On 2013-07-26 19:19, H Hartley Sweeten wrote:
On Friday, July 26, 2013 10:59 AM, Shaun Laing wrote:
Please make sure to copy the mailing list... ;-)
I'm sorry Hartley,
I'm really new at this so there's probably something I'm doing wrong.
Do you have time to help me figure this out? Here's how I'm building
this module:
shaun@rivendell:~/src/linux/staging$ make O=../staging-build C=1
CF=-D__CHECK_ENDIAN__ drivers/staging/comedi/drivers/dt9812.o
Ah.. That's the difference.
I usually don't pass the CF=-D__CHECK_ENDIAN__ option.
With that option it appears there are a number of comedi drivers that
have some endian issues.
Your patch does fix these four in the dt9812 driver:
drivers/staging/comedi/drivers/dt9812.c:739:18: warning: cast to restricted __le16
drivers/staging/comedi/drivers/dt9812.c:746:19: warning: cast to restricted __le16
drivers/staging/comedi/drivers/dt9812.c:753:27: warning: cast to restricted __le16
drivers/staging/comedi/drivers/dt9812.c:760:18: warning: cast to restricted __le32
So, FWIW:
Reviewed-by: H Hartley Sweeten <hsweeten@xxxxxxxxxxxxxxxxxxx>
These endian warnings still exist in the driver:
drivers/staging/comedi/drivers/dt9812.c:260:17: warning: incorrect type in assignment (different base types)
drivers/staging/comedi/drivers/dt9812.c:260:17: expected unsigned int [unsigned] [usertype] cmd
drivers/staging/comedi/drivers/dt9812.c:260:17: got restricted __le32 [usertype] <noident>
drivers/staging/comedi/drivers/dt9812.c:261:39: warning: incorrect type in assignment (different base types)
drivers/staging/comedi/drivers/dt9812.c:261:39: expected unsigned short [unsigned] [usertype] address
drivers/staging/comedi/drivers/dt9812.c:261:39: got restricted __le16 [usertype] <noident>
drivers/staging/comedi/drivers/dt9812.c:263:40: warning: incorrect type in assignment (different base types)
drivers/staging/comedi/drivers/dt9812.c:263:40: expected unsigned short [unsigned] [usertype] numbytes
drivers/staging/comedi/drivers/dt9812.c:263:40: got restricted __le16 [usertype] <noident>
drivers/staging/comedi/drivers/dt9812.c:284:17: warning: incorrect type in assignment (different base types)
drivers/staging/comedi/drivers/dt9812.c:284:17: expected unsigned int [unsigned] [usertype] cmd
drivers/staging/comedi/drivers/dt9812.c:284:17: got restricted __le32 [usertype] <noident>
drivers/staging/comedi/drivers/dt9812.c:308:17: warning: incorrect type in assignment (different base types)
drivers/staging/comedi/drivers/dt9812.c:308:17: expected unsigned int [unsigned] [usertype] cmd
drivers/staging/comedi/drivers/dt9812.c:308:17: got restricted __le32 [usertype] <noident>
drivers/staging/comedi/drivers/dt9812.c:329:17: warning: incorrect type in assignment (different base types)
drivers/staging/comedi/drivers/dt9812.c:329:17: expected unsigned int [unsigned] [usertype] cmd
drivers/staging/comedi/drivers/dt9812.c:329:17: got restricted __le32 [usertype] <noident>
It appears they can be "fixed" by changing the types in the struct dt9812_flash_data
and struct dt9812_usb_cmd. But I'm not positive if this is the "correct" fix. Hopefully
someone with a bit more endian knowledge will comment. If they do, and this is the
right fix, feel free to add this chunk to your patch.
Regards,
Hartley
@@ -188,8 +188,8 @@ enum {
};
struct dt9812_flash_data {
- u16 numbytes;
- u16 address;
+ __le16 numbytes;
+ __le16 address;
};
#define DT9812_MAX_NUM_MULTI_BYTE_RDS \
@@ -230,7 +230,7 @@ struct dt9812_rmw_multi {
};
struct dt9812_usb_cmd {
- u32 cmd;
+ __le32 cmd;
union {
struct dt9812_flash_data flash_data_info;
struct dt9812_read_multi read_multi_info;
Yes, your additions look correct, Hartley.
--
-=( Ian Abbott @ MEV Ltd. E-mail: <abbotti@xxxxxxxxx> )=-
-=( Tel: +44 (0)161 477 1898 FAX: +44 (0)161 718 3587 )=-
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
