On Fri, Mar 9, 2018 at 4:38 PM, Linus Torvalds <torva...@linux-foundation.org> wrote: > On Fri, Mar 9, 2018 at 4:32 PM, Andrew Morton <a...@linux-foundation.org> > wrote: >> >> I wonder which gcc versions actually accept Kees's addition.
Ah, my old nemesis, gcc 4.4.4. *sob* > Note that we already do have this pattern, as seen by: > > git grep -2 __builtin_choose_expr | grep -2 __builtin_constant_p > > which show drivers/pinctrl/intel/pinctrl-intel.h, so the pattern > already exists current kernels and _works_ - it apparently just > doesn't work in slightly more complicated cases. Fun. Yeah, so all the PIN_GROUP() callers have either a literal or an array name for that argument, so the argument to __builtin_constant_p() isn't complex. git grep '\bPIN_GROUP\b' | egrep -v '(1|2|3|true|false)\)' > It's one reason why I wondered if simplifying the expression to have > just that single __builtin_constant_p() might not end up working.. Yeah, it seems like it doesn't bail out as "false" for complex expressions given to __builtin_constant_p(). If no magic solution, then which of these? - go back to my original "multi-eval max only for constants" macro (meh) - add gcc version checks around this and similarly for -Wvla in the future (eww) - raise gcc version (yikes) -Kees -- Kees Cook Pixel Security<div class="gmail_extra"><br><div class="gmail_quote">On Fri, Mar 9, 2018 at 4:38 PM, Linus Torvalds <span dir="ltr"><<a href="mailto:torva...@linux-foundation.org"; target="_blank">torva...@linux-foundation.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On Fri, Mar 9, 2018 at 4:32 PM, Andrew Morton <<a href="mailto:a...@linux-foundation.org";>a...@linux-foundation.org</a>> wrote:<br> ><br> > I wonder which gcc versions actually accept Kees's addition.<br> <br> </span>Note that we already do have this pattern, as seen by:<br> <br> git grep -2 __builtin_choose_expr | grep -2 __builtin_constant_p<br> <br> which show drivers/pinctrl/intel/pinctrl-<wbr>intel.h, so the pattern<br> already exists current kernels and _works_ - it apparently just<br> doesn't work in slightly more complicated cases.<br> <br> It's one reason why I wondered if simplifying the expression to have<br> just that single __builtin_constant_p() might not end up working..<br> <span class="HOEnZb"><font color="#888888"><br> Linus<br> </font></span></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature">Kees Cook<br>Pixel Security</div> </div> -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
