This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] compile: Use libcc1.so->libcc1.so.0
- From: Pedro Alves <palves at redhat dot com>
- To: Jan Kratochvil <jan dot kratochvil at redhat dot com>, Phil Muldoon <pmuldoon at redhat dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Thu, 23 Apr 2015 12:59:39 +0100
- Subject: Re: [PATCH] compile: Use libcc1.so->libcc1.so.0
- Authentication-results: sourceware.org; auth=none
- References: <20150421213616 dot 14023 dot 38329 dot stgit at host1 dot jankratochvil dot net> <55380F04 dot 9050909 at redhat dot com> <20150423052909 dot GA18986 at host1 dot jankratochvil dot net> <5538CF08 dot 60801 at redhat dot com> <20150423114745 dot GA5268 at host1 dot jankratochvil dot net>
On 04/23/2015 12:47 PM, Jan Kratochvil wrote:
> On Thu, 23 Apr 2015 12:52:56 +0200, Phil Muldoon wrote:
>> On 23/04/15 06:29, Jan Kratochvil wrote:
>>>
>>> So you request forward/backward compatibilities, specifically:
>>>
>>> (1) Do you request future gdb-7.10 is compatible with existing gcc-5.x?
>>>
>>> (2) Do you request future gcc-6.0 is compatible with existing gdb-7.9?
>>>
>>> With an answer for (1) and (2) we can decide on how to implement it.
>>
>> Both! ;)
>
> While (1) could be possibly useful personally I do not find
> the compatibility (2) useful.
>
>
>> I don't think a version change merits that. And the change is tiny:
>> just one more parameter for a function. You could avoid it by having
>> two public methods exported in the vtable: foo (old params), foo (old
>> params, new params) and then re-factoring out the old function to
>> foo_worker_1 and have the two "foo" functions call foo_worker_1 with
>> the new parameter or NULL in its place.
>
> I do not see so clear how to implement it.
>
> If GDB changes:
> gdb/compile/compile-c-support.c:110: context = (*func) (GCC_FE_VERSION_0, GCC_C_FE_VERSION_0);
> ->
> gdb/compile/compile-c-support.c:110: context = (*func) (GCC_FE_VERSION_1, GCC_C_FE_VERSION_0);
> then compatibiity (1) is violated. Besides that you said no new API version
> should be introduced.
Here's what I suggest. I'm not sure whether it should be GCC_FE_VERSION_0 or
GCC_C_FE_VERSION_0 that should be bumped in this case, but, assuming the former,
you'd solve this by first making (and documenting) the vtable of v1 compatible
with v0. Then have gdb do:
context = (*func) (GCC_FE_VERSION_1, GCC_C_FE_VERSION_0);
if (context == NULL)
context = (*func) (GCC_FE_VERSION_0, GCC_C_FE_VERSION_0);
For as long as gdb supports both v1 and v0, places that call the
v1-only functions must check that they're talking to a v1 context
before calling through the function pointer, of course.
On the plugin side do something like:
if (version == 0 || version == 1)
{
// v1 is compatible with v0, and we're
supporting v0 for a while.
context->version = version;
return context;
}
else
{
// error.
}
Thanks,
Pedro Alves