From: Oleg Nesterov <oleg@redhat.com>
Date: Wed, 1 Dec 2010 16:38:08 -0500
Subject: [fs] exec: make argv/envp memory visible to oom-killer
Message-id: <20101201163808.GC1758@redhat.com>
Patchwork-id: 29770
O-Subject: [RHEL5.6 PATCH 2/3] bz625694: exec: make argv/envp memory visible
	to oom-killer
Bugzilla: 625694
CVE: CVE-2010-4243

https://bugzilla.redhat.com/show_bug.cgi?id=625694

Upstream commit 3c77f845722158206a7209c45ccddc264d19319c
Author: Oleg Nesterov <oleg@redhat.com>
Date:   Tue Nov 30 20:55:34 2010 +0100

    exec: make argv/envp memory visible to oom-killer

    Brad Spengler published a local memory-allocation DoS that
    evades the OOM-killer (though not the virtual memory RLIMIT):
    http://www.grsecurity.net/~spender/64bit_dos.c

    execve()->copy_strings() can allocate a lot of memory, but
    this is not visible to oom-killer, nobody can see the nascent
    bprm->mm and take it into account.

    With this patch get_arg_page() increments current's MM_ANONPAGES
    counter every time we allocate the new page for argv/envp. When
    do_execve() succeds or fails, we change this counter back.

    Technically this is not 100% correct, we can't know if the new
    page is swapped out and turn MM_ANONPAGES into MM_SWAPENTS, but
    I don't think this really matters and everything becomes correct
    once exec changes ->mm or fails.

    Reported-by: Brad Spengler <spender@grsecurity.net>
    Reviewed-and-discussed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
    Signed-off-by: Oleg Nesterov <oleg@redhat.com>
    Cc: stable@kernel.org
    Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Compared to upstream:

	RHEL5 doesn't have MM_ANONPAGES, and mm->anon_rss field can't
	help. Oom-killer's badness() takes mm->total_vm into account
	and nothing else. So acct_arg_size() has to play with this
	counter too.

	Note: we take mmap_sem to update ->total_vm, but the usage
	of this counter is racy in rhel kernels. For example, do_brk()
	changes it lockless.

Signed-off-by: Oleg Nesterov <oleg@redhat.com>

diff --git a/fs/exec.c b/fs/exec.c
index 3fe0248..d6f4971 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -179,6 +179,21 @@ exit:
 
 #ifdef CONFIG_MMU
 
+static void acct_arg_size(struct linux_binprm *bprm, unsigned long pages)
+{
+	struct mm_struct *mm = current->mm;
+	long diff = (long)(pages - bprm->vma_pages);
+
+	if (!mm || !diff)
+		return;
+
+	bprm->vma_pages = pages;
+
+	down_write(&mm->mmap_sem);
+	mm->total_vm += diff;
+	up_write(&mm->mmap_sem);
+}
+
 struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
 		int write)
 {
@@ -201,6 +216,8 @@ struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
 		struct rlimit *rlim = current->signal->rlim;
 		unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start;
 
+		acct_arg_size(bprm, size / PAGE_SIZE);
+
 		/*
 		 * Limit to 1/4-th the stack size for the argv+env strings.
 		 * This ensures that:
@@ -289,6 +306,10 @@ static bool valid_arg_len(struct linux_binprm *bprm, long len)
 
 #else
 
+static inline void acct_arg_size(struct linux_binprm *bprm, unsigned long pages)
+{
+}
+
 struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos,
 		int write)
 {
@@ -1057,6 +1078,7 @@ int flush_old_exec(struct linux_binprm * bprm)
 	/*
 	 * Release all of the old mmap stuff
 	 */
+	acct_arg_size(bprm, 0);
 	retval = exec_mmap(bprm->mm);
 	if (retval)
 		goto mmap_failed;
@@ -1435,8 +1457,10 @@ out:
 		security_bprm_free(bprm);
 
 out_mm:
-	if (bprm->mm)
-		mmput (bprm->mm);
+	if (bprm->mm) {
+		acct_arg_size(bprm, 0);
+		mmput(bprm->mm);
+	}
 
 out_file:
 	if (bprm->file) {
diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h
index 523963c..b05117c 100644
--- a/include/linux/binfmts.h
+++ b/include/linux/binfmts.h
@@ -43,6 +43,7 @@ struct linux_binprm{
 	unsigned long loader, exec;
 #ifndef __GENKSYMS__
 #ifdef CONFIG_MMU
+	unsigned long vma_pages;
 	struct vm_area_struct *vma;
 #endif
 #endif