Re: [2.6 patch] SCSI seagate.c: remove SEAGATE_USE_ASM

2007-01-22 Thread Alan
On Sun, 21 Jan 2007 20:13:00 +0100 Adrian Bunk [EMAIL PROTECTED] wrote: Using assembler code for performance in drivers might have been a good idea 15 years ago when this code was written, but with today's compilers that's unlikely to be an advantage. Besides this, it also hurts the

Re: [2.6 patch] SCSI seagate.c: remove SEAGATE_USE_ASM

2007-01-22 Thread Adrian Bunk
On Mon, Jan 22, 2007 at 03:18:41PM +, Alan wrote: On Sun, 21 Jan 2007 20:13:00 +0100 Adrian Bunk [EMAIL PROTECTED] wrote: Using assembler code for performance in drivers might have been a good idea 15 years ago when this code was written, but with today's compilers that's unlikely

Re: [2.6 patch] SCSI seagate.c: remove SEAGATE_USE_ASM

2007-01-22 Thread Alan
The C codepaths are essentially untested on this driver. Has any part of this driver ever be tested with kernel 2.6? Or compiled with gcc 4? The C code paths have never been tested at all, the asm ones certainly worked in late 2.4, but I don't; have an ISA box any more. - To unsubscribe

[2.6 patch] SCSI seagate.c: remove SEAGATE_USE_ASM

2007-01-21 Thread Adrian Bunk
Using assembler code for performance in drivers might have been a good idea 15 years ago when this code was written, but with today's compilers that's unlikely to be an advantage. Besides this, it also hurts the readability. Simply use the C code that was already there as an alternative.

[2.6 patch] SCSI seagate.c: remove SEAGATE_USE_ASM

2007-01-05 Thread Adrian Bunk
Using assembler code for performance in drivers might have been a good idea 15 years ago when this code was written, but with today's compilers that's unlikely to be an advantage. Besides this, it also hurts the readability. Simply use the C code that was already there as an alternative.