[PATCH] D80415: [AIX] Add '-bcdtors:all:0:s' to linker to gather static init functions
Hubert Tong via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Fri May 22 15:02:19 PDT 2020
hubert.reinterpretcast added inline comments.
================
Comment at: clang/test/Driver/aix-ld.cpp:2
+// Check powerpc-ibm-aix7.1.0.0, 32-bit. 'bcdtors' and argument order.
+// RUN: %clangxx -no-canonical-prefixes %s 2>&1 -### \
+// RUN: -Wl,-bnocdtors \
----------------
stevewan wrote:
> hubert.reinterpretcast wrote:
> > I am wondering if it makes sense to put this into `aix-ld.c`. It helps to demonstrate that the link step behaviour is not determined by the source file name.
> >
> > For example, in `hexagon-toolchain-elf.c`:
> > ```
> > 182 // RUN: %clangxx -### -target hexagon-unknown-elf \
> > 183 // RUN: -ccc-install-dir %S/Inputs/hexagon_tree/Tools/bin \
> > 184 // RUN: -mcpu=hexagonv60 \
> > 185 // RUN: %s 2>&1 \
> > 186 // RUN: | FileCheck -check-prefix=CHECK031 %s
> > ```
> Yes, it makes sense to show that, and that was my original design. I moved this to a separate file to avoid getting a warning because of the inherited `CHECK-LD32-ARG-ORDER-NOT: warning:`. I'm now thinking of two ways to incorporate this into `aix-ld.c`,
> 1. Simply remove the warning check like what's done in the example you've showed.
> 2. Change the line `CHECK-LD32-ARG-ORDER-NOT: warning:` into something like,
>
> ```
> // CHECK-LD32-ARG-ORDER-NOT: warning:
> // CHECK-LD32-ARG-ORDER: warning: treating 'c' input as 'c++' when in C++ mode, this behavior is deprecated [-Wdeprecated]
> // CHECK-LD32-ARG-ORDER-NOT: warning:
> ```
> Do we have a preference for one way over another?
I think not adding something extra is better than adding something extra that's not particularly relevant. The intended behaviour for what we're testing is the same even if the deprecated behaviour is finally actually removed (and maybe with the removal of or a change to the warning message).
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D80415/new/
https://reviews.llvm.org/D80415
More information about the cfe-commits
mailing list