[cfe-commits] r153854 - /cfe/trunk/lib/Driver/Tools.cpp
dblaikie at gmail.com
Wed Apr 4 09:54:24 PDT 2012
On Tue, Apr 3, 2012 at 11:13 PM, Bob Wilson <bob.wilson at apple.com> wrote:
> On Apr 1, 2012, at 4:47 PM, David Blaikie wrote:
>> On Sun, Apr 1, 2012 at 4:03 PM, Bob Wilson <bob.wilson at apple.com> wrote:
>>> Author: bwilson
>>> Date: Sun Apr 1 18:03:29 2012
>>> New Revision: 153854
>>> URL: http://llvm.org/viewvc/llvm-project?rev=153854&view=rev
>>> Ignore unused --serialize-diagnostics option when using llvm-gcc.
>>> llvm-gcc doesn't handle --serialize-diagnostics so when compiling i386
>>> kernel/kext code with -Werror, you get an error about that option being
>>> unused. Claim the argument to prevent this from breaking builds.
>>> Modified: cfe/trunk/lib/Driver/Tools.cpp
>>> URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Driver/Tools.cpp?rev=153854&r1=153853&r2=153854&view=diff
>>> --- cfe/trunk/lib/Driver/Tools.cpp (original)
>>> +++ cfe/trunk/lib/Driver/Tools.cpp Sun Apr 1 18:03:29 2012
>>> @@ -3619,6 +3619,9 @@
>>> assert(Inputs.size() == 1 && "Unexpected number of inputs!");
>>> + // Silence warning about unused --serialize-diagnostics
>>> + Args.ClaimAllArgs(options::OPT__serialize_diags);
>> Should this be in the big table at Tools.cpp:3208?
> I'm hardly an expert on the driver, but no, I don't think so. That big table is a list of cc1 options that are not supported by llvm-gcc and need to be filtered out of the cc1 command line. This is for a driver option that is never passed to cc1 regardless. It doesn't need to be filtered out -- it just needs to be claimed to avoid unwanted warnings.
Hmm, fair enough - I'm certainly not an expert in this area either &
only made the suggestion based on a few heuristics. Among other things
I would've expected this not to be the only option that needs to be
handled this way, so it's a bit surprising that it's treated alone
More information about the cfe-commits