* Re: [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance [not found] <88.C4.11914.9D4A9225@epcpsbge6.samsung.com> @ 2013-09-10 0:52 ` Jaegeuk Kim 2013-09-11 3:13 ` Gu Zheng 2013-09-11 5:37 ` Gu Zheng 0 siblings, 2 replies; 7+ messages in thread From: Jaegeuk Kim @ 2013-09-10 0:52 UTC (permalink / raw) To: chao2.yu; +Cc: shu.tan, linux-fsdevel, linux-kernel, linux-f2fs-devel Hi, At first, thank you for the report and please follow the email writing rules. :) Anyway, I agree to the below issue. One thing that I can think of is that we don't need to use the spin_lock, since we don't care about the exact lock number, but just need to get any not-collided number. So, how about removing the spin_lock? And how about using a random number? Thanks, 2013-09-06 (금), 09:48 +0000, Chao Yu: > Hi Kim: > > I think there is a performance problem: when all sbi->fs_lock is > holded, > > then all other threads may get the same next_lock value from > sbi->next_lock_num in function mutex_lock_op, > > and wait to get the same lock at position fs_lock[next_lock], it > unbalance the fs_lock usage. > > It may lost performance when we do the multithread test. > > > > Here is the patch to fix this problem: > > > > Signed-off-by: Yu Chao <chao2.yu@samsung.com> > > diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h > > old mode 100644 > > new mode 100755 > > index 467d42d..983bb45 > > --- a/fs/f2fs/f2fs.h > > +++ b/fs/f2fs/f2fs.h > > @@ -371,6 +371,7 @@ struct f2fs_sb_info { > > struct mutex fs_lock[NR_GLOBAL_LOCKS]; /* blocking FS > operations */ > > struct mutex node_write; /* locking node writes > */ > > struct mutex writepages; /* mutex for > writepages() */ > > + spinlock_t spin_lock; /* lock for > next_lock_num */ > > unsigned char next_lock_num; /* round-robin global > locks */ > > int por_doing; /* recovery is doing > or not */ > > int on_build_free_nids; /* build_free_nids is > doing */ > > @@ -533,15 +534,19 @@ static inline void mutex_unlock_all(struct > f2fs_sb_info *sbi) > > > > static inline int mutex_lock_op(struct f2fs_sb_info *sbi) > > { > > - unsigned char next_lock = sbi->next_lock_num % > NR_GLOBAL_LOCKS; > > + unsigned char next_lock; > > int i = 0; > > > > for (; i < NR_GLOBAL_LOCKS; i++) > > if (mutex_trylock(&sbi->fs_lock[i])) > > return i; > > > > - mutex_lock(&sbi->fs_lock[next_lock]); > > + spin_lock(&sbi->spin_lock); > > + next_lock = sbi->next_lock_num % NR_GLOBAL_LOCKS; > > sbi->next_lock_num++; > > + spin_unlock(&sbi->spin_lock); > > + > > + mutex_lock(&sbi->fs_lock[next_lock]); > > return next_lock; > > } > > > > diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c > > old mode 100644 > > new mode 100755 > > index 75c7dc3..4f27596 > > --- a/fs/f2fs/super.c > > +++ b/fs/f2fs/super.c > > @@ -657,6 +657,7 @@ static int f2fs_fill_super(struct super_block *sb, > void *data, int silent) > > mutex_init(&sbi->cp_mutex); > > for (i = 0; i < NR_GLOBAL_LOCKS; i++) > > mutex_init(&sbi->fs_lock[i]); > > + spin_lock_init(&sbi->spin_lock); > > mutex_init(&sbi->node_write); > > sbi->por_doing = 0; > > spin_lock_init(&sbi->stat_lock); > > (END) > > > > > > -- Jaegeuk Kim Samsung ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance 2013-09-10 0:52 ` [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance Jaegeuk Kim @ 2013-09-11 3:13 ` Gu Zheng 2013-09-11 5:37 ` Gu Zheng 1 sibling, 0 replies; 7+ messages in thread From: Gu Zheng @ 2013-09-11 3:13 UTC (permalink / raw) To: jaegeuk.kim Cc: chao2.yu, shu.tan, linux-fsdevel, linux-kernel, linux-f2fs-devel Hi Jaegeuk, On 09/10/2013 08:52 AM, Jaegeuk Kim wrote: > Hi, > > At first, thank you for the report and please follow the email writing > rules. :) > > Anyway, I agree to the below issue. > One thing that I can think of is that we don't need to use the > spin_lock, since we don't care about the exact lock number, but just > need to get any not-collided number. Agree, but if all the locks are held, IMO, we need to balance the following threads to wait for each not-collided number lock, though complete balance is unreachable. > > So, how about removing the spin_lock? Yeah, in this case, spin_lock is a bit heavy cost. > And how about using a random number? Now NR_GLOBAL_LOCKS is 8, it seems that random can not offer an balance number as we expected. Regards, Gu > Thanks, > > 2013-09-06 (금), 09:48 +0000, Chao Yu: >> Hi Kim: >> >> I think there is a performance problem: when all sbi->fs_lock is >> holded, >> >> then all other threads may get the same next_lock value from >> sbi->next_lock_num in function mutex_lock_op, >> >> and wait to get the same lock at position fs_lock[next_lock], it >> unbalance the fs_lock usage. >> >> It may lost performance when we do the multithread test. >> >> >> >> Here is the patch to fix this problem: >> >> >> >> Signed-off-by: Yu Chao <chao2.yu@samsung.com> >> >> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h >> >> old mode 100644 >> >> new mode 100755 >> >> index 467d42d..983bb45 >> >> --- a/fs/f2fs/f2fs.h >> >> +++ b/fs/f2fs/f2fs.h >> >> @@ -371,6 +371,7 @@ struct f2fs_sb_info { >> >> struct mutex fs_lock[NR_GLOBAL_LOCKS]; /* blocking FS >> operations */ >> >> struct mutex node_write; /* locking node writes >> */ >> >> struct mutex writepages; /* mutex for >> writepages() */ >> >> + spinlock_t spin_lock; /* lock for >> next_lock_num */ >> >> unsigned char next_lock_num; /* round-robin global >> locks */ >> >> int por_doing; /* recovery is doing >> or not */ >> >> int on_build_free_nids; /* build_free_nids is >> doing */ >> >> @@ -533,15 +534,19 @@ static inline void mutex_unlock_all(struct >> f2fs_sb_info *sbi) >> >> >> >> static inline int mutex_lock_op(struct f2fs_sb_info *sbi) >> >> { >> >> - unsigned char next_lock = sbi->next_lock_num % >> NR_GLOBAL_LOCKS; >> >> + unsigned char next_lock; >> >> int i = 0; >> >> >> >> for (; i < NR_GLOBAL_LOCKS; i++) >> >> if (mutex_trylock(&sbi->fs_lock[i])) >> >> return i; >> >> >> >> - mutex_lock(&sbi->fs_lock[next_lock]); >> >> + spin_lock(&sbi->spin_lock); >> >> + next_lock = sbi->next_lock_num % NR_GLOBAL_LOCKS; >> >> sbi->next_lock_num++; >> >> + spin_unlock(&sbi->spin_lock); >> >> + >> >> + mutex_lock(&sbi->fs_lock[next_lock]); >> >> return next_lock; >> >> } >> >> >> >> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c >> >> old mode 100644 >> >> new mode 100755 >> >> index 75c7dc3..4f27596 >> >> --- a/fs/f2fs/super.c >> >> +++ b/fs/f2fs/super.c >> >> @@ -657,6 +657,7 @@ static int f2fs_fill_super(struct super_block *sb, >> void *data, int silent) >> >> mutex_init(&sbi->cp_mutex); >> >> for (i = 0; i < NR_GLOBAL_LOCKS; i++) >> >> mutex_init(&sbi->fs_lock[i]); >> >> + spin_lock_init(&sbi->spin_lock); >> >> mutex_init(&sbi->node_write); >> >> sbi->por_doing = 0; >> >> spin_lock_init(&sbi->stat_lock); >> >> (END) >> >> >> >> >> >> > ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance 2013-09-10 0:52 ` [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance Jaegeuk Kim 2013-09-11 3:13 ` Gu Zheng @ 2013-09-11 5:37 ` Gu Zheng 2013-09-11 13:22 ` [f2fs-dev] [PATCH] " Kim Jaegeuk 2013-09-12 2:40 ` [f2fs-dev][PATCH] " 俞超 1 sibling, 2 replies; 7+ messages in thread From: Gu Zheng @ 2013-09-11 5:37 UTC (permalink / raw) To: jaegeuk.kim Cc: chao2.yu, shu.tan, linux-fsdevel, linux-kernel, linux-f2fs-devel Hi Jaegeuk, Chao, On 09/10/2013 08:52 AM, Jaegeuk Kim wrote: > Hi, > > At first, thank you for the report and please follow the email writing > rules. :) > > Anyway, I agree to the below issue. > One thing that I can think of is that we don't need to use the > spin_lock, since we don't care about the exact lock number, but just > need to get any not-collided number. IMHO, just moving sbi->next_lock_num++ before mutex_lock(&sbi->fs_lock[next_lock]) can avoid unbalance issue mostly. IMO, the case two or more threads increase sbi->next_lock_num in the same time is really very very little. If you think it is not rigorous, change next_lock_num to atomic one can fix it. What's your opinion? Regards, Gu > > So, how about removing the spin_lock? > And how about using a random number? > Thanks, > > 2013-09-06 (금), 09:48 +0000, Chao Yu: >> Hi Kim: >> >> I think there is a performance problem: when all sbi->fs_lock is >> holded, >> >> then all other threads may get the same next_lock value from >> sbi->next_lock_num in function mutex_lock_op, >> >> and wait to get the same lock at position fs_lock[next_lock], it >> unbalance the fs_lock usage. >> >> It may lost performance when we do the multithread test. >> >> >> >> Here is the patch to fix this problem: >> >> >> >> Signed-off-by: Yu Chao <chao2.yu@samsung.com> >> >> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h >> >> old mode 100644 >> >> new mode 100755 >> >> index 467d42d..983bb45 >> >> --- a/fs/f2fs/f2fs.h >> >> +++ b/fs/f2fs/f2fs.h >> >> @@ -371,6 +371,7 @@ struct f2fs_sb_info { >> >> struct mutex fs_lock[NR_GLOBAL_LOCKS]; /* blocking FS >> operations */ >> >> struct mutex node_write; /* locking node writes >> */ >> >> struct mutex writepages; /* mutex for >> writepages() */ >> >> + spinlock_t spin_lock; /* lock for >> next_lock_num */ >> >> unsigned char next_lock_num; /* round-robin global >> locks */ >> >> int por_doing; /* recovery is doing >> or not */ >> >> int on_build_free_nids; /* build_free_nids is >> doing */ >> >> @@ -533,15 +534,19 @@ static inline void mutex_unlock_all(struct >> f2fs_sb_info *sbi) >> >> >> >> static inline int mutex_lock_op(struct f2fs_sb_info *sbi) >> >> { >> >> - unsigned char next_lock = sbi->next_lock_num % >> NR_GLOBAL_LOCKS; >> >> + unsigned char next_lock; >> >> int i = 0; >> >> >> >> for (; i < NR_GLOBAL_LOCKS; i++) >> >> if (mutex_trylock(&sbi->fs_lock[i])) >> >> return i; >> >> >> >> - mutex_lock(&sbi->fs_lock[next_lock]); >> >> + spin_lock(&sbi->spin_lock); >> >> + next_lock = sbi->next_lock_num % NR_GLOBAL_LOCKS; >> >> sbi->next_lock_num++; >> >> + spin_unlock(&sbi->spin_lock); >> >> + >> >> + mutex_lock(&sbi->fs_lock[next_lock]); >> >> return next_lock; >> >> } >> >> >> >> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c >> >> old mode 100644 >> >> new mode 100755 >> >> index 75c7dc3..4f27596 >> >> --- a/fs/f2fs/super.c >> >> +++ b/fs/f2fs/super.c >> >> @@ -657,6 +657,7 @@ static int f2fs_fill_super(struct super_block *sb, >> void *data, int silent) >> >> mutex_init(&sbi->cp_mutex); >> >> for (i = 0; i < NR_GLOBAL_LOCKS; i++) >> >> mutex_init(&sbi->fs_lock[i]); >> >> + spin_lock_init(&sbi->spin_lock); >> >> mutex_init(&sbi->node_write); >> >> sbi->por_doing = 0; >> >> spin_lock_init(&sbi->stat_lock); >> >> (END) >> >> >> >> >> >> > ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [f2fs-dev] [PATCH] f2fs: optimize fs_lock for better performance 2013-09-11 5:37 ` Gu Zheng @ 2013-09-11 13:22 ` Kim Jaegeuk 2013-09-12 2:40 ` [f2fs-dev][PATCH] " 俞超 1 sibling, 0 replies; 7+ messages in thread From: Kim Jaegeuk @ 2013-09-11 13:22 UTC (permalink / raw) To: Gu Zheng Cc: Jaegeuk Kim, linux-fsdevel, 谭姝, linux-kernel, linux-f2fs-devel Hi Gu, 2013/9/11 Gu Zheng <guz.fnst@cn.fujitsu.com>: > Hi Jaegeuk, Chao, > > On 09/10/2013 08:52 AM, Jaegeuk Kim wrote: > >> Hi, >> >> At first, thank you for the report and please follow the email writing >> rules. :) >> >> Anyway, I agree to the below issue. >> One thing that I can think of is that we don't need to use the >> spin_lock, since we don't care about the exact lock number, but just >> need to get any not-collided number. > > IMHO, just moving sbi->next_lock_num++ before mutex_lock(&sbi->fs_lock[next_lock]) > can avoid unbalance issue mostly. > IMO, the case two or more threads increase sbi->next_lock_num in the same time is > really very very little. If you think it is not rigorous, change next_lock_num to > atomic one can fix it. > What's your opinion? As your opinion, I think it is enough to replace it with simple sbi->next_lock_num++. Thanks, > > Regards, > Gu > >> >> So, how about removing the spin_lock? >> And how about using a random number? > >> Thanks, >> >> 2013-09-06 (금), 09:48 +0000, Chao Yu: >>> Hi Kim: >>> >>> I think there is a performance problem: when all sbi->fs_lock is >>> holded, >>> >>> then all other threads may get the same next_lock value from >>> sbi->next_lock_num in function mutex_lock_op, >>> >>> and wait to get the same lock at position fs_lock[next_lock], it >>> unbalance the fs_lock usage. >>> >>> It may lost performance when we do the multithread test. >>> >>> >>> >>> Here is the patch to fix this problem: >>> >>> >>> >>> Signed-off-by: Yu Chao <chao2.yu@samsung.com> >>> >>> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h >>> >>> old mode 100644 >>> >>> new mode 100755 >>> >>> index 467d42d..983bb45 >>> >>> --- a/fs/f2fs/f2fs.h >>> >>> +++ b/fs/f2fs/f2fs.h >>> >>> @@ -371,6 +371,7 @@ struct f2fs_sb_info { >>> >>> struct mutex fs_lock[NR_GLOBAL_LOCKS]; /* blocking FS >>> operations */ >>> >>> struct mutex node_write; /* locking node writes >>> */ >>> >>> struct mutex writepages; /* mutex for >>> writepages() */ >>> >>> + spinlock_t spin_lock; /* lock for >>> next_lock_num */ >>> >>> unsigned char next_lock_num; /* round-robin global >>> locks */ >>> >>> int por_doing; /* recovery is doing >>> or not */ >>> >>> int on_build_free_nids; /* build_free_nids is >>> doing */ >>> >>> @@ -533,15 +534,19 @@ static inline void mutex_unlock_all(struct >>> f2fs_sb_info *sbi) >>> >>> >>> >>> static inline int mutex_lock_op(struct f2fs_sb_info *sbi) >>> >>> { >>> >>> - unsigned char next_lock = sbi->next_lock_num % >>> NR_GLOBAL_LOCKS; >>> >>> + unsigned char next_lock; >>> >>> int i = 0; >>> >>> >>> >>> for (; i < NR_GLOBAL_LOCKS; i++) >>> >>> if (mutex_trylock(&sbi->fs_lock[i])) >>> >>> return i; >>> >>> >>> >>> - mutex_lock(&sbi->fs_lock[next_lock]); >>> >>> + spin_lock(&sbi->spin_lock); >>> >>> + next_lock = sbi->next_lock_num % NR_GLOBAL_LOCKS; >>> >>> sbi->next_lock_num++; >>> >>> + spin_unlock(&sbi->spin_lock); >>> >>> + >>> >>> + mutex_lock(&sbi->fs_lock[next_lock]); >>> >>> return next_lock; >>> >>> } >>> >>> >>> >>> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c >>> >>> old mode 100644 >>> >>> new mode 100755 >>> >>> index 75c7dc3..4f27596 >>> >>> --- a/fs/f2fs/super.c >>> >>> +++ b/fs/f2fs/super.c >>> >>> @@ -657,6 +657,7 @@ static int f2fs_fill_super(struct super_block *sb, >>> void *data, int silent) >>> >>> mutex_init(&sbi->cp_mutex); >>> >>> for (i = 0; i < NR_GLOBAL_LOCKS; i++) >>> >>> mutex_init(&sbi->fs_lock[i]); >>> >>> + spin_lock_init(&sbi->spin_lock); >>> >>> mutex_init(&sbi->node_write); >>> >>> sbi->por_doing = 0; >>> >>> spin_lock_init(&sbi->stat_lock); >>> >>> (END) >>> >>> >>> >>> >>> >>> >> > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. Consolidate legacy IT systems to a single system of record for IT > 2. Standardize and globalize service processes across IT > 3. Implement zero-touch automation to replace manual, redundant tasks > http://pubads.g.doubleclick.net/gampad/clk?id=51271111&iu=/4140/ostg.clktrk > _______________________________________________ > Linux-f2fs-devel mailing list > Linux-f2fs-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel ^ permalink raw reply [flat|nested] 7+ messages in thread
* RE: [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance 2013-09-11 5:37 ` Gu Zheng 2013-09-11 13:22 ` [f2fs-dev] [PATCH] " Kim Jaegeuk @ 2013-09-12 2:40 ` 俞超 2013-09-12 3:17 ` [f2fs-dev][PATCH V2] " Gu Zheng 2013-09-12 3:18 ` [f2fs-dev][PATCH] " Gu Zheng 1 sibling, 2 replies; 7+ messages in thread From: 俞超 @ 2013-09-12 2:40 UTC (permalink / raw) To: 'Gu Zheng', jaegeuk.kim Cc: shu.tan, linux-fsdevel, linux-kernel, linux-f2fs-devel Hi Gu > -----Original Message----- > From: Gu Zheng [mailto:guz.fnst@cn.fujitsu.com] > Sent: Wednesday, September 11, 2013 1:38 PM > To: jaegeuk.kim@samsung.com > Cc: chao2.yu@samsung.com; shu.tan@samsung.com; > linux-fsdevel@vger.kernel.org; linux-kernel@vger.kernel.org; > linux-f2fs-devel@lists.sourceforge.net > Subject: Re: [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance > > Hi Jaegeuk, Chao, > > On 09/10/2013 08:52 AM, Jaegeuk Kim wrote: > > > Hi, > > > > At first, thank you for the report and please follow the email writing > > rules. :) > > > > Anyway, I agree to the below issue. > > One thing that I can think of is that we don't need to use the > > spin_lock, since we don't care about the exact lock number, but just > > need to get any not-collided number. > > IMHO, just moving sbi->next_lock_num++ before > mutex_lock(&sbi->fs_lock[next_lock]) > can avoid unbalance issue mostly. > IMO, the case two or more threads increase sbi->next_lock_num in the same > time is really very very little. If you think it is not rigorous, change > next_lock_num to atomic one can fix it. > What's your opinion? > > Regards, > Gu I did the test sbi->next_lock_num++ compare with the atomic one, And I found performance of them is almost the same under a small number thread racing. So as your and Kim's opinion, it's enough to use "sbi->next_lock_num++" to fix this issue. Thanks for the advice. > > > > > So, how about removing the spin_lock? > > And how about using a random number? > > > Thanks, > > > > 2013-09-06 (금), 09:48 +0000, Chao Yu: > >> Hi Kim: > >> > >> I think there is a performance problem: when all sbi->fs_lock is > >> holded, > >> > >> then all other threads may get the same next_lock value from > >> sbi->next_lock_num in function mutex_lock_op, > >> > >> and wait to get the same lock at position fs_lock[next_lock], it > >> unbalance the fs_lock usage. > >> > >> It may lost performance when we do the multithread test. > >> > >> > >> > >> Here is the patch to fix this problem: > >> > >> > >> > >> Signed-off-by: Yu Chao <chao2.yu@samsung.com> > >> > >> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h > >> > >> old mode 100644 > >> > >> new mode 100755 > >> > >> index 467d42d..983bb45 > >> > >> --- a/fs/f2fs/f2fs.h > >> > >> +++ b/fs/f2fs/f2fs.h > >> > >> @@ -371,6 +371,7 @@ struct f2fs_sb_info { > >> > >> struct mutex fs_lock[NR_GLOBAL_LOCKS]; /* blocking FS > >> operations */ > >> > >> struct mutex node_write; /* locking node > writes > >> */ > >> > >> struct mutex writepages; /* mutex for > >> writepages() */ > >> > >> + spinlock_t spin_lock; /* lock for > >> next_lock_num */ > >> > >> unsigned char next_lock_num; /* round-robin > global > >> locks */ > >> > >> int por_doing; /* recovery is doing > >> or not */ > >> > >> int on_build_free_nids; /* build_free_nids is > >> doing */ > >> > >> @@ -533,15 +534,19 @@ static inline void mutex_unlock_all(struct > >> f2fs_sb_info *sbi) > >> > >> > >> > >> static inline int mutex_lock_op(struct f2fs_sb_info *sbi) > >> > >> { > >> > >> - unsigned char next_lock = sbi->next_lock_num % > >> NR_GLOBAL_LOCKS; > >> > >> + unsigned char next_lock; > >> > >> int i = 0; > >> > >> > >> > >> for (; i < NR_GLOBAL_LOCKS; i++) > >> > >> if (mutex_trylock(&sbi->fs_lock[i])) > >> > >> return i; > >> > >> > >> > >> - mutex_lock(&sbi->fs_lock[next_lock]); > >> > >> + spin_lock(&sbi->spin_lock); > >> > >> + next_lock = sbi->next_lock_num % NR_GLOBAL_LOCKS; > >> > >> sbi->next_lock_num++; > >> > >> + spin_unlock(&sbi->spin_lock); > >> > >> + > >> > >> + mutex_lock(&sbi->fs_lock[next_lock]); > >> > >> return next_lock; > >> > >> } > >> > >> > >> > >> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c > >> > >> old mode 100644 > >> > >> new mode 100755 > >> > >> index 75c7dc3..4f27596 > >> > >> --- a/fs/f2fs/super.c > >> > >> +++ b/fs/f2fs/super.c > >> > >> @@ -657,6 +657,7 @@ static int f2fs_fill_super(struct super_block > >> *sb, void *data, int silent) > >> > >> mutex_init(&sbi->cp_mutex); > >> > >> for (i = 0; i < NR_GLOBAL_LOCKS; i++) > >> > >> mutex_init(&sbi->fs_lock[i]); > >> > >> + spin_lock_init(&sbi->spin_lock); > >> > >> mutex_init(&sbi->node_write); > >> > >> sbi->por_doing = 0; > >> > >> spin_lock_init(&sbi->stat_lock); > >> > >> (END) > >> > >> > >> > >> > >> > >> > > > > > = ^ permalink raw reply [flat|nested] 7+ messages in thread
* [f2fs-dev][PATCH V2] f2fs: optimize fs_lock for better performance 2013-09-12 2:40 ` [f2fs-dev][PATCH] " 俞超 @ 2013-09-12 3:17 ` Gu Zheng 2013-09-12 3:18 ` [f2fs-dev][PATCH] " Gu Zheng 1 sibling, 0 replies; 7+ messages in thread From: Gu Zheng @ 2013-09-12 3:17 UTC (permalink / raw) To: jaegeuk.kim Cc: 俞超, shu.tan, linux-fsdevel, linux-kernel, linux-f2fs-devel From: Yu Chao <chao2.yu@samsung.com> There is a performance problem: when all sbi->fs_lock are holded, then all the following threads may get the same next_lock value from sbi->next_lock_num in function mutex_lock_op, and wait for the same lock(fs_lock[next_lock]), it may cause performance reduce. So we move the sbi->next_lock_num++ before getting lock, this will average the following threads if all sbi->fs_lock are holded. v1-->v2: Drop the needless spin_lock as Jaegeuk suggested. Suggested-by: Jaegeuk Kim <jaegeuk.kim@samsung.com> Signed-off-by: Yu Chao <chao2.yu@samsung.com> Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com> --- fs/f2fs/f2fs.h | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 608f0df..7fd99d8 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -544,15 +544,15 @@ static inline void mutex_unlock_all(struct f2fs_sb_info *sbi) static inline int mutex_lock_op(struct f2fs_sb_info *sbi) { - unsigned char next_lock = sbi->next_lock_num % NR_GLOBAL_LOCKS; + unsigned char next_lock; int i = 0; for (; i < NR_GLOBAL_LOCKS; i++) if (mutex_trylock(&sbi->fs_lock[i])) return i; + next_lock = sbi->next_lock_num++ % NR_GLOBAL_LOCKS; mutex_lock(&sbi->fs_lock[next_lock]); - sbi->next_lock_num++; return next_lock; } -- 1.7.7 ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance 2013-09-12 2:40 ` [f2fs-dev][PATCH] " 俞超 2013-09-12 3:17 ` [f2fs-dev][PATCH V2] " Gu Zheng @ 2013-09-12 3:18 ` Gu Zheng 1 sibling, 0 replies; 7+ messages in thread From: Gu Zheng @ 2013-09-12 3:18 UTC (permalink / raw) To: 俞超 Cc: jaegeuk.kim, shu.tan, linux-fsdevel, linux-kernel, linux-f2fs-devel Hi Chao, On 09/12/2013 10:40 AM, 俞超 wrote: > Hi Gu > >> -----Original Message----- >> From: Gu Zheng [mailto:guz.fnst@cn.fujitsu.com] >> Sent: Wednesday, September 11, 2013 1:38 PM >> To: jaegeuk.kim@samsung.com >> Cc: chao2.yu@samsung.com; shu.tan@samsung.com; >> linux-fsdevel@vger.kernel.org; linux-kernel@vger.kernel.org; >> linux-f2fs-devel@lists.sourceforge.net >> Subject: Re: [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance >> >> Hi Jaegeuk, Chao, >> >> On 09/10/2013 08:52 AM, Jaegeuk Kim wrote: >> >>> Hi, >>> >>> At first, thank you for the report and please follow the email writing >>> rules. :) >>> >>> Anyway, I agree to the below issue. >>> One thing that I can think of is that we don't need to use the >>> spin_lock, since we don't care about the exact lock number, but just >>> need to get any not-collided number. >> >> IMHO, just moving sbi->next_lock_num++ before >> mutex_lock(&sbi->fs_lock[next_lock]) >> can avoid unbalance issue mostly. >> IMO, the case two or more threads increase sbi->next_lock_num in the same >> time is really very very little. If you think it is not rigorous, change >> next_lock_num to atomic one can fix it. >> What's your opinion? >> >> Regards, >> Gu > > I did the test sbi->next_lock_num++ compare with the atomic one, > And I found performance of them is almost the same under a small number thread racing. > So as your and Kim's opinion, it's enough to use "sbi->next_lock_num++" to fix this issue. Good, but it seems that your replay patch is out of format, and it's hard for Jaegeuk to merge. I'll format it, see the following thread. Thanks, Gu > > Thanks for the advice. >> >>> >>> So, how about removing the spin_lock? >>> And how about using a random number? >> >>> Thanks, >>> >>> 2013-09-06 (금), 09:48 +0000, Chao Yu: >>>> Hi Kim: >>>> >>>> I think there is a performance problem: when all sbi->fs_lock is >>>> holded, >>>> >>>> then all other threads may get the same next_lock value from >>>> sbi->next_lock_num in function mutex_lock_op, >>>> >>>> and wait to get the same lock at position fs_lock[next_lock], it >>>> unbalance the fs_lock usage. >>>> >>>> It may lost performance when we do the multithread test. >>>> >>>> >>>> >>>> Here is the patch to fix this problem: >>>> >>>> >>>> >>>> Signed-off-by: Yu Chao <chao2.yu@samsung.com> >>>> >>>> diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h >>>> >>>> old mode 100644 >>>> >>>> new mode 100755 >>>> >>>> index 467d42d..983bb45 >>>> >>>> --- a/fs/f2fs/f2fs.h >>>> >>>> +++ b/fs/f2fs/f2fs.h >>>> >>>> @@ -371,6 +371,7 @@ struct f2fs_sb_info { >>>> >>>> struct mutex fs_lock[NR_GLOBAL_LOCKS]; /* blocking FS >>>> operations */ >>>> >>>> struct mutex node_write; /* locking node >> writes >>>> */ >>>> >>>> struct mutex writepages; /* mutex for >>>> writepages() */ >>>> >>>> + spinlock_t spin_lock; /* lock for >>>> next_lock_num */ >>>> >>>> unsigned char next_lock_num; /* round-robin >> global >>>> locks */ >>>> >>>> int por_doing; /* recovery is doing >>>> or not */ >>>> >>>> int on_build_free_nids; /* build_free_nids is >>>> doing */ >>>> >>>> @@ -533,15 +534,19 @@ static inline void mutex_unlock_all(struct >>>> f2fs_sb_info *sbi) >>>> >>>> >>>> >>>> static inline int mutex_lock_op(struct f2fs_sb_info *sbi) >>>> >>>> { >>>> >>>> - unsigned char next_lock = sbi->next_lock_num % >>>> NR_GLOBAL_LOCKS; >>>> >>>> + unsigned char next_lock; >>>> >>>> int i = 0; >>>> >>>> >>>> >>>> for (; i < NR_GLOBAL_LOCKS; i++) >>>> >>>> if (mutex_trylock(&sbi->fs_lock[i])) >>>> >>>> return i; >>>> >>>> >>>> >>>> - mutex_lock(&sbi->fs_lock[next_lock]); >>>> >>>> + spin_lock(&sbi->spin_lock); >>>> >>>> + next_lock = sbi->next_lock_num % NR_GLOBAL_LOCKS; >>>> >>>> sbi->next_lock_num++; >>>> >>>> + spin_unlock(&sbi->spin_lock); >>>> >>>> + >>>> >>>> + mutex_lock(&sbi->fs_lock[next_lock]); >>>> >>>> return next_lock; >>>> >>>> } >>>> >>>> >>>> >>>> diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c >>>> >>>> old mode 100644 >>>> >>>> new mode 100755 >>>> >>>> index 75c7dc3..4f27596 >>>> >>>> --- a/fs/f2fs/super.c >>>> >>>> +++ b/fs/f2fs/super.c >>>> >>>> @@ -657,6 +657,7 @@ static int f2fs_fill_super(struct super_block >>>> *sb, void *data, int silent) >>>> >>>> mutex_init(&sbi->cp_mutex); >>>> >>>> for (i = 0; i < NR_GLOBAL_LOCKS; i++) >>>> >>>> mutex_init(&sbi->fs_lock[i]); >>>> >>>> + spin_lock_init(&sbi->spin_lock); >>>> >>>> mutex_init(&sbi->node_write); >>>> >>>> sbi->por_doing = 0; >>>> >>>> spin_lock_init(&sbi->stat_lock); >>>> >>>> (END) >>>> >>>> >>>> >>>> >>>> >>>> >>> >> >> >> = > > ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2013-09-12 3:22 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <88.C4.11914.9D4A9225@epcpsbge6.samsung.com>
2013-09-10 0:52 ` [f2fs-dev][PATCH] f2fs: optimize fs_lock for better performance Jaegeuk Kim
2013-09-11 3:13 ` Gu Zheng
2013-09-11 5:37 ` Gu Zheng
2013-09-11 13:22 ` [f2fs-dev] [PATCH] " Kim Jaegeuk
2013-09-12 2:40 ` [f2fs-dev][PATCH] " 俞超
2013-09-12 3:17 ` [f2fs-dev][PATCH V2] " Gu Zheng
2013-09-12 3:18 ` [f2fs-dev][PATCH] " Gu Zheng
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®