Re: the s_lock_stuck on perform_spin_delay

Robert Haas <robertmhaas(at)gmail(dot)com> writes:

> On Mon, Jan 22, 2024 at 12:13 PM Andy Fan <zhihuifan1213(at)163(dot)com> wrote:
>> > On Mon, Jan 22, 2024 at 11:58 AM Andy Fan <zhihuifan1213(at)163(dot)com> wrote:
>> >> I get your point! Acquiring an already held spinlock in quickdie is
>> >> unlikely to happen, but since our existing infrastructure can handle it,
>> >> then there is no reason to bypass it.
>> >
>> > No, the existing infrastructure cannot handle that at all.
>>
>> Actually I mean we can handle it without 0003. am I still wrong?
>> Without the 0003, if we acquiring the spin lock which is held by
>> ourself already. VerifyNoSpinLocksHeld in SpinLockAcquire should catch
>> it.
>
> But that's only going to run in assert-only builds. The whole point of
> the patch set is to tell developers that there are bugs in the code
> that need fixing, not to catch problems that actually occur in
> production.

I see. As to this aspect, then yes.

--
Best Regards
Andy Fan