Re: [PATCH] bms_prev_member() can read beyond the end of the array of allocated words - Mailing list pgsql-hackers

> On Aug 15, 2025, at 12:37 AM, David Rowley <dgrowleyml@gmail.com> wrote:
>
> On Fri, 15 Aug 2025 at 15:24, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>
>> David Rowley <dgrowleyml@gmail.com> writes:
>>> I'm happy to push Greg's v5 patch if you have no counterarguments.
>>
>> In the end this isn't something I find worth arguing about.  If
>> you prefer v5, sure.  I do suggest though that if we're installing
>> Asserts at all, defending against prevbit < -1 is worth doing.

Agreed, that's a reasonable addition.

> Agreed about defending against prevbit < -1. I added an Assert for
> that. Technically, that Assert could be up above the if (a == NULL)
> check, but I didn't think it mattered that much and opted to keep both
> Asserts together. The difference being that bms_prev_member(NULL, -2)
> will return -2 rather than Assert fail. I'm not too worried about
> that, but if you feel strongly differently, I can adjust what I just
> pushed.

Thanks for adding that Assert.  I don't think it's worth relocating at
this time.

I'll post the addition of tests in a new thread.  I hope you both have
interest and time to review that as well.

> David

Thanks David for pushing the commit! :)

best.

-greg