All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Tang Junhui <tang.junhui@zte.com.cn>,
	Coly Li <colyli@suse.de>, Jens Axboe <axboe@kernel.dk>
Subject: [PATCH 4.4 61/66] bcache: correct cache_dirty_target in __update_writeback_rate()
Date: Sun, 24 Sep 2017 22:31:56 +0200	[thread overview]
Message-ID: <20170924202923.073549828@linuxfoundation.org> (raw)
In-Reply-To: <20170924202920.581603259@linuxfoundation.org>

4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Tang Junhui <tang.junhui@zte.com.cn>

commit a8394090a9129b40f9d90dcb7f4a49d60c727ca6 upstream.

__update_write_rate() uses a Proportion-Differentiation Controller
algorithm to control writeback rate. A dirty target number is used in
this PD controller to control writeback rate. A larger target number
will make the writeback rate smaller, on the versus, a smaller target
number will make the writeback rate larger.

bcache uses the following steps to calculate the target number,
1) cache_sectors = all-buckets-of-cache-set * buckets-size
2) cache_dirty_target = cache_sectors * cached-device-writeback_percent
3) target = cache_dirty_target *
(sectors-of-cached-device/sectors-of-all-cached-devices-of-this-cache-set)

The calculation at step 1) for cache_sectors is incorrect, which does
not consider dirty blocks occupied by flash only volume.

A flash only volume can be took as a bcache device without cached
device. All data sectors allocated for it are persistent on cache device
and marked dirty, they are not touched by bcache writeback and garbage
collection code. So data blocks of flash only volume should be ignore
when calculating cache_sectors of cache set.

Current code does not subtract dirty sectors of flash only volume, which
results a larger target number from the above 3 steps. And in sequence
the cache device's writeback rate is smaller then a correct value,
writeback speed is slower on all cached devices.

This patch fixes the incorrect slower writeback rate by subtracting
dirty sectors of flash only volumes in __update_writeback_rate().

(Commit log composed by Coly Li to pass checkpatch.pl checking)

Signed-off-by: Tang Junhui <tang.junhui@zte.com.cn>
Reviewed-by: Coly Li <colyli@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/md/bcache/writeback.c |    3 ++-
 drivers/md/bcache/writeback.h |   19 +++++++++++++++++++
 2 files changed, 21 insertions(+), 1 deletion(-)

--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -21,7 +21,8 @@
 static void __update_writeback_rate(struct cached_dev *dc)
 {
 	struct cache_set *c = dc->disk.c;
-	uint64_t cache_sectors = c->nbuckets * c->sb.bucket_size;
+	uint64_t cache_sectors = c->nbuckets * c->sb.bucket_size -
+				bcache_flash_devs_sectors_dirty(c);
 	uint64_t cache_dirty_target =
 		div_u64(cache_sectors * dc->writeback_percent, 100);
 
--- a/drivers/md/bcache/writeback.h
+++ b/drivers/md/bcache/writeback.h
@@ -14,6 +14,25 @@ static inline uint64_t bcache_dev_sector
 	return ret;
 }
 
+static inline uint64_t  bcache_flash_devs_sectors_dirty(struct cache_set *c)
+{
+	uint64_t i, ret = 0;
+
+	mutex_lock(&bch_register_lock);
+
+	for (i = 0; i < c->nr_uuids; i++) {
+		struct bcache_device *d = c->devices[i];
+
+		if (!d || !UUID_FLASH_ONLY(&c->uuids[i]))
+			continue;
+	   ret += bcache_dev_sectors_dirty(d);
+	}
+
+	mutex_unlock(&bch_register_lock);
+
+	return ret;
+}
+
 static inline unsigned offset_to_stripe(struct bcache_device *d,
 					uint64_t offset)
 {

  parent reply	other threads:[~2017-09-24 20:36 UTC|newest]

Thread overview: 69+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-24 20:30 [PATCH 4.4 00/66] 4.4.89-stable review Greg Kroah-Hartman
2017-09-24 20:30 ` [PATCH 4.4 01/66] ipv6: accept 64k - 1 packet length in ip6_find_1stfragopt() Greg Kroah-Hartman
2017-09-24 20:30 ` [PATCH 4.4 02/66] ipv6: add rcu grace period before freeing fib6_node Greg Kroah-Hartman
2017-09-24 20:30 ` [PATCH 4.4 03/66] ipv6: fix sparse warning on rt6i_node Greg Kroah-Hartman
2017-09-24 20:30 ` [PATCH 4.4 04/66] qlge: avoid memcpy buffer overflow Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 05/66] Revert "net: phy: Correctly process PHY_HALTED in phy_stop_machine()" Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 06/66] tcp: initialize rcv_mss to TCP_MIN_MSS instead of 0 Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 07/66] Revert "net: use lib/percpu_counter API for fragmentation mem accounting" Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 08/66] Revert "net: fix percpu memory leaks" Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 09/66] gianfar: Fix Tx flow control deactivation Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 10/66] ipv6: fix memory leak with multiple tables during netns destruction Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 11/66] ipv6: fix typo in fib6_net_exit() Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 12/66] f2fs: check hot_data for roll-forward recovery Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 13/66] x86/fsgsbase/64: Report FSBASE and GSBASE correctly in core dumps Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 14/66] md/raid5: release/flush io in raid5_do_work() Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 15/66] nfsd: Fix general protection fault in release_lock_stateid() Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 16/66] mm: prevent double decrease of nr_reserved_highatomic Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 17/66] tty: improve tty_insert_flip_char() fast path Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 18/66] tty: improve tty_insert_flip_char() slow path Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 19/66] tty: fix __tty_insert_flip_char regression Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 20/66] Input: i8042 - add Gigabyte P57 to the keyboard reset table Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 21/66] MIPS: math-emu: <MAX|MAXA|MIN|MINA>.<D|S>: Fix quiet NaN propagation Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 22/66] MIPS: math-emu: <MAX|MAXA|MIN|MINA>.<D|S>: Fix cases of both inputs zero Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 23/66] MIPS: math-emu: <MAX|MIN>.<D|S>: Fix cases of both inputs negative Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 24/66] MIPS: math-emu: <MAXA|MINA>.<D|S>: Fix cases of input values with opposite signs Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 25/66] MIPS: math-emu: <MAXA|MINA>.<D|S>: Fix cases of both infinite inputs Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 26/66] MIPS: math-emu: MINA.<D|S>: Fix some cases of infinity and zero inputs Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 27/66] [PATCH - RESEND] crypto: AF_ALG - remove SGL terminator indicator when chaining Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 28/66] ext4: fix incorrect quotaoff if the quota feature is enabled Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 29/66] ext4: fix quota inconsistency during orphan cleanup for read-only mounts Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 30/66] powerpc: Fix DAR reporting when alignment handler faults Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 31/66] block: Relax a check in blk_start_queue() Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 32/66] md/bitmap: disable bitmap_resize for file-backed bitmaps Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 33/66] skd: Avoid that module unloading triggers a use-after-free Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 34/66] skd: Submit requests to firmware before triggering the doorbell Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 35/66] scsi: zfcp: fix queuecommand for scsi_eh commands when DIX enabled Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 36/66] scsi: zfcp: add handling for FCP_RESID_OVER to the fcp ingress path Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 37/66] scsi: zfcp: fix capping of unsuccessful GPN_FT SAN response trace records Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 38/66] scsi: zfcp: fix passing fsf_req to SCSI trace on TMF to correlate with HBA Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 39/66] scsi: zfcp: fix missing trace records for early returns in TMF eh handlers Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 40/66] scsi: zfcp: fix payload with full FCP_RSP IU in SCSI trace records Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 41/66] scsi: zfcp: trace HBA FSF response by default on dismiss or timedout late response Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 42/66] scsi: zfcp: trace high part of "new" 64 bit SCSI LUN Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 43/66] scsi: megaraid_sas: Check valid aen class range to avoid kernel panic Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 44/66] scsi: megaraid_sas: Return pended IOCTLs with cmd_status MFI_STAT_WRONG_STATE in case adapter is dead Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 45/66] scsi: storvsc: fix memory leak on ring buffer busy Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 46/66] scsi: sg: remove save_scat_len Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 47/66] scsi: sg: use standard lists for sg_requests Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 48/66] scsi: sg: off by one in sg_ioctl() Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 49/66] scsi: sg: factor out sg_fill_request_table() Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 50/66] scsi: sg: fixup infoleak when using SG_GET_REQUEST_TABLE Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 51/66] scsi: qla2xxx: Fix an integer overflow in sysfs code Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 52/66] ftrace: Fix selftest goto location on error Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 53/66] tracing: Apply trace_clock changes to instance max buffer Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 54/66] ARC: Re-enable MMU upon Machine Check exception Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 55/66] PCI: shpchp: Enable bridge bus mastering if MSI is enabled Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 56/66] media: v4l2-compat-ioctl32: Fix timespec conversion Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 57/66] media: uvcvideo: Prevent heap overflow when accessing mapped controls Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 58/66] bcache: initialize dirty stripes in flash_dev_run() Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 59/66] bcache: Fix leak of bdev reference Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 60/66] bcache: do not subtract sectors_to_gc for bypassed IO Greg Kroah-Hartman
2017-09-24 20:31 ` Greg Kroah-Hartman [this message]
2017-09-24 20:31 ` [PATCH 4.4 62/66] bcache: Correct return value for sysfs attach errors Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 63/66] bcache: fix for gc and write-back race Greg Kroah-Hartman
2017-09-24 20:31 ` [PATCH 4.4 64/66] bcache: fix bch_hprint crash and improve output Greg Kroah-Hartman
2017-09-24 20:32 ` [PATCH 4.4 65/66] ftrace: Fix memleak when unregistering dynamic ops when tracing disabled Greg Kroah-Hartman
2017-09-24 20:32 ` [PATCH 4.4 66/66] mac80211: flush hw_roc_start work before cancelling the ROC Greg Kroah-Hartman
2017-09-25  1:04 ` [PATCH 4.4 00/66] 4.4.89-stable review Guenter Roeck
2017-09-25 23:12 ` Shuah Khan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170924202923.073549828@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=axboe@kernel.dk \
    --cc=colyli@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tang.junhui@zte.com.cn \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.