linux.git/drivers/staging/cxt1e1/linux.c, branch v3.10

staging: cxt1e1: buffer overflow in do_del_chan()

2013-01-25T19:21:26+00:00

If we don't restrict "cp.channum" to 3 digits then the sprintf() will
overflow.  I've added a check and changed the sprintf() to snprintf().

Signed-off-by: Dan Carpenter 
Signed-off-by: Greg Kroah-Hartman

staging: cxt1e1: solve coding style problem

2012-09-04T21:27:42+00:00

remove the spaces and replace with tabs at the beginning of a line
and also remove space between function call and its open paranthesis

Signed-off-by: Devendra Naga 
Signed-off-by: Greg Kroah-Hartman

staging: cxt1e1: use kernel's way of returning error codes

2012-09-04T21:27:41+00:00

The error codes the kernel functions return are -ve numbers, convert this
function to follow the other kernel functions

Signed-off-by: Devendra Naga 
Signed-off-by: Greg Kroah-Hartman

staging: Add module.h to more drivers implicitly using it.

2011-10-31T23:32:10+00:00

This 2nd batch of implicit module.h users only appeared when we
removed the unnecessary module.h from include/linux/miscdevice.h
[The 1st batch is already present in Greg's staging tree.]

Signed-off-by: Paul Gortmaker

staging: irq: Remove IRQF_DISABLED

2011-09-16T18:09:38+00:00

This flag is a NOOP and can be removed now.

Signed-off-by: Yong Zhang 
Signed-off-by: Greg Kroah-Hartman

staging: Fix some incorrect use of positive error codes.

2011-01-21T20:35:50+00:00

Use -E... instead of just E... in a few places where negative error
codes are expected by a functions callers.  These were found by grepping
with coccinelle & then inspecting by hand to determine which were bugs.

The staging/cxt1e1 driver appears to intentionally use positive E...
error codes in some places, and negative -E... error codes in others,
making it hard to know which is intended where - very likely I missed
some problems in that driver.

Signed-off-by: Ralph Loader 
Signed-off-by: Greg Kroah-Hartman

Staging: cxt1e1: rename global log_level variable

2010-12-10T18:46:30+00:00

The driver should not have such a generic global variable name.

Reported-by: Zimny Lech 
Cc: Bob Beers 
Signed-off-by: Greg Kroah-Hartman

Staging: cxt1e1: multiple definition of max_mtu fix

2010-10-19T17:12:14+00:00

eliminate multiple definition of max_mtu by renaming
 with cxt1e1_ prefix, similarly rename max_mru

Signed-off-by: Bob Beers 
Signed-off-by: Greg Kroah-Hartman

Staging: cxt1e1: remove code guarded by GENERIC_HDLC_VERSION

2010-06-18T22:20:59+00:00

Current linux kernel has GENERIC_HDLC_VERSION equal to 4, no need to
have dead code in-kernel which was there for ancient kernel versions.

Signed-off-by: Jiri Kosina 
Signed-off-by: Greg Kroah-Hartman

Staging: cxt1e1: remove all code dependent on LINUX_VERSION_CODE

2010-06-18T22:16:21+00:00

Remove all code which is dead for in-kernel driver due to being
ifdefed by LINUX_VERSION_CODE.

While at it, also remove surrounding code which is commented out,
or '#if 1' nops.

Signed-off-by: Jiri Kosina 
Signed-off-by: Greg Kroah-Hartman