This is the mail archive of the gdb-testers@sourceware.org mailing list for the GDB 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]

[binutils-gdb] Change function signature passed to clone


*** TEST RESULTS FOR COMMIT ba4dd7c4a1d99c62a1c2edd68f511a82f8fe041e ***

Author: Yao Qi <yao.qi@linaro.org>
Branch: master
Commit: ba4dd7c4a1d99c62a1c2edd68f511a82f8fe041e

Change function signature passed to clone

I see the following compile error with an old bfin-uclinux gcc to
build GDBserver,

 cc1: warnings being treated as errors
 gdb/gdbserver/../nat/linux-ptrace.c: In function 'linux_fork_to_function':
 gdb/gdbserver/../nat/linux-ptrace.c:283: error: passing argument 1 of 'clone' from incompatible pointer type

in glibc, clone's prototype is like this, and in uClibc, it is the same,

       int clone(int (*fn)(void *), void *child_stack,
                 int flags, void *arg, ...
                 /* pid_t *ptid, struct user_desc *tls, pid_t *ctid */ );

so this patch changes function signature from 'void (*function) (gdb_byte *)'
to 'int (*function) (void *)'.

Note that I find Pedro advised to change argument type from 'void *'
to 'gdb_byte *' during the patch review
https://sourceware.org/ml/gdb-patches/2013-08/msg00611.html  however,
I think fix compile error can justify the change back to 'void *'.

gdb:

2016-01-12  Yao Qi  <yao.qi@linaro.org>

	* nat/linux-ptrace.c (linux_fork_to_function): Change type
	of argument 'function'.
	(linux_grandchild_function): Change return type to 'int'.
	Change child_stack's type to 'void *'.
	(linux_child_function): Likewise.


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