From: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
To: Oleg Nesterov <oleg@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>, Anton Arapov <anton@redhat.com>,
"Frank Ch. Eigler" <fche@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
William Cohen <wcohen@redhat.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] uprobes: Ignore unsupported instructions in uprobe_mmap
Date: Fri, 3 Aug 2012 17:43:42 +0530 [thread overview]
Message-ID: <20120803121342.GD3748@linux.vnet.ibm.com> (raw)
In-Reply-To: <20120802135313.GA4334@redhat.com>
> OK, lets start with dup_mmap:
>
> // retval == 0
>
> if (file && uprobe_mmap(tmp))
> goto out;
>
> out:
> up_write(&mm->mmap_sem);
> flush_tlb_mm(oldmm);
> up_write(&oldmm->mmap_sem);
> return retval;
>
> Given that retval == 0, what do you think dup_mmap() returns if
> uprobe_mmap() fails? And note that we didn't copy all vmas.
> OK, at least this can't crash (afaics), and easy to fix.
>
>
> But mmap_region() is worse, much worse. It simply can _not_ fail
> after uprobe_mmap (of course, I am not saying this is unfixable)
> without the crash. And note that the crash is "delayed". And btw,
> like dup_mmap(), mmap_region() doesn't return the error too.
>
> Srikar, I strongly believe this horror must not exist. Either
> we should teach mmap_region() and dup_mmap() (and vma_adjust!)
> to fail correctly, or we should ignore the error code.
>
> It is that simple, isn't it?
I think you would have thought of this approach already but just wanted
to check if below is fine with you.
diff --git a/kernel/fork.c b/kernel/fork.c
index f00e319..78bfd94 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -456,10 +456,10 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
if (tmp->vm_ops && tmp->vm_ops->open)
tmp->vm_ops->open(tmp);
- if (retval)
- goto out;
+ if (!retval && file)
+ retval = uprobe_mmap(tmp);
- if (file && uprobe_mmap(tmp))
+ if (retval)
goto out;
}
/* a new mm has just been created */
diff --git a/mm/mmap.c b/mm/mmap.c
index 4fe2697..91d36fb 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1355,9 +1355,12 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
} else if ((flags & MAP_POPULATE) && !(flags & MAP_NONBLOCK))
make_pages_present(addr, addr + len);
- if (file && uprobe_mmap(vma))
+ if (file) {
+ error = uprobe_mmap(vma)
/* matching probes but cannot insert */
- goto unmap_and_free_vma;
+ if (error)
+ goto unmap_and_free_vma;
+ }
return addr;
Basically, I am setting the return value of uprobe_mmap() so that
mmap_region and do_fork() fail. This still needs the fix in
uprobe_mmap() to ignore unsupported instructions.
I am completely _okay_ with not setting the return values as proposed by you.
Just that setting return values and the fix in uprobe_mmap() might help
in minor issues
- We either probe all probes where possible or intimate to the user that
the requested operation wasnt successful.
- If valid probes follow a probe with invalid instruction, we still
allow valid probes.
- If get_user_pages()/set_swbp() fail because of genuine reasons like
ENOMEM, then we dont retry to place probes on the subsequent vmas.
--
thanks and regards
Srikar
next prev parent reply other threads:[~2012-08-03 12:14 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-28 16:31 [PATCH] uprobes: mmap_region() corrupts mm->mm_rb if uprobe_mmap() fails Oleg Nesterov
2012-07-28 16:34 ` Oleg Nesterov
2012-07-30 13:22 ` William Cohen
2012-07-31 6:47 ` Srikar Dronamraju
2012-07-31 12:48 ` Oleg Nesterov
2012-07-31 13:25 ` Oleg Nesterov
2012-08-02 10:05 ` [PATCH] uprobes: Ignore unsupported instructions in uprobe_mmap Srikar Dronamraju
2012-08-02 13:53 ` Oleg Nesterov
2012-08-02 16:42 ` Srikar Dronamraju
2012-08-02 17:48 ` Oleg Nesterov
2012-08-03 12:13 ` Srikar Dronamraju [this message]
2012-08-03 13:38 ` Oleg Nesterov
2012-08-02 14:17 ` Oleg Nesterov
2012-08-02 16:54 ` Srikar Dronamraju
2012-08-02 17:53 ` Oleg Nesterov
2012-08-03 1:20 ` Srikar Dronamraju
2012-08-03 13:47 ` Oleg Nesterov
2012-08-03 17:46 ` [PATCH] uprobes: mmap_region() corrupts mm->mm_rb if uprobe_mmap() fails Srikar Dronamraju
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=20120803121342.GD3748@linux.vnet.ibm.com \
--to=srikar@linux.vnet.ibm.com \
--cc=anton@redhat.com \
--cc=fche@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=oleg@redhat.com \
--cc=peterz@infradead.org \
--cc=wcohen@redhat.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;
as well as URLs for NNTP newsgroup(s).