filesystem
should be done synchronously.But the ceph-sync-mod
Signed-off-by: Jianpeng Ma
---
fs/ceph/caps.c | 21 +++--
fs/ceph/super.h | 2 +-
2 files changed, 8 insertions(+), 15 deletions(-)
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index da0f9b8..6d2655c 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -147,7 +147,7 @@ void ceph
Because in func queue_con_delay, it will call con->ops->get(con).
So if we cancel the work successfully,we must call con->ops->put(con).
Otherwise some resource can't free like 'struct ceph_mds_session'.
Signed-off-by: Jianpeng Ma
---
net/ceph/messenger.c | 3 ++-
1 file changed, 2 insertions(+)
Signed-off-by: Jianpeng Ma
---
fs/ceph/addr.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 3e68ac1..25fa8b8 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -677,8 +677,8 @@ static int ceph_writepages_start(struct address_space
[ 1121.231883] BUG: sleeping function called from invalid context at
kernel/rwsem.c:20
[ 1121.231935] in_atomic(): 1, irqs_disabled(): 0, pid: 9831, name: mv
[ 1121.231971] 1 lock held by mv/9831:
[ 1121.231973] #0:
(&(&ci->i_ceph_lock)->rlock){+.+...},at:[]
ceph_getxattr+0x58/0x1d0 [ceph]
[ 1
>On Tue, 18 Jun 2013, majianpeng wrote:
>> [ 1121.231883] BUG: sleeping function called from invalid context at
>> kernel/rwsem.c:20
>> [ 1121.231935] in_atomic(): 1, irqs_disabled(): 0, pid: 9831, name: mv
>> [ 1121.231971] 1 lock held by mv/9831:
>> [ 1121.
[ 1121.231883] BUG: sleeping function called from invalid context at
kernel/rwsem.c:20
[ 1121.231935] in_atomic(): 1, irqs_disabled(): 0, pid: 9831, name: mv
[ 1121.231971] 1 lock held by mv/9831:
[ 1121.231973] #0: (&(&ci->i_ceph_lock)->rlock){+.+...},
at:[] ceph_getxattr+0x58/0x1d0 [ceph]
[ 1
Either in vfs_write or io_submit,it call file_start/end_write.
The different between file_start/end_write and sb_start/end_write is
file_ only handle regular file.But i think in ceph_aio_write,it only
for regular file.
Signed-off-by: Jianpeng Ma
---
fs/ceph/file.c | 2 --
1 file changed, 2 delet
When cleanup extra stat if this object was a notifier,it should free the
notifier.
Signed-off-by: Jianpeng Ma
---
drivers/gpu/drm/nouveau/nouveau_abi16.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c
b/drivers/gpu/drm/nouveau/nouv
Hi all,
I wanted to monitor /proc/partitions using inotify.Although now i know it
can't catch the event of write.
But for read,it can't catch. The code is:
int main()
{
int fd = inotify_init();
int wd;
char buff[4096];
int ret;
if (fd < 0)
>On Tue, May 14, 2013 at 12:49:44PM +0100, majianpeng wrote:
>> Signed-off-by: Jianpeng Ma
>> ---
>> mm/kmemleak.c | 8 +---
>> 1 file changed, 1 insertion(+), 7 deletions(-)
>>
>> diff --git a/mm/kmemleak.c b/mm/kmemleak.c
>> index b1525db..f0ec
On 05/16/2013 04:41 PM, Peter Zijlstra wrote:
> On Thu, May 16, 2013 at 09:16:45AM +0800, majianpeng wrote:
>
>>> There isn't. What you typically want to do is annotate the lock site.
>>> In particular it looks like mutex_lock_all() is the offensive piece of
>>&g
On 05/15/2013 04:28 PM, Peter Zijlstra wrote:
> On Wed, May 15, 2013 at 02:58:53PM +0800, majianpeng wrote:
>> When mounted the f2fs, kernel will print the following messages:
>>
>> [ 105.533038] =
>> [ 105.533065] [ INFO:
On 05/15/2013 03:21 PM, Libo Chen wrote:
> On 2013/5/15 14:58, majianpeng wrote:
>> By adding some messages, i found this problem because the gcc
>> optimizing. For those codes:
>>>>for (i = 0; i < NR_GLOBAL_LOCKS; i++)
>>>>mutex_in
When mounted the f2fs, kernel will print the following messages:
[ 105.533038] =
[ 105.533065] [ INFO: possible recursive locking detected ]
[ 105.533088] 3.10.0-rc1+ #101 Not tainted
[ 105.533105] -
[ 105
If the scan-areas are adjacent,it can merge in order to reduce memomy.
And using pr_warn instead of pr_warning.
Signed-off-by: Jianpeng Ma
---
mm/kmemleak.c | 26 +++---
1 file changed, 19 insertions(+), 7 deletions(-)
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index f0ece93
Signed-off-by: Jianpeng Ma
---
mm/kmemleak.c | 8 +---
1 file changed, 1 insertion(+), 7 deletions(-)
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index b1525db..f0ece93 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -1225,22 +1225,16 @@ static void scan_gray_list(void)
* from inside
Jianpeng Ma (3):
mm/kmemleak.c: Use %u to print ->checksum.
mm/kmemleak.c: Use list_for_each_entry_safe to reconstruct function
scan_gray_list.
mm/kmemleak.c: Merge the consecutive scan-areas.
mm/kmemleak.c | 36 +---
1 file changed, 21 insertions(+),
Signed-off-by: Jianpeng Ma
---
mm/kmemleak.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/kmemleak.c b/mm/kmemleak.c
index c8d7f31..b1525db 100644
--- a/mm/kmemleak.c
+++ b/mm/kmemleak.c
@@ -386,7 +386,7 @@ static void dump_object_info(struct kmemleak_object *object)
In commit 523c81135,it used "__refdata" on event_class_ftrace_##call.
It will cause kmemleak to misjudge because when loading module it did
not add '.ref.data' into kmemleak-scan-area.
Signed-off-by: Jianpeng Ma
---
kernel/module.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff
Hi all,
Some time ago, I mentioned there are some problems on x86-32 system about
handling md-block-device which size is larger than 16TB.
And i send a patch.But there are no concern with it.
The website of is: http://www.spinics.net/lists/linux-fsdevel/msg55672.html.
Except the wrapping pr
Because alloc 'struct buffer_head' using kmem_cache_zalloc,so it doesn't
need to call 'init_buffer(bh, NULL, NULL)' and other set-zero-operation.
Signed-off-by: Jianpeng Ma
---
fs/buffer.c | 2 --
fs/jbd/journal.c | 2 --
fs/jbd2/journal.c | 2 --
3 files changed, 6 deletions(-)
diff --g
On 04/19/2013 10:48 PM, Jan Kara wrote:
> On Fri 19-04-13 10:29:36, majianpeng wrote:
>> Because alloc struct buffer_head using kmem_cache_zalloc,so it's not need to
>> call 'init_buffer(bh, NULL, NULL)'.
>>
>> Signed-off-by: Jianpeng Ma
>> ---
&
Because alloc struct buffer_head using kmem_cache_zalloc,so it's not need to
call 'init_buffer(bh, NULL, NULL)'.
Signed-off-by: Jianpeng Ma
---
fs/buffer.c | 2 --
fs/jbd/journal.c | 1 -
fs/jbd2/journal.c | 1 -
3 files changed, 4 deletions(-)
diff --git a/fs/buffer.c b/fs/buffer.c
inde
Hi Jens,
How about this patch? It was a long time.
Thanks!
Jianpeng Ma
> Thare's a race between elevator switching and normal io operation.
> Because the allocation of struct elevator_queue and struct elevator_data
> don't in a atomic operation.So there are have chance to use NULL
> ->
Signed-off-by: Jianpeng Ma
---
fs/block_dev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/block_dev.c b/fs/block_dev.c
index aae187a..f0328f1 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1559,7 +1559,7 @@ static ssize_t blkdev_aio_read(struct kiocb *iocb, const
After allocating a inode,it don't chain in superblock.At this time,it
won't be used by others.So it make no sense to protect there.
Signed-off-by: Jianpeng Ma
---
fs/inode.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/fs/inode.c b/fs/inode.c
index f5f7c06..1d7b0dd 100644
--- a/fs/inode.c
For direct-write and normal-write operations, after write they will call
fuse_invalidate_attr.
So it's make no sense to call file_update_time before writing.
In function file_update_time, there is other thing which update version of
inode.For this i am not sure.
Signed-off-by: Jianpeng Ma
---
f
If fc->big_writes == 0, for write-operation it only do 4k in func
fuse_fill_write_pages.
So when alloc fuse_req,we should consider this restrict in order to
do additional operations.
The additional operations are memset-opeartion in func fuse_requenst_alloc and
alloc more space if nr_pages is lar
Thare's a race between elevator switching and normal io operation.
Because the allocation of struct elevator_queue and struct elevator_data
don't in a atomic operation.So there are have chance to use NULL
->elevator_data.
For example:
Thread A:
>On Sun, Feb 03 2013, majianpeng wrote:
>> Hi all,
>> When I wanted to do discard operations,but i set the openflag was
>> O_RDONLY,it returned a EBADF rather than EACCES or EPERM.
>> I searched the code and found:
>> >case BLKDISCARD:
>> >c
Hi all,
When I wanted to do discard operations,but i set the openflag was
O_RDONLY,it returned a EBADF rather than EACCES or EPERM.
I searched the code and found:
>case BLKDISCARD:
>case BLKSECDISCARD: {
> uint64_t range[2];
> if (!(mode & FMODE_WRITE))
>
>On Mon, Jan 14, 2013 at 10:15:58AM +0800, majianpeng wrote:
>
>[..]
>> >Index: linux-2.6/block/blk-throttle.c
>> >===
>> >--- linux-2.6.orig/block/blk-throttle.c 2012-10-18 01:52:28.
[maybe you received more, I'm very sorry for that]
>On Fri, Jan 11, 2013 at 03:26:37PM +0100, Jens Axboe wrote:
>> On 2013-01-11 10:11, majianpeng wrote:
>> > In func tg_may_dispatch,
>> >> if (throtl_slice_used(td, tg, rw))
>> >> throtl
>On Fri, Jan 11, 2013 at 03:26:37PM +0100, Jens Axboe wrote:
>> On 2013-01-11 10:11, majianpeng wrote:
>> > In func tg_may_dispatch,
>> >> if (throtl_slice_used(td, tg, rw))
>> >> throtl_start_new_slice(td, tg, rw);
>> > ...
&g
>On Fri, Jan 11, 2013 at 03:26:37PM +0100, Jens Axboe wrote:
>> On 2013-01-11 10:11, majianpeng wrote:
>> > In func tg_may_dispatch,
>> >> if (throtl_slice_used(td, tg, rw))
>> >> throtl_start_new_slice(td, tg, rw);
>> > ...
&g
In func tg_may_dispatch,
>if (throtl_slice_used(td, tg, rw))
> throtl_start_new_slice(td, tg, rw);
...
>if (tg_with_in_bps_limit(td, tg, bio, &bps_wait)
> && tg_with_in_iops_limit(td, tg, bio, &iops_wait)) {
In funcs tg_with_in_(bps/iops)_limit, it used the slice_start to c
Because only mergeable rq can add rqhash.So it does not make sense to
judge rq_mergeable(rq) in func elv_rqhash_find.
Signed-off-by: Jianpeng Ma
---
block/elevator.c |5 -
1 file changed, 5 deletions(-)
diff --git a/block/elevator.c b/block/elevator.c
index 9edba1b..d5901a4 100644
--- a
Because only mergeable rq can add rqhash.So it does not make sense to
judge rq_mergeable(rq) in func elv_rqhash_find.
Signed-off-by: Jianpeng Ma
---
block/elevator.c |5 -
1 file changed, 5 deletions(-)
diff --git a/block/elevator.c b/block/elevator.c
index 9edba1b..d5901a4 100644
--- a
In commit 975927b942c932,it add blk_rq_pos to sort rq when flushing.
Although this commit was used for the situation which blk_plug handled
multi devices on the same time like md device.
I think there must be some situations like this but only single
device.
So remove the should_sort judgement.
Bec
I met a odd prblem:read /proc/partitions may return zero.
I wrote a file test.c:
int main()
{
char buff[4096];
int ret;
int fd;
printf("pid=%d\n",getpid());
while (1) {
fd = open("/proc/partitions", O_RDONLY);
if (fd < 0) {
On 2012-07-27 22:21 Jeff Moyer Wrote:
>majianpeng writes:
>
>> When exec bio_alloc, the bi_rw is zero.But after calling bio_add_page,
>> it will use bi_rw.
>> Fox example, in functiion __bio_add_page,it will call merge_bvec_fn().
>> The merge_bvec_fn of raid456 w
When exec bio_alloc, the bi_rw is zero.But after calling bio_add_page,
it will use bi_rw.
Fox example, in functiion __bio_add_page,it will call merge_bvec_fn().
The merge_bvec_fn of raid456 will use the bi_rw to judge the merge.
>> if ((bvm->bi_rw & 1) == WRITE)
>> return biovec->bv_le
83 matches
Mail list logo
