David Howells <dhowells@xxxxxxxxxx> wrote: > + data = kmap(page); > + old_fs = get_fs(); > + set_fs(KERNEL_DS); > + ret = file->f_op->write( > + file, (const void __user *) data, PAGE_SIZE, > + &pos); > + set_fs(old_fs); > + kunmap(page); Hmmm... Now I've thought about it some more, I'm not sure this is a good idea. Can this run a system out of kmap() resources? These are run by a dynamic thread pool, where each thread in the pool can do a kmap and issue a write, but the write may get delayed for a significant amount of time for a number of reasons - such as block allocation, data fetching and memory allocation. David -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
