Menu
▾
▴
nanovm-users — Discussions regarding the usage of the NanoVM
You can subscribe to this list here.
| 2007 |
Jan
|
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(1) |
Sep
(1) |
Oct
|
Nov
(1) |
Dec
|
|---|
|
From: Thomas W. <tow...@ci...> - 2007-11-20 15:28:51
|
=0AHi, the NanoVMTool.jar in the current Nanovm_plugin.zip does not match to the l= atest NVM (nanovm-20070707.zip). This causes an ERR:K (NVMFILE: wrong nvm file version). Please do an update= to CVS /nanovm/eclipse-plugin/org.eclipse.nanovm.tool/lib/ and recreate Na= novm_plugin.zip. The file version at /nanovm/tool/ in the CVS is o.k.! You can check the rig= ht version with 'java -jar NanoVmTool.jar'. V1.5 fits, V1.4 does not fit. Thank You Thomas=0A=0A=0A-- =0Ahttp://www.westropolis.de - Kultur fuer alle, Kritik = von allen. Die Plattform fuer Kultur im Web. =0A |
|
From: Till H. / L. <li...@ha...> - 2007-09-16 19:47:57
|
Hi, a user just informed me that he can't build the NanoVMTool because he's missing the files UploaderCtBot.java, UploaderNvmCom1.java and UploaderNvmCom2.java I couldn't find them as well and they seem to be missing in the CVS. Can the author of these files please upload them to the CVS? Thanks, Till |
|
From: Till H. / L. <li...@ha...> - 2007-08-08 17:42:44
|
Hi,
i cannot reproduce your problem. My test run of your example gave me:
local method call from method 2 to 1
Remove 2 args from stack
Allocating space for 2 local(s) and 2 stack elements - 2 args
HEAP: request to steal 36 bytes
0/(sp:13) - 1a (26): iload_0 (5, 0x00000005)
1/(sp:14) - 1b (27): iload_1 (2, 0x00000002)
2/(sp:15) - 68 (104): imul(5,2) = 10
3/(sp:14) - ac (172): ireturn: stack base depth: 5
Return from method with 2 local(s) and 2 stack elements - 2 args
HEAP: request to unsteal 36 bytes
ireturn val: 10
As you see it returns 10 which is what i'd expect it to return. So it's
some incompatibility on your side. The fact that you have trouble with some of
the feature bits in the magic header id suggests that your version
of the NanoVMTool and the vm itself aren't matching. You might want
to try the CVS version.
So in order to check this, please verify that you can reproduce this behaviour
with the CVS version as well. If the problem still perists, please post the
java file of your demo together with the class file created out of this and
the nvm file. This way we can figure out which step is making the difference
(the java to class compilation, the conversion or the actual vm run).
Till
Am Mittwoch 08 August 2007 schrieb lounnaci mohamed:
> hello Mr
>
> when testing nanovm, i found some problem. the
> major one is when calling a method from the main method, the virtual
> machine will not be able to read the local variables of the called fct
> from Locals, they are always 0.
> a second one when using arrays, the VMTool generate a bad magic : 10 00 00
> be This is my test class:
>
> public class VMfile {
>
> public static int a=7;
> public static int b=3;
> public static int c=2;
>
> private static int multiply(int y,int z)
> {
> return y*z;
> }
> public static void main(String[] args) {
> int
> f=-5;
>
> int sum=a+b+f;
> int div=sum/c;
> int Res=multiply(sum,div);
> }
>
> }
>
> and this is the Debug:
>
> ->Start UART
> ->Get Magic 0x000000be
> ->vm_init() with 17 static fields
> ->heap_init()
> ->HEAP: request to steal 36 bytes
> ->calling clinit 0
> ->Running method 0
> ->Allocating space for 0 local(s) and 1 stack elements - 0 args
> ->HEAP: request to steal 2 bytes
> ->0/(sp:17) - 10 (16): bipush #7
> ->2/(sp:18) - b3 (179): putstatic #0000 -> 0007
> ->5/(sp:17) - 06 (6): iconst_3
> ->6/(sp:18) - b3 (179): putstatic #0001 -> 0003
> ->9/(sp:17) - 05 (5): iconst_2
> ->10/(sp:18) - b3 (179): putstatic #0002 -> 0002
> ->13/(sp:17) - b1 (177): return: stack base depth: 0
> ->HEAP: request to unsteal 2 bytes
> ->Running method
> 3
> ->Allocating space for 5 local(s) and 2 stack elements - 1 args
> ->HEAP: request to steal 16 bytes
> ->0/(sp:22) - 10 (16): bipush #-5
> ->2/(sp:23) - 3c (60): istore_1 (-5)
> ->3/(sp:22) - b2 (178): getstatic #0000
> ->6/(sp:23) - b2 (178): getstatic #0001
> ->9/(sp:24) - 60 (96): iadd(7,3) = 10
> ->10/(sp:23) - 1b (27): iload_1 (-5, 0xfffb)
> ->11/(sp:24) - 60 (96): iadd(10,-5) = 5
> ->12/(sp:23) - 3d (61): istore_2 (5)
> ->13/(sp:22) - 1c (28): iload_2 (5, 0x0005)
> ->14/(sp:23) - b2 (178): getstatic #0002
> ->17/(sp:24) - 6c (108): idiv(5,2) = 2
> ->18/(sp:23) - 3e (62): istore_3 (2)
> ->19/(sp:22) - 1c (28): iload_2 (5, 0x0005)
> ->20/(sp:23) - 1d (29): iload_3 (2, 0x0002)
> ->21/(sp:24) - b8 (184): invokestatic #0002
> ->local method call from method 3 to 2
> ->Remove 2 args from stack
> ->Allocating space for 2 local(s) and 2 stack elements - 2 args
> ->HEAP:
> request to steal 18 bytes
> ->0/(sp:283) - 1a (26): iload_0 (0, 0x0000)
> //First problem ->1/(sp:284) - 1b (27): iload_1 (0, 0x0000)
> ->2/(sp:285) - 68 (104): imul(0,0) = 0
> ->3/(sp:284) - ac (172): ireturn: stack base depth: 261
> ->Return from method with 2 local(s) and 2 stack elements - 2 args
> ->HEAP: request to unsteal 18 bytes
> ->ireturn val: 0
> ->24/(sp:535) - 36 (54): istore 4 (0)
> ->26/(sp:534) - b1 (177): return: stack base depth: 512
> ->Return from method with 5 local(s) and 2 stack elements - 1 args
> ->HEAP: request to unsteal 22 bytes
> ->-29/(sp:782) - 00 (0):
> nop //second
> problem ->-28/(sp:782) - 01 (1): ERR:M
>
> Regards
> lounnaci mohamed
>
>
>
>
> ___________________________________________________________________________
>__ Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo!
> Mail
|
|
From: Nils S. <ni...@ni...> - 2007-02-18 13:38:24
|
The NanoVM Team today announced the immediate availability of the NanoVM 1.5, a release containing several new features: * c't-Bot support * Floating point is now functional * 31-bit constant values * Several new op-codes implemented (lookupswitch, dup*, astore*, aload*, ifnull, ifnonnull and floatingpoint op-codes) * Built-in nanovm.lang.Math library * Built-in nanovm.util.Formatter library for formatted output of floating point, boolean and integer values with java-like syntax * Compatible with Java 5.0 * Some checks of nvm bytecode compatibility with NanoVM compile options * Bytecode may be optinal stored in flash memory Enjoy, NanoVM Team |