On 04/22/2009 02:09 PM, Matthew Wilcox wrote: > On Wed, Apr 22, 2009 at 10:32:59AM +0300, Boaz Harrosh wrote: >> These are actually aligned access it might be worth sacrificing a cast >> to be32/64 for sake of speed. > > "for sake of speed"? How often do you think we ask a device how large > it is? OK, that was the wrong choice of words, on my part. I meant for sake of "nobleness". I calculated as a programmer that these are aligned do I make the extra effort of stating that in code, or I get lazy because it does not matter? > How much overhead do you think is incurred by the unaligned code > if the data happens to be aligned? > Well for BE systems we are already order of magnitude faster by just using the accessors, so I guess we are already well in the "plus" ;) This is such a small matter, sorry to bother you about it. Just that it's a programming style I'm constantly debating with myself, feel free to ignore it. The patch looks very good to me as it is. Thanks Boaz -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
