aboutsummaryrefslogtreecommitdiffstats
path: root/fs/proc
diff options
context:
space:
mode:
authorEric W. Biederman <ebiederm@xmission.com>2006-10-02 05:17:04 -0400
committerLinus Torvalds <torvalds@g5.osdl.org>2006-10-02 10:57:12 -0400
commit0804ef4b0de7121261f77c565b20a11ac694e877 (patch)
treeff12e3b999dc2ce66d97fce5d76cd7df073c0d5c /fs/proc
parent2bc2d61a9638dab670d8361e928d1a5a291173ef (diff)
[PATCH] proc: readdir race fix (take 3)
The problem: An opendir, readdir, closedir sequence can fail to report process ids that are continually in use throughout the sequence of system calls. For this race to trigger the process that proc_pid_readdir stops at must exit before readdir is called again. This can cause ps to fail to report processes, and it is in violation of posix guarantees and normal application expectations with respect to readdir. Currently there is no way to work around this problem in user space short of providing a gargantuan buffer to user space so the directory read all happens in on system call. This patch implements the normal directory semantics for proc, that guarantee that a directory entry that is neither created nor destroyed while reading the directory entry will be returned. For directory that are either created or destroyed during the readdir you may or may not see them. Furthermore you may seek to a directory offset you have previously seen. These are the guarantee that ext[23] provides and that posix requires, and more importantly that user space expects. Plus it is a simple semantic to implement reliable service. It is just a matter of calling readdir a second time if you are wondering if something new has show up. These better semantics are implemented by scanning through the pids in numerical order and by making the file offset a pid plus a fixed offset. The pid scan happens on the pid bitmap, which when you look at it is remarkably efficient for a brute force algorithm. Given that a typical cache line is 64 bytes and thus covers space for 64*8 == 200 pids. There are only 40 cache lines for the entire 32K pid space. A typical system will have 100 pids or more so this is actually fewer cache lines we have to look at to scan a linked list, and the worst case of having to scan the entire pid bitmap is pretty reasonable. If we need something more efficient we can go to a more efficient data structure for indexing the pids, but for now what we have should be sufficient. In addition this takes no additional locks and is actually less code than what we are doing now. Also another very subtle bug in this area has been fixed. It is possible to catch a task in the middle of de_thread where a thread is assuming the thread of it's thread group leader. This patch carefully handles that case so if we hit it we don't fail to return the pid, that is undergoing the de_thread dance. Thanks to KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> for providing the first fix, pointing this out and working on it. [oleg@tv-sign.ru: fix it] Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Cc: Jean Delvare <jdelvare@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'fs/proc')
-rw-r--r--fs/proc/base.c104
1 files changed, 35 insertions, 69 deletions
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 89c20d9d50bf..b18f3773dd43 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2142,72 +2142,43 @@ out_no_task:
2142} 2142}
2143 2143
2144/* 2144/*
2145 * Find the first tgid to return to user space. 2145 * Find the first task with tgid >= tgid
2146 * 2146 *
2147 * Usually this is just whatever follows &init_task, but if the users
2148 * buffer was too small to hold the full list or there was a seek into
2149 * the middle of the directory we have more work to do.
2150 *
2151 * In the case of a short read we start with find_task_by_pid.
2152 *
2153 * In the case of a seek we start with &init_task and walk nr
2154 * threads past it.
2155 */ 2147 */
2156static struct task_struct *first_tgid(int tgid, unsigned int nr) 2148static struct task_struct *next_tgid(unsigned int tgid)
2157{ 2149{
2158 struct task_struct *pos; 2150 struct task_struct *task;
2159 rcu_read_lock(); 2151 struct pid *pid;
2160 if (tgid && nr) {
2161 pos = find_task_by_pid(tgid);
2162 if (pos && thread_group_leader(pos))
2163 goto found;
2164 }
2165 /* If nr exceeds the number of processes get out quickly */
2166 pos = NULL;
2167 if (nr && nr >= nr_processes())
2168 goto done;
2169
2170 /* If we haven't found our starting place yet start with
2171 * the init_task and walk nr tasks forward.
2172 */
2173 for (pos = next_task(&init_task); nr > 0; --nr) {
2174 pos = next_task(pos);
2175 if (pos == &init_task) {
2176 pos = NULL;
2177 goto done;
2178 }
2179 }
2180found:
2181 get_task_struct(pos);
2182done:
2183 rcu_read_unlock();
2184 return pos;
2185}
2186 2152
2187/*
2188 * Find the next task in the task list.
2189 * Return NULL if we loop or there is any error.
2190 *
2191 * The reference to the input task_struct is released.
2192 */
2193static struct task_struct *next_tgid(struct task_struct *start)
2194{
2195 struct task_struct *pos;
2196 rcu_read_lock(); 2153 rcu_read_lock();
2197 pos = start; 2154retry:
2198 if (pid_alive(start)) 2155 task = NULL;
2199 pos = next_task(start); 2156 pid = find_ge_pid(tgid);
2200 if (pid_alive(pos) && (pos != &init_task)) { 2157 if (pid) {
2201 get_task_struct(pos); 2158 tgid = pid->nr + 1;
2202 goto done; 2159 task = pid_task(pid, PIDTYPE_PID);
2160 /* What we to know is if the pid we have find is the
2161 * pid of a thread_group_leader. Testing for task
2162 * being a thread_group_leader is the obvious thing
2163 * todo but there is a window when it fails, due to
2164 * the pid transfer logic in de_thread.
2165 *
2166 * So we perform the straight forward test of seeing
2167 * if the pid we have found is the pid of a thread
2168 * group leader, and don't worry if the task we have
2169 * found doesn't happen to be a thread group leader.
2170 * As we don't care in the case of readdir.
2171 */
2172 if (!task || !has_group_leader_pid(task))
2173 goto retry;
2174 get_task_struct(task);
2203 } 2175 }
2204 pos = NULL;
2205done:
2206 rcu_read_unlock(); 2176 rcu_read_unlock();
2207 put_task_struct(start); 2177 return task;
2208 return pos;
2209} 2178}
2210 2179
2180#define TGID_OFFSET (FIRST_PROCESS_ENTRY + (1 /* /proc/self */))
2181
2211/* for the /proc/ directory itself, after non-process stuff has been done */ 2182/* for the /proc/ directory itself, after non-process stuff has been done */
2212int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir) 2183int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir)
2213{ 2184{
@@ -2223,29 +2194,24 @@ int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir)
2223 filp->f_pos++; 2194 filp->f_pos++;
2224 nr++; 2195 nr++;
2225 } 2196 }
2226 nr -= 1;
2227 2197
2228 /* f_version caches the tgid value that the last readdir call couldn't 2198 tgid = filp->f_pos - TGID_OFFSET;
2229 * return. lseek aka telldir automagically resets f_version to 0. 2199 for (task = next_tgid(tgid);
2230 */
2231 tgid = filp->f_version;
2232 filp->f_version = 0;
2233 for (task = first_tgid(tgid, nr);
2234 task; 2200 task;
2235 task = next_tgid(task), filp->f_pos++) { 2201 put_task_struct(task), task = next_tgid(tgid + 1)) {
2236 int len; 2202 int len;
2237 ino_t ino; 2203 ino_t ino;
2238 tgid = task->pid; 2204 tgid = task->pid;
2205 filp->f_pos = tgid + TGID_OFFSET;
2239 len = snprintf(buf, sizeof(buf), "%d", tgid); 2206 len = snprintf(buf, sizeof(buf), "%d", tgid);
2240 ino = fake_ino(tgid, PROC_TGID_INO); 2207 ino = fake_ino(tgid, PROC_TGID_INO);
2241 if (filldir(dirent, buf, len, filp->f_pos, ino, DT_DIR) < 0) { 2208 if (filldir(dirent, buf, len, filp->f_pos, ino, DT_DIR) < 0) {
2242 /* returning this tgid failed, save it as the first
2243 * pid for the next readir call */
2244 filp->f_version = tgid;
2245 put_task_struct(task); 2209 put_task_struct(task);
2246 break; 2210 goto out;
2247 } 2211 }
2248 } 2212 }
2213 filp->f_pos = PID_MAX_LIMIT + TGID_OFFSET;
2214out:
2249 return 0; 2215 return 0;
2250} 2216}
2251 2217