Compiling stuff for Prizm

[Kristaba]

Hey guys
I saw nobody here posted a real tutorial about how to build stuff for SH3 (and later for Prizm itself)

Simon Lothar and Andreas Bertheussen, two casio hacker, made some years ago a good reverse engineering of fx9860 calculator, and they wrote some tools
for, and some informations about how to use GCC to make G1A Addin :
http://sourceforge.net/apps/trac/fxsdk/wiki/UsingGCC

I advise you to subscribe and download the Kpit GNU tools (most recent and up-to-date) at http://www.kpitgnutools.com/index.php
They have a strange policy : they allow to download freely the code (because GCC is under GNU GPL license of course ) but you need to subscribe to download the binaries, so I think it could be a good thing to distribute ourselves the binaries from their source in the future...

The link I give is about fx9860, so they are some thing to do before use this for the Prizm :
* In addin.ld change "rom  : o = 0x00300200" by "rom  : o = 0x00300000"
* crt.s need to be modified (we need to to do a reverse-engineering on the Prizm initilization code)
* a g3a wrapper

Edit :
Okay, so I disassembled Converter addin this morning (downloaded on casio.edu) to try to isolate the init stuff.
The complete disassembled sources (all conv.g3a), with only some labels to understand more the code (because 200 lines of assembly without any label to indicate constant values, loops and functions is really hard to read) is here http://pastebin.com/xT8XNAhH

I rewrote the code that seems to be the init stuff, and as I expected it's very similar to the fx9860 crt0.s written by SimLo and/or Andreas, but it seems call only _GLibAddinAplExecutionCheck after the ram is initialized, and immediately jump to _main...
So it's either the Prizm OS does a part of the initialization that was did by the addin code in fx9860 OS, or a part of the Converter code I thought it's the "_main" code is in fact init code.
Maybe Simon Lothar already isolate the init stuff, so, if you see this, please say what you think about that...
crt0.s : http://pastebin.com/ccE3NPhZ
my addin.ld : http://pastebin.com/YMrdDHWS

(and really sorry for my poor english )

[JosJuice]

Great! I can't really check this out now, since I don't have a Prizm yet... But it certainly does look nice. Is this mostly for assembly, or is it for C?

* In addin.ld change "rom  : o = 0x00300200" by "rom  : o = 0x00300000"

Shouldn't this be 0x00307000 on the Prizm?

[Kristaba]

Great! I can't really check this out now, since I don't have a Prizm yet... But it certainly does look nice. Is this mostly for assembly, or is it for C?

* In addin.ld change "rom  : o = 0x00300200" by "rom  : o = 0x00300000"

Shouldn't this be 0x00307000 on the Prizm?

Yeah, I know the problem because I haven't my one yet too
It's a fully GCC so you can compile a lot of languages, mainly asm and C, but also C++ (with some change to remove the runtime), Ada, Fortan, etc...

And no, it seems Prizm set the MMU to 0x300000 as the first code address (Andreas wrote this in Casio Kingdom same days ago)

Edit 02/16/11 :
First post updated

[DJ Omnimaga]

Hmm nice, hopefully this could be useful, since both calcs are similar processor-wise.

[SimonLothar]

I attached the current version of insight (see directory projects\insight\insight.g3a). The source and the tools to build the g3a are included, too. Only one exception: the BIN-directory of the legacy CASIO fx-9860 SDK, which is needed, is not included.

EDIT:
New version: http://www.omnimaga.org/index.php?topic=6608.msg128206#msg128206
version 1.01 removed.

[DJ Omnimaga]

Nice, I just checked and I like it so far.

[SimonLothar]

Insight version 1.02:
RTC read and set; battery voltage; new syscalls; a new library fxCG_TOOLS.LIB; 256 color bitmap to fxCG icon converter; memory search; memory backup.

Outdated: update here http://www.omnimaga.org/index.php?topic=6608.msg132980#msg132980

[DJ Omnimaga]

Cool, thanks for the update.

[JonimusPrime]

I actually have my own built toolchain I built based on GCC as I didn't want to register to dl the one from that KPIT site. It can be found at http://jonimoose.net/prizm/sh3-unknown-elf/ the two tars are identical one just used xz compression to be smaller.

[DJ Omnimaga]

Cool to hear Jonimus.

[SimonLothar]

Insight version 1.03:
a little display demo: moving and bouncing objects; a HTMLHELP-file with some information concerning syscalls of fx-9860- and fx-CG-systems (to be continued).

EDIT: ah, yes. @cfxm: and a new pair of my cute icons, which everyone loves.

EDIT(2011-04-03):
Updated: http://www.omnimaga.org/index.php?topic=6608.msg139453#msg139453
version 1.03 removed

[fxdev]

I totally like the display routine!

[DJ Omnimaga]

For a bit I thought the display thing was separate from Insight and I couldn't find the file, so I had to sign up on Casio site to download the 9860G SDK and stuff, but then I realized it was part of Insight. X.x

Anyway this looks nice. I'm glad to see more showcases of SH3 ASM programs.

EDIT: Here's a screenshot (well... sort-of). Don't use Internets Exploder to ruin it.

[SimonLothar]

I totally like the display routine!

So? Then if you like it crowded, try this...

@DJ_O: it is C, not assembler!

Code: Select All | Copy To Clipboard

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144

static unsigned int lastrandom=0x12345678; unsigned short random( int seed = 0 ){     if (seed) lastrandom=seed;     lastrandom = ( 0x41C64E6D*lastrandom ) + 0x3039;     return ( lastrandom >> 16 ); } #define NSHAPES 100 // int MoveOne( TBdispFillArea a[], TShapeProps sp[], int idx ){ int vy1, vy2, j, k; int xoverlap, yoverlap, overlap; int xd, yd; a[idx].x1 += sp[idx].dx; if ( a[idx].x1 >= 384 - sp[idx].wx){ sp[idx].dx = -sp[idx].dx; a[idx].x1 += sp[idx].dx; sp[idx].dx = -1;  // cold wall return 0; } if ( a[idx].x1 <= 1){ sp[idx].dx = -sp[idx].dx; a[idx].x1 += sp[idx].dx; sp[idx].dx = 3; // hot wall return 0; } for (j=0;j<NSHAPES;j++){ overlap = 0; if ( j != idx ){ xd = a[idx].x1 - a[j].x1; yd = a[idx].y1 - a[j].y1; if ( xd * xd + yd * yd < 100 ) overlap = 1; if ( overlap ) break; } } if ( overlap ){ sp[idx].dx = -sp[idx].dx; a[idx].x1 += sp[idx].dx; k = sp[idx].dx; sp[idx].dx = sp[j].dx; sp[j].dx = -k; return 0; } a[idx].y1 += sp[idx].dy; if ( a[idx].y1 >= 216 - sp[idx].wy){ sp[idx].dy = -sp[idx].dy; a[idx].y1 += sp[idx].dy; return 0; } if ( a[idx].y1 <= 1){ sp[idx].dy = -sp[idx].dy; a[idx].y1 += sp[idx].dy; return 0; } for (j=0;j<NSHAPES;j++){ overlap = 0; if ( j != idx ){ xd = a[idx].x1 - a[j].x1; yd = a[idx].y1 - a[j].y1; if ( xd * xd + yd * yd < 100 ) overlap = 1; if ( overlap ) break; } } if ( overlap ){ sp[idx].dy = -sp[idx].dy; a[idx].y1 += sp[idx].dy; k = sp[idx].dy; sp[idx].dy = sp[j].dy; sp[j].dy = -k; return 0; } k = 0; if (sp[idx].dy<0) k -= sp[idx].dy; else k += sp[idx].dy; if (sp[idx].dx<0) k -= sp[idx].dx; else k += sp[idx].dx; switch (k){ case 2 : sp[idx].color = 0x001F ; break; case 3 : sp[idx].color = 0x03FF ; break; case 4 : sp[idx].color = 0x07E0 ; break; case 5 : sp[idx].color = 0xFFC0 ; break; case 6 : sp[idx].color = 0xF800 ; break; } a[idx].mode = 0; Bdisp_AreaClr( &sp[idx].saved, 1, -1 ); a[idx].mode = 1; a[idx].x2 = a[idx].x1 + sp[idx].wx; a[idx].y2 = a[idx].y1 + sp[idx].wy; Bdisp_AreaClr( &a[idx], 1, sp[idx].color ); memcpy( &sp[idx].saved, &a[idx], sizeof( TBdispFillArea ) ); /* vy1 = a[idx].y1 - sp[idx].wy / 2; vy2 = a[idx].y2 + sp[idx].wy / 2; if ( vy1 < 0 ) vy1 = 0; if ( vy2 > 215 ) vy2 = 215; Bdisp_PutDisp_DD_stripe( vy1, vy2 ); */ return 1; } #define V 1 void F7_Handler(){ TBdispFillArea ar[NSHAPES]; TShapeProps mp[NSHAPES]; int ticks, i, iwork; ticks = RTC_GetTicks(); random( ticks ); while (!RTC_Elapsed_ms( ticks, 100 )); for ( i=0;i<NSHAPES;i++){ ar[i]. mode = 1; memset( &mp[i], 0, sizeof( TShapeProps ) ); mp[i].wx = 10; mp[i].wy = 10; ar[i].x1 = 11 * ((i >> 2)+1); mp[i].color = random(); ar[i].y1 = 10 + 11 * ( i & 0x3 ); iwork = random(); iwork = ( 3 * iwork ) / 0x10000 + 1; if ( random() & 0x8000 ) iwork = -iwork; mp[i].dx = iwork; iwork = random(); iwork = ( 3 * iwork ) / 0x10000 + 1; if ( random() & 0x8000 ) iwork = -iwork; mp[i].dy = iwork; ar[i].x2 = ar[i].x1 + mp[i].wx; ar[i].y2 = ar[i].y1 + mp[i].wy; } while ( !(*(unsigned short*)0xA44B0000) ){ for (i=0;i<NSHAPES;i++){ MoveOne( ar, mp, i ); }    Bdisp_PutDisp_DD(); } }

Attachment removed. Actual version included here:
http://www.omnimaga.org/index.php?topic=6608.msg164247#msg164247

[DJ Omnimaga]

Yeah sorry I meant SH3 C and ASM. I am aware that this is C. I tend to just say ASM sometimes when it comes to calculator hacking, having been used to a platform with no decent C capabilities for years (the TI-83 Plus series). ASM here would be in reference to Qwerty.55 and z80man, although they use hexadecimal rather than SH3 mnemonics.