Re: FreeTDS port to VMS V9.x on x86?

On 6/24/25 7:38 PM, Arne Vajhøj wrote:

On 6/24/2025 12:25 PM, Arne Vajhøj wrote:

I don't like the #elif HAVE_VSNPRINTF code in tds_vasprintf.
>
:-)
>
The algorithm is basically to try vsnprintf with buffers
increasing in size 512, 1024, 1536, ... until there is space.
>
That is very inefficient compared to just use the return value.
>
And since it does not use va_copy, then only the first call to
vsnprintf will work.

 Even if va_copy was called and we ignore the option of
using vsnprintf return value, then +512 is not good
compared to *2. +512 with a 1 GB CLOB value would mean
2 million calls to vsnprintf.
 (I have not checked if TDS actually allow that long
statements, but +512 is a problem)

TDS packet length is negotiated but must be between 512 and 32,767 bytes:
https://learn.microsoft.com/en-us/openspecs/windows_protocols/ms-tds/c1cddd03-b448-470a-946a-9b1b908f27a7
And vasprintf appears to be used in FreeTDS only for query strings and
commands, not for data.
You aren't wrong, but improving the vasprintf implementation isn't my
highest priority; feel free to take a crack at it and submit a PR.  To
work against current sources, you either need to use autoconf with a git
checkout or get a nightly snapshot from <https://www.freetds.org>.