gcc-8 now checks that the return result of the resolver function
is a function pointer matching the signature of the resolved function,
which causes cmake's test to fail. The actual implementation already did
this right.
Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxxxx>
---
CMakeLists.txt | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
This is needed for FC28, Leon can you add it to your existing PR on
this same topic?
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 0a5fe5ae3b240d..a69d64c584c6b7 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -208,7 +208,8 @@ CHECK_C_SOURCE_COMPILES("
static void do_entry(void) {}
void entry(void) __attribute__((ifunc(\"resolve_entry\")));
- static void *resolve_entry(void) {return &do_entry;}
+ typedef void (*fn_t)(void);
+ static fn_t resolve_entry(void) {return &do_entry;}
int main(int argc,const char *argv[]) { entry(); }"
HAVE_FUNC_ATTRIBUTE_IFUNC
--
2.17.0
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
