Re: pgsql: Revert online data checksum transitions

This is the goto example:
https://markshroyer.com/2012/06/c-both-true-and-false/

On Wed, Sep 16, 2026 at 12:39 PM Bryan Green <dbryan(dot)green(at)gmail(dot)com> wrote:

> If we are pulling in stdbool.h and this is C11...that is undefined
> behavior and because it wasn't validly initialized it is an
> indeterminate-value read...which means for optimization that the compiler
> is within its right to assume it is a 0 or 1 and could test it in different
> ways depending on the surrounding code (nonzero test, compare to 1, etc).
>
> On Wed, Sep 16, 2026 at 12:34 PM Jacob Champion <
> jacob(dot)champion(at)enterprisedb(dot)com> wrote:
>
>> On Wed, Sep 16, 2026 at 10:14 AM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> > However, what I'm seeing as I trace through the crash on
>> > longfin is that nulls[19] has a value that's not either 0 or 1 (not
>> > too surprising), and somehow heap_compute_data_size is interpreting
>> > that as "true" while fill_val is reading it as "false". According
>> > to my understanding of the C standard, fill_val is in the wrong here.
>>
>> Is the _Bool representation allowed to contain anything but zero/false
>> or one/true? I thought that was undefined behavior. (IIRC, mixing C++
>> bool with our C89-era bool definition suffered from similar UB
>> excitement.)
>>
>> --Jacob
>>
>>
>>