XC8 Compilers - Discussion

[pumper]

I installed the compiler and crack and compiler works fine but when i compile the code the licence type is : License type: w
sorry for the spell

[Ichan]

Read here.

-ichan

[max]

is there any tutorial how to start with xc8 compiler using mplab ide 8.xx

regards

[avrlover]

here you can get xc8 tutorials go to subtopic Mid-range PIC C Programming
http://www.gooligum.com.au/tutorials.html

[micropcb]

I know this topic is about xc8 but felt I should share this.
Here I found a nice intro to Creating a new XC16 project in MPLAB X
http://batchloaf.wordpress.com/2012/08/27/creating-a-new-xc16-project-in-mplab-x/


Cheers.

[ptr]

I know this topic is about xc8 but felt I should share this.

Why did you not post this here:
http://www.sonsivri.to/forum/index.php?topic=45708.0

[micropcb]

Why did you not post this here:
http://www.sonsivri.to/forum/index.php?topic=45708.0

Sorry to have missed that. I was trying to get started with mplabx and xc8 but thought the procedure was very similar to that for xc8.

[Manuel]

FYI:
 -> MPLAB® XC8 Compiler v1.21

No Info Still available on update contents...

It still Work!

Manuel.

[alichan]

Hello.

File is working with current XC8 version v1.21.

Provided my 'metal' i attach it anyway. I got some strange behaviours with other files i found in this thread (surely due to outdated), but this one worked nicely.

[Catcatcat]

update

MPLAB® XC8 Compiler v1.30

[alichan]

Hello

I tested the patch with the 1.30 version and:

-the messages about free mode, bla bla, are suppressed as expected.
-generated code (hex) is the same both in free mode as with the patch applied (last one published in this thread, working with 1.21).
 Reports from XC8 are identical. Also i compared both codes manually in a hex editor with no diferences.

Since the code is quite simple (just blinking a led) i can't do any further affirmation, because perhaps there is nothing to optimize so the generated hex files are the same in both cases.

Resuming: i don't know if the patch is still working and the PRO mode is really enabled. Looks it's not enabled.

I'll do more checks.

[metal]

this is not true; I compiled few minutes ago a code and it produces the following in free mode:

Memory Summary:
    Program space        used   195h (   405) of  2000h words   (  4.9%)
    Data space           used    23h (    35) of   170h bytes   (  9.5%)
    EEPROM space         used     0h (     0) of   100h bytes   (  0.0%)
    Data stack space     used     0h (     0) of    60h bytes   (  0.0%)
    Configuration bits   used     1h (     1) of     1h word    (100.0%)
    ID Location space    used     0h (     0) of     4h bytes   (  0.0%)

while in PRO mode it produces the following:

Memory Summary:
    Program space        used   13Dh (   317) of  2000h words   (  3.9%)
    Data space           used    20h (    32) of   170h bytes   (  8.7%)
    EEPROM space         used     0h (     0) of   100h bytes   (  0.0%)
    Data stack space     used     0h (     0) of    60h bytes   (  0.0%)
    Configuration bits   used     1h (     1) of     1h word    (100.0%)
    ID Location space    used     0h (     0) of     4h bytes   (  0.0%)

Compiler version is 1.30, actually I tested the PRO code on real HW and it works. I feel like this is a waste of time to test again.

[alichan]

Hello

As i said i had to perform more checks.

Result: patch still is working in free, std and pro mode for XC8 v1.30  

Checked generated asm files, hex files and compiler output to console.

FREE:

Code:

Memory Summary:
    Program space        used    5Dh (    93) of   800h words   (  4.5%)
    Data space           used     Bh (    11) of    80h bytes   (  8.6%)
    EEPROM space         None available
    Data stack space     used     0h (     0) of    70h bytes   (  0.0%)
    Configuration bits   used     1h (     1) of     2h words   ( 50.0%)
    ID Location space    used     0h (     0) of     4h bytes   (  0.0%)

STD:

Code:

Memory Summary:
    Program space        used    56h (    86) of   800h words   (  4.2%)
    Data space           used     Bh (    11) of    80h bytes   (  8.6%)
    EEPROM space         None available
    Data stack space     used     0h (     0) of    70h bytes   (  0.0%)
    Configuration bits   used     1h (     1) of     2h words   ( 50.0%)
    ID Location space    used     0h (     0) of     4h bytes   (  0.0%)

PRO:

Code:

Memory Summary:
    Program space        used    3Eh (    62) of   800h words   (  3.0%)
    Data space           used     Bh (    11) of    80h bytes   (  8.6%)
    EEPROM space         None available
    Data stack space     used     0h (     0) of    70h bytes   (  0.0%)
    Configuration bits   used     1h (     1) of     2h words   ( 50.0%)
    ID Location space    used     0h (     0) of     4h bytes   (  0.0%)

BTW, surfing the web i found this guy checking the optimizations www.t4f.org/articles/optimization-of-microchip-pic-xc8-compiler-in-free-and-pro-mode/, it's interesting. I have to agree with him after my own checks, the code generated in free mode is.. weird and suspicious.


Regards.

[Catcatcat]

Medicine works is 100% tested. A degree of optimization depends on the program and features utilized.

[jamesbond]

xc8 Compiler V1.31 available

[alichan]

v1.31: path tested, still working

Results:

Code:

free
----
    Program space        used    57h (    87) of   800h words   (  4.2%)
    Data space           used     Eh (    14) of    80h bytes   ( 10.9%)
    EEPROM space         None available
    Data stack space     used     0h (     0) of    70h bytes   (  0.0%)
    Configuration bits   used     1h (     1) of     2h words   ( 50.0%)
    ID Location space    used     0h (     0) of     4h bytes   (  0.0%)

std
---
    Program space        used    4Ch (    76) of   800h words   (  3.7%)
    Data space           used     Eh (    14) of    80h bytes   ( 10.9%)
    EEPROM space         None available
    Data stack space     used     0h (     0) of    70h bytes   (  0.0%)
    Configuration bits   used     1h (     1) of     2h words   ( 50.0%)
    ID Location space    used     0h (     0) of     4h bytes   (  0.0%)

pro
---
    Program space        used    1Ch (    28) of   800h words   (  1.4%)
    Data space           used     Dh (    13) of    80h bytes   ( 10.2%)
    EEPROM space         None available
    Data stack space     used     0h (     0) of    70h bytes   (  0.0%)
    Configuration bits   used     1h (     1) of     2h words   ( 50.0%)
    ID Location space    used     0h (     0) of     4h bytes   (  0.0%)

[Manuel]

FYI: released a new version Windows (x86/x64) MPLAB® XC8 Compiler v1.32

CAN NOT 'njoy! :-(

Patch anymore working...

Manuel.

[metal]

FYI: released a new version Windows (x86/x64) MPLAB® XC8 Compiler v1.32

CAN NOT 'njoy! :-(

Patch anymore working...

Manuel.

during/after installing, where does it mention that this executable is installing x64 executables?

I am asking because when I use "File for Windows" to check if this application is x64 or not, it reports the following:

Code:

file.exe "C:\Program Files (x86)\Microchip\xc8\v1.32\bin\picc18.exe"
PE32 executable for MS Windows (console) Intel 80386 32-bit

file.exe "C:\Program Files (x86)\Microchip\xc8\v1.32\bin\picc.exe"
PE32 executable for MS Windows (console) Intel 80386 32-bit

file.exe "C:\Program Files (x86)\Microchip\xc8\v1.32\bin\xc8.exe"
PE32 executable for MS Windows (console) Intel 80386 32-bit

am I missing sth here?

Your post sounds as if the patch is not working:

CAN NOT 'njoy! :-(

Patch anymore working...

Manuel.

It's annoying to post this way, what am I supposed to understand from this sentence: Patch anymore working..., you made me uninstall v1.31, download and install v1.32, test some codes to see if the license is working or not, here is what xc8 says about license (last line):

Code:

C:\Program Files (x86)\Microchip\xc8\v1.32\bin>xc8 --help
Microchip MPLAB XC8 C Compiler V1.32
Part Support Version: 1.32 (A)
Copyright (C) 2014 Microchip Technology Inc.
License type: Node Configuration

Please check before posting so you don't waste other members' time, even if you are right, add more lines so we understand what you are trying to say..

[Manuel]

Agree. Sorry!

about x32/x64...the same executable is suitable for both OS.

Anyhow, previous version location place: http://www.microchip.com/pagehandler/en-us/devtools/dev-tools-parts.html
direct link: http://ww1.microchip.com/downloads/en/DeviceDoc/xc8-v1.31-win.exe

once installed at compiling time you receive the following...
"Running this compiler in PRO mode, with Omniscient Code Generation enabled,
produces code which is typically 40% smaller than in Free mode.
The MPLAB XC8 PRO compiler output for this code could be 493 words smaller.
See http://microchip.com for more information."

so patch is out of work..

Manuel.

[Catcatcat]

Regardless of the option selected compilation v32 works only in Free Mode, the virus mutated , you need a new drug

[petarp]

Hi,
Solution for new release is:
 1. Open "C:\Program Files\Microchip\xc8\v1.32\bin\picc.exe" file with hex editor.
 2. Find these bytes "74 16 83 EC 08 56" and replace first two bytes with "90" ("90 90 83 EC 08 56").
 3. Find these bytes "74 15 83 C3" and replace first byte with "EB" ("EB 15 83 C3").
 4. Do same with "C:\Program Files\Microchip\xc8\v1.32\bin\picc18.exe" file.
 5. And use latest "xclm" file from here.

They check for modified xclm file as is in latest XC32 release.

Regards.
 

[Wannabe]

Hi,
Solution for new release is:
 1. Open "C:\Program Files\Microchip\xc8\v1.32\bin\picc.exe" file with hex editor.
 2. Find these bytes "74 16 83 EC 08 56" and replace first two bytes with "90" ("90 90 83 EC 08 56").
 3. Find these bytes "74 15 83 C3" and replace first byte with "EB" ("EB 15 83 C3").
 4. Do same with "C:\Program Files\Microchip\xc8\v1.32\bin\picc18.exe" file.
 5. And use latest "xclm" file from here.

They check for modified xclm file as is in latest XC32 release.

Regards.
 

I took the liberty of making a patch using your findings. All credits go to you, petarp
As always, antivirus may say something about the patch. I've included the dup2-project.
Link for patch included in my next message below.

[Catcatcat]

Many thanks petarp! It works!

[Wannabe]

It seems the file cgpic.exe and cgpic18.exe perform the same checks for xclm.exe .
Unsure if it has an impact on compilation.
I have included a new XC8 Patcher that take care of this, just in case.

[microcris]

Hello everyone

Hi,
Solution for new release is:
 1. Open "C:\Program Files\Microchip\xc8\v1.32\bin\picc.exe" file with hex editor.
 2. Find these bytes "74 16 83 EC 08 56" and replace first two bytes with "90" ("90 90 83 EC 08 56").
 3. Find these bytes "74 15 83 C3" and replace first byte with "EB" ("EB 15 83 C3").
 4. Do same with "C:\Program Files\Microchip\xc8\v1.32\bin\picc18.exe" file.
 5. And use latest "xclm" file from here.

They check for modified xclm file as is in latest XC32 release.

Regards.
 

How do you found those bytes, the right bytes to change?

I'm on linux and, obviously, the bytes to change are different.  

Like you said, they check for the xclm authenticity. If I make a

strace ./picc18 --help

among many things I get:

write(1, "\n", 1)     = 1
open("/opt/microchip/xc8/v1.32/bin/xclm", O_RDONLY) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\2\0\3\0\1\0\0\0@\245\4\0104\0\0\0"..., 16384) = 16384

then I can make a objdump:

Code:

objdump -D -S -r -j .text ./picc18

and I get this (posting only a small part)

./picc18:     file format elf32-i386


Disassembly of section .text:

08048130 <.text>:
 8048130:   31 ed                   xor    %ebp,%ebp
 8048132:   5e                      pop    %esi
 8048133:   89 e1                   mov    %esp,%ecx
 8048135:   83 e4 f0                and    $0xfffffff0,%esp
 8048138:   50                      push   %eax
 8048139:   54                      push   %esp
 804813a:   52                      push   %edx
 804813b:   68 90 f2 07 08          push   $0x807f290
 8048140:   68 d0 f2 07 08          push   $0x807f2d0
 8048145:   51                      push   %ecx
 8048146:   56                      push   %esi
 8048147:   68 80 fa 04 08          push   $0x804fa80
 804814c:   e8 af 6b 03 00          call   0x807ed00

But still, I can't find a relation between the bytes in the objdump and the place where the file is opened and where they make a decision if the file is original or not.

Does anyone can help in this ?