On Fri, Aug 13, 2010 at 03:38:40PM +0200, Dan Carpenter wrote:
> On Wed, Aug 11, 2010 at 10:46:47PM -0400, Jarod Wilson wrote:
> > On Tue, Aug 10, 2010 at 08:11:20AM +0200, Dan Carpenter wrote:
> > > We never use control_req so I removed it.
> > >
> > > Signed-off-by: Dan Carpenter <error27@xxxxxxxxx>
> >
> > Leftover cruft from when lirc_imon also supported the devices now covered
> > by the ir-core imon driver.
> >
>
> Speak of left over stuff, it's weird that I didn't notice this before
> but gcc complains about an unitialized variable in
> imon_incoming_packet().
>
> drivers/staging/lirc/lirc_imon.c: In function ‘imon_incoming_packet’:
> drivers/staging/lirc/lirc_imon.c:661: warning: ‘chunk_num’ may be used
> uninitialized in this function
>
> I don't know how to fix that, but it looks important.
Ew. Yeah, that doesn't look so hot like it is right now. The old lirc_imon
driver had chunk_num = buf[7], and made much more extensive use of
chunk_num. Simply removing chunk_num and using buf[7] should be fine.
Signed-off-by: Jarod Wilson <jarod@xxxxxxxxxx>
---
drivers/staging/lirc/lirc_imon.c | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/staging/lirc/lirc_imon.c b/drivers/staging/lirc/lirc_imon.c
index 6649325..2351c88 100644
--- a/drivers/staging/lirc/lirc_imon.c
+++ b/drivers/staging/lirc/lirc_imon.c
@@ -599,7 +599,7 @@ static void imon_incoming_packet(struct imon_context *context,
struct device *dev = context->driver->dev;
int octet, bit;
unsigned char mask;
- int i, chunk_num;
+ int i;
/*
* just bail out if no listening IR client
@@ -658,7 +658,7 @@ static void imon_incoming_packet(struct imon_context *context,
}
}
- if (chunk_num == 10) {
+ if (buf[7] == 10) {
if (context->rx.count) {
submit_data(context);
context->rx.count = 0;
--
Jarod Wilson
jarod@xxxxxxxxxx
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
