so lsr,ror,lsr,ror,lsr,ror,lsr,ror,lsr vs mov,andi,andi,swap,add
its good i didnt go to sleep because buddy just got here
Posted by renesis at 15:29 | permalink | 0 comments
lordpil's blog
one you do alot of swaps and masks and movs
the other you do that with variation or a bunch of rotates
so now you have 3 diff ways to code for space or speed or whatever gets you off
lsr ror are both single cycle single word
tired
mov is single cycle single word, andi is same, swap is too
Posted by renesis at 15:23 | permalink | 0 comments
but you have to copy the reg
and recover your addy byte
mine is like 9 opcodes and everything just ends up formatted
man you guys are fucked up
i swear if i start avrstudio just to start racing little bits of assembly
how does that change shit
hello you made two diff methods your way
Posted by renesis at 15:18 | permalink | 0 comments
it packs and unpacks better
fine do it that way
lsr temp2,temp
ror temp2,temp
ror temp2,temp
ror temp2,temp
ror temp2,temp
thats how you do that
actually that doesnt work
but thats basically what you do
if avr was baddass as fuck itd work tho
know he remember theres a swap
phonetic typo ftw
so mask swap and add
Posted by renesis at 15:13 | permalink | 0 comments
okay unless youre doing fault detection
or parity shit or something
you know that byte 0 is byte 0 because it came first
wow thats way fail
youre way before way better
you rotate from addy byte into data byte
23:36 < danielson> Nah, like <port # nibble, high data nibble><low data nibble, 0000>
23:40 < danielson> Like [ 1 | <0-7, port designator> | <data high nibble> ] [ 0 | 0 | 0 | 0 | <data low nibble> ]
2nd way sucks
?
what?
wtf are you on fucking crack
[ 1 | <0-7, port designator> | <data high nibble> ] [ <data low nibble> | 0 | 0 | 0 | 0 ]
Posted by renesis at 15:08 | permalink | 0 comments
im tired
i dont even get your application
but its seems like 2b worth of data
wait what
[8b data][4b addy]
why are you splitting data
youre prob just going to have to put it back together
dude get the data, store, get the addy, do
youre like, get the addy and data, rotate high bits into the high nibble of the 2nd byte
Posted by renesis at 15:03 | permalink | 0 comments
low,low2...
start those at r2
yeah i dunno why he wants too
r0 and r1 are where the mul opcode pops results out
8b x 8b multiplication
so i dont use those as general registers
also i keep most my variables in sram
so i hardly ever use low regs, and most shit dont use more then 2 temp regs
anyway, it doesnt matter on an avr much
you have alot of registers and pointers and decent sram, do it how you want
Posted by renesis at 14:58 | permalink | 0 comments
damn
temp,temp2..temp6
starting at 16
i usually use z pointer in my programs, sometimes ill use all 6 reg pairs as pointers
if youre doing circular buffers you can prob use that
Posted by renesis at 14:53 | permalink | 0 comments
dx^: yes!
i just dropped it because i didnt wanna go to school
also that other teacher could be an asshole
for how many?
will they deliver to us
they give tooling/unit costs in their quotes?
oh wow
yeh
link to tech page
Posted by renesis at 14:48 | permalink | 0 comments
i did pwn the trig midterm
\o/
wtf
why would i be taking a midterm in taig
dx^: yay
Posted by renesis at 14:03 | permalink | 0 comments
Top | Add to Technorati Favorites
© 2007 lordpil. XHTML 1.0! CSS! Site design by GNAA Blog Engine by pbx | MULTI2 | ian hanschen | lolwat