Just checking, will this will make it into the next release?
Thanks, Anton
On Thu, Jun 30, 2011 at 7:34 AM, Detlev Zundel dzu@denx.de wrote:
Hi Anton,
On Wed, Jun 29, 2011 at 6:04 AM, Detlev Zundel dzu@denx.de wrote:
Hi Anton,
Previously reading or writing zero full sectors (reading the end of one sector and the beginning of the next for example) was special cased and involved stack allocating a second sector buffer. This change uses the same code path for this case as well as when there are a non-zero number of full sectors to access. The result is easier to read and reduces the maximum stack used.
It's non-trivial to prove that your change is equivalent and unfortunately I do not have enough time to do this. If your tests work, than this is certainly a good indication ;) The one thing I'd like to be sure is that the previous code looks like it used the stack for whole sectors but copied only parts of this to the provided address pointer. Your new code looks like it always writes whole sectors to the provided pointer. Is this safe even if the caller allocated space without overhead for whole sectors?
Thanks for the reviews by the way. My new version of the code still bounces partial sector reads (both at the beginning and end of the range) through a stack allocated sector buffer. The portion that is writing directly to the users buffer is only used for reading the full sectors. The middle section (in the "if (sectors > 0)" block) is reading only as many sectors as are specified by (byte_len / SECTOR_SIZE). byte_len, buf and sector at this point in the function have been updated by the first block that deals with reading the unaligned start of the data (if it exists).
Also, I have tested this code on a Tegra board using ext2ls and ext2load of a kernel image.
Thanks for the explanation. The new code certainly reads cleaner so
Acked-by: Detlev Zundel dzu@denx.de
Thanks Detlev
-- The management question ... is not _whether_ to build a pilot system and throw it away. You _will_ do that. The only question is whether to plan in advance to build a throwaway, or to promise to deliver the throwaway to customers. - Fred Brooks, "The Mythical Man Month" -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-40 Fax: (+49)-8142-66989-80 Email: dzu@denx.de