Does it really matter which instruction the compiler chooses? Both constructs
consume the same amount of memory and execution time.
I understand your concern that potentially more time and memory could be used
if the compiler resisted the use of inc and the register was below R16, but
unless
Muahahahahha... Welcome to the asylum.
(Sorry, couldn't resist.)
Gre7g
From: Schnarrenberger, Mathias [EMAIL PROTECTED]
To: Weddington, Eric [EMAIL PROTECTED]; avr-gcc-list@nongnu.org
Sent: Wednesday, November 5, 2008 8:35:50 AM
Subject: AW: [avr-gcc-list]
From: Charalampos Alexopoulos [EMAIL PROTECTED]
To: avr-gcc-list@nongnu.org
Sent: Wednesday, October 29, 2008 12:12:00 PM
Subject: Re: [avr-gcc-list] How to handle interrupt vectors ?
I was asking for information about interrupt vector's not for ISR. Yes i
have seen the code for the ISR and
From: Dave N6NZ [EMAIL PROTECTED]
To: Gre7g Luterman [EMAIL PROTECTED]
Cc: avr-gcc-list@nongnu.org
Sent: Thursday, October 9, 2008 8:56:07 PM
Subject: Re: [avr-gcc-list] Link script question
I would take very slightly different approach. Make that a section of
it's own, and put
From: Dave N6NZ [EMAIL PROTECTED]
To: Gre7g Luterman [EMAIL PROTECTED]
Cc: avr-gcc-list@nongnu.org
Sent: Friday, October 10, 2008 11:02:21 AM
Subject: Re: [avr-gcc-list] Link script question
That would work, of course, but I'd have to designate a constant
address for the block, right
I freakin' hate Yahoo! mail. Here's the attachment I was trying to include.
/* Default linker script, for normal executables */
OUTPUT_FORMAT(elf32-avr,elf32-avr,elf32-avr)
OUTPUT_ARCH(avr:1)
MEMORY
{
text (rx) : ORIGIN = 0x00, LENGTH = 0x7D00
changes (rx) : ORIGIN =
Freakin' awesome. Thanks, tons!
Gre7g
- Original Message
From: Dave Hylands [EMAIL PROTECTED]
To: Gre7g Luterman [EMAIL PROTECTED]
Cc: avr-gcc-list@nongnu.org
Sent: Friday, October 10, 2008 2:12:21 PM
Subject: Re: [avr-gcc-list] Link script question
Gu Gre7g,
Try this version
I've got an avrgcc application that uses free Flash blocks to hold user code.
To maximize the space available for the user, I start this area at the
beginning of the first block following my application code. To locate this, I
added the following to my linker script:
--- mega32.orig 2008-10-09
- Original Message
From: Ruud Vlaming [EMAIL PROTECTED]
To: avr-gcc-list@nongnu.org
Sent: Friday, September 12, 2008 8:23:23 AM
Subject: Re: [avr-gcc-list] cbr and negative numbers
So have a look at my OS for the AVR, sources included (GPLv3)
http://www.femtoos.org/
;-)
There's an OS
--- David Brown [EMAIL PROTECTED] wrote:
You really should do some basic search, and read the
avr gcc and
avr-libc documentation before posting - you might
not find all your
answers, but it will cover most things.
http://www.nongnu.org/avr-libc/user-manual/FAQ.html#faq_reg_usage
--- Javier Almansa Sobrino [EMAIL PROTECTED]
wrote:
Hi everyone. I'm trying to simulate and trace a
program developed with
avr-gcc, but I'm not able to see the value of any
local variable with
ddd or avr-gdb.
This is a common problem that is not simply solved.
gcc is optimizing your code
Writing this sounds intriguing and amusing to me. If
you need more coders, add me to the list. I'd be
interested in contributing.
Gre7g
--- Paulo Marques [EMAIL PROTECTED] wrote:
Quoting Weddington, Eric
[EMAIL PROTECTED]:
So, I don't think I'm reinventing the wheel here.
This is getting
--- Paulo Marques [EMAIL PROTECTED] wrote:
Actually after a few more digging, we can do this
all in plain C, by
using a smaller reciprocal multiplier:
snipped
Wow, that's freakin' inspired! And you're right it
does optimize nicely. Here's a 20 byte version that
executes in 12 cycles:
--- Dmitry K. [EMAIL PROTECTED] wrote:
Alas, the usage of div() function is more space and
more time
expansive. The reason is that div() uses an another
function:
__divmodhi4(), i.e. 16-bits.
Compare two programs:
prog1: 150 bytes, 238 clocks
prog2: 212 bytes, 316 clocks
Well if
--- Eric Weddington [EMAIL PROTECTED] wrote:
The best thing to do is to go ask the TinyOS people
directly:
http://www.tinyos.net/scoop/special/support#mailing-lists
But if you find out, could you report back here? I'd
be curious to know and others may be too. It could be
kind'a neat.
Gre7g
AVR is a little-endian processor family, so the first
byte in memory of a multi-byte value is the LSB. That
would be by0 in your example since union will not
re-arrange anything on you. by3 would be the MSB.
HTH,
Gre7g
--- Royce Pereira [EMAIL PROTECTED] wrote:
Hi,
Sorry for my rusty C,
--- Royce Pereira [EMAIL PROTECTED] wrote:
What does the following warning mean?
warning: comparison is always true due to limited
range of data type
It means the if will always happen.
#define CHK_1307 0x55
uint8 twi_read(void)
if(twi_read() != ~CHK_1307) ok= 0;
CHK_1307
Can anyone tell me what I'm doing wrong? I'm creating
a .hex file with avr-objcopy.exe (like I've always
done in the past), but one of my sections is
disappearing.
From cygwin, I can type:
$ avr-objdump.exe -V
GNU objdump 2.17 + coff-avr-patch (20050630)
Copyright 2005 Free Software Foundation,
--- Gre7g Luterman [EMAIL PROTECTED] wrote:
$ avr-objcopy.exe -j .update RIO-15C/temp1.elf
temp.hex
BFD: RIO-15C/temp1.elf: warning: Empty loadable
segment detected, is this intentional ?
snipped
Sorry, I neglected to specify the file format:
$ avr-objcopy.exe -O ihex -j .update RIO-15C
Yes, I've had this problem as well, and it is a PITA!
The core of the problem is that AVR Studio can't
handle loading .elf files with multiple sections of
Flash.
The workaround is no joy either. What I do, is I use
gcc to compile the files to .hex and .eep, burn them
into the AVR, and then set
--- Tobias Pflug [EMAIL PROTECTED] wrote:
int test;
int *p;
int main()
{
test = 0;
p = test;
...
}
(gdb) print p
$1 = (int *) 0x80
(gdb) print test
$2 = (int *) 0x804020
1. Do an avr-objdump to figure out which address is
correct, 0x80 or 0x804020.
2. Look through
--- Parthasaradhi Nayani [EMAIL PROTECTED]
wrote:
Hello Group,
I am writing an assembly language function in which
I need to get the address of a sub-routine into a
register pair. How can I do it? A sample code is
given below.
func1:
nop
nop
ldi r24, lo8(label1)
ldi
--- Eric Weddington [EMAIL PROTECTED] wrote:
Can you provide your real test case instead of the
degenerate case above?
Huh? But that's the whole point of making a simple
test case. That way I don't have to make you slog
through hundreds of pages of code and I don't have to
put my customer's
I suppose I could be misunderstanding how the clobber
list of an inline asm statement is supposed to work,
but it doesn't seem to be working for me. I specify
that I'm using r28 and r29, but the code following my
asm statement continues to use Y as if it was
untouched.
To demonstrate, I've
I haven't used that part of the toolset, so I can't
speak authoritatively, but a quick websearch shows the
libusb0 is a SourceForge project. Try downloading it
at http://tinyurl.com/yo5nrq and see if that makes
things happier.
Gre7g
--- David L. Gordon [EMAIL PROTECTED] wrote:
Hello I am
--- Schwichtenberg, Knut
[EMAIL PROTECTED] wrote:
Are you using the default linker script? If so the
problem is based
there.
The following copy is from the avr5.x file:
} text
.data : AT (ADDR (.text) + SIZEOF (.text))
{
Which means: The .data segment starts after the
I'm using the latest WinAVR and I'm trying to fit my
code into the tiny atmega8515. The code does fit, but
I am trying to force certain sections of code to be
hard-coded to certain locations.
I am trying to do this with the --section-start option
when linking. This technique worked fine with the
--- Dave Hylands [EMAIL PROTECTED] wrote:
The times I've use section start, I've used an
address like 0x1B00
rather than just 1B00.
Many thanks, Dave, but I just tried changing that and
I still get the exact same error. At least it's
reassuring that I'm not the only one using
28 matches
Mail list logo
