Bug in __copy_user
I have started to look a little bit at the LTP tests.
And one of the testcases that fails (actually it doesn't fail as it supposed to do) is the syscall getsockopt.
I think the failure is due to the copy_to_user(0, from, 4) call returns 0, which I wouldn't expect when the destination pointer is NULL.
I think the problem is in the __copy_user function in arch/mips/lib/memcpy.
It tries to handle the exception, which we get because the destination pointer is NULL, by returning the number of uncopied bytes in $a2 to the caller.
But in this case the length is only 4 bytes, and the copying is done by a single 'sw'. The problem is the length ($a2) is decreased by 4 before the 'sw' is executed.
The 'sw' fails and __copy_user terminates, but returns with $a2 = 0 (instead 4).
I thing the following patch will solve the problem.
/Carsten
--
_ _ ____ ___ Carsten Langgaard Mailto:[email protected] |\ /|||___)(___ MIPS Denmark Direct: +45 4486 5527 | \/ ||| ____) Lautrupvang 4B Switch: +45 4486 5555 TECHNOLOGIES 2750 Ballerup Fax...: +45 4486 5556 Denmark http://www.mips.com
Index: arch/mips/lib/memcpy.S
===================================================================
RCS file: /home/repository/sw/linux-2.4.18/arch/mips/lib/memcpy.S,v retrieving revision 1.1.1.1
mips/lib/memcpy.S +1 -1
@@ -248,8 +248,8 @@ 1: EXC( LOAD t0, 0(src), l_exc) ADD src, src, NBYTES - SUB len, len, NBYTES EXC( STORE t0, 0(dst), s_exc) + SUB len, len, NBYTES bne rem, len, 1b ADD dst, dst, NBYTES
Quoted 18 lines · Carsten Langgaard
I have started to look a little bit at the LTP tests.
And one of the testcases that fails (actually it doesn't fail as it supposed to do) is the syscall getsockopt.
I think the failure is due to thecopy_to_user(0, from, 4)call returns 0, which I wouldn't expect when the destination pointer is NULL.I think the problem is in the
__copy_userfunction inarch/mips/lib/memcpy.
It tries to handle the exception, which we get because the destination pointer is NULL, by returning the number of uncopied bytes in $a2 to the caller.
But in this case the length is only 4 bytes, and the copying is done by a single 'sw'. The problem is the length ($a2) is decreased by 4 before the 'sw' is executed.
The 'sw' fails and__copy_userterminates, but returns with $a2 = 0 (instead 4).
Quoted 1 line
I thing the following patch will solve the problem.
Been here, done that. In fact, I posted the following patch few days ago to the list:less_than_4units:
/*
* rem = len % NBYTES
*/
beq rem, len, copy_bytes
nop
1:EXC( LOAD t0, 0(src), l_exc)
ADD src, src, NBYTES
SUB len, len, NBYTES
-EXC( STORE t0, 0(dst), s_exc) +EXC( STORE t0, 0(dst), s_exc_p1u)
bne rem, len, 1b
ADD dst, dst, NBYTES
This patch also solves the problem for mips in 2.4/2.5 kernel. My question was about the patch for mips64 and mips in 2.2 kernel.
Shall memcpy.S from 2.4/2.5 be backported to 2.2 and mips64?
Regards,
Gleb.
Gleb,
Quoted 4 lines · Gleb O. Raiko
This patch also solves the problem for mips in 2.4/2.5 kernel. My question was about the patch for mips64 and mips in 2.2 kernel.
Shall
memcpy.Sfrom 2.4/2.5 be backported to 2.2 and mips64?
I think that would be the prefered solution as it'll be easier to maintain.
I've not received any 2.2 patches in a very long, long time - anybody actually still using it?
Ralf
Quoted 8 lines · Ralf Baechle
Gleb,
Quoted 3 lines · Gleb O. Raiko
This patch also solves the problem for mips in 2.4/2.5 kernel. My question was about the patch for mips64 and mips in 2.2 kernel.
Quoted 6 lines
Quoted 1 line
Shall
memcpy.Sfrom 2.4/2.5 be backported to 2.2 and mips64?I think that would be the prefered solution as it'll be easier to maintain.
I've not received any 2.2 patches in a very long, long time - anybody actually still using it?
Yes, I am still using 2.2. Considering pathes, I've got patches for r3k, which doesn't work with a standard 2.2 kernel. But those patches are backports from 2.4. So, I don't think they shall be applied. From other hand, I don't plan develop 'normal' patches for 2.2.
Regards,
Gleb.
From: "Ralf Baechle" <[email protected]>
Quoted 11 lines · Gleb O. Raiko
Quoted 5 lines
This patch also solves the problem for mips in 2.4/2.5 kernel. My question was about the patch for mips64 and mips in 2.2 kernel.
Shall
memcpy.Sfrom 2.4/2.5 be backported to 2.2 and mips64?I think that would be the prefered solution as it'll be easier to maintain.
I've not received any 2.2 patches in a very long, long time - anybody actually still using it?
Only a few thousand (or tens of thousands) of Sony Playstation 2 users. ;-) Most of them are running the wierd 2.2.1.?? Sony branch, but there are a few poor deluded souls tryng to update to 2.2.20...
Kevin K.
Quoted 4 lines · Kevin D. Kissell
Only a few thousand (or tens of thousands) of Sony Playstation 2 users. ;-) Most of them are running the wierd 2.2.1.?? Sony branch, but there are a few poor deluded souls tryng to update to 2.2.20...
Just in time when the Linux world has already mostly abandoned 2.2 ...
Ralf
From: "Ralf Baechle" <[email protected]>
Quoted 8 lines · Kevin D. Kissell
Quoted 5 lines
Only a few thousand (or tens of thousands) of Sony Playstation 2 users. ;-) Most of them are running the wierd 2.2.1.?? Sony branch, but there are a few poor deluded souls tryng to update to 2.2.20...
Just in time when the Linux world has already mostly abandoned 2.2 ...
Well, like I said last week, we need to get the PS2 onto 2.4 ASAP, for a whole bunch of reasons...
Kevin K.
* Kevin D. Kissell <[email protected]> on Mon, Jun 24, 2002:
Quoted 8 lines
From: "Ralf Baechle" <[email protected]>
Quoted 7 lines · Kevin D. Kissell
Quoted 5 lines
Only a few thousand (or tens of thousands) of Sony Playstation 2 users. ;-) Most of them are running the wierd 2.2.1.?? Sony branch, but there are a few poor deluded souls tryng to update to 2.2.20...
Quoted 5 lines
Quoted 1 line
Just in time when the Linux world has already mostly abandoned 2.2 ...
Well, like I said last week, we need to get the PS2 onto 2.4 ASAP, for a whole bunch of reasons...
ps2hacking.sourceforge.net - broken port of Sony's 2.2.1 kernel to 2.4.8 (or so).
M. R.