Capsicumize uudecode/b64decode

Authored by allanjude on Aug 26 2016, 4:28 AM.



Special handling for -o, open the fd before entering the sandbox, then set the mode on it later (the intended mode is specified in the input)

Diff Detail

allanjude retitled this revision from to Capsicumize uudecode/b64decode.Aug 26 2016, 4:28 AM
allanjude updated this object.
allanjude edited the test plan for this revision. (Show Details)
allanjude added reviewers: emaste, oshogbo.
allanjude added inline comments.Aug 26 2016, 4:29 AM

missing space


leftover debugging

Some comments.
Later in a day I will review this more ;)


This can fail.


malloc can fail.


strdup as well.


This can be done in oneline.

allanjude updated this revision to Diff 19805.Aug 29 2016, 11:33 PM

Updated with feedback from oshogbo

bapt accepted this revision.Sep 18 2016, 5:43 PM
This revision is now accepted and ready to land.Sep 18 2016, 5:43 PM
cem added a comment.Sep 18 2016, 5:53 PM

Just some high level comments. I think if the diff is smaller, it's easier to review the rest.


Why have this at all?

We keep a dirfd around which can open outfile anyway — no need to open the file before entering sandbox.

It duplicates the safe-output-file checking logic below, except for potential time-of-check vs time-of-use attacks. IMO, this whole function can be removed.


I think in general we should preopen an array of fds rather than FILE objects, then fdopen as needed. (The concern is that the overhead of FILE objects vs fds is meaningful.)

One potential cost is the size of the stdio buffering. I guess I would be a little shocked if FILE prefetched file contents or allocated much memory on open. But even so a FILE is bulkier than just an fd.


I think you could avoid a lot of unnecessary code churn of this patch by setting globals infile, infp, (as is the existing pattern in this code) rather than passing these things as parameters.

allanjude updated this revision to Diff 21338.Oct 13 2016, 2:44 AM

Use new capsicum_helpers.h

Greatly reduce the diff as suggested by cem

Address a bug where input is provided via stdin

This revision now requires review to proceed.Oct 13 2016, 2:44 AM
allanjude added inline comments.Oct 13 2016, 2:45 AM

should be place limits on this directory FD?


do we need to limit FDs opened within the sandbox?

cem added inline comments.Oct 13 2016, 3:16 AM

Yep. Whatever rights you want openated fds to have, plus CAP_LOOKUP.


I think openat from a restricted dirfd inherits dirfd's restrictions. Not 100% sure of this.

Even if so, maybe child fds don't need CAP_LOOKUP.

cem added a comment.Oct 13 2016, 3:18 AM

LGTM. I'd like to see restriction on cwdfd and openated child fds (if they don't inherit restrictions).


And rights for fstatat and unlinkat.

This revision is now accepted and ready to land.Jun 12 2017, 7:22 PM