linux-toradex.git/arch/mips/kernel, branch v2.6.28.9

Rename old_readdir to sys_old_readdir

2009-01-18T18:43:51+00:00

commit e55380edf68796d75bf41391a781c68ee678587d upstream.

This way it matches the generic system call name convention.

Signed-off-by: Heiko Carstens 
Signed-off-by: Greg Kroah-Hartman

MIPS: Fix incorrect use of loose in vpe.c

2008-12-05T18:19:23+00:00

It should be 'lose', not 'loose'.

Signed-off-by: Nick Andrew 
Signed-off-by: Ralf Baechle

MIPS: Return ENOSYS from sys32_syscall on 64bit kernels like elsewhere.

2008-12-04T17:47:28+00:00

When the o32 errno was changed to ENOSYS, we forgot to update the code
for 64bit kernels.

Signed-off-by: David Daney 
Acked-by: Atsushi Nemoto 
Signed-off-by: Ralf Baechle

MIPS: 64-bit: vmsplice needs to use the compat wrapper for o32 and N32.

2008-12-04T17:47:28+00:00

Signed-off-by: Ralf Baechle

MIPS: o32: Fix number of arguments to splice(2).

2008-12-04T17:47:28+00:00

The syscall code was assuming splice only takes 4 arguments so no stack
arguments were being copied from the userspace stack to the kernel stack.
As the result splice was likely to fail with EINVAL.

Signed-off-by: Ralf Baechle

MIPS: Fix potential DOS by untrusted user app.

2008-12-04T17:47:26+00:00

On a 64 bit kernel if an o32 syscall was made with a syscall number less
than 4000, we would read the function from outside of the bounds of the
syscall table.  This led to non-deterministic behavior including system
crashes.

While we were at it we reworked the 32 bit version as well to use fewer
instructions.  Both 32 and 64 bit versions are use the same code now.

Signed-off-by: Vlad Malov 
Signed-off-by: David Daney 
Signed-off-by: Ralf Baechle

MIPS: csrc-r4k: Fix spelling mistake.

2008-11-20T19:42:33+00:00

Signed-off-by: Ralf Baechle

MIPS: Sort out CPU type to name translation.

2008-10-30T14:44:34+00:00

As noticed by David Daney , the old long switch
statement did not comply with the Linux C coding style.  It was also yet
another place of code to be changed when adding a new processor type
leading to annoying bugs for example in /proc/cpuinfo.

Fixed by moving the setting of the CPU type string into the core of the
probing code and a few BUG_ON() test to ensure the CPU probing code indeed
did its job and removing multiple now redundant tests.

Signed-off-by: Ralf Baechle

MIPS: Probe for watch registers on cores of all vendors, not just MTI.

2008-10-30T14:44:34+00:00

Signed-off-by: Ralf Baechle 
Acked-by: David Daney

MIPS: Switch FPU emulator trap to BREAK instruction.

2008-10-30T14:44:34+00:00

Arguably using the address error handler has always been ugly.  But with
processors that handle unaligned loads and stores in hardware the
current mechanism ceases to work so switch it to a BREAK instruction and
allocate break code 514 to the FPU emulator.

Yoichi Yuasa provided a build fix for CONFIG_BUG=n.

Signed-off-by: Ralf Baechle 
Signed-off-by: Yoichi Yuasa