This is the mail archive of the
mailing list for the glibc project.
Re: [PATCH 0/2] nptl: Update struct pthread_unwind_buf
On Sun, Feb 25, 2018 at 4:53 AM, H.J. Lu <email@example.com> wrote:
> On Sun, Feb 25, 2018 at 3:59 AM, Florian Weimer <firstname.lastname@example.org> wrote:
>> * H. J. Lu:
>>> On Sun, Feb 25, 2018 at 1:26 AM, Florian Weimer <email@example.com> wrote:
>>>> * H. J. Lu:
>>>>> What have I done wrong?
>>>> My understanding (shared by Carlos as far as I know) is that you do
>>>> not need to restore the shadow stack pointer when unwinding for
>>>> cancellation or thread exit.
>>> It may be true for thread exit. Are you saying that a function will never
>>> return after cancellation point?
>> That's certainly the intent. Cancellation is supposed to be
> The question is what happens after cancellation. Can a function return
> after cancellation?
Let me rephrase. Can a function, which contains cancellation points,
return to its caller after cancellation happened?