On Fri, Feb 21, 2025 at 9:28 AM Andres Freund <and...@anarazel.de> wrote:
> The patch curently uses a hardcoded 6 for the length of MAX_BACKENDS. Does
> anybody have a good idea for how to either
>
> a) derive 6 from MAX_BACKENDS in a way that can be used in a C array size
Do we even check the binary digits? BTW I see a place in lwlock.c
that still talks about 2^23-1, looks like it is out of date. Hmmm, I
wonder if it would be better to start by declaring how many bits we
want to use, given that is our real constraint. And then:
#define PROCNUMBER_BITS 18
#define MAX_BACKENDS ((1 << PROCNUMBER_BITS) - 1)
#define PROCNUMBER_CHARS DECIMAL_DIGITS_FOR_BITS(PROCNUMBER_BITS)
... with a little helper ported to preprocessor hell from Hacker's
Delight magic[1] for that last bit. See attached. But if that's a
bit too nuts...
> b) Use a static assert to check that it fits?
Right, easy stuff like sizeof(CppString2(MAX_BACKENDS)) - 1 can only
work if the token is a decimal number. I suppose you could just use
constants:
#define MAX_BACKENDS 0x3ffff
#define PROCNUMBER_BITS 18
#define PROCNUMBER_CHARS 6
... and then use the previous magic to statically assert their
relationship inside one translation unit, to keep it out of a widely
included header.
[1]
https://lemire.me/blog/2021/06/03/computing-the-number-of-digits-of-an-integer-even-faster/
#include <stdio.h>
#define DECIMAL_DIGITS_TABLE(i) \
(i == 0 ? 9 : \
i == 1 ? 99 : \
i == 2 ? 999 : \
i == 3 ? 9999 : \
i == 4 ? 99999 : \
i == 5 ? 999999 : \
i == 6 ? 9999999 : \
i == 7 ? 99999999 : \
i == 8 ? 999999999 : 0 )
#define DECIMAL_DIGITS_FOR_BITS(bits) \
(1 + \
((9 * (bits)) >> 5) + \
(((1 << (bits)) - 1) > DECIMAL_DIGITS_TABLE((9 * (bits)) >> 5)))
#define PROCNUMBER_BITS 18
#define MAX_BACKENDS ((1 << PROCNUMBER_BITS) - 1)
#define PROCNUMBER_CHARS DECIMAL_DIGITS_FOR_BITS(PROCNUMBER_BITS)
int
main()
{
printf("PROCNUMBER_BITS = %d\n", PROCNUMBER_BITS);
printf("PROCNUMBER_CHARS = %d\n", PROCNUMBER_CHARS);
printf("MAX_BACKENDS = %x\n", MAX_BACKENDS);
for (unsigned int i = 1; i <= 32; ++i)
printf("bits = %u, max = %u, digits = %d\n",
i,
(1 << i) - 1,
DECIMAL_DIGITS_FOR_BITS(i));
}
