Re: [libvirt] [PATCH] lxcFreezeContainer: avoid test-after-deref of never-NULL pointer
Eric Blake wrote:
On 05/17/2010 11:53 AM, Jim Meyering wrote:
>> Very good idea.
>> Here's a patch to do that.
>> With this, "make sc_avoid_if_before_free" (part of make syntax-check)
>> will now prevent any new useless checks.
>> The above was the sole offender.
>
> Actually, this is just the tip of the iceberg.
> A couple minutes work have shown that there are many more:
>
> Running this shows there are potentially at least 100
> candidate functions:
>
> aid free|grep '^vi'
aid? Cool alias/shell script on just your machine?
It's a program from the id-utils package:
http://www.gnu.org/software/id-utils/
You run "mkid" to create an index (or "make ID" in
an automake-enabled project), then commands like
aid, lid, gid, aid, eid, etc. do useful things.
These tools used to be essential, but "git grep" and fast
disks have come a long ways to rendering "gid" unnecessary.
$ aid --help
Usage: aid [OPTION]... PATTERN...
Query ID database and report results.
By default, output consists of multiple lines, each line containing the
matched identifier followed by the list of file names in which it occurs.
-f, --file=FILE file name of ID database
-i, --ignore-case match PATTERN case insensitively
-l, --literal match PATTERN as a literal string
-r, --regexp match PATTERN as a regular expression
-w, --word match PATTERN as a delimited word
-s, --substring match PATTERN as a substring
Note: If PATTERN contains extended regular expression meta-
characters, it is interpreted as a regular expression substring.
Otherwise, PATTERN is interpreted as a literal word.
-k, --key=STYLE STYLE is one of `token', `pattern' or `none'
-R, --result=STYLE STYLE is one of `filenames', `grep', `edit' or
`none'
-S, --separator=STYLE STYLE is one of `braces', `space' or `newline' and
only applies to file names when `--result=filenames'
The above STYLE options control how query results are presented.
Defaults are --key=token --result=filenames --separator=space
-F, --frequency=FREQ find tokens that occur FREQ times, where FREQ
is a range expressed as `N..M'. If N is omitted, it
defaults to 1, if M is omitted it defaults to MAX_USHRT
-a, --ambiguous=LEN find tokens whose names are ambiguous for LEN chars
-x, --hex only find numbers expressed as hexadecimal
-d, --decimal only find numbers expressed as decimal
-o, --octal only find numbers expressed as octal
By default, searches match numbers of any radix.
--help display this help and exit
--version output version information and exit
Report bugs to bug-idutils(a)gnu.org