[PATCH try #4] proc: fix PAGE_SIZE limit of /proc/$PID/cmdline
From: Alexey Dobriyan
Date: Fri May 08 2015 - 08:28:18 EST
/proc/$PID/cmdline truncates output at PAGE_SIZE. It is easy to see with
$ cat /proc/self/cmdline $(seq 1037) 2>/dev/null
However, command line size was never limited to PAGE_SIZE but to 128 KB and
relatively recently limitation was removed altogether.
People noticed and ask questions:
http://stackoverflow.com/questions/199130/how-do-i-increase-the-proc-pid-cmdline-4096-byte-limit
seq file interface is not OK, because it kmalloc's for whole output and
open + read(, 1) + sleep will pin arbitrary amounts of kernel memory.
To not do that, limit must be imposed which is incompatible with
arbitrary sized command lines.
I apologize for hairy code, but this it direct consequence of command line
layout in memory and hacks to support things like "init [3]".
The loops are "unrolled" otherwise it is either macros which hide
control flow or functions with 7-8 arguments with equal line count.
There should be real setproctitle(2) or something.
Signed-off-by: Alexey Dobriyan <adobriyan@xxxxxxxxx>
Tested-by: Jarod Wilson <jarod@xxxxxxxxxx>
Acked-by: Jarod Wilson <jarod@xxxxxxxxxx>
---
fs/proc/base.c | 203 ++++++++++++++++++++++++++++++++++++++++++++++++++++++---
1 file changed, 194 insertions(+), 9 deletions(-)
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -196,18 +196,203 @@ static int proc_root_link(struct dentry *dentry, struct path *path)
return result;
}
-static int proc_pid_cmdline(struct seq_file *m, struct pid_namespace *ns,
- struct pid *pid, struct task_struct *task)
+static ssize_t proc_pid_cmdline_read(struct file *file, char __user *buf,
+ size_t count, loff_t *pos)
{
+ struct task_struct *tsk;
+ struct mm_struct *mm;
+ char *page;
+ unsigned long arg_start, arg_end, env_start, env_end;
+ unsigned long len1, len2, len;
+ unsigned long p;
+ char c;
+ ssize_t rv;
+
+ BUG_ON(*pos < 0);
+
+ tsk = get_proc_task(file_inode(file));
+ if (!tsk)
+ return -ESRCH;
+ mm = get_task_mm(tsk);
+ put_task_struct(tsk);
+ if (!mm)
+ return 0;
+ if (!mm->arg_end) {
+ rv = 0;
+ goto out_mmput;
+ }
+
+ page = (char *)__get_free_page(GFP_TEMPORARY);
+ if (!page) {
+ rv = -ENOMEM;
+ goto out_mmput;
+ }
+
+ down_read(&mm->mmap_sem);
+ arg_start = mm->arg_start;
+ arg_end = mm->arg_end;
+ env_start = mm->env_start;
+ env_end = mm->env_end;
+ up_read(&mm->mmap_sem);
+
+ BUG_ON(arg_start > arg_end);
+ BUG_ON(env_start > env_end);
+
+ len1 = arg_end - arg_start;
+ len2 = env_end - env_start;
+
/*
- * Rely on struct seq_operations::show() being called once
- * per internal buffer allocation. See single_open(), traverse().
+ * Inherently racy -- command line shares address space
+ * with code and data.
*/
- BUG_ON(m->size < PAGE_SIZE);
- m->count += get_cmdline(task, m->buf, PAGE_SIZE);
- return 0;
+ rv = access_remote_vm(mm, arg_end - 1, &c, 1, 0);
+ if (rv <= 0)
+ goto out_free_page;
+
+ rv = 0;
+
+ if (c == '\0') {
+ /* Command line (set of strings) occupies whole ARGV. */
+ if (len1 <= *pos)
+ goto out_free_page;
+
+ p = arg_start + *pos;
+ len = len1 - *pos;
+ while (count > 0 && len > 0) {
+ unsigned int _count;
+ int nr_read;
+
+ _count = min3(count, len, PAGE_SIZE);
+ nr_read = access_remote_vm(mm, p, page, _count, 0);
+ if (nr_read < 0)
+ rv = nr_read;
+ if (nr_read <= 0)
+ goto out_free_page;
+
+ if (copy_to_user(buf, page, nr_read)) {
+ rv = -EFAULT;
+ goto out_free_page;
+ }
+
+ p += nr_read;
+ len -= nr_read;
+ buf += nr_read;
+ count -= nr_read;
+ rv += nr_read;
+ }
+ } else {
+ /*
+ * Command line (1 string) occupies ARGV and maybe
+ * extends into ENVP.
+ */
+ if (len1 + len2 <= *pos)
+ goto skip_argv_envp;
+ if (len1 <= *pos)
+ goto skip_argv;
+
+ p = arg_start + *pos;
+ len = len1 - *pos;
+ while (count > 0 && len > 0) {
+ unsigned int _count, l;
+ int nr_read;
+ bool final;
+
+ _count = min3(count, len, PAGE_SIZE);
+ nr_read = access_remote_vm(mm, p, page, _count, 0);
+ if (nr_read < 0)
+ rv = nr_read;
+ if (nr_read <= 0)
+ goto out_free_page;
+
+ /*
+ * Command line can be shorter than whole ARGV
+ * even if last "marker" byte says it is not.
+ */
+ final = false;
+ l = strnlen(page, nr_read);
+ if (l < nr_read) {
+ nr_read = l;
+ final = true;
+ }
+
+ if (copy_to_user(buf, page, nr_read)) {
+ rv = -EFAULT;
+ goto out_free_page;
+ }
+
+ p += nr_read;
+ len -= nr_read;
+ buf += nr_read;
+ count -= nr_read;
+ rv += nr_read;
+
+ if (final)
+ goto out_free_page;
+ }
+skip_argv:
+ /*
+ * Command line (1 string) occupies ARGV and
+ * extends into ENVP.
+ */
+ if (len1 <= *pos) {
+ p = env_start + *pos - len1;
+ len = len1 + len2 - *pos;
+ } else {
+ p = env_start;
+ len = len2;
+ }
+ while (count > 0 && len > 0) {
+ unsigned int _count, l;
+ int nr_read;
+ bool final;
+
+ _count = min3(count, len, PAGE_SIZE);
+ nr_read = access_remote_vm(mm, p, page, _count, 0);
+ if (nr_read < 0)
+ rv = nr_read;
+ if (nr_read <= 0)
+ goto out_free_page;
+
+ /* Find EOS. */
+ final = false;
+ l = strnlen(page, nr_read);
+ if (l < nr_read) {
+ nr_read = l;
+ final = true;
+ }
+
+ if (copy_to_user(buf, page, nr_read)) {
+ rv = -EFAULT;
+ goto out_free_page;
+ }
+
+ p += nr_read;
+ len -= nr_read;
+ buf += nr_read;
+ count -= nr_read;
+ rv += nr_read;
+
+ if (final)
+ goto out_free_page;
+ }
+skip_argv_envp:
+ ;
+ }
+
+out_free_page:
+ free_page((unsigned long)page);
+out_mmput:
+ mmput(mm);
+ if (rv > 0)
+ *pos += rv;
+ return rv;
}
+static const struct file_operations proc_pid_cmdline_ops = {
+ .read = proc_pid_cmdline_read,
+ .llseek = generic_file_llseek,
+};
+
static int proc_pid_auxv(struct seq_file *m, struct pid_namespace *ns,
struct pid *pid, struct task_struct *task)
{
@@ -2560,7 +2745,7 @@ static const struct pid_entry tgid_base_stuff[] = {
#ifdef CONFIG_HAVE_ARCH_TRACEHOOK
ONE("syscall", S_IRUSR, proc_pid_syscall),
#endif
- ONE("cmdline", S_IRUGO, proc_pid_cmdline),
+ REG("cmdline", S_IRUGO, proc_pid_cmdline_ops),
ONE("stat", S_IRUGO, proc_tgid_stat),
ONE("statm", S_IRUGO, proc_pid_statm),
REG("maps", S_IRUGO, proc_pid_maps_operations),
@@ -2906,7 +3091,7 @@ static const struct pid_entry tid_base_stuff[] = {
#ifdef CONFIG_HAVE_ARCH_TRACEHOOK
ONE("syscall", S_IRUSR, proc_pid_syscall),
#endif
- ONE("cmdline", S_IRUGO, proc_pid_cmdline),
+ REG("cmdline", S_IRUGO, proc_pid_cmdline_ops),
ONE("stat", S_IRUGO, proc_tid_stat),
ONE("statm", S_IRUGO, proc_pid_statm),
REG("maps", S_IRUGO, proc_tid_maps_operations),
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/