In article <[EMAIL PROTECTED]>, Christoph Lameter <[EMAIL PROTECTED]> wrote: >Also check with Philip Hazel [EMAIL PROTECTED] who has done a significant >amount of research on that issue for exim. >The locking code in exim is probably the newest, most up to date code I know.
I just read the code in exim, and it does exactly the same as my code :). Only exim relies (in certain occasions, AFAIK) on the st_nlink value from stat(), which is know to be incorrect sometimes over NFS due to caching. Otherwise exim & liblockfile use the same algorithm. I see that exim also uses locking with fcntl() AFTER using a lockfile. I should probably also use that for maillock(). I must say, the exim code is extremely well commented, and the comments are very coherent. I'll probably use some of it as a base for documenting liblockfile. Mike. -- | Miquel van Smoorenburg | "I need more space" "Well, why not move to Texas" | | [EMAIL PROTECTED] | "No, on my account, stupid." "Stupid? Uh-oh.." | | PGP fingerprint: FE 66 52 4F CD 59 A5 36 7F 39 8B 20 F1 D6 74 02 | -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
