Jean McCormack wrote:
> be_mount.c:
>     line 348: I'm confused as to why you do the be_get_uuid call. uu  is 
> never used after it and
>        it's not clear there's any side effect. So why do so? If there's 
> a reason a comment would help.
> 
>     line 351: Do you need the if(gen_tmp_altroot) free(tmp_altroot); 
> code here?
>     line 477: ditto
> 
>    line 545: Aren't we supposed to use strncmp for security reasons?
> 
>    line 598: ditto
> 
>    line 645: Shouldn't these really be be_errno_t?
> 
>    line 679: Nit (there always has to be at least one nit). Remove the 
> get, it's already on the line above.
> 
>    line 828- 841: What happens if the strlcpy actually truncates dir?
> 
>    line 851: Is this really none? Maybe something here or in the returns 
> about tmp_mp.
> 
>    line 858  & 861: shouldn't this really be be_errno_t?  (Actually this 
> is going to be a generic comment
>        since I see it many places)
> 
>    line 867: should this be strncpy?
> 
>    line 940 & 942: should this be strncmp?
> 
>    line 1447: Huh?
> 
> 
>    line 1895: do we need to free altroot?
> 
>    line 2175: Maybe file a bug on this?
>   
> 
> be_utils.c
>     general: same comment as above for the return types. The comment 
> says be_errno_t but it returns an int and ret/err are ints. I
>        know it's really the same thing but it would be nice to see the 
> cleanup done sometime. Maybe a low level bug?
> 
>    line 1710:  continaer_ds   -> container_ds

fixed...

> 
>    line 1754: Is there a bug open on this?

No, running BE management commands is currently not supported within a zone.

> 
>     line 1755: furture -> future
> 
>     line 2478: containder ->container

#...@$^$%@ fingers... oh well at least they're consistently wrong ;-)

> 
>    cleanup:  I believe a close(fd) should be there.

what line?

Thanks!
-evan

Reply via email to