RE: [tip: sched/core] sched/psi: Correct overly pessimistic size calculation

[David Laight]

From: Miles Chen
> Sent: 13 September 2019 23:43
> The following commit has been merged into the sched/core branch of tip:
> 
> Commit-ID:     4adcdcea717cb2d8436bef00dd689aa5bc76f11b
> Gitweb:        https://git.kernel.org/tip/4adcdcea717cb2d8436bef00dd689aa5bc76f11b
> Author:        Miles Chen <miles.chen@xxxxxxxxxxxx>
> AuthorDate:    Thu, 12 Sep 2019 18:34:52 +08:00
> Committer:     Ingo Molnar <mingo@xxxxxxxxxx>
> CommitterDate: Fri, 13 Sep 2019 07:49:28 +02:00
> 
> sched/psi: Correct overly pessimistic size calculation
> 
> When passing a equal or more then 32 bytes long string to psi_write(),
> psi_write() copies 31 bytes to its buf and overwrites buf[30]
> with '\0'. Which makes the input string 1 byte shorter than
> it should be.
> 
> Fix it by copying sizeof(buf) bytes when nbytes >= sizeof(buf).
> 
> This does not cause problems in normal use case like:
> "some 500000 10000000" or "full 500000 10000000" because they
> are less than 32 bytes in length.
> 
> 	/* assuming nbytes == 35 */
> 	char buf[32];
> 
> 	buf_size = min(nbytes, (sizeof(buf) - 1)); /* buf_size = 31 */
> 	if (copy_from_user(buf, user_buf, buf_size))
> 		return -EFAULT;
> 
> 	buf[buf_size - 1] = '\0'; /* buf[30] = '\0' */

Wouldn't it be better to also allow the user to pass
an unterminated string?
So leave the '-1' on the assignment to buf_size, but remove
it from the last line.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)