random thought: sync() / fsync() / fdatasync() are like execution-serializing barriers. a lighter barrier type might be useful for things like package managers: only guarantee that all future writes happen after all previous ones, maybe with an option for specifying constraints
-
Show this thread
-
Replying to @tehjh
Seems similar to SYNC_FILE_RANGE_WAIT_BEFORE, but it's a quirky API... http://man7.org/linux/man-pages/man2/sync_file_range.2.html …
1 reply 0 retweets 1 like
Replying to @fagiolinux @tehjh
"This system call is extremely dangerous" ;) The metadata interactions make it a bit hard to use. I wish the docs were a bit more precise about what it can and what it can not be used for. Reading source doesn't guarantee behaviour won't change...
11:26 PM - 14 Apr 2018
0 replies
0 retweets
2 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.