Re: O_DIRECT to md raid 6 is slow
On 2012-08-15 08:49 Andy Lutomirski Wrote:
>If I do:
># dd if=/dev/zero of=/dev/md0p1 bs=8M
>then iostat -m 5 says:
>
>avg-cpu: %user %nice %system %iowait %steal %idle
> 0.000.00 26.88 35.270.00 37.85
>
>Device:tpsMB_read/sMB_wrtn/sMB_readMB_wrtn
>sdb 265.20 1.1654.79 5273
>sdc 266.20 1.4754.73 7273
>sdd 264.20 1.3854.54 6272
>sdf 286.00 1.8454.74 9273
>sde 266.60 1.0454.75 5273
>sdg 265.00 1.0254.74 5273
>md0 55808.00 0.00 218.00 0 1090
>
>If I do:
># dd if=/dev/zero of=/dev/md0p1 bs=8M oflag=direct
>then iostat -m 5 says:
>avg-cpu: %user %nice %system %iowait %steal %idle
> 0.000.00 11.70 12.940.00 75.36
>
>Device:tpsMB_read/sMB_wrtn/sMB_readMB_wrtn
>sdb 831.00 8.5830.42 42152
>sdc 832.80 8.0529.99 40149
>sdd 832.00 9.1029.78 45148
>sdf 838.40 9.1129.72 45148
>sde 828.80 7.9129.79 39148
>sdg 850.80 8.0030.18 40150
>md01012.60 0.00 101.27 0506
>
>It looks like md isn't recognizing that I'm writing whole stripes when
>I'm in O_DIRECT mode.
>
kernel version?
>--Andy
>
>--
>Andy Lutomirski
>AMA Capital Management, LLC
>--
>To unsubscribe from this list: send the line "unsubscribe linux-raid" in
>the body of a message to majord...@vger.kernel.org
>More majordomo info at
>http://vger.kernel.org/majordomo-info.htmlN�Р骒r��yb�X�肚�v�^�)藓{.n�+�伐�{��赙zXФ�≤�}��财�z�&j:+v�����赙zZ+��+zf＂�h���~i���z� �wア�?�ㄨ��&�)撷f��^j谦y�m��@A�a囤�
> 0鹅h���i
Re: Re: O_DIRECT to md raid 6 is slow
On 2012-08-15 09:12 Andy Lutomirski Wrote:
>Ubuntu's 3.2.0-27-generic. I can test on a newer kernel tomorrow.
I guess maybe miss the blk_plug function.
Can you add this patch and retest.
Move unplugging for direct I/O from around ->direct_IO() down to
do_blockdev_direct_IO(). This implicitly adds plugging for direct
writes.
CC: Li Shaohua
Acked-by: Jeff Moyer
Signed-off-by: Wu Fengguang
---
fs/direct-io.c |5 +
mm/filemap.c |4
2 files changed, 5 insertions(+), 4 deletions(-)
--- linux-next.orig/mm/filemap.c 2012-08-05 16:24:47.859465122 +0800
+++ linux-next/mm/filemap.c 2012-08-05 16:24:48.407465135 +0800
@@ -1412,12 +1412,8 @@ generic_file_aio_read(struct kiocb *iocb
retval = filemap_write_and_wait_range(mapping, pos,
pos + iov_length(iov, nr_segs) - 1);
if (!retval) {
- struct blk_plug plug;
-
- blk_start_plug(&plug);
retval = mapping->a_ops->direct_IO(READ, iocb,
iov, pos, nr_segs);
- blk_finish_plug(&plug);
}
if (retval > 0) {
*ppos = pos + retval;
--- linux-next.orig/fs/direct-io.c 2012-07-07 21:46:39.531508198 +0800
+++ linux-next/fs/direct-io.c 2012-08-05 16:24:48.411465136 +0800
@@ -1062,6 +1062,7 @@ do_blockdev_direct_IO(int rw, struct kio
unsigned long user_addr;
size_t bytes;
struct buffer_head map_bh = { 0, };
+ struct blk_plug plug;
if (rw & WRITE)
rw = WRITE_ODIRECT;
@@ -1177,6 +1178,8 @@ do_blockdev_direct_IO(int rw, struct kio
PAGE_SIZE - user_addr / PAGE_SIZE);
}
+ blk_start_plug(&plug);
+
for (seg = 0; seg < nr_segs; seg++) {
user_addr = (unsigned long)iov[seg].iov_base;
sdio.size += bytes = iov[seg].iov_len;
@@ -1235,6 +1238,8 @@ do_blockdev_direct_IO(int rw, struct kio
if (sdio.bio)
dio_bio_submit(dio, &sdio);
+ blk_finish_plug(&plug);
+
/*
* It is possible that, we return short IO due to end of file.
* In that case, we need to release all the pages we got hold on.
--
Re: Re: scsi-mq + open-iscsi support patches..?
Hi all， I used target-pending/scsi-mq to test intel ahci，but it can't work.I modify some places, now it can work.But still met hung task or oops. Who can receive my patch for that? Thanks! Jianpeng Ma >On Sat, 2013-11-02 at 16:10 +, Jayamohan Kallickal wrote: > > > >> >> On a related note, some iscsi vendor has been hitting a crash with >> >> your tree. >> >> >I got an email from Jayamohan recently, but the OOPs did not appear to be >> >scsi-mq related.. >> >> I do see the crash on my Ubuntu VM running on VirtualBox but don't see the >> crash on a Standalone system. >> Attaching the screenshot >> > >Hi Jayamohan, > >Care to enable the virtual serial port for this VM in order to grab the >full dmesg output..? > >Also, an idea of what SATA emulation is being used here would be helpful >as well. (eg: lspci -vvv for the ATA device on the running system) > >--nab > >-- >To unsubscribe from this list: send the line "unsubscribe linux-scsi" in >the body of a message to majord...@vger.kernel.org >More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Re: [PATCH 3/3] md/raid5: For stripe with R5_ReadNoMerge, we replace REQ_FLUSH with REQ_NOMERGE.
>On Wed, 30 Oct 2013 10:41:46 +0800 majianpeng wrote: > >> 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/md/raid5.c >> index f8b9068..8e0aeaa 100644 >> --- a/drivers/md/raid5.c >> +++ b/drivers/md/raid5.c >> @@ -772,7 +772,7 @@ static void ops_run_io(struct stripe_head *sh, struct >> stripe_head_state *s) >> bi->bi_sector = (sh->sector >> + rdev->data_offset); >> if (test_bit(R5_ReadNoMerge, &sh->dev[i].flags)) >> -bi->bi_rw |= REQ_FLUSH; >> +bi->bi_rw |= REQ_NOMERGE; >> >> bi->bi_vcnt = 1; >> bi->bi_io_vec[0].bv_len = STRIPE_SIZE; > >Thanks. > >I've queued this up pending Jens' acceptance of the other patches. > >By the way, your patch file has a carriage return as well as a newline at the >end of each line. >"patch" can still handle them but "git am" doesn't like it. > >If you are able to send the patches as plain text with just a NEWLINE at the >end of each line, rather than base64 encodes with carriage returns, it would >help a little. > Sorroy for that, i'll pay attention to this. Thanks! Jianpeng Ma >Thanks, >NeilBrown >
