From: Byungchul Park <byungchul.park@lge.com> To: Theodore Ts'o <tytso@mit.edu> Cc: torvalds@linux-foundation.org, damien.lemoal@opensource.wdc.com, linux-ide@vger.kernel.org, adilger.kernel@dilger.ca, linux-ext4@vger.kernel.org, mingo@redhat.com, linux-kernel@vger.kernel.org, peterz@infradead.org, will@kernel.org, tglx@linutronix.de, rostedt@goodmis.org, joel@joelfernandes.org, sashal@kernel.org, daniel.vetter@ffwll.ch, chris@chris-wilson.co.uk, duyuyang@gmail.com, johannes.berg@intel.com, tj@kernel.org, willy@infradead.org, david@fromorbit.com, amir73il@gmail.com, bfields@fieldses.org, gregkh@linuxfoundation.org, kernel-team@lge.com, linux-mm@kvack.org, akpm@linux-foundation.org, mhocko@kernel.org, minchan@kernel.org, hannes@cmpxchg.org, vdavydov.dev@gmail.com, sj@kernel.org, jglisse@redhat.com, dennis@kernel.org, cl@linux.com, penberg@kernel.org, rientjes@google.com, vbabka@suse.cz, ngupta@vflare.org, linux-block@vger.kernel.org, paolo.valente@linaro.org, josef@toxicpanda.com, linux-fsdevel@vger.kernel.org, viro@zeniv.linux.org.uk, jack@suse.cz, jack@suse.com, jlayton@kernel.org, dan.j.williams@intel.com, hch@infradead.org, djwong@kernel.org, dri-devel@lists.freedesktop.org, airlied@linux.ie, rodrigosiqueiramelo@gmail.com, melissa.srw@gmail.com, hamohammed.sa@gmail.com, 42.hyeyoo@gmail.com Subject: Re: [PATCH RFC v6 00/21] DEPT(Dependency Tracker) Date: Tue, 10 May 2022 09:32:13 +0900 [thread overview] Message-ID: <20220510003213.GD6047@X58A-UD3R> (raw) In-Reply-To: <YnmVgVQ7usoXnJ1N@mit.edu> On Mon, May 09, 2022 at 06:28:17PM -0400, Theodore Ts'o wrote: > Oh, one other problem with DEPT --- it's SLOW --- the overhead is > enormous. Using kvm-xfstests[1] running "kvm-xfstests smoke", here > are some sample times: Yes, right. DEPT has never been optimized. It rather turns on CONFIG_LOCKDEP and even CONFIG_PROVE_LOCKING when CONFIG_DEPT gets on because of porting issue. I have no choice but to rely on those to develop DEPT out of tree. Of course, that's what I don't like. Plus, for now, I'm focusing on removing false positives. Once it's considered settled down, I will work on performance optimizaition. But it should still keep relying on Lockdep CONFIGs and adding additional overhead on it until DEPT can be developed in the tree. > LOCKDEP DEPT > Time to first test 49 seconds 602 seconds > ext4/001 2 s 22 s > ext4/003 2 s 8 s > ext4/005 0 s 7 s > ext4/020 1 s 8 s > ext4/021 11 s 17 s > ext4/023 0 s 83 s > generic/001 4 s 76 s > generic/002 0 s 11 s > generic/003 10 s 19 s > > There are some large variations; in some cases, some xfstests take 10x > as much time or more to run. In fact, when I first started the > kvm-xfstests run with DEPT, I thought something had hung and that > tests would never start. (In fact, with gce-xfstests the default > watchdog "something has gone terribly wrong with the kexec" had fired, > and I didn't get any test results using gce-xfstests at all. If DEPT > goes in without any optimizations, I'm going to have to adjust the > watchdogs timers for gce-xfstests.) Thank you for informing it. I will go for the optimization as well. > The bottom line is that at the moment, between the false positives, > and the significant overhead imposed by DEPT, I would suggest that if > DEPT ever does go in, that it should be possible to disable DEPT and > only use the existing CONFIG_PROVE_LOCKING version of LOCKDEP, just > because DEPT is S - L - O - W. > > [1] https://github.com/tytso/xfstests-bld/blob/master/Documentation/kvm-quickstart.md > > - Ted > > P.S. Darrick and I both have disabled using LOCKDEP by default > because it slows down ext4 -g auto testing by a factor 2, and xfs -g > auto testing by a factor of 3. So the fact that DEPT is a factor of > 2x to 10x or more slower than LOCKDEP when running various xfstests > tests should be a real concern. DEPT is tracking way more objects than Lockdep so it's inevitable to be slower, but let me try to make it have the similar performance to Lockdep. Byungchul
WARNING: multiple messages have this Message-ID (diff)
From: Byungchul Park <byungchul.park@lge.com> To: Theodore Ts'o <tytso@mit.edu> Cc: hamohammed.sa@gmail.com, jack@suse.cz, peterz@infradead.org, daniel.vetter@ffwll.ch, amir73il@gmail.com, david@fromorbit.com, dri-devel@lists.freedesktop.org, chris@chris-wilson.co.uk, bfields@fieldses.org, linux-ide@vger.kernel.org, adilger.kernel@dilger.ca, joel@joelfernandes.org, 42.hyeyoo@gmail.com, cl@linux.com, will@kernel.org, duyuyang@gmail.com, sashal@kernel.org, paolo.valente@linaro.org, damien.lemoal@opensource.wdc.com, willy@infradead.org, hch@infradead.org, airlied@linux.ie, mingo@redhat.com, djwong@kernel.org, vdavydov.dev@gmail.com, rientjes@google.com, dennis@kernel.org, linux-ext4@vger.kernel.org, linux-mm@kvack.org, ngupta@vflare.org, johannes.berg@intel.com, jack@suse.com, dan.j.williams@intel.com, josef@toxicpanda.com, rostedt@goodmis.org, linux-block@vger.kernel.org, linux-fsdevel@vger.kernel.org, jglisse@redhat.com, viro@zeniv.linux.org.uk, tglx@linutronix.de, mhocko@kernel.org, vbabka@suse.cz, melissa.srw@gmail.com, sj@kernel.org, rodrigosiqueiramelo@gmail.com, kernel-team@lge.com, gregkh@linuxfoundation.org, jlayton@kernel.org, linux-kernel@vger.kernel.org, penberg@kernel.org, minchan@kernel.org, hannes@cmpxchg.org, tj@kernel.org, akpm@linux-foundation.org, torvalds@linux-foundation.org Subject: Re: [PATCH RFC v6 00/21] DEPT(Dependency Tracker) Date: Tue, 10 May 2022 09:32:13 +0900 [thread overview] Message-ID: <20220510003213.GD6047@X58A-UD3R> (raw) In-Reply-To: <YnmVgVQ7usoXnJ1N@mit.edu> On Mon, May 09, 2022 at 06:28:17PM -0400, Theodore Ts'o wrote: > Oh, one other problem with DEPT --- it's SLOW --- the overhead is > enormous. Using kvm-xfstests[1] running "kvm-xfstests smoke", here > are some sample times: Yes, right. DEPT has never been optimized. It rather turns on CONFIG_LOCKDEP and even CONFIG_PROVE_LOCKING when CONFIG_DEPT gets on because of porting issue. I have no choice but to rely on those to develop DEPT out of tree. Of course, that's what I don't like. Plus, for now, I'm focusing on removing false positives. Once it's considered settled down, I will work on performance optimizaition. But it should still keep relying on Lockdep CONFIGs and adding additional overhead on it until DEPT can be developed in the tree. > LOCKDEP DEPT > Time to first test 49 seconds 602 seconds > ext4/001 2 s 22 s > ext4/003 2 s 8 s > ext4/005 0 s 7 s > ext4/020 1 s 8 s > ext4/021 11 s 17 s > ext4/023 0 s 83 s > generic/001 4 s 76 s > generic/002 0 s 11 s > generic/003 10 s 19 s > > There are some large variations; in some cases, some xfstests take 10x > as much time or more to run. In fact, when I first started the > kvm-xfstests run with DEPT, I thought something had hung and that > tests would never start. (In fact, with gce-xfstests the default > watchdog "something has gone terribly wrong with the kexec" had fired, > and I didn't get any test results using gce-xfstests at all. If DEPT > goes in without any optimizations, I'm going to have to adjust the > watchdogs timers for gce-xfstests.) Thank you for informing it. I will go for the optimization as well. > The bottom line is that at the moment, between the false positives, > and the significant overhead imposed by DEPT, I would suggest that if > DEPT ever does go in, that it should be possible to disable DEPT and > only use the existing CONFIG_PROVE_LOCKING version of LOCKDEP, just > because DEPT is S - L - O - W. > > [1] https://github.com/tytso/xfstests-bld/blob/master/Documentation/kvm-quickstart.md > > - Ted > > P.S. Darrick and I both have disabled using LOCKDEP by default > because it slows down ext4 -g auto testing by a factor 2, and xfs -g > auto testing by a factor of 3. So the fact that DEPT is a factor of > 2x to 10x or more slower than LOCKDEP when running various xfstests > tests should be a real concern. DEPT is tracking way more objects than Lockdep so it's inevitable to be slower, but let me try to make it have the similar performance to Lockdep. Byungchul
next prev parent reply other threads:[~2022-05-10 0:33 UTC|newest] Thread overview: 105+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-05-04 8:17 [PATCH RFC v6 00/21] DEPT(Dependency Tracker) Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 01/21] llist: Move llist_{head,node} definition to types.h Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 01/21] llist: Move llist_{head, node} " Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 02/21] dept: Implement Dept(Dependency Tracker) Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 13:29 ` kernel test robot 2022-05-21 3:24 ` Hyeonggon Yoo 2022-05-21 3:24 ` Hyeonggon Yoo 2022-05-04 8:17 ` [PATCH RFC v6 03/21] dept: Apply Dept to spinlock Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 04/21] dept: Apply Dept to mutex families Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 05/21] dept: Apply Dept to rwlock Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 06/21] dept: Apply Dept to wait_for_completion()/complete() Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 07/21] dept: Apply Dept to seqlock Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-21 5:25 ` Hyeonggon Yoo 2022-05-21 5:25 ` Hyeonggon Yoo 2022-05-24 6:00 ` Byungchul Park 2022-05-24 6:00 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 08/21] dept: Apply Dept to rwsem Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 09/21] dept: Add proc knobs to show stats and dependency graph Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 10/21] dept: Introduce split map concept and new APIs for them Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 11/21] dept: Apply Dept to wait/event of PG_{locked,writeback} Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 11/21] dept: Apply Dept to wait/event of PG_{locked, writeback} Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 12/21] dept: Apply SDT to swait Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 13/21] dept: Apply SDT to wait(waitqueue) Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 14/21] locking/lockdep, cpu/hotplus: Use a weaker annotation in AP thread Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 15/21] dept: Distinguish each syscall context from another Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 16/21] dept: Distinguish each work " Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 11:23 ` Sergey Shtylyov 2022-05-04 11:23 ` Sergey Shtylyov 2022-05-04 8:17 ` [PATCH RFC v6 17/21] dept: Disable Dept within the wait_bit layer by default Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 18/21] dept: Disable Dept on struct crypto_larval's completion for now Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 19/21] dept: Differentiate onstack maps from others of different tasks in class Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 20/21] dept: Do not add dependencies between events within scheduler and sleeps Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 8:17 ` [PATCH RFC v6 21/21] dept: Unstage wait when tagging a normal sleep wait Byungchul Park 2022-05-04 8:17 ` Byungchul Park 2022-05-04 18:17 ` [PATCH RFC v6 00/21] DEPT(Dependency Tracker) Linus Torvalds 2022-05-04 18:17 ` Linus Torvalds 2022-05-06 0:11 ` Byungchul Park 2022-05-06 0:11 ` Byungchul Park 2022-05-07 7:20 ` Hyeonggon Yoo 2022-05-07 7:20 ` Hyeonggon Yoo 2022-05-09 0:16 ` Byungchul Park 2022-05-09 0:16 ` Byungchul Park 2022-05-09 20:47 ` Steven Rostedt 2022-05-09 20:47 ` Steven Rostedt 2022-05-09 23:38 ` Byungchul Park 2022-05-09 23:38 ` Byungchul Park 2022-05-10 14:12 ` Steven Rostedt 2022-05-10 14:12 ` Steven Rostedt 2022-05-10 23:26 ` Byungchul Park 2022-05-10 23:26 ` Byungchul Park 2022-05-10 11:18 ` Hyeonggon Yoo 2022-05-10 11:18 ` Hyeonggon Yoo 2022-05-10 23:39 ` Byungchul Park 2022-05-10 23:39 ` Byungchul Park 2022-05-11 10:04 ` Hyeonggon Yoo 2022-05-11 10:04 ` Hyeonggon Yoo 2022-05-19 10:11 ` Catalin Marinas 2022-05-19 10:11 ` Catalin Marinas 2022-05-23 2:43 ` Byungchul Park 2022-05-23 2:43 ` Byungchul Park 2022-05-09 1:22 ` Byungchul Park 2022-05-09 1:22 ` Byungchul Park 2022-05-09 21:05 ` Theodore Ts'o 2022-05-09 21:05 ` Theodore Ts'o 2022-05-09 22:28 ` Theodore Ts'o 2022-05-09 22:28 ` Theodore Ts'o 2022-05-10 0:32 ` Byungchul Park [this message] 2022-05-10 0:32 ` Byungchul Park 2022-05-10 1:32 ` Theodore Ts'o 2022-05-10 1:32 ` Theodore Ts'o 2022-05-10 5:37 ` Byungchul Park 2022-05-10 5:37 ` Byungchul Park 2022-05-11 1:16 ` Byungchul Park 2022-05-11 1:16 ` Byungchul Park 2022-05-12 5:25 ` [REPORT] syscall reboot + umh + firmware fallback Byungchul Park 2022-05-12 5:25 ` Byungchul Park 2022-05-12 9:15 ` Tejun Heo 2022-05-12 9:15 ` Tejun Heo 2022-05-12 11:18 ` Byungchul Park 2022-05-12 11:18 ` Byungchul Park 2022-05-12 13:56 ` Theodore Ts'o 2022-05-12 13:56 ` Theodore Ts'o 2022-05-23 1:10 ` Byungchul Park 2022-05-23 1:10 ` Byungchul Park 2022-05-12 16:41 ` Tejun Heo 2022-05-12 16:41 ` Tejun Heo
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=20220510003213.GD6047@X58A-UD3R \ --to=byungchul.park@lge.com \ --cc=42.hyeyoo@gmail.com \ --cc=adilger.kernel@dilger.ca \ --cc=airlied@linux.ie \ --cc=akpm@linux-foundation.org \ --cc=amir73il@gmail.com \ --cc=bfields@fieldses.org \ --cc=chris@chris-wilson.co.uk \ --cc=cl@linux.com \ --cc=damien.lemoal@opensource.wdc.com \ --cc=dan.j.williams@intel.com \ --cc=daniel.vetter@ffwll.ch \ --cc=david@fromorbit.com \ --cc=dennis@kernel.org \ --cc=djwong@kernel.org \ --cc=dri-devel@lists.freedesktop.org \ --cc=duyuyang@gmail.com \ --cc=gregkh@linuxfoundation.org \ --cc=hamohammed.sa@gmail.com \ --cc=hannes@cmpxchg.org \ --cc=hch@infradead.org \ --cc=jack@suse.com \ --cc=jack@suse.cz \ --cc=jglisse@redhat.com \ --cc=jlayton@kernel.org \ --cc=joel@joelfernandes.org \ --cc=johannes.berg@intel.com \ --cc=josef@toxicpanda.com \ --cc=kernel-team@lge.com \ --cc=linux-block@vger.kernel.org \ --cc=linux-ext4@vger.kernel.org \ --cc=linux-fsdevel@vger.kernel.org \ --cc=linux-ide@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=melissa.srw@gmail.com \ --cc=mhocko@kernel.org \ --cc=minchan@kernel.org \ --cc=mingo@redhat.com \ --cc=ngupta@vflare.org \ --cc=paolo.valente@linaro.org \ --cc=penberg@kernel.org \ --cc=peterz@infradead.org \ --cc=rientjes@google.com \ --cc=rodrigosiqueiramelo@gmail.com \ --cc=rostedt@goodmis.org \ --cc=sashal@kernel.org \ --cc=sj@kernel.org \ --cc=tglx@linutronix.de \ --cc=tj@kernel.org \ --cc=torvalds@linux-foundation.org \ --cc=tytso@mit.edu \ --cc=vbabka@suse.cz \ --cc=vdavydov.dev@gmail.com \ --cc=viro@zeniv.linux.org.uk \ --cc=will@kernel.org \ --cc=willy@infradead.org \ /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: linkBe 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.