Re: [PATCH -tip 1/3] Pass mm->flags to binfmt core_dump for bitflag consistency

[Masami Hiramatsu <mhiramat@redhat.com>]

Ingo Molnar wrote:
>
> * Andrew Morton<akpm@linux-foundation.org>  wrote:
>
>> On Fri, 13 Nov 2009 17:52:27 -0500
>> Masami Hiramatsu<mhiramat@redhat.com>  wrote:
>>
>>> Pass mm->flags to binfmt core_dump for bitflag consistency.
>>> Since mm->flags bit flags is not protected by locks, it will be
>>> changed while dumping core. This patch copies mm->flags to a
>>> mm_flags local variable at the beginning of do_coredump() and
>>> use it while dumping. mm_flags also includes dump_filter which
>>> filters elf sections from core file in elf_core_dump().
>>> So, this patch also passes mm_flags to each binfmt->core_dump().
>>
>> I can kind-of guess the answer, but it would be much more reliable if
>> we were to hear this from yourself:
>>
>>    Why did you write this patch?  What problem is being observed?
>
> i'm not Masami so i'm only guessing that while writing the tracepoint a
> race got noticed but that otherwise there's no big practical effect,
> 'just' a cleanliness problem fixed.

Right, I'd like to add a tracepoint of coredump event with
its information. And also, this patch may fix a small
dumpable inconsistency issue below code

---
1787         if (mm->core_state || !get_dumpable(mm)) {  <- (1)
1788                 up_write(&mm->mmap_sem);
1789                 put_cred(cred);
1790                 goto fail;
1791         }
1792
[...]
1798         if (get_dumpable(mm) == 2) {    /* Setuid core dump mode */ <-(2)
1799                 flag = O_EXCL;          /* Stop rewrite attacks */
1800                 cred->fsuid = 0;        /* Dump root private */
1801         }

Since dumpable bits are not protected by lock, there is a
chance to change these bits between (1) and (2).

This patch copies mm->flags to a local variable and check the variable
for consistency.

Thank you,

-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America), Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com