 | | Re: fast firmware downloader
|
|
(...) The URL is: (URL) in the nick of time? The original firmdl.c was dated 10/3/98. It is currently 10/3/99 (PDT at least). This was not planned! Like firmdl.c, firmdl3.c is a single file joined from smaller ones for simplicity. I will link a .tgz (...) (25 years ago, 4-Oct-99, to lugnet.robotics.rcx.legos, lugnet.robotics.rcx)
|
| |
| | Re: couldn't keep myself from it ;-)
|
|
(...) Whoa. Not so fast. My firmdl3 does the job in 44 seconds, not 53 seconds, so I think firmdl4 is a step backwards. :) Just hold on a bit while I finish packing it up. It is almost done, I promise. The new version (firmdl3) will be MPL. If you (...) (25 years ago, 4-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | couldn't keep myself from it ;-)
|
|
Hi Kekoa, I herewith jettison fastdl. - added code to remove parity and disable complements to your stub - split up firmdl2 into nice little modules - figured out how to deal with complement-disabled packets - optimized delays, fixed numerous little (...) (25 years ago, 4-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: fast firmware downloader
|
|
(...) Oops, I meant complements, not checksums, throughout the above paragraph. After reviewing my notes, it seems that the ROM properly obeys the complement flag when receiving but not when sending. When the ROM is sending a "long" message, it (...) (25 years ago, 3-Oct-99, to lugnet.robotics.rcx.legos, lugnet.robotics.rcx)
|
| |
| | RE: fast firmware downloader
|
|
(...) This is what I like so much about the "Extreme Team"...usually more than one way to skin a cat. I must have missed those posts, Kekoa. I try to use as much as possible of the ROM code in the brick too. The next step for pbFORTH is to link (...) (25 years ago, 3-Oct-99, to lugnet.robotics.rcx.legos, lugnet.robotics.rcx)
|
| |
| | Re: fast firmware downloader
|
|
(...) Just for the record, I implemented a dual-speed downloader a few months ago and mailed John Tamplin with information about it with regards to the JVM port that seems to have died. I used a different strategy to hack the dual speed downloads. (...) (25 years ago, 3-Oct-99, to lugnet.robotics.rcx.legos, lugnet.robotics.rcx)
|
| |
| | egcs-1.1.1 patch, legOS projects structure
|
|
Hi, this night is proving productive. I went ahead and tweaked egcs to produce RCX-specific interrupt handler prologues and epilogues. Apply this patch against egcs-1.1.1/gcc/confi...0/h8300.c, and you will be able to create interrupt handlers the (...) (25 years ago, 3-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: fast firmware downloader - and compilers
|
|
(...) Not at all ;-) fast.srec works like this: - load new image, storing it immediately behind fast.srec - fix stack pointer, you'll start new firmware anyways - relocate a small routine immediately below SP - call that routine - relocate the (...) (25 years ago, 3-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | RE: fast firmware downloader - and compilers
|
|
I'll have to try fastdl at some point. My pbFORTH image is assembled to live at 0x8000 and be loaded just like "real" firmware. If I use fastdl, how do I have to change my firmware assembly output to accomodate it? On the subject of configurable (...) (25 years ago, 3-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: fast firmware downloader
|
|
(...) Erm, hrm, well... the reason I didn't ship it yet is that it won't compile with any released version of legOS. In order to gain full configurability, I had to set up a seperate directory structure, with its own include tree and stripped down (...) (25 years ago, 3-Oct-99, to lugnet.robotics.rcx.legos)
|
