This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug target/69318] [6 regression] ICE in symtab_node::verify with -fabi-version=7 -Wabi=8 -m32


https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69318

tbsaunde at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tbsaunde at gcc dot gnu.org

--- Comment #2 from tbsaunde at gcc dot gnu.org ---
So, the direct cause is that we have two symbols
_Z18IndirectExternCallIPFviiEiEvT_T0_S3_ one for each of the template
instantiations.

Now the interesting bit is that if you compile this code as
g++ test.cc -m32 -fabi-version=7 (note there is no -Wabi=N)
then you get
cc1plus: error: ÃâËvoid IndirectExternCall(F, T, T) [with F = void
(__attribute__((stdcall, regparm(3))) *)(int, int); T = int]Ãââ conflicts with
a previous declaration
test.cc:4:6: note: previous declaration ÃâËvoid IndirectExternCall(F, T, T)
[with F = void (*)(int, int); T = int]Ãââ
 void IndirectExternCall(F f, T t1, T t2) { // { dg-warning "mangled name" }
      ^~~~~~~~~~~~~~~~~~

test.cc:4:6: note: a later -fabi-version= (or =0) avoids this error with a
change in mangling

which is a rather laime error, but at least it isn't an ICE I guess.

if you compile it as
g++ test.cc -m32 -fabi-version=7 -fabi-compat-version=8
then you get the same ICE as with -Wabi=8

So I suspect somehow the aliases from -fabi-compat-version are preventing us
from just erroring out.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]