In message <[EMAIL PROTECTED]>, Bruce Evans writes:
>This was decreed but not agreed to. I don't use devfs and don't plan
>to use it until it works at least as well as specfs (if this is
>possible). I have noticed about 10 minor bugs in it despite only running
>it to test it every 6 months or so. Examples:
Why don't you report such findings then ?
>- file time updates are broken in several ways.
>- stat() doesn't work right on devfs directories (st_size and st_blocks
> zero for all devfs files, but should be nonzero for directories).
Where is this requirement from ?
>- pathconf() doesn't work right on devfs directrories (_PC_NAME_MAX and
> _PC_PATH_MAX are unsupported, but must be supported for directories).
OK, that's a bug.
>- acd devices still have insecure modes (world readable).
That is not a fault in DEVFS but in the driver.
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
[EMAIL PROTECTED] | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message