| Message ID | 20180103140325.63175-4-colyli@suse.de (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 01/03/2018 06:03 AM, Coly Li wrote: > Member devices of struct cache_set is used to reference all attached > bcache devices to this cache set. If it is treated as array of pointers, > size of devices[] is indicated by member nr_uuids of struct cache_set. > > nr_uuids is calculated in drivers/md/super.c:bch_cache_set_alloc(), > bucket_bytes(c) / sizeof(struct uuid_entry) > Bucket size is determined by user space tool "make-bcache", by default it > is 1024 sectors (defined in bcache-tools/make-bcache.c:main()). So default > nr_uuids value is 4096 from the above calculation. > > Every time when bcache code iterates bcache devices of a cache set, all > the 4096 pointers are checked even only 1 bcache device is attached to the > cache set, that's a wast of time and unncessary. > > This patch adds a member devices_max_used to struct cache_set. Its value > is 1 + the maximum used index of devices[] in a cache set. When iterating > all valid bcache devices of a cache set, use c->devices_max_used in > for-loop may reduce a lot of useless checking. > > Personally, my motivation of this patch is not for performance, I use it > in bcache debugging, which helps me to narrow down the scape to check > valid bcached devices of a cache set. Oh, OK. I was going to argue that I didn't think this was worth it, but I can see how it would be nice for inspecting memory and debug code. > > Signed-off-by: Coly Li <colyli@suse.de> Reviewed-by: Michael Lyle <mlyle@lyle.org>
On 01/03/2018 03:03 PM, Coly Li wrote: > Member devices of struct cache_set is used to reference all attached > bcache devices to this cache set. If it is treated as array of pointers, > size of devices[] is indicated by member nr_uuids of struct cache_set. > > nr_uuids is calculated in drivers/md/super.c:bch_cache_set_alloc(), > bucket_bytes(c) / sizeof(struct uuid_entry) > Bucket size is determined by user space tool "make-bcache", by default it > is 1024 sectors (defined in bcache-tools/make-bcache.c:main()). So default > nr_uuids value is 4096 from the above calculation. > > Every time when bcache code iterates bcache devices of a cache set, all > the 4096 pointers are checked even only 1 bcache device is attached to the > cache set, that's a wast of time and unncessary. > > This patch adds a member devices_max_used to struct cache_set. Its value > is 1 + the maximum used index of devices[] in a cache set. When iterating > all valid bcache devices of a cache set, use c->devices_max_used in > for-loop may reduce a lot of useless checking. > > Personally, my motivation of this patch is not for performance, I use it > in bcache debugging, which helps me to narrow down the scape to check > valid bcached devices of a cache set. > > Signed-off-by: Coly Li <colyli@suse.de> > --- > drivers/md/bcache/bcache.h | 1 + > drivers/md/bcache/btree.c | 2 +- > drivers/md/bcache/super.c | 9 ++++++--- > drivers/md/bcache/writeback.h | 2 +- > 4 files changed, 9 insertions(+), 5 deletions(-) > > diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h > index 843877e017e1..83c569942bd0 100644 > --- a/drivers/md/bcache/bcache.h > +++ b/drivers/md/bcache/bcache.h > @@ -488,6 +488,7 @@ struct cache_set { > int caches_loaded; > > struct bcache_device **devices; > + unsigned devices_max_used; > struct list_head cached_devs; > uint64_t cached_dev_sectors; > struct closure caching; > diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c > index 81e8dc3dbe5e..bf0d7978bc3d 100644 > --- a/drivers/md/bcache/btree.c > +++ b/drivers/md/bcache/btree.c > @@ -1678,7 +1678,7 @@ static void bch_btree_gc_finish(struct cache_set *c) > > /* don't reclaim buckets to which writeback keys point */ > rcu_read_lock(); > - for (i = 0; i < c->nr_uuids; i++) { > + for (i = 0; i < c->devices_max_used; i++) { > struct bcache_device *d = c->devices[i]; > struct cached_dev *dc; > struct keybuf_key *w, *n; > diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c > index b4d28928dec5..064efd869017 100644 > --- a/drivers/md/bcache/super.c > +++ b/drivers/md/bcache/super.c > @@ -721,6 +721,9 @@ static void bcache_device_attach(struct bcache_device *d, struct cache_set *c, > d->c = c; > c->devices[id] = d; > > + if (id >= c->devices_max_used) > + c->devices_max_used = id + 1; > + > closure_get(&c->caching); > } > > @@ -1261,7 +1264,7 @@ static int flash_devs_run(struct cache_set *c) > struct uuid_entry *u; > > for (u = c->uuids; > - u < c->uuids + c->nr_uuids && !ret; > + u < c->uuids + c->devices_max_used && !ret; > u++) > if (UUID_FLASH_ONLY(u)) > ret = flash_dev_run(c, u); > @@ -1427,7 +1430,7 @@ static void __cache_set_unregister(struct closure *cl) > > mutex_lock(&bch_register_lock); > > - for (i = 0; i < c->nr_uuids; i++) > + for (i = 0; i < c->devices_max_used; i++) > if (c->devices[i]) { > if (!UUID_FLASH_ONLY(&c->uuids[i]) && > test_bit(CACHE_SET_UNREGISTERING, &c->flags)) { > @@ -1490,7 +1493,7 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb) > c->bucket_bits = ilog2(sb->bucket_size); > c->block_bits = ilog2(sb->block_size); > c->nr_uuids = bucket_bytes(c) / sizeof(struct uuid_entry); > - > + c->devices_max_used = 0; > c->btree_pages = bucket_pages(c); > if (c->btree_pages > BTREE_MAX_PAGES) > c->btree_pages = max_t(int, c->btree_pages / 4, > diff --git a/drivers/md/bcache/writeback.h b/drivers/md/bcache/writeback.h > index a9e3ffb4b03c..1d284f3d0363 100644 > --- a/drivers/md/bcache/writeback.h > +++ b/drivers/md/bcache/writeback.h > @@ -21,7 +21,7 @@ static inline uint64_t bcache_flash_devs_sectors_dirty(struct cache_set *c) > > mutex_lock(&bch_register_lock); > > - for (i = 0; i < c->nr_uuids; i++) { > + for (i = 0; i < c->devices_max_used; i++) { > struct bcache_device *d = c->devices[i]; > > if (!d || !UUID_FLASH_ONLY(&c->uuids[i])) > Hmm. This just cries out for using IDA ... but maybe for a later patchset. Reviewed-by: Hannes Reinecke <hare@suse.com> Cheers, Hannes
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h index 843877e017e1..83c569942bd0 100644 --- a/drivers/md/bcache/bcache.h +++ b/drivers/md/bcache/bcache.h @@ -488,6 +488,7 @@ struct cache_set { int caches_loaded; struct bcache_device **devices; + unsigned devices_max_used; struct list_head cached_devs; uint64_t cached_dev_sectors; struct closure caching; diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c index 81e8dc3dbe5e..bf0d7978bc3d 100644 --- a/drivers/md/bcache/btree.c +++ b/drivers/md/bcache/btree.c @@ -1678,7 +1678,7 @@ static void bch_btree_gc_finish(struct cache_set *c) /* don't reclaim buckets to which writeback keys point */ rcu_read_lock(); - for (i = 0; i < c->nr_uuids; i++) { + for (i = 0; i < c->devices_max_used; i++) { struct bcache_device *d = c->devices[i]; struct cached_dev *dc; struct keybuf_key *w, *n; diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c index b4d28928dec5..064efd869017 100644 --- a/drivers/md/bcache/super.c +++ b/drivers/md/bcache/super.c @@ -721,6 +721,9 @@ static void bcache_device_attach(struct bcache_device *d, struct cache_set *c, d->c = c; c->devices[id] = d; + if (id >= c->devices_max_used) + c->devices_max_used = id + 1; + closure_get(&c->caching); } @@ -1261,7 +1264,7 @@ static int flash_devs_run(struct cache_set *c) struct uuid_entry *u; for (u = c->uuids; - u < c->uuids + c->nr_uuids && !ret; + u < c->uuids + c->devices_max_used && !ret; u++) if (UUID_FLASH_ONLY(u)) ret = flash_dev_run(c, u); @@ -1427,7 +1430,7 @@ static void __cache_set_unregister(struct closure *cl) mutex_lock(&bch_register_lock); - for (i = 0; i < c->nr_uuids; i++) + for (i = 0; i < c->devices_max_used; i++) if (c->devices[i]) { if (!UUID_FLASH_ONLY(&c->uuids[i]) && test_bit(CACHE_SET_UNREGISTERING, &c->flags)) { @@ -1490,7 +1493,7 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb) c->bucket_bits = ilog2(sb->bucket_size); c->block_bits = ilog2(sb->block_size); c->nr_uuids = bucket_bytes(c) / sizeof(struct uuid_entry); - + c->devices_max_used = 0; c->btree_pages = bucket_pages(c); if (c->btree_pages > BTREE_MAX_PAGES) c->btree_pages = max_t(int, c->btree_pages / 4, diff --git a/drivers/md/bcache/writeback.h b/drivers/md/bcache/writeback.h index a9e3ffb4b03c..1d284f3d0363 100644 --- a/drivers/md/bcache/writeback.h +++ b/drivers/md/bcache/writeback.h @@ -21,7 +21,7 @@ static inline uint64_t bcache_flash_devs_sectors_dirty(struct cache_set *c) mutex_lock(&bch_register_lock); - for (i = 0; i < c->nr_uuids; i++) { + for (i = 0; i < c->devices_max_used; i++) { struct bcache_device *d = c->devices[i]; if (!d || !UUID_FLASH_ONLY(&c->uuids[i]))
Member devices of struct cache_set is used to reference all attached bcache devices to this cache set. If it is treated as array of pointers, size of devices[] is indicated by member nr_uuids of struct cache_set. nr_uuids is calculated in drivers/md/super.c:bch_cache_set_alloc(), bucket_bytes(c) / sizeof(struct uuid_entry) Bucket size is determined by user space tool "make-bcache", by default it is 1024 sectors (defined in bcache-tools/make-bcache.c:main()). So default nr_uuids value is 4096 from the above calculation. Every time when bcache code iterates bcache devices of a cache set, all the 4096 pointers are checked even only 1 bcache device is attached to the cache set, that's a wast of time and unncessary. This patch adds a member devices_max_used to struct cache_set. Its value is 1 + the maximum used index of devices[] in a cache set. When iterating all valid bcache devices of a cache set, use c->devices_max_used in for-loop may reduce a lot of useless checking. Personally, my motivation of this patch is not for performance, I use it in bcache debugging, which helps me to narrow down the scape to check valid bcached devices of a cache set. Signed-off-by: Coly Li <colyli@suse.de> --- drivers/md/bcache/bcache.h | 1 + drivers/md/bcache/btree.c | 2 +- drivers/md/bcache/super.c | 9 ++++++--- drivers/md/bcache/writeback.h | 2 +- 4 files changed, 9 insertions(+), 5 deletions(-)