Re: Non-reproducible AIO failure
Konstantin Knizhnik <knizhnik@garret.ru>
From: Konstantin Knizhnik <knizhnik@garret.ru>
To: Dmitry Mityugov <dmitry.mityugov@gmail.com>
Cc: Thomas Munro <thomas.munro@gmail.com>, Andres Freund
<andres@anarazel.de>, Nico Williams <nico@cryptonector.com>,
Alexander Lakhin <exclusion@gmail.com>, Daniel Gustafsson <daniel@yesql.se>,
PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>,
rmt@lists.postgresql.org
Date: 2025-09-03T18:50:42Z
Lists: pgsql-hackers
On 03/09/2025 8:37 PM, Dmitry Mityugov wrote:
> Quite inspiring discussion. The patch is brilliantly good but it adds
> a bunch of explicit type casts, and it's not always easy to remember
> what cast to use in a particular case, and that may eventually lead to
> errors in the future. Just wanted to add that when 64-bit code is
> generated, uint8s are probably aligned to 64-bit boundaries for
> structures that contain other 64-bit members like pointers, and bit
> fields are not (and they may not have an address at all). To align bit
> fields properly, anonymous bit fields probably can be used, or unions
> containing the bitfields together with uint8s, to avoid the explicit
> casts, like in this example:
>
> struct Foo {
> union {
> char a:8;
> char aa;
> };
> char b:8;
> int* c;
> };
>
> Regards,
>
Size of PgAioHandle is144 bytes. I wonder how critical for us is to save
9 bytes for it (3 bytes vs 3 integers)?
Why not to use normal enums instead of bitfields and uint8 with type casts?