[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[RP] [PATCH] ratpoison segfaults when getting RP_COMMAND_REQUEST without
From: |
Bernhard R. Link |
Subject: |
[RP] [PATCH] ratpoison segfaults when getting RP_COMMAND_REQUEST without RP_COMMAND. |
Date: |
Thu, 23 Nov 2006 16:30:40 +0100 |
User-agent: |
Mutt/1.5.9i |
When ratpoison gets a RP_COMMAND_REQUEST for a window without a
RP_COMMAND, it segfaults, because ratpoison returns NULL instead
of a cmdret structure. As cmdret_new is static in action.c, this
needs one of:
- new cmdret_fatal_failure() in action.c usable in events.c
(as the attached patch implements)
- cmdret_new non-static and used from events.c
- handle NULL from execute_remote_command special
- something else
So it is (at least almost) a design decision.
Shell I apply the attached patch or prepare another solution?
Hochachtungsvoll,
Bernhard R. Link
It's not a really bad bug. Whoever can DoS ratpoison by this
needs to have access to the root window, thus access to X and
may to much uglier things. But in case some other attribute
adds a attribute calles RP_COMMAND_REQUEST to the root window
by bad luck in naming, having ratpoison just segfault is not
so nice...
P.S: I've applied the following as "documentation and bugfixes":
2006-11-23 Bernhard R. Link <address@hidden>
* doc/ratpoison.texi (set winfmt): document number between % and format.
* doc/ratpoison.1 (set winfmt): complete format char description,
document number between % and format.
2006-11-21 Bernhard R. Link <address@hidden>
* add MAINTAINERCLEANFILES variables to sub Makefile.ams
* add depcomp to top-level Makefile.am's MAINTAINERCLEANFILES
* add fdl.texi to doc/Makefile.am's EXTRA_DIST
nosegfault.diff
Description: Text document
- [RP] [PATCH] ratpoison segfaults when getting RP_COMMAND_REQUEST without RP_COMMAND.,
Bernhard R. Link <=