[PATCH 06/11] restart: rename 'ctx->tasks_arr' to 'ctx->tasks'

Oren Laadan orenl at cs.columbia.edu
Mon Feb 7 09:21:27 PST 2011


In preparation for next (super)-patch.

Signed-off-by: Oren Laadan <orenl at cs.columbia.edu>
---
 restart.c |   44 ++++++++++++++++++++++----------------------
 1 files changed, 22 insertions(+), 22 deletions(-)

diff --git a/restart.c b/restart.c
index 05d101b..966f7a1 100644
--- a/restart.c
+++ b/restart.c
@@ -141,7 +141,7 @@ struct ckpt_ctx {
 	struct ckpt_pids *copy_arr;
 	__s32 *vpids_arr;
 
-	struct task *tasks_arr;
+	struct task *tasks;
 	int tasks_nr;
 	int tasks_max;
 	int tasks_pid;
@@ -521,8 +521,8 @@ static void exit_ctx(struct ckpt_ctx *ctx)
 {
 	if (ctx->freezer)
 		free(ctx->freezer);
-	if (ctx->tasks_arr)
-		free(ctx->tasks_arr);
+	if (ctx->tasks)
+		free(ctx->tasks);
 	if (ctx->pids_arr)
 		free(ctx->pids_arr);
 	if (ctx->copy_arr)
@@ -641,13 +641,13 @@ int cr_restart(struct cr_restart_args *args)
 	 * setup devpts, root-dir and /proc if necessary, ...
 	 */
 	if (ctx.args->mnt_pty)
-		ctx.tasks_arr[0].flags |= TASK_NEWPTS;
+		ctx.tasks[0].flags |= TASK_NEWPTS;
 	if (ctx.args->mntns)
-		ctx.tasks_arr[0].flags |= TASK_NEWNS;
+		ctx.tasks[0].flags |= TASK_NEWNS;
 	if (ctx.args->root)
-		ctx.tasks_arr[0].flags |= TASK_NEWROOT;
+		ctx.tasks[0].flags |= TASK_NEWROOT;
 
-	if (ctx.args->pidns && ctx.tasks_arr[0].pid != 1) {
+	if (ctx.args->pidns && ctx.tasks[0].pid != 1) {
 		ckpt_dbg("new pidns without init\n");
 		if (global_send_sigint == -1)
 			global_send_sigint = SIGINT;
@@ -656,12 +656,12 @@ int cr_restart(struct cr_restart_args *args)
 		 * the coordinator should set up the filesystems and
 		 * not the first process in the application process tree.
 		 */
-		ctx.tasks_arr[0].flags &=
-			~(TASK_NEWPTS | TASK_NEWROOT |TASK_NEWNS);
+		ctx.tasks[0].flags &=
+			~(TASK_NEWPTS | TASK_NEWROOT | TASK_NEWNS);
 		ret = ckpt_coordinator_pidns(&ctx);
 	} else if (ctx.args->pidns) {
 		ckpt_dbg("new pidns with init\n");
-		ctx.tasks_arr[0].flags |= TASK_NEWPID | TASK_NEWNS;
+		ctx.tasks[0].flags |= TASK_NEWPID | TASK_NEWNS;
 		if (global_send_sigint == -1)
 			global_send_sigint = SIGKILL;
 		ret = ckpt_coordinator(&ctx);
@@ -1039,7 +1039,7 @@ static int ckpt_coordinator(struct ckpt_ctx *ctx)
 	pid_t root_pid;
 	int ret;
 
-	root_pid = ckpt_fork_child(ctx, &ctx->tasks_arr[0]);
+	root_pid = ckpt_fork_child(ctx, &ctx->tasks[0]);
 	if (root_pid < 0)
 		return -1;
 	global_child_pid = root_pid;
@@ -1069,7 +1069,7 @@ static int ckpt_coordinator(struct ckpt_ctx *ctx)
 		ret = 0;
 	}
 
-	if (ctx->args->pidns && ctx->tasks_arr[0].pid != 1) {
+	if (ctx->args->pidns && ctx->tasks[0].pid != 1) {
 		/* Report success/failure to the parent */
 		if (ret < 0)
 			ret = ctx->error;
@@ -1104,7 +1104,7 @@ static int ckpt_coordinator(struct ckpt_ctx *ctx)
 
 static inline struct task *ckpt_init_task(struct ckpt_ctx *ctx)
 {
-	return (&ctx->tasks_arr[0]);
+	return (&ctx->tasks[0]);
 }
 
 /*
@@ -1123,8 +1123,8 @@ static int ckpt_build_tree(struct ckpt_ctx *ctx)
 	 * placeholder tasks (each session id may have at most one)
 	 */
 	ctx->tasks_max = ctx->pids_nr * 4;
-	ctx->tasks_arr = malloc(sizeof(*ctx->tasks_arr) * ctx->tasks_max);
-	if (!ctx->tasks_arr) {
+	ctx->tasks = malloc(sizeof(*ctx->tasks) * ctx->tasks_max);
+	if (!ctx->tasks) {
 		ckpt_perror("malloc tasks array");
 		return -1;
 	}
@@ -1135,7 +1135,7 @@ static int ckpt_build_tree(struct ckpt_ctx *ctx)
 
 	/* assign a creator to each task */
 	for (i = 0; i < ctx->tasks_nr; i++) {
-		task = &ctx->tasks_arr[i];
+		task = &ctx->tasks[i];
 		if (task->creator)
 			continue;
 		if (ckpt_set_creator(ctx, task) < 0)
@@ -1145,7 +1145,7 @@ static int ckpt_build_tree(struct ckpt_ctx *ctx)
 #ifdef CHECKPOINT_DEBUG
 	ckpt_dbg("====== TASKS\n");
 	for (i = 0; i < ctx->tasks_nr; i++) {
-		task = &ctx->tasks_arr[i];
+		task = &ctx->tasks[i];
 		ckpt_dbg("\t[%d] pid %d ppid %d sid %d creator %d",
 			 i, task->pid, task->ppid, task->sid,
 			 task->creator->pid);
@@ -1180,7 +1180,7 @@ static int ckpt_setup_task(struct ckpt_ctx *ctx, pid_t pid, pid_t ppid)
 	if (hash_lookup(ctx, pid))  /* already handled */
 		return 0;
 
-	task = &ctx->tasks_arr[ctx->tasks_nr++];
+	task = &ctx->tasks[ctx->tasks_nr++];
 
 	task->flags = TASK_GHOST;
 
@@ -1305,7 +1305,7 @@ static int ckpt_init_tree(struct ckpt_ctx *ctx)
 
 	/* populate with known tasks */
 	for (i = 0; i < pids_nr; i++) {
-		task = &ctx->tasks_arr[i];
+		task = &ctx->tasks[i];
 
 		task->flags = 0;
 
@@ -1582,7 +1582,7 @@ static int ckpt_set_creator(struct ckpt_ctx *ctx, struct task *task)
 static int ckpt_placeholder_task(struct ckpt_ctx *ctx, struct task *task)
 {
 	struct task *session = hash_lookup(ctx, task->sid);
-	struct task *holder = &ctx->tasks_arr[ctx->tasks_nr++];
+	struct task *holder = &ctx->tasks[ctx->tasks_nr++];
 	pid_t pid;
 
 	if (ctx->tasks_nr > ctx->tasks_max) {
@@ -2492,7 +2492,7 @@ static int assign_vpids(struct ckpt_ctx *ctx)
 	}
 
 	for (tidx = 0, hidx = 0, vidx = 0; tidx < ctx->pids_nr; tidx++) {
-		task = &ctx->tasks_arr[tidx];
+		task = &ctx->tasks[tidx];
 		depth = ctx->pids_arr[tidx].depth;
 
 		task->vidx = vidx;
@@ -2532,7 +2532,7 @@ static int ckpt_read_vpids(struct ckpt_ctx *ctx)
 		if (ctx->pids_arr[i].depth < 0) {
 			ckpt_err("Invalid depth %d for pid %d",
 				 ctx->pids_arr[i].depth,
-				 ctx->tasks_arr[i].pid);
+				 ctx->tasks[i].pid);
 			errno = -EINVAL;
 			return -1;
 		}
-- 
1.7.1



More information about the Containers mailing list