* Theodore Ts'o <ty...@mit.edu>, 2017-08-21, 23:51:
$ gzip -d underwrite.ext2.gz
$ valgrind /sbin/e2fsck -f -y underwrite.ext2 > /dev/null
==6645== Memcheck, a memory error detector
==6645== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==6645== Using Valgrind-3.13.0 and LibVEX; rerun with -h for copyright info
==6645== Command: /sbin/e2fsck -f -y underwrite.ext2
==6645==
e2fsck 1.43.5 (04-Aug-2017)
==6645== Invalid write of size 1
==6645== at 0x48349C0: memset (in
/usr/lib/valgrind/vgpreload_memcheck-x86-linux.so)
==6645== by 0x4892A2E: memset (string3.h:90)
==6645== by 0x4892A2E: raw_read_blk (unix_io.c:250)
[...]
I'm not able to reproduce the problem. Can you double check whether you
uploaded the correct image?
Yes, it's the correct one.
<tytso@callcc> {/build/e2fsprogs-maint/e2fsck}
1280% sha1sum /tmp/underwrite.ext2
32f671c8659cd0ac2f2d273b45669642d016344a /tmp/underwrite.ext2
That's the same file I have. :)
<tytso@callcc> {/build/e2fsprogs-maint/e2fsck}
1281% valgrind /sbin/e2fsck -f -y /tmp/foo.img > /dev/null
[...]
==7619== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
Hmm... What is your architecture? I can reproduce the bug on i386, but not on
amd64.
--
Jakub Wilk
