Re: [PATCH] grep: fall back to interpreter mode if JIT fails

From: Carlo Arenas <carenas@gmail.com>
Date: Dec 17, 2022, 02:50 UTC
Message-ID: <CAPUEspjqgSOS4KKw2nTaRYdiCFo4Ok6OfmKrqb+Mjq+oXn5nsg@mail.gmail.com>
In-Reply-To: <xmqqlen7kksr.fsf@gitster.g>

On Fri, Dec 16, 2022 at 3:09 PM Junio C Hamano <gitster@pobox.com> wrote:

Show 9 quoted lines

>
> Mathias Krause <minipli@grsecurity.net> writes:
>
> > ... However, from a user's point of view a fall back to
> > interpreter mode might still be desired in this case, as a failing
> > 'git grep' is simply not acceptable, IMHO.
>
> "git grep" that silently produces a wrong result (by falling back
> after a problem is detected) would not be acceptable, either.

except that an error at this point only invalidates the use of JIT, so calling pcre2_jit_match() is invalid but calling pcre2_match() is not.

the later is setup to be used later by the code that is added,

> Receiving BADOPTION could be a sign that there is something wrong in
> the input, not from the end-user but from the code, in which case
> stopping with BUG() may be a more appropriate?

The way PCRE handles this kind of errors internally is to instruct pcre2_match() to use the interpreter.

While a BUG() might be a way to ensure the code is using the right set of options I would expect that the failure will be reported by pcre2_compile instead, with the only cases left, only being interna to PCRE (ex: JIT can't yet support a feature the interpreter has)

Carlo