[TODO] Memory waste on generated FFI code
Reported by Tilman Sauerbeck | December 9th, 2007 @ 02:06 PM | in 1.1
atm we're allocating 4K for the buffers in which we stuff the generated FFI code.
This bug is a reminder that this needs to fixed one day ;)
FWIW, 64 bytes per function seem to be enough on this Linux/x86 box.
We'd want to use a memory pool there to keep memory fragmentation down.
Comments and changes to this ticket
-
Tilman Sauerbeck December 16th, 2007 @ 03:25 PM
- → Assigned user changed from “” to “Tilman Sauerbeck”
I'm working on a nice fix ;)
-
Eero Saynatkari January 20th, 2008 @ 01:29 PM
- → Title changed from “Memory waste on generated FFI code” to “[TODO] Memory waste on generated FFI code”
This can be just flat-out reduced from 4k. To optimize, need a few runs to determine. Tilman was working on a memory pool earlier, I think?
-
Tilman Sauerbeck January 21st, 2008 @ 08:58 AM
- → Assigned user cleared.
Doesn't look like I'll work on this anytime soon :>
-
Ryan Davis February 29th, 2008 @ 03:20 PM
- → Assigned user changed from “” to “Evan Phoenix”
-
Ryan Davis February 29th, 2008 @ 02:58 PM
- → Milestone changed from “1.0 preview” to “1.1”
-
Eero Saynatkari March 31st, 2008 @ 04:38 PM
- → State changed from “open” to “resolved”
- → Assigned user changed from “Evan Phoenix” to “Eero Saynatkari”
Libffi should have gotten rid of this.
Please Login or create a free account to add a new comment.
You can update this ticket by sending an email to from your email client. (help)
Create your profile
Help contribute to this project by taking a few moments to create your personal profile. Create your profile »
Evan Phoenix
Ragnar Dahlén