linux-kernel - [RFC dlm/next 03/12] dlm: add struct net to dlm_new

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240814143414.1877505-4-aahringo@redhat.com>
Date: Wed, 14 Aug 2024 10:34:05 -0400
From: Alexander Aring <aahringo@...hat.com>
To: teigland@...hat.com
Cc: gfs2@...ts.linux.dev,
	song@...nel.org,
	yukuai3@...wei.com,
	agruenba@...hat.com,
	mark@...heh.com,
	jlbec@...lplan.org,
	joseph.qi@...ux.alibaba.com,
	gregkh@...uxfoundation.org,
	rafael@...nel.org,
	akpm@...ux-foundation.org,
	linux-kernel@...r.kernel.org,
	linux-raid@...r.kernel.org,
	ocfs2-devel@...ts.linux.dev,
	lucien.xin@...il.com,
	aahringo@...hat.com
Subject: [RFC dlm/next 03/12] dlm: add struct net to dlm_new_lockspace()

To prepare a namespace separation for each DLM lockspaces context we add
a struct net parameter that the user tells us in which net-namespace
the lockspace should be created. We are using net-namespace here because
a DLM lockspaces context acts like a per cluster node separation and the
created per node sockets need to be separated by their net-namespaces
anyway. It just fits that the DLM lockspaces are also separated by a per
"network entity".

This patch only prepares for such parameter for a functionality that
does not exist yet. It does not have any effect. If there will be
support for such handling the DLM user need to activate it anyway as the
applied parameter for now is the "&init_net" instance that is the
default namespace which we are currently using.

Signed-off-by: Alexander Aring <aahringo@...hat.com>
---
 drivers/md/md-cluster.c | 3 ++-
 fs/dlm/lockspace.c      | 5 +++--
 fs/gfs2/lock_dlm.c      | 6 +++---
 fs/ocfs2/stack_user.c   | 2 +-
 include/linux/dlm.h     | 9 +++++++--
 5 files changed, 16 insertions(+), 9 deletions(-)

diff --git a/drivers/md/md-cluster.c b/drivers/md/md-cluster.c
index 1d0db62f0351..cc1c93370510 100644
--- a/drivers/md/md-cluster.c
+++ b/drivers/md/md-cluster.c
@@ -896,7 +896,8 @@ static int join(struct mddev *mddev, int nodes)
 
 	memset(str, 0, 64);
 	sprintf(str, "%pU", mddev->uuid);
-	ret = dlm_new_lockspace(str, mddev->bitmap_info.cluster_name,
+	ret = dlm_new_lockspace(&init_net, str,
+				mddev->bitmap_info.cluster_name,
 				DLM_LSFL_SOFTIRQ, LVB_SIZE, &md_ls_ops, mddev,
 				&ops_rv, &cinfo->lockspace);
 	if (ret)
diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c
index 00d37125bc44..2dd37a2e718d 100644
--- a/fs/dlm/lockspace.c
+++ b/fs/dlm/lockspace.c
@@ -663,8 +663,9 @@ static int __dlm_new_lockspace(const char *name, const char *cluster,
 	return error;
 }
 
-int dlm_new_lockspace(const char *name, const char *cluster, uint32_t flags,
-		      int lvblen, const struct dlm_lockspace_ops *ops,
+int dlm_new_lockspace(struct net *net, const char *name, const char *cluster,
+		      uint32_t flags, int lvblen,
+		      const struct dlm_lockspace_ops *ops,
 		      void *ops_arg, int *ops_result,
 		      dlm_lockspace_t **lockspace)
 {
diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c
index fa5134df985f..6c5dce57a2ee 100644
--- a/fs/gfs2/lock_dlm.c
+++ b/fs/gfs2/lock_dlm.c
@@ -1328,9 +1328,9 @@ static int gdlm_mount(struct gfs2_sbd *sdp, const char *table)
 	 * create/join lockspace
 	 */
 
-	error = dlm_new_lockspace(fsname, cluster, flags, GDLM_LVB_SIZE,
-				  &gdlm_lockspace_ops, sdp, &ops_result,
-				  &ls->ls_dlm);
+	error = dlm_new_lockspace(&init_net, fsname, cluster, flags,
+				  GDLM_LVB_SIZE, &gdlm_lockspace_ops, sdp,
+				  &ops_result, &ls->ls_dlm);
 	if (error) {
 		fs_err(sdp, "dlm_new_lockspace error %d\n", error);
 		goto fail_free;
diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c
index 77edcd70f72c..23611eba58ef 100644
--- a/fs/ocfs2/stack_user.c
+++ b/fs/ocfs2/stack_user.c
@@ -984,7 +984,7 @@ static int user_cluster_connect(struct ocfs2_cluster_connection *conn)
 	conn->cc_private = lc;
 	lc->oc_type = NO_CONTROLD;
 
-	rc = dlm_new_lockspace(conn->cc_name, conn->cc_cluster_name,
+	rc = dlm_new_lockspace(&init_net, conn->cc_name, conn->cc_cluster_name,
 			       DLM_LSFL_NEWEXCL, DLM_LVB_LEN,
 			       &ocfs2_ls_ops, conn, &ops_rv, &fsdlm);
 	if (rc) {
diff --git a/include/linux/dlm.h b/include/linux/dlm.h
index bacda9898f2b..ecab5c197a7f 100644
--- a/include/linux/dlm.h
+++ b/include/linux/dlm.h
@@ -11,9 +11,9 @@
 #ifndef __DLM_DOT_H__
 #define __DLM_DOT_H__
 
+#include <net/net_namespace.h>
 #include <uapi/linux/dlm.h>
 
-
 struct dlm_slot {
 	int nodeid; /* 1 to MAX_INT */
 	int slot;   /* 1 to MAX_INT */
@@ -43,6 +43,11 @@ struct dlm_lockspace_ops {
  *
  * Create/join a lockspace.
  *
+ * net: the net namespace context pointer where the lockspace belongs to.
+ * DLM lockspaces can be separated according to net namespaces. As DLM
+ * requires networking communication this net namespace can be used to
+ * have a own DLM lockspace on each network entity e.g. a DLM node.
+ *
  * name: lockspace name, null terminated, up to DLM_LOCKSPACE_LEN (not
  *   including terminating null).
  *
@@ -82,7 +87,7 @@ struct dlm_lockspace_ops {
  * lockspace: handle for dlm functions
  */
 
-int dlm_new_lockspace(const char *name, const char *cluster,
+int dlm_new_lockspace(struct net *net, const char *name, const char *cluster,
 		      uint32_t flags, int lvblen,
 		      const struct dlm_lockspace_ops *ops, void *ops_arg,
 		      int *ops_result, dlm_lockspace_t **lockspace);
-- 
2.43.0