[PATCH] Fix a crash that occurs when PWD is invalid.
echristo at gmail.com
Mon Dec 9 14:50:50 PST 2013
On Mon, Dec 9, 2013 at 1:58 PM, Filip Pizlo <fpizlo at apple.com> wrote:
> On Dec 9, 2013, at 12:42 PM, Andrew Trick <atrick at apple.com> wrote:
> On Dec 9, 2013, at 12:01 PM, Eric Christopher <echristo at gmail.com> wrote:
> Out of curiosity why isn't it able to get a compilation dir?
> Probably one of these.
> The getcwd() function will fail if:
> [EACCES] Read or search permission was denied for a component
> of the pathname.
> [ENOENT] A component of the pathname no longer exists.
> Phil might know for sure.
> I'm not sure, actually. ;-) When I hit the bug it was probably ENOENT
> caused by weirdness in our JSC test harness.
> with $PWD set to anything remotely sensible. Moreover, it's my
> understanding that libraries and frameworks in general never assume anything
> about the sanity of $PWD unless they are asked to do relative path IO. This
> isn't just a JSC bug: if the claim is that LLVM is a JIT library then it's
> probably best if LLVM gracefully handles not having a sane $PWD.
That's not quite what I'm talking about, FWIW I agree that you
shouldn't have to have an environment at all :)
Basically the lines of code in MCDwarf.cpp are supposed to be when
generating dwarf for assembly files and I wasn't under the impression
you were compiling .s files. :)
That code should probably have some decent fallback in it, but I don't
think a null DW_AT_compilation_dir is the correct one. It should omit
the attribute if it can't get a valid value.
> On Mon, Dec 9, 2013 at 11:41 AM, Andrew Trick <atrick at apple.com> wrote:
> MCJIT needs to be able to run in hostile environments, even when PWD
> is invalid. There's no need to crash MCJIT in this case.
> If we do want to guarantee valid CompilationDir, that should be done
> only for clients of getCompilationDir(). This is as simple as checking
> for an empty string.
> The obvious fix is to simply leave MCContext's CompilationDir empty
> when PWD can't be determined. The only behavioral difference would be
> that EmitGenDwarfInfo would output an empty string when recording the
> working dir for the assembly output.
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
More information about the llvm-commits