The group you are posting to is a Usenet group. Messages posted to this group will make your email address visible to anyone on the Internet.
Your reply message has not been sent.
Your post was successful
 |
Newsgroups: comp.lang.lisp
From: Peter Keller <psil...@merlin.cs.wisc.edu>
Date: 29 Apr 2010 17:27:14 GMT
Local: Thurs, Apr 29 2010 1:27 pm
Subject: Re: adjustable arrays
RG <rNOSPA...@flownet.com> wrote: I do not deny that, have any references? The nice thing about lisp is I > In article <4bd9afe9$0$9898$80265...@spool.cs.wisc.edu>, > Peter Keller <psil...@merlin.cs.wisc.edu> wrote: >> The context is that when writing a data array to a client, the buffer >> management layer needs to wrap a header around the data before sending it. >> I thought about not doing the wrapping at all--just write the small header >> then the payload, but then that leads to alternating writes of a very >> small amount of data and then a large amount of data. That oscillation >> sucks from a network scheduling/bandwidth perspective. I need to write >> as many bytes as I can as often as I can. > Seems to me then that what you want is not a growable array but rather can just go back into that layer and bash the code into something else with relative ease. >> I guess at this point, I'll just assemble the arrays naively and if the Speaking of memcpy, in lisp what is the equivalent of memcpy? replace? >> profiler tells me there is a problem there, I'll look into it deeper. > That sounds like a fine plan. Premature optimization and all that. Hrm... I should use that in a few places I stupidly wrote the byte copies with a do loop. Thanks for the oblique suggestsion. :) -pete You must Sign in before you can post messages.
To post a message you must first join this group.
Please update your nickname on the subscription settings page before posting.
You do not have the permission required to post.
| |||||||||||||