http://llvm.org/bugs/show_bug.cgi?id=9232
Ted Kremenek <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED --- Comment #18 from Ted Kremenek <[email protected]> 2011-03-16 16:13:47 CDT --- (In reply to comment #17) > Created an attachment (id=6324) --> (http://llvm.org/bugs/attachment.cgi?id=6324) [details] > updated diff against TOT Applied in r127758. > > I'm sorry, I thought that I had tried this with a clean script and it > still failed. I was especially thrown off by the error, which made it > seem like '-Xclang' is an unsupported option, when in reality it was > '-Xclang -o' that got parsed as unknown option. Could you check that > code-path once more? Is it printing the correct offending arg? Everything is working fine for me with your patch. > > > I did it mostly because I was lazy and it got the job done. That should teach > me ... No worries. If you want to disable --analyzer-display-progress, the way to do this is to: 1) add an option to scan-build, which sets an environment variable read by ccc-analyzer. 2) ccc-analyzer consults the environment variable to determine if it should not add -analyzer-display-progress to the clang -cc1 line. There are numerous examples of this kind of argument passing in scan-build and ccc-analyzer. > > I'd also like to give the user a knob to turn off the (usually benign) > DeadStore reports. Perhaps I'll get to it later this week. We recently landed changes in clang to disable/enable specific checks. We know need to forward those controls to scan-build. -- Configure bugmail: http://llvm.org/bugs/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug. _______________________________________________ LLVMbugs mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/llvmbugs
