Archive-Team
/
zfs
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases 103 Wiki Activity

Call commit callbacks from the tail of the list 

Our zfs backed Lustre MDT had soft lockups while under heavy metadata
workloads while handling transaction callbacks from osd_zfs.

The problem is zfs is not taking advantage of the fast path in
Lustre's trans callback handling, where Lustre will skip the calls
to ptlrpc_commit_replies() when it already saw a higher transaction
number.

This patch corrects this, it also has a positive impact on metadata
performance on Lustre with osd_zfs, plus some cleanup in the headers.

A similar issue for ext4/ldiskfs is described on:
https://jira.hpdd.intel.com/browse/LU-6527

Reviewed-by: Olaf Faaland <faaland1@llnl.gov>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Li Dongyang <dongyang.li@anu.edu.au>
Closes #6986
This commit is contained in:
lidongyang 2017-12-23 05:19:51 +11:00 committed by Brian Behlendorf
parent 44b61ea506
commit 823d48bfb1
3 changed files with 6 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
include/sys/dmu.h
View File

@ -748,11 +748,16 @@ void dmu_tx_mark_netfree(dmu_tx_t *tx);
* to stable storage and will also be called if the dmu_tx is aborted. * to stable storage and will also be called if the dmu_tx is aborted.
* If there is any error which prevents the transaction from being committed to * If there is any error which prevents the transaction from being committed to
* disk, the callback will be called with a value of error != 0. * disk, the callback will be called with a value of error != 0.
*
* When multiple callbacks are registered to the transaction, the callbacks
* will be called in reverse order to let Lustre, the only user of commit
* callback currently, take the fast path of its commit callback handling.
*/ */
typedef void dmu_tx_callback_func_t(void *dcb_data, int error); typedef void dmu_tx_callback_func_t(void *dcb_data, int error);
void dmu_tx_callback_register(dmu_tx_t *tx, dmu_tx_callback_func_t *dcb_func, void dmu_tx_callback_register(dmu_tx_t *tx, dmu_tx_callback_func_t *dcb_func,
void *dcb_data); void *dcb_data);
void dmu_tx_do_callbacks(list_t *cb_list, int error);
/* /*
* Free up the data blocks for a defined range of a file. If size is * Free up the data blocks for a defined range of a file. If size is

4
include/sys/dmu_tx.h
View File

@ -145,10 +145,6 @@ uint64_t dmu_tx_get_txg(dmu_tx_t *tx);
struct dsl_pool *dmu_tx_pool(dmu_tx_t *tx); struct dsl_pool *dmu_tx_pool(dmu_tx_t *tx);
void dmu_tx_wait(dmu_tx_t *tx); void dmu_tx_wait(dmu_tx_t *tx);
void dmu_tx_callback_register(dmu_tx_t *tx, dmu_tx_callback_func_t *dcb_func,
void *dcb_data);
void dmu_tx_do_callbacks(list_t *cb_list, int error);
/* /*
* These routines are defined in dmu_spa.h, and are called by the SPA. * These routines are defined in dmu_spa.h, and are called by the SPA.
*/ */

2
module/zfs/dmu_tx.c
View File

@ -1197,7 +1197,7 @@ dmu_tx_do_callbacks(list_t *cb_list, int error)
{ {
dmu_tx_callback_t *dcb; dmu_tx_callback_t *dcb;
while ((dcb = list_head(cb_list)) != NULL) { while ((dcb = list_tail(cb_list)) != NULL) {
list_remove(cb_list, dcb); list_remove(cb_list, dcb);
dcb->dcb_func(dcb->dcb_data, error); dcb->dcb_func(dcb->dcb_data, error);
kmem_free(dcb, sizeof (dmu_tx_callback_t)); kmem_free(dcb, sizeof (dmu_tx_callback_t));