xfs
[Top] [All Lists]

Re: [PATCH V2] xfsprogs: Fix possible unallocated memory access in fiema

To: Eric Sandeen <sandeen@xxxxxxxxxxx>
Subject: Re: [PATCH V2] xfsprogs: Fix possible unallocated memory access in fiemap
From: Mark Tinguely <tinguely@xxxxxxx>
Date: Thu, 21 Feb 2013 09:28:08 -0600
Cc: hch@xxxxxxxxxxxxx, xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <5102F4BE.6060009@xxxxxxxxxxx>
References: <1355319059-12111-1-git-send-email-lczerner@xxxxxxxxxx> <20121212231245.GY16353@dastard> <alpine.LFD.2.00.1212131024460.17970@localhost> <20121213230608.GI16353@dastard> <5102F4BE.6060009@xxxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:9.0) Gecko/20120122 Thunderbird/9.0
On 01/25/13 15:10, Eric Sandeen wrote:
(Based on original patch by Lukas Czerner&  comments by Dave Chinner)

Currently we could access unallocated memory in fiemap because we're
using uninitialized variable 'fiemap' in fiemap_f(). In fact this has
been spotted on x390s machine where xfs_io would segfault.

The problem happens in the for cycle which seems to be intended to
compute the header item spacing. However at that point the fiemap
structure has just been allocated and does not contain any extents
yet, so it is entirely useless and it never actually worked.

This patch delays the format calculation until the first batch
of extents has come in for analysis.

Signed-off-by: Eric Sandeen<sandeen@xxxxxxxxxx>
---

A working calculated format is a great improvement.

Reviewed-by: Mark Tinguely <tinguely@xxxxxxx>

<Prev in Thread] Current Thread [Next in Thread>
  • Re: [PATCH V2] xfsprogs: Fix possible unallocated memory access in fiemap, Mark Tinguely <=