On Wed, Feb 11 2015, Dan Carpenter <dan.carpenter@xxxxxxxxxx> wrote: > Wow! So exciting! :) Glad you like it :) >> While developing this, I found it useful to only enable that specific >> check (both to get smatch run faster and to get less noise in the >> output), so there's also a few unrelated patches in the printf branch >> implementing that feature. > > It probably would have been better to say --enable="check_free" or > whatever. Still, I'll merge this as-is and change it later if anyone > gets around to it. > Well, I wanted to make it easy to do a range assuming one was implementing a set of related tests (--enable=40-44), but it should be rather easy to support both (just look at whether the first char of each 'range' is a letter or digit). Anyway, I purposely didn't add text to --help. >> sparse currently ignores attribute((format)), so the list of printf functions >> has been extracted with a perl script and hard-coded. Even if sparse >> understood attribute((format)), I wouldn't know how to set up a hook for >> 'call of function with this or that attribute'. > > Also probably people don't use that attribute all the time. I've been wanting to suggest adding -Wsuggest-attribute=format to the kernel's build flags for some time, but haven't gotten around to it. It also requires a somewhat recent gcc (4.8+, I think). My perl script also only picks up those with that attribute. Maybe I should simply generate a list of any function declaring a va_list on the stack and compare the two lists... > You'd have to do a add_hook(&match_call_info, FUNCTION_CALL_HOOK) and > look at the function symbol. OK. But since this works for now I'm not going to try to patch sparse to recognize the attribute. >> I don't think it's ready to be merged upstream (and whether that will >> even happen is of course entirely up to Dan), but now it's out there for >> people to play with. I have already sent patches for the four %p bugs >> found, but there may be a few more lurking in arch/<not x86>/ - I don't >> know how to pursuade the build system to go there. > > None of these patches have a Signed-off-by line. That's not really on purpose, but OTOH they were also mostly RFC. I don't usually care about the legalese - most of the stuff I put online (github or elsewhere) can be used by anyone for anything. Rasmus -- To unsubscribe from this list: send the line "unsubscribe smatch" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
