home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.acorn.tech
- Path: sparky!uunet!pipex!pavo.csi.cam.ac.uk!idg10
- From: idg10@cl.cam.ac.uk (I.D. Griffiths)
- Subject: Re: ARM risc speed?
- Message-ID: <1993Jan26.121706.4309@infodev.cam.ac.uk>
- Sender: news@infodev.cam.ac.uk (USENET news)
- Nntp-Posting-Host: stonea.cl.cam.ac.uk
- Reply-To: idg10@phx.cam.ac.uk
- Organization: U of Cambridge Computer Lab, UK
- References: <1993Jan20.151326.23097@infodev.cam.ac.uk> <1993Jan20.163935.29452@dcs.warwick.ac.uk> <1993Jan21.114022.5930@cs.nott.ac.uk> <2804@eagle.ukc.ac.uk>
- Date: Tue, 26 Jan 1993 12:17:06 GMT
- Lines: 39
-
- In article <2804@eagle.ukc.ac.uk>, spt1@ukc.ac.uk (S.P.Thomas) writes:
- |>....
- |> In a similar vain, one of the optimisations I do is to cause the
- |> code to be generated such that the starting addresses of the
- |> code sequences that can be jumped to are aligned to have an address
- |> of &XXXXXX4. This maximises the number of consequtive (sp?) sequential
- |> memory cycles when using MEMC1a. I have three main questions.
- |>
- |> 1) The effect of this alignment is quite significant on an ARM2
- |> machine. Would the effect be as significant on an ARM3 (or ARM250,
- |> although I suspect the answer is yes, in this case)?
-
- Yes, very - the ARM3 fetches stuff into the cache 4 words at a time!
-
-
- |> 2) In the next generation of memory controllers, are different rules
- |> likely to apply? If so (I suspect this is highly likely), can anyone
- |> give me an idea what they might be?
-
- Very likely. Don't know any details however. ARM ltd might be able to help
- you on that one, I would guess that information on the Arm610 is available..
-
-
- |> 3) Is it possible that certain orderings of instructions are "better"
- |> than others (ie, faster), even though they achieve the same effect? For
- |> example
- |>
- |> ADR ad1,blk1 ADR ad1,blk1
- |> LDMIA ad1,{r0-r7} compared ADR ad2,blk2
- |> ADR ad1,blk2 with LDMIA ad1,{r0-r7}
- |> STMIA ad1,{r0-r7} STMIA ad2,{r0-r7}
- |>
- |> These sort of sequences occur a great deal in the code I'm generating.
- |> If there is a difference, what are the rules?
-
- The rule is to run both and time them. :-)
-
-
- Ian Griffiths
-