Re: loop-AES problems with RedHat 7.2

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> > I try to build loop-AES-v1.4f (latest) on a RedHat 7.2 Linux i686.
> > Unfortunately the "make tests" failes:
> >
> > # make tests
> > dd if=/dev/zero of=test-file1 bs=1024 count=10
> > 10+0 records in
> > 10+0 records out
> > cp test-file1 test-file3
> > echo 12345678901234567890 | losetup -p 0 -e AES128 /dev/loop7 test-file1
> > ioctl: LOOP_SET_STATUS: Invalid argument
> > make: *** [tests] Error 1
>
>You probably have 2 loop.o modules in /lib/modules/`uname -r`/ tree.
>
>Can you try "make modules_install" after compiling your kernel. "make
>modules_install" is supposed to remove the old wrong loop.o module if
>CONFIG_BLK_DEV_LOOP is N.

I don't have "CONFIG_BLK_DEV_LOOP is N", I have "# CONFIG_BLK_DEV_LOOP is 
not set" but I suppose that is ok.

Anyway, I will try "make modules_install" after the kernel compilation:

# find /lib/modules/ -name "loop.o"
/lib/modules/2.4.9-13/kernel/drivers/block/loop.o
/lib/modules/2.4.9-13custom/block/loop.o

(2.4.9-13custom was created by "make" with loop-AES.)

# cd /usr/src/linux-2.4.9-13
# make modules_install
[...]

But the "loop.o" didn't got removed:

# find /lib/modules/ -name "loop.o"
/lib/modules/2.4.9-13/kernel/drivers/block/loop.o
/lib/modules/2.4.9-13custom/block/loop.o

So since it still is there I removed it by hand:

# mv /lib/modules/2.4.9-13/kernel/drivers/block/loop.o /tmp/

# find /lib/modules/ -name "loop.o"
/lib/modules/2.4.9-13custom/block/loop.o

No "loop.o" is loaded so far:

# lsmod
Module                  Size  Used by
nfs                    79872   4  (autoclean)
lockd                  53168   1  (autoclean) [nfs]
sunrpc                 64816   1  (autoclean) [nfs lockd]
binfmt_misc             6448   1
autofs                 11584   2  (autoclean)
3c59x                  26496   1
ipchains               39392   0
usb-uhci               21696   0  (unused)
usbcore                51808   1  [usb-uhci]
ext3                   62480   7
jbd                    41056   7  [ext3]

Now for the "make tests" again:

# make tests
dd if=/dev/zero of=test-file1 bs=1024 count=10
10+0 records in
10+0 records out
cp test-file1 test-file3
echo 12345678901234567890 | losetup -p 0 -e AES128 /dev/loop7 test-file1
/dev/loop7: No such device or address
make: *** [tests] Error 1

:-(

But then... :-) I read Paul Hilton's comment "I discovered that one of the 
crypto modules was not being loaded automatically, loading them manually 
solved the problem.".

So I tried that:

# insmod /lib/modules/2.4.9-13custom/block/loop.o

# lsmod
Module                  Size  Used by
loop                   41296   0  (unused)
nfs                    79872   5  (autoclean)
lockd                  53168   1  (autoclean) [nfs]
sunrpc                 64816   1  (autoclean) [nfs lockd]
binfmt_misc             6448   1
autofs                 11584   2  (autoclean)
3c59x                  26496   1
ipchains               39392   0
usb-uhci               21696   0  (unused)
usbcore                51808   1  [usb-uhci]
ext3                   62480   7
jbd                    41056   7  [ext3]

# make tests
dd if=/dev/zero of=test-file1 bs=1024 count=10
10+0 records in
10+0 records out
cp test-file1 test-file3
echo 12345678901234567890 | losetup -p 0 -e AES128 /dev/loop7 test-file1
dd if=/dev/zero of=/dev/loop7 bs=1024 count=10 conv=notrunc
10+0 records in
10+0 records out
dd if=/dev/loop7 of=test-file4 bs=10240 count=1
1+0 records in
1+0 records out
losetup -d /dev/loop7
md5sum test-file1 >test-file2
echo "d1ebb2873aec9cee47835c45ebdbb507  test-file1" | cmp test-file2 -
cmp test-file3 test-file4
echo 12345678901234567890 | losetup -p 0 -e AES192 /dev/loop7 test-file1
dd if=/dev/zero of=/dev/loop7 bs=1024 count=10 conv=notrunc
10+0 records in
10+0 records out
dd if=/dev/loop7 of=test-file4 bs=10240 count=1
1+0 records in
1+0 records out
losetup -d /dev/loop7
md5sum test-file1 >test-file2
echo "1ed138ff36b688fd46031cb7b8a24e55  test-file1" | cmp test-file2 -
cmp test-file3 test-file4
echo 12345678901234567890 | losetup -p 0 -e AES256 /dev/loop7 test-file1
dd if=/dev/zero of=/dev/loop7 bs=1024 count=10 conv=notrunc
10+0 records in
10+0 records out
dd if=/dev/loop7 of=test-file4 bs=10240 count=1
1+0 records in
1+0 records out
losetup -d /dev/loop7
md5sum test-file1 >test-file2
echo "818fa5f242bd2ff55244acbde0e9fd8f  test-file1" | cmp test-file2 -
cmp test-file3 test-file4
rm -f test-file[1234]
*** Test results ok ***

Whoa, it seem to be ok! :-)

The question is, do I dare to use it since the "6 out of 29 hunks FAILED" 
with "make" and loop-AES seem to me a bit scary... but the "Test results ok" 
above looks nice anyway...


Best regards,

     Jonap

_________________________________________________________________
Hämta MSN Explorer kostnadsfritt på http://explorer.msn.se

-
Linux-crypto:  cryptography in and on the Linux system
Archive:       http://mail.nl.linux.org/linux-crypto/



[Index of Archives]     [Kernel]     [Linux Crypto]     [Gnu Crypto]     [Gnu Classpath]     [Netfilter]     [Bugtraq]
  Powered by Linux