In fact, my original goal was simply to bind a pre-allocated memory block to the acl_ctx so it could use that memory persistently, without reallocating and freeing it on each build. I wasn’t trying to specify any particular memory or object management strategy beyond that.

On 12/11/2025 9:46 AM, Stephen Hemminger wrote:
Rather than custom allocators, I did a couple of quick AI queries about
alternatives. It looks like there are some big global gains possible
here:


Reply via email to