Skip to contentSkip to conversation

Need correction for Acard SCSI Linux driver posted on Linuxwebsite. v2.6.12

DISCUSSION James Bottomley 1 participant 1 message Feb 18, 2007 1 message not in the archive SCSI

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): 048701c750fa$fcf13220$6200a8c0@jameshsu

James Bottomley

author maintainer
#1 lore raw
Quoted 1 line · jameshsu

Resend

I'm afraid you can't send something of this size to linux-scsi ... there's a 100k limit on the mail reflector; that's why people tend to do smaller sizes or multiple patches.

Quoted 2 lines

Please let me know if this submition is proper.
appreciate for your help!

It would still be nice to have a diff rather than the actual files.

However, looking through there are still one or two fundamental problems:

mDelay(3000); // wait for some ATAPI device ready

Can't be in the driver ... that's a busy wait for 3s which will trip all of the soft lockup detectors if we ever hit it. Since this is supposed to be a SCSI driver, what ATAPI devices are you thinking of? There's only one SCSI<->ATA bridge I know and it was used for an obscure DVD device.

Also, the type Obfuscation:

typedef unsigned char    	UCHAR;
typedef unsigned short   	USHORT;
typedef unsigned int     	UINT;
typedef unsigned long    	ULONG;
typedef unsigned char *  	PUCHAR;
typedef unsigned short * 		PUSHORT;
typedef unsigned int *   	PUINT;
typedef unsigned long *  	PULONG;
typedef void *           	PVOID;
typedef unsigned long    	ULONG_PTR;

typedef unsigned short IOPORT;

is really contrary to the linux coding style, particularly when it buys nothing other than a bigger diff in the driver.

Most of this is covered in the Documentation/CodingStyle file in the kernel tree.

James