Messages in this thread | | | Date | Fri, 25 Feb 2000 04:00:12 +0000 | From | "Andrew Morton" <> | Subject | read_proc_t API |
| |
Please tell me if I've negotiated this twisty maze correctly:
The function which is passed into create_proc_read_entry() has the following API:
ret= read_proc( char *page_buffer, char **my_first_byte, off_t virtual_start, int length, int *eof, void *data);
Where:
The caller has a virtual view of the read operation. It is trying to read the window of data between virtual_start and virtual_start+length
Virtual data:
----------------------------------------------------------- | | | | | | |a|b|c|d|e|f|g|h|i|j|k|l|m|n| | | | | | | | | | ----------------------------------------------------------- ^ virtual_start ^ virtual_start+length
But the read_proc function is passed a little page into which it has to put _some_ of the data at *virtual_start:
Page buffer:
------------------------------- | | |a|b|c|d|e|f|g| | | | | | | ------------------------------- ^ my_first_byte ^my_first_byte+ret
So read_proc()'s responsibility is to grab _some_ of the data which starts at 'a' (up to 'length') and to put it _somewhere_ in the page_buffer. It puts a pointer to the first byte into *my_first_byte and returns the number of bytes written to the caller.
If the byte 'g' was the last one which read_proc() wants to produce then it sets *eof to 1.
I think I've convinced myself. Now, what does 'data' do?
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |