From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 4AC4214F78 for ; Tue, 27 Feb 2024 01:49:21 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=170.10.129.124 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708998564; cv=none; b=Yl8SSXk131WK4SbF7FyQTy7QYJ9y1u5IhpO2Ob6xjOYPNg7027ri8Bf1IrZA6HNSgx79ngFLvpPkEX7fIFBPbTSe3G4ZubPStLz5BDlDu/jfLubN5Ke9DM1WOfw6GUFua4O0EyIaH3LaPWAk6UE9GBPQZsE7PzcOL/+6TfSWnXE= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708998564; c=relaxed/simple; bh=N7ab6x8fum7CaAYD2aWdOfw+6FjhM0DtD/VHd24PAjo=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=V6RIkTRzJ8uUmlrx8woAHfojF+j+K6WYlGXEfnu4zLaxsPxMTz7MbblxGVWJUfqVK4HtKxH9McMTMmXvkuVRPSjFCAFsj43h2DE/xX/Ybg1rh81pOuox6xTY/+e/PB5/AyKlKT73FzVjjwS1IDlHyp0mh0wdlm+YjWDm2sMhI8g= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com; spf=pass smtp.mailfrom=redhat.com; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b=UbbdYBXp; arc=none smtp.client-ip=170.10.129.124 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=redhat.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="UbbdYBXp" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1708998561; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=73ZVlNyHBo8gy7vjcvbH56F5zAGTQV9p32FgIrdTcSo=; b=UbbdYBXptbD5gdbYWkKM+PmTFWK1cRux4H/z7SE5ceLZBUeuPxPUybIXghMv+4FRHqEH/t NW1/aCjOAev7PueE1u3uevM9sbFg8XolOEoTPWSLX2gZdKctnEWrwjf7Ms4EgIZtFE6gSJ oRILtKJTBjFlvmXmQkfODWFFdc47CM8= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-195-N5xNEH42NxOuHwZuAMyoaA-1; Mon, 26 Feb 2024 20:49:19 -0500 X-MC-Unique: N5xNEH42NxOuHwZuAMyoaA-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 6BDCF85A597 for ; Tue, 27 Feb 2024 01:49:19 +0000 (UTC) Received: from fs-i40c-03.fast.eng.rdu2.dc.redhat.com (fs-i40c-03.fast.eng.rdu2.dc.redhat.com [10.6.23.54]) by smtp.corp.redhat.com (Postfix) with ESMTP id 629922166B5E; Tue, 27 Feb 2024 01:49:19 +0000 (UTC) From: Alexander Aring To: teigland@redhat.com Cc: gfs2@lists.linux.dev, aahringo@redhat.com Subject: [PATCHv3 v6.8-rc6 14/18] dlm: make requestqueue handling non sleepable Date: Mon, 26 Feb 2024 20:49:05 -0500 Message-ID: <20240227014909.93945-15-aahringo@redhat.com> In-Reply-To: <20240227014909.93945-1-aahringo@redhat.com> References: <20240227014909.93945-1-aahringo@redhat.com> Precedence: bulk X-Mailing-List: gfs2@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.4.1 on 10.11.54.6 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="US-ASCII"; x-default=true This patch removes the ls_requestqueue_wait and convert the ls_requestqueue_mutex to a rw lock. Instead of calling wait_event() in dlm processing which waits until all messages are processed and allow new message processing after recovering is done, this patch is using a bitflag to signal when a message should be saved for future or not. When recovery processes all saved messages we will clear this bit again and allow new messages to processed directly. Signed-off-by: Alexander Aring --- fs/dlm/dlm_internal.h | 5 ++--- fs/dlm/lock.c | 16 ++++++++++++++-- fs/dlm/lockspace.c | 4 +--- fs/dlm/member.c | 5 +++++ fs/dlm/requestqueue.c | 41 ++++++++--------------------------------- 5 files changed, 30 insertions(+), 41 deletions(-) diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h index 13225b28fa98..4cb1f38067d3 100644 --- a/fs/dlm/dlm_internal.h +++ b/fs/dlm/dlm_internal.h @@ -625,9 +625,7 @@ struct dlm_ls { struct rw_semaphore ls_in_recovery; /* block local requests */ struct rw_semaphore ls_recv_active; /* block dlm_recv */ struct list_head ls_requestqueue;/* queue remote requests */ - atomic_t ls_requestqueue_cnt; - wait_queue_head_t ls_requestqueue_wait; - struct mutex ls_requestqueue_mutex; + rwlock_t ls_requestqueue_lock; struct dlm_rcom *ls_recover_buf; int ls_recover_nodeid; /* for debugging */ unsigned int ls_recover_locks_in; /* for log info */ @@ -687,6 +685,7 @@ struct dlm_ls { #define LSFL_UEVENT_WAIT 7 #define LSFL_CB_DELAY 9 #define LSFL_NODIR 10 +#define LSFL_RECV_MSG_BLOCKED 11 /* much of this is just saving user space pointers associated with the lock that we pass back to the user lib with an ast */ diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 267a18bcf2c6..10cebb418931 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -4749,20 +4749,32 @@ static void _receive_message(struct dlm_ls *ls, const struct dlm_message *ms, static void dlm_receive_message(struct dlm_ls *ls, const struct dlm_message *ms, int nodeid) { - if (dlm_locking_stopped(ls)) { +try_again: + read_lock(&ls->ls_requestqueue_lock); + if (test_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags)) { /* If we were a member of this lockspace, left, and rejoined, other nodes may still be sending us messages from the lockspace generation before we left. */ if (WARN_ON_ONCE(!ls->ls_generation)) { + read_unlock(&ls->ls_requestqueue_lock); log_limit(ls, "receive %d from %d ignore old gen", le32_to_cpu(ms->m_type), nodeid); return; } + read_unlock(&ls->ls_requestqueue_lock); + write_lock(&ls->ls_requestqueue_lock); + /* recheck because we hold writelock now */ + if (!test_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags)) { + write_unlock_bh(&ls->ls_requestqueue_lock); + goto try_again; + } + dlm_add_requestqueue(ls, nodeid, ms); + write_unlock(&ls->ls_requestqueue_lock); } else { - dlm_wait_requestqueue(ls); _receive_message(ls, ms, 0); + read_unlock(&ls->ls_requestqueue_lock); } } diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 42866190af38..0df0f07fa092 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -554,9 +554,7 @@ static int new_lockspace(const char *name, const char *cluster, init_rwsem(&ls->ls_in_recovery); init_rwsem(&ls->ls_recv_active); INIT_LIST_HEAD(&ls->ls_requestqueue); - atomic_set(&ls->ls_requestqueue_cnt, 0); - init_waitqueue_head(&ls->ls_requestqueue_wait); - mutex_init(&ls->ls_requestqueue_mutex); + rwlock_init(&ls->ls_requestqueue_lock); spin_lock_init(&ls->ls_clear_proc_locks); /* Due backwards compatibility with 3.1 we need to use maximum diff --git a/fs/dlm/member.c b/fs/dlm/member.c index be7909ead71b..707cebcdc533 100644 --- a/fs/dlm/member.c +++ b/fs/dlm/member.c @@ -642,6 +642,11 @@ int dlm_ls_stop(struct dlm_ls *ls) set_bit(LSFL_RECOVER_STOP, &ls->ls_flags); new = test_and_clear_bit(LSFL_RUNNING, &ls->ls_flags); ls->ls_recover_seq++; + + /* activate requestqueue and stop processing */ + write_lock(&ls->ls_requestqueue_lock); + set_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags); + write_unlock(&ls->ls_requestqueue_lock); spin_unlock(&ls->ls_recover_lock); /* diff --git a/fs/dlm/requestqueue.c b/fs/dlm/requestqueue.c index c05940afd063..9b646026df46 100644 --- a/fs/dlm/requestqueue.c +++ b/fs/dlm/requestqueue.c @@ -48,10 +48,7 @@ void dlm_add_requestqueue(struct dlm_ls *ls, int nodeid, memcpy(&e->request, ms, sizeof(*ms)); memcpy(&e->request.m_extra, ms->m_extra, length); - atomic_inc(&ls->ls_requestqueue_cnt); - mutex_lock(&ls->ls_requestqueue_mutex); list_add_tail(&e->list, &ls->ls_requestqueue); - mutex_unlock(&ls->ls_requestqueue_mutex); } /* @@ -71,16 +68,14 @@ int dlm_process_requestqueue(struct dlm_ls *ls) struct dlm_message *ms; int error = 0; - mutex_lock(&ls->ls_requestqueue_mutex); - + write_lock(&ls->ls_requestqueue_lock); for (;;) { if (list_empty(&ls->ls_requestqueue)) { - mutex_unlock(&ls->ls_requestqueue_mutex); + clear_bit(LSFL_RECV_MSG_BLOCKED, &ls->ls_flags); error = 0; break; } - e = list_entry(ls->ls_requestqueue.next, struct rq_entry, list); - mutex_unlock(&ls->ls_requestqueue_mutex); + e = list_first_entry(&ls->ls_requestqueue, struct rq_entry, list); ms = &e->request; @@ -93,41 +88,23 @@ int dlm_process_requestqueue(struct dlm_ls *ls) e->recover_seq); dlm_receive_message_saved(ls, &e->request, e->recover_seq); - - mutex_lock(&ls->ls_requestqueue_mutex); list_del(&e->list); - if (atomic_dec_and_test(&ls->ls_requestqueue_cnt)) - wake_up(&ls->ls_requestqueue_wait); kfree(e); if (dlm_locking_stopped(ls)) { log_debug(ls, "process_requestqueue abort running"); - mutex_unlock(&ls->ls_requestqueue_mutex); error = -EINTR; break; } + write_unlock(&ls->ls_requestqueue_lock); schedule(); + write_lock(&ls->ls_requestqueue_lock); } + write_unlock(&ls->ls_requestqueue_lock); return error; } -/* - * After recovery is done, locking is resumed and dlm_recoverd takes all the - * saved requests and processes them as they would have been by dlm_recv. At - * the same time, dlm_recv will start receiving new requests from remote nodes. - * We want to delay dlm_recv processing new requests until dlm_recoverd has - * finished processing the old saved requests. We don't check for locking - * stopped here because dlm_ls_stop won't stop locking until it's suspended us - * (dlm_recv). - */ - -void dlm_wait_requestqueue(struct dlm_ls *ls) -{ - wait_event(ls->ls_requestqueue_wait, - atomic_read(&ls->ls_requestqueue_cnt) == 0); -} - static int purge_request(struct dlm_ls *ls, struct dlm_message *ms, int nodeid) { __le32 type = ms->m_type; @@ -158,17 +135,15 @@ void dlm_purge_requestqueue(struct dlm_ls *ls) struct dlm_message *ms; struct rq_entry *e, *safe; - mutex_lock(&ls->ls_requestqueue_mutex); + write_lock(&ls->ls_requestqueue_lock); list_for_each_entry_safe(e, safe, &ls->ls_requestqueue, list) { ms = &e->request; if (purge_request(ls, ms, e->nodeid)) { list_del(&e->list); - if (atomic_dec_and_test(&ls->ls_requestqueue_cnt)) - wake_up(&ls->ls_requestqueue_wait); kfree(e); } } - mutex_unlock(&ls->ls_requestqueue_mutex); + write_unlock(&ls->ls_requestqueue_lock); } -- 2.43.0