At least superficially this is not difficult to fix on windows. I have built LLVM34 something like 165 success 1 failed, on windows today and will attempt clang soon. I have to look at the structure of
Who prioritizes these issues? Do they get reviewed first before getting promoted to the actual official to-do list?
I’m not aware of any official review process.
And please create a bug report to track this issue. Ideally with a little bit more info. I’m guessing that he’s trying to pass something like *.cpp in order to compile all cpp files?
There is already PR17098 for it.
I think so. Surprising that is not already enabled. OK. This is a good example. Do we need the ability to handle wild cards on the command line? If so I will go ahead. So is there a code review on check ins? It is interesting there is no consensus process. Maybe there is in the negative sense. I guess if no one says DONT DO IT, then that’s a go Step one is to create a bug report.
This discussion is really out of sync
The bug report for this issue exist http://llvm.org/bugs/show_bug.cgi?id=17098
There isn’t a strict review process for all reported issues, some are discussed at length, some are fixed, some are rejected etc.
There is a strict review process for commits, nothing gets committed until it’s been reviewed. cfe-commits is used for reviews or Phabricator if you prefer web interface.