Stateless and I stumbled upon this issue while discussing the
semantics of read, accepting a size_t but only being able to return
ssize_t, effectively lacking the ability to report successful
reads > SSIZE_MAX.
The discussion went along and we came to the topic of input-based
memory allocations. Basically, it was possible for the argument
to a memory-allocation-function to overflow, leading to a segfault
later.
The OpenBSD-guys came up with the ingenious reallocarray-function,
and I implemented it as ereallocarray, which automatically returns
on error.
Read more about it here[0].
A simple testcase is this (courtesy to stateless):
$ sbase-strings -n (2^(32|64) / 4)
This will segfault before this patch and properly return an OOM-
situation afterwards (thanks to the overflow-check in reallocarray).
[0]: http://www.openbsd.org/cgi-bin/man.cgi/OpenBSD-current/man3/calloc.3
1) don't mix declarations and code (leave recursion alone for now as I
plan on changing/using recurse)
2) change **argv to *argv[]
3) check for error on fork()
- Make globals static
- Fix a comment
- Change some data types
- Rearrange struct members from largest to smallest
(no affect due to small structs, good practice)
Before removing bit fields:
$ size find
text data bss dec hex filename
16751 968 48 17767 4567 find
After removing bit fields:
$ size find
text data bss dec hex filename
16527 968 68 17563 449b find
This is an example where bit fields uses more memory
than integers or char. There is going to be only one
gflags struct, so the waste in instructions is bigger
than the space saved by bit fields. In the case of Permarg,
Sizearg, Execarg there is only one bit field, so at least
one unsigned is used, so there is no any gain.