Re: [PATCH v2 5/7] linux: Remove SIGCANCEL/SIGSETXID handling on sigprocmask

[Zack Weinberg <zackw at panix dot com>]

On Thu, Dec 12, 2019 at 2:05 PM Adhemerval Zanella
<adhemerval.zanella@linaro.org> wrote:
> On 12/12/2019 14:59, Zack Weinberg wrote:
> > On Thu, Dec 12, 2019 at 8:44 AM Adhemerval Zanella
> > <adhemerval.zanella@linaro.org> wrote:
> >> On 12/12/2019 10:12, Florian Weimer wrote:
> >>> * Adhemerval Zanella:
> >>>
> >>>> On 12/12/2019 09:54, Florian Weimer wrote:
> >>>>> * Adhemerval Zanella:
> >>>>>
> >>>>>> The sigfillset already does it, and this is the canonical way to operate
> >>>>>> on sigset_t.  The only way to actually broke this assumption is if caller
> >>>>>> initialize sigset with memset or something similar, i.e, bypassing glibc
> >>>>>> (and again this is not a valid construction).
> >
> > I think it would be appropriate for us to guarantee that `memset(s, 0,
> > sizeof(sigset_t))` has the same effect as `sigemptyset(s)`, because I
> > bet there is real code that does that, probably without realizing it's
> > technically wrong (e.g. by using memset to wipe an entire struct
> > sigaction and then not bothering to do a separate sigemptyset on
> > sa_mask, or by statically allocating a sigset_t and assuming that
> > zero-initialization will produce the same effect as sigemptyset).
> >
> > But that argument doesn't apply to `memset(s, 0xFF, sizeof(sigset_t))`.
>
> In this case we will to either keep the current semantic or remove any
> filter in sigprocmark, sigemptyset, sigfillset, sigaddset, and sigdelset.

I think I didn't explain myself clearly enough; I am _not_ pointing
out a problem with your proposed patch.  We don't support blocking
SIGSETXID and SIGCANCEL, but we don't care if someone tries to unblock
them, because they never get blocked in the first place.  So
`memset(s, 0, sizeof(sigset_t))` currently _does_ have the same effect
as `sigemptyset(s)` and your patch wouldn't change that.  And
`memset(s, 0xFF, sizeof(sigset_t))` currently _doesn't_ have the same
effect as `sigfillset(s)` and, again, your patch wouldn't change that.
And I think we need to support `memset(s, 0, sizeof(sigset_t))` but
not `memset(s, 0xFF, sizeof(sigset_t))`.

zw