public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Ingo Molnar <mingo@elte.hu>,
	Peter Zijlstra <peterz@infradead.org>,
	Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>,
	Anton Arapov <anton@redhat.com>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	linux-kernel@vger.kernel.org
Subject: [PATCH 6/7] uprobes: Fix uprobe_copy_insn() race with itself
Date: Sun, 30 Sep 2012 21:42:24 +0200	[thread overview]
Message-ID: <20120930194224.GA11347@redhat.com> (raw)
In-Reply-To: <20120930194119.GA11278@redhat.com>

install_breakpoint() is called under mm->mmap_sem, this protects
set_swbp() but not uprobe_copy_insn(). Two or more different tasks
can call install_breakpoint()->uprobe_copy_insn() at the same time,
this leads to numerous problems if UPROBE_COPY_INSN is not set.

Just for example, the second copy_insn() can corrupt the already
analyzed/fixuped uprobe->arch.insn and race with handle_swbp().

This patch simply adds uprobe->copy_mutex to serialize this code.
We could probably reuse ->consumer_rwsem, but this would mean that
consumer->handler() can not use mm->mmap_sem, not good.

Note: this is another temporary ugly hack until we move this logic
into uprobe_register().

Signed-off-by: Oleg Nesterov <oleg@redhat.com>
---
 kernel/events/uprobes.c |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 5c0c1b0..8410388 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -89,6 +89,7 @@ struct uprobe {
 	struct rb_node		rb_node;	/* node in the rb tree */
 	atomic_t		ref;
 	struct rw_semaphore	consumer_rwsem;
+	struct mutex		copy_mutex;	/* TODO: kill me and UPROBE_COPY_INSN */
 	struct list_head	pending_list;
 	struct uprobe_consumer	*consumers;
 	struct inode		*inode;		/* Also hold a ref to inode */
@@ -444,6 +445,7 @@ static struct uprobe *alloc_uprobe(struct inode *inode, loff_t offset)
 	uprobe->inode = igrab(inode);
 	uprobe->offset = offset;
 	init_rwsem(&uprobe->consumer_rwsem);
+	mutex_init(&uprobe->copy_mutex);
 
 	/* add to uprobes_tree, sorted on inode:offset */
 	cur_uprobe = insert_uprobe(uprobe);
@@ -578,6 +580,10 @@ static int uprobe_copy_insn(struct uprobe *uprobe, struct file *file,
 	if (uprobe->flags & UPROBE_COPY_INSN)
 		return ret;
 
+	mutex_lock(&uprobe->copy_mutex);
+	if (uprobe->flags & UPROBE_COPY_INSN)
+		goto out;
+
 	ret = copy_insn(uprobe, file);
 	if (ret)
 		goto out;
@@ -598,6 +604,8 @@ static int uprobe_copy_insn(struct uprobe *uprobe, struct file *file,
 	uprobe->flags |= UPROBE_COPY_INSN;
 	ret = 0;
  out:
+	mutex_unlock(&uprobe->copy_mutex);
+
 	return ret;
 }
 
-- 
1.5.5.1


  parent reply	other threads:[~2012-09-30 19:41 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-30 19:41 [PATCH 0/7] uprobes: register/unregister bugfixes Oleg Nesterov
2012-09-30 19:41 ` [PATCH 1/7] uprobes/x86: Only rep+nop can be emulated correctly Oleg Nesterov
2012-10-06  7:20   ` Srikar Dronamraju
2012-09-30 19:42 ` [PATCH 2/7] uprobes: Don't return success if alloc_uprobe() fails Oleg Nesterov
2012-10-06  7:25   ` Srikar Dronamraju
2012-09-30 19:42 ` [PATCH 3/7] uprobes: Do not delete uprobe if uprobe_unregister() fails Oleg Nesterov
2012-10-06  8:48   ` Srikar Dronamraju
2012-09-30 19:42 ` [PATCH 4/7] uprobes: Fix handle_swbp() vs unregister() + register() race Oleg Nesterov
2012-10-02 18:42   ` Oleg Nesterov
2012-10-06  9:33   ` Srikar Dronamraju
2012-10-06 17:25     ` Oleg Nesterov
2012-10-06 17:37       ` Srikar Dronamraju
2012-10-06 18:53         ` Oleg Nesterov
2012-10-07  7:12           ` Srikar Dronamraju
2012-09-30 19:42 ` [PATCH 5/7] uprobes: Introduce uprobe_copy_insn() Oleg Nesterov
2012-10-06  9:45   ` Srikar Dronamraju
2012-10-06 17:10     ` Oleg Nesterov
2012-10-06 17:38       ` Srikar Dronamraju
2012-10-06 18:59         ` Oleg Nesterov
2012-10-07  7:14           ` Srikar Dronamraju
2012-09-30 19:42 ` Oleg Nesterov [this message]
2012-10-06  9:52   ` [PATCH 6/7] uprobes: Fix uprobe_copy_insn() race with itself Srikar Dronamraju
2012-09-30 19:42 ` [PATCH 7/7] uprobes: Fix the racy uprobe->flags manipulation Oleg Nesterov
2012-10-04  8:57   ` Anton Arapov
2012-10-06  9:54   ` Srikar Dronamraju
2012-09-30 19:44 ` [PATCH 0/7] uprobes: register/unregister bugfixes Oleg Nesterov
2012-10-01 12:55   ` Srikar Dronamraju
2012-10-01 14:03     ` Oleg Nesterov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20120930194224.GA11347@redhat.com \
    --to=oleg@redhat.com \
    --cc=ananth@in.ibm.com \
    --cc=anton@redhat.com \
    --cc=bigeasy@linutronix.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=peterz@infradead.org \
    --cc=srikar@linux.vnet.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox