bug-libtool
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [libtool 2.2.11a] testsuite: 48 69 92 failed [cygwin]


From: Charles Wilson
Subject: Re: [libtool 2.2.11a] testsuite: 48 69 92 failed [cygwin]
Date: Sat, 19 Jun 2010 17:07:23 -0400
User-agent: Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.8.1.23) Gecko/20090812 Thunderbird/2.0.0.23 Mnenhy/0.7.6.666

On 6/19/2010 4:07 PM, Ralf Wildenhues wrote:
> Thank you for pursuing this!
> 
>> The first patch attached fixes (1), and enables this test to pass on cygwin.
> 
> That change is definitely ok!  I think you should commit this first, and
> separately from the rest.

OK, I will commit part (1) shortly.

> I think this is also ok.  I don't quite understand why the issue (2) is
> not exposed for us on Cygwin.  Please leave 48 hours for others to
> comment.

Again, I'll test on cygwin/mingw/linux before posting a "real" patch for
this part (2) -- and then wait a few days from that point.

I believe the reason we are not presently bitten on cygwin is because
g++ is being "nice" with regards to some ODR violations. Which isn't to
say that all pei-386 compilers (e.g. ones that must worry about DLLs)
will be "nice" in this way. Or that g++ will always be so nice.

But I'm not sure that MSVC is as nice as cygwin-g++ is (or as most ELF
c++ ABIs seem to be).  And using explicit declspec is clearly the
correct mechanism on this platform, regardless of all the auto-import
"help" that the GNU toolchain provides on mingw/cygwin. (All that "help"
is really a fancy way of cheating to make our lives easier. When it
works; fortunately that's most of the time.)

> Please add a short comment to this end though.

I'll add appropriate verbiage to the .at file (as part of the (2) patch).

>> As Dave put it, C++ is complicated.  Which solution should I propose on
>> libtool-patches? (I'll work up full changelog entries, test on linux &
>> mingw, etc, once it's decided which path to pursue).
> 
> That would be nice.  My take is that you can commit this, given the
> testing, and given that you add a link to this thread or the cygwin
> discussion and put a comment in the code why the common lib is needed.

Will add this to the .at file (as part of the (2) patch, which
implements the common lib stuff).

--
Chuck





reply via email to

[Prev in Thread] Current Thread [Next in Thread]