[commit] r2181 - cpu/x86/pc/olpc
by repository service
Author: quozl
Date: Thu Mar 24 22:59:24 2011
New Revision: 2181
URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2181
Log:
OLPC XO-1 - return to using EC version that was used for Q2E45
Modified:
cpu/x86/pc/olpc/versions.fth
Modified: cpu/x86/pc/olpc/versions.fth
==============================================================================
--- cpu/x86/pc/olpc/versions.fth Thu Mar 24 03:52:07 2011 (r2180)
+++ cpu/x86/pc/olpc/versions.fth Thu Mar 24 22:59:24 2011 (r2181)
@@ -2,10 +2,10 @@
\ The overall firmware revision
macro: FW_MAJOR E
-macro: FW_MINOR 45r
+macro: FW_MINOR 45s
\ The EC microcode
-macro: EC_VERSION e34
+macro: EC_VERSION e35
\ Alternate command for getting EC microcode, for testing new versions.
\ Temporarily uncomment the line and modify the path as necessary
8 years, 3 months
[commit] r2180 - cpu/x86/pc/olpc
by repository service
Author: quozl
Date: Thu Mar 24 03:52:07 2011
New Revision: 2180
URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2180
Log:
OLPC XO-1 - disable menu dev.laptop.org #10778
Modified:
cpu/x86/pc/olpc/fw.bth
cpu/x86/pc/olpc/versions.fth
Modified: cpu/x86/pc/olpc/fw.bth
==============================================================================
--- cpu/x86/pc/olpc/fw.bth Sat Mar 12 02:04:51 2011 (r2179)
+++ cpu/x86/pc/olpc/fw.bth Thu Mar 24 03:52:07 2011 (r2180)
@@ -665,7 +665,7 @@
unfreeze visible ( no-banner? )
if banner then \ Reissue banner if it was suppressed
- blue-letters ." Type 'help' for more information." black-letters
+ blue-letters ." Type 'help' for more information." black-letters cancel
cr cr
quit
@@ -683,6 +683,11 @@
fload ${BP}/cpu/x86/pc/olpc/diskspeed.fth \ Mass storage speed test
+\ #10778 disable menu
+warning @ warning off
+: menu ;
+warning !
+
\ This helps with TeraTerm, which sends ESC-O as the arrow key prefix
also hidden also keys-forth definitions
warning @ warning off
Modified: cpu/x86/pc/olpc/versions.fth
==============================================================================
--- cpu/x86/pc/olpc/versions.fth Sat Mar 12 02:04:51 2011 (r2179)
+++ cpu/x86/pc/olpc/versions.fth Thu Mar 24 03:52:07 2011 (r2180)
@@ -2,7 +2,7 @@
\ The overall firmware revision
macro: FW_MAJOR E
-macro: FW_MINOR 45q
+macro: FW_MINOR 45r
\ The EC microcode
macro: EC_VERSION e34
8 years, 3 months
Re: [openfirmware] Using Open Firmware as a boot manager
by Mark Morgan Lloyd
Mitch Bradley wrote:
>>> Booting Windows is not supported in the PC demonstration build, but is
>>> possible with a fair amount of effort. It requires ACPI and numerous
>>> other accomodations for Windows' use of conventional BIOS. We support
>>> it on the One Laptop Per Child systems.
>>
>> At present things I'm likely to want to boot are Linux and DOS, although
>> I'm currently trying to evaluate ReactOS and Sanos so I can report back
>> to another group of developers.
>
> DOS and ReactOS are going to have the same problem as Windows, namely
> that they expect to do conventional 16-bit BIOS calls. The OFW demo
> version does not support that.
>
> Are you considering using OFW on both x86 and ARM systems, or is the x86
> work just for exploration?
x86 really just for exploration, but I could say much the same for ARM.
>> Bearing in mind that I've not really used Forth since doing my Masters
>> [mumble] years ago, and that my use of Sun's implementation was mostly
>> pretty trivial, I think I need an idiot's guide as to what can be done
>> from the command line e.g. how to get aliases populated. At the very
>> least so that I can ask questions intelligently.
>
>
> http://www.firmworks.com/QuickRef.html
Thanks, I'd missed that and had been working through the IEEE doc.
>> I've really only got OFW working today, I tried (on some different
>> systems) a few months ago but I suspect that the "Making a Prototype
>> Floppy Image with Linux Commands" didn't work properly for me.
>>
>> Now that I am able to run successfully on a PC, is there a standard
>> config that I can use which will allow me to dump boot sectors etc. on
>> BIOS-supported discs (i.e. preferably both IDE and SCSI)?
>>
>>>> I'm hoping to eventually put it on some ARM-based development boards as
>>>> an interactive layer before they boot Linux (etc.) from disc.
>>>
>>> My current work at One Laptop Per Child is ARM-based, so the OFW ARM
>>> port is being actively maintained and enhanced. I have versions that
>>> you can load from uboot and others that run directly with no
>>> underlying firmware. The latter, of course, requires that you take
>>> responsibility for the low-level issues of memory controller and DRAM
>>> initialization.
>>>
>>> If you are a consumer of the board, then it's probably better to use
>>> whatever firmware is already there as the initial loader, using OFW as
>>> the boot manager as you say. If you are developing your own hardware,
>>> it starts to make sense to use OFW from the ground up.
>>>
>>> Feel free to contact me directly if you need additional guidance. At
>>> this point I'm more interested in the ARM version than the x86
>>> version, so the sooner you get onto ARM, the more help I'll be able to
>>> offer.
>>
>> Bear in mind that I'm doing this on spec,
>
> Effectively, so am I. The existence of other open source firmware
> solutions such as uboot and coreboot (formerly linuxbios) and
> bootloaders like GRUB have made it difficult bordering on impossible to
> make money with Open Firmware, except via consulting, and the market for
> that is rather limited.
Although I think that OFW is about the only one that has the potential
for doing things like listing the partition table (and associated code).
--
Mark Morgan Lloyd
markMLl .AT. telemetry.co .DOT. uk
[Opinions above are the author's, not those of his employers or colleagues]
8 years, 4 months
[commit] r2179 - dev/olpc/kb3700
by repository service
Author: rsmith
Date: Sat Mar 12 02:04:51 2011
New Revision: 2179
URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2179
Log:
* XO 1 and 1.5: Add bat-recover tool from batman.fth
- This makes doing bat-recover on a lot of batteries that have charge
balance a lot easier since you don't have to reload batman.fth every
reboot.
Modified:
dev/olpc/kb3700/battery.fth
Modified: dev/olpc/kb3700/battery.fth
==============================================================================
--- dev/olpc/kb3700/battery.fth Sat Mar 12 02:04:49 2011 (r2178)
+++ dev/olpc/kb3700/battery.fth Sat Mar 12 02:04:51 2011 (r2179)
@@ -672,6 +672,46 @@
h# 10 invert and ec-debugflag4 bat-b!
;
+: bat-recover-info@ ( -- voltage ACR current temp )
+ ds-bank-buf 6 h# 0c 1w-read ( )
+ ds-bank-buf c@ 8 << ( voltage_msb )
+ ds-bank-buf 1 + c@ or d# 16 << d# 16 >>a ( voltage )
+ ds-bank-buf 4 + c@ 8 << ( voltage ACR_msb )
+ ds-bank-buf 5 + c@ or d# 16 << d# 16 >>a ( voltage ACR )
+ ds-bank-buf 4 h# 18 1w-read ( voltage ACR )
+ ds-bank-buf 2 + c@ 8 << ( voltage ACR current_msb )
+ ds-bank-buf 3 + c@ or d# 16 << d# 16 >>a ( voltage ACR current )
+ swap ( voltage current ACR )
+ ds-bank-buf c@ 8 << ( voltage ACR current temp_msb )
+ ds-bank-buf 1 + c@ or d# 16 << d# 16 >>a ( voltage ACR current temp )
+;
+
+4 value bon
+2 value boff
+
+: bat-recover ( -- )
+ batman-init?
+ bg-acr@
+ begin
+ bat-enable-charge
+ bon ms
+ bat-disable-charge
+ bat-recover-info@
+ bg-temp>degc >sd.dd type ." C "
+ to bg-last-acr
+ bg-I>mA >sd.dd type ." mA "
+ bg-v_avg + 2 / dup to bg-v_avg
+ bg-V>V >sd.ddd type ." V "
+ dup bg-last-acr swap -
+ bg-acr>mAh >sd.dd type ." mAh "
+ cr
+ boff ms
+ false
+ until
+ drop
+ bat-disable-charge
+;
+
fload ${BP}/dev/olpc/kb3700/batstat.fth
\ LICENSE_BEGIN
8 years, 4 months
[commit] r2178 - dev/olpc/kb3700
by repository service
Author: rsmith
Date: Sat Mar 12 02:04:49 2011
New Revision: 2178
URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2178
Log:
* Fix typo in mppt-limit@
Modified:
dev/olpc/kb3700/eccmdcom.fth
Modified: dev/olpc/kb3700/eccmdcom.fth
==============================================================================
--- dev/olpc/kb3700/eccmdcom.fth Thu Mar 3 19:32:13 2011 (r2177)
+++ dev/olpc/kb3700/eccmdcom.fth Sat Mar 12 02:04:49 2011 (r2178)
@@ -18,7 +18,7 @@
: autowack-off ( -- ) 0 h# 33 ec-cmd-b! ;
: mppt-active@ ( -- b ) h# 3d ec-cmd-b@ ;
-: mppt-limit@ ( -- b ) h# 3e ec-cmd-b! ;
+: mppt-limit@ ( -- b ) h# 3e ec-cmd-b@ ;
: mppt-limit! ( b -- ) h# 3f ec-cmd-b! ;
: mppt-off ( -- ) h# 40 ec-cmd ;
: mppt-on ( -- ) h# 41 ec-cmd ;
8 years, 4 months
Using Open Firmware as a boot manager
by Mark Morgan Lloyd
Given a copy of OFW booted from floppy on a PC, is there a comparatively
straightforward way of using it to boot an OS or loader from other block
devices or over the LAN, in the way that I'm used to on Sun kit?
I'm hoping to eventually put it on some ARM-based development boards as
an interactive layer before they boot Linux (etc.) from disc.
--
Mark Morgan Lloyd
markMLl .AT. telemetry.co .DOT. uk
[Opinions above are the author's, not those of his employers or colleagues]
8 years, 4 months
[commit] r2177 - cpu/arm
by repository service
Author: wmb
Date: Thu Mar 3 19:32:13 2011
New Revision: 2177
URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2177
Log:
ARM assembler - Changed the "[ifdef] armv7" guarding the use of the
movw instruction in (set) to a runtime value "use-movw?" (default
value false), so the system can be compiled to run anywhere, but
it can later generate architecture-specific code for special purposes.
Modified:
cpu/arm/assem.fth
Modified: cpu/arm/assem.fth
==============================================================================
--- cpu/arm/assem.fth Thu Mar 3 07:14:41 2011 (r2176)
+++ cpu/arm/assem.fth Thu Mar 3 19:32:13 2011 (r2177)
@@ -44,6 +44,8 @@
defer asm-set-relocation-bit
also arm-assembler definitions
+false value use-movw?
+
: asm, ( n -- ) here /l asm-allot asm! ;
previous definitions
@@ -1162,15 +1164,15 @@
true asm-const ( reg# op )
then
else ( reg# imm imm )
-[ifdef] armv7
- 0 1.0000 within if ( reg# imm )
- set-imm16 0300.0000 ( reg# op ) \ movw rN,#<imm16>
- else ( reg# imm )
- false asm-const ( reg# op )
- then
-[else]
- drop false asm-const ( reg# op )
-[then]
+ use-movw? if ( reg# imm imm )
+ 1.0000 u< if ( reg# imm )
+ set-imm16 0300.0000 ( reg# op ) \ movw rN,#<imm16>
+ else ( reg# imm )
+ false asm-const ( reg# op )
+ then ( reg# op )
+ else ( reg# imm imm )
+ drop false asm-const ( reg# op )
+ then ( reg# op )
then ( reg# op )
then ( reg# op )
iop rd-field !op
8 years, 4 months
[commit] r2176 - cpu/arm forth/lib
by repository service
Author: tooch
Date: Thu Mar 3 07:14:41 2011
New Revision: 2176
URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2176
Log:
ARM - Add sqrt, some FP, vector, ARMv7 instructions. Change trace on/off
trigger in armsim. Remove movw optimization in (set) -- it breaks ARMv4.
Modified:
cpu/arm/armsim.c
cpu/arm/assem.fth
cpu/arm/code.fth
cpu/arm/disassem.fth
cpu/arm/kerncode.fth
cpu/arm/kernel.bth
cpu/arm/sqroot.fth
forth/lib/loclabel.fth
Modified: cpu/arm/armsim.c
==============================================================================
--- cpu/arm/armsim.c Mon Feb 28 18:36:42 2011 (r2175)
+++ cpu/arm/armsim.c Thu Mar 3 07:14:41 2011 (r2176)
@@ -39,10 +39,10 @@
void regdump(u32 instruction, u32 last_pc, u8 cr)
{
- printf(" 0 %08x 1 %08x 2 %08x 3 %08x\n", r[0], r[1], r[2], r[3]);
- printf(" 4 %08x 5 %08x 6 %08x 7 %08x\n", r[4], r[5], r[6], r[7]);
- printf(" 8 %08x 9 %08x a %08x b %08x\n", r[8], r[9], r[10], r[11]);
- printf(" c %08x d %08x e %08x f %08x\n", r[12], r[13], r[14], r[15]);
+ printf(" r0 %08x r1 %08x r2 %08x r3 %08x\n", r[0], r[1], r[2], r[3]);
+ printf(" r4 %08x r5 %08x r6 %08x base %08x\n", r[4], r[5], r[6], r[7]);
+ printf(" r8 %08x up %08x tos %08x rp %08x\n", r[8], r[9], r[10], r[11]);
+ printf(" ip %08x sp %08x lr %08x pc %08x\n", r[12], r[13], r[14], r[15]);
printf("pc %08x lpc %08x i %08x ", PC - 8, last_pc, instruction);
if (cr)
putchar('\n');
@@ -800,8 +800,7 @@
case 0x11: INSTR("eor"); RD = RN ^ IMM32; UPCC(RD); break;
case 0x12: INSTR("sub"); SBB(RD, RN, IMM32, 1); break;
case 0x13: INSTR("rsb"); SBB(RD, IMM32, RN, 1); break;
-case 0x14: /* if (instruction == 0xe2809020) trace = 0; */
- INSTR("add"); ADC(RD, RN, IMM32, 0); break;
+case 0x14: INSTR("add"); ADC(RD, RN, IMM32, 0); break;
case 0x15: INSTR("adc"); ADC(RD, RN, IMM32, C); break;
case 0x16: INSTR("sbc"); SBB(RD, IMM32, RN, C); break;
case 0x17: INSTR("rsc"); SBB(RD, IMM32, RN, C); break;
@@ -810,9 +809,12 @@
case 0x0: INSTR("nop"); break;
case 0x1:
INSTR("wrc");
- if (RN == -2) {
+ if (UFIELD(19, 4) == 0xf) { // "wrc pc"
printf("Tracing on\n");
trace = 1;
+ } else if (UFIELD(19, 4) == 0xe) { // "wrc lr"
+ printf("Tracing off\n");
+ trace = 0;
} else if (RN == -1) {
// trace = 1;
// printf("find %x %x %x %x\n",r[2], r[1], r[0], r[3]);
Modified: cpu/arm/assem.fth
==============================================================================
--- cpu/arm/assem.fth Mon Feb 28 18:36:42 2011 (r2175)
+++ cpu/arm/assem.fth Thu Mar 3 07:14:41 2011 (r2176)
@@ -13,8 +13,8 @@
[then]
\needs land : land and ;
-\needs cindex fload ${BP}/forth/lib/parses1.fth
-\needs lex fload ${BP}/forth/lib/lex.fth
+\needs cindex fload ${BP}/forth/lib/parses1.fth
+\needs lex fload ${BP}/forth/lib/lex.fth
\needs 2nip : 2nip ( n1 n2 n3 n4 -- n3 n4 ) 2swap 2drop ;
vocabulary arm-assembler also arm-assembler definitions
@@ -149,7 +149,9 @@
next-cons constant adt-reg \ r0, r1, ..., pc.
next-cons constant adt-coreg \ c0, c1, ...
next-cons constant adt-coproc \ p0, p1, ...
-next-cons constant adt-xpsr \ cpsr, spsr.
+next-cons constant adt-dreg \ d0, d1, ...
+next-cons constant adt-sreg \ s0, s1, ...
+next-cons constant adt-xpsr \ cpsr, spsr, fpscr, ...
next-cons constant adt-shift \ Shift op in Shifter Operands.
next-cons constant adt-rrx
next-cons constant adt-immed \ #immediate_value.
@@ -240,6 +242,10 @@
rem$ " !#*+,-[]^_{}`" lex 0= if ( field$ )
0 0 2swap 0 ( rem$ field$ delim )
then ( rem$ field$ delim )
+ \ Horrible special case for L#
+ dup ascii # = 2over " l" $= and if
+ drop 1+ 0
+ then
is adr-delim 2swap set-rem$ ( field$ )
dup 0<> adr-delim or if ( field$ )
true ( field$ true )
@@ -268,6 +274,14 @@
: coreg: ( n "name" -- ) create , does> @ adt-coreg ;
: coregs: ( 10x"name" -- ) 10 0 do i coreg: loop ;
+\ Define the VFP/SIMD double-precision registers.
+: dreg: ( n "name" -- ) create , does> @ adt-dreg ;
+: dregs: ( 20x"name" -- ) 20 0 do i dreg: loop ;
+
+\ Define the VFP/SIMD single-precision registers.
+: sreg: ( n "name" -- ) create , does> @ adt-sreg ;
+: sregs: ( 20x"name" -- ) 20 0 do i sreg: loop ;
+
: range-error ( n msg$ -- ) type .d cr abort ;
: expecting ( $ -- ) ." Expecting " ad-error ;
@@ -388,7 +402,9 @@
r> set-field
;
-: ?register ( adt -- ) adt-reg <> " register" ?expecting ;
+: ?register ( adt -- ) adt-reg <> " register" ?expecting ;
+: ?dregister ( adt -- ) adt-dreg <> " dregister" ?expecting ;
+: ?sregister ( adt -- ) adt-sreg <> " sregister" ?expecting ;
: get-immediate ( -- n )
get-whatever adt-immed <> " immediate" ?expecting
@@ -413,6 +429,28 @@
: get-r12 ( -- ) d# 12 get-rn ;
: get-r16 ( -- ) d# 16 get-rn ;
+: ?dregister ( adt -- ) adt-dreg <> " doubleword register" ?expecting ;
+: get-dregister ( -- dreg )
+ require-field $asm-execute case
+ adt-sreg of 0000.0100 xop endof
+ adt-dreg of endof
+ " floating-point register" expecting
+ endcase
+;
+
+: set-vdfield ( n lo-D? -- )
+ \ Reg[5] is encoded as Vd[15-12]:D[22] OR as D[22]:Vd[15-12]. Sigh.
+ if
+ dup 1 and d# 22 lshift iop
+ 1 >> d# 12 lshift iop
+ else
+ dup h# 10 land d# 18 lshift iop
+ h# 0f land d# 12 lshift iop
+ then
+;
+: get-d12 ( -- ) get-dregister true set-vdfield ;
+
+
: expecting-reg/immed ( -- ) " register or immediate" expecting ;
: get-shiftr# ( -- )
\ Back over a real delimiter, then get the next thing.
@@ -539,6 +577,9 @@
!op
;
+: amode-vmrs ( -- ) init-operands get-r12 get-whatever ?psr iop !op ;
+: amode-vmsr ( -- ) init-operands get-whatever ?psr iop get-r12 !op ;
+
: (amode-mul) ( -- ) init-operands get-r16 get-r00 get-r08 ;
: amode-mul ( -- ) (amode-mul) !op ;
: amode-mla ( -- ) (amode-mul) get-r12 !op ;
@@ -600,7 +641,47 @@
!op
;
-\ rd, [rn, <immed12>] {!}
+: amode-vlsm ( need-r16? -- )
+ init-operands
+ if
+ get-r16 ( )
+ adr-delim ascii ! = if ( )
+ flip-w ( )
+ require-field " ," ?expecting drop ( )
+ then ( )
+
+ \ There should be a comma on the end of the register.
+ adr-delim ascii , <> " ," ?expecting
+ then
+
+ \ The next thing up should be an open brace for the register list.
+ get-whatever adt-delimiter <> " {" ?expecting
+ ascii { <> " {" ?expecting
+
+ \ Start with dreg-list "0, 0" and we'll update as we go along.
+ d# 32 0 ( first last )
+ begin adr-delim ascii } <> while
+ get-whatever case ( first last value adt )
+
+ adt-dreg of ( first last dreg )
+ \ Update first, last
+ tuck max -rot min swap ( first' last' )
+ \ Check the delimiter for - meaning a range.
+ adr-delim ascii - = if ( first' last' )
+ get-whatever ?dregister ( first' last' dreg )
+ max ( first' last'' )
+ then
+ endof
+
+ " register or }" expecting
+ endcase
+ repeat ( first last )
+
+ \ Encode the resulting Vd, imm8 fields
+ 1+ over - 2* iop false set-vdfield !op
+;
+
+\ rd, [rn, <immed12>] {!}
\ rd, [rn, +-rm] {!}
\ rd, [rn, +-rm, <shift>] {!}
\ rd, [rn], <immed12>
@@ -711,6 +792,27 @@
p? if {!} then
;
+: get-imm8 ( -- )
+ \ Get the offset for v[ldr|str] instructions
+ get-whatever case
+ adt-delimiter of
+ case
+ ascii + of get-r00 endof
+ ascii - of flip-u get-r00 endof
+ " +, -, or number" expecting
+ endcase
+ endof
+
+ adt-immed of
+ \ If the value is negative, switch things around.
+ dup 0< if negate flip-u then
+ d# 10 ?#bits 2 >>a iop
+ endof
+
+ " immediate value" expecting
+ endcase
+;
+
\ rd, [rn, <immed8>] {!}
\ rd, [rn, +-rm] {!}
\ rd, [rn], <immed8>
@@ -735,7 +837,6 @@
: amode-ldrex ( -- )
init-operands get-r12 ['] get-off0 get-ea !op
;
-
: amode-copr ( -- ) \ Co-processors: mcr, mrc
\ p, #, r, c, c, #
init-operands
@@ -778,6 +879,14 @@
!op
;
+: amode-vldst ( -- ) \ vldr, vstr instructions
+ init-operands
+ \ Set the add offset and 64-bit width as defaults.
+ 0080.0100 iop
+ get-d12 ['] get-imm8 get-ea
+ !op
+;
+
\ ----------------
: next-2? ( -- $ true | false )
@@ -843,11 +952,19 @@
: {cond/s} ( opcode -- ) {cond} {s} ;
: {uncond} ( opcode -- ) is newword ;
-: parse-inc ( l-flag -- )
- \ Parse the increment tag for ldm and stm. There MUST be a two letter
- \ code to specify the increment option so we bail if we don't get one
- \ of the eight possible codes. l-flag true specifies ldm, vice stm.
- 0= >r next-2? 0= " increment specifier" ?expecting
+: parse-inc ( default$ l-flag -- )
+ \ Parse the increment tag for ldm and stm.
+ \ If default is -1 then there MUST be a two letter code to specify
+ \ the increment option: we bail if we don't get one of the eight
+ \ possible codes. If default$ is non-null we'll use that instead
+ \ (see vldm et al.).
+ \ l-flag true specifies ldm, vice stm.
+ 0= >r
+ next-2? 0= if ( default$ )
+ ?dup 0= if drop true " increment specifier" ?expecting then
+ else ( default$ incr-spec$ )
+ 2swap 2drop
+ then
\ Correct tags have an even index from sindex.
" daiadbibfafdeaed" sindex dup 1 land " increment specifier" ?expecting
@@ -918,6 +1035,9 @@
: spsr ( -- n1 n2 ) 00400000 adt-xpsr ;
: cpsr ( -- n1 n2 ) 00000000 adt-xpsr ;
+: fpsid ( -- n1 n2 ) 00000000 adt-xpsr ;
+: fpscr ( -- n1 n2 ) 00010000 adt-xpsr ;
+: fpexc ( -- n1 n2 ) 00080000 adt-xpsr ;
psrs: _c _x _cx _s _cs _xs _cxs _f _cf _xf _cxf _sf _csf _xsf _cxsf
1 psr: _ctl
@@ -927,70 +1047,85 @@
coprocs: p0 p1 p2 p3 p4 p5 p6 p7 p8 p9 p10 p11 p12 p13 p14 p15
coregs: cr0 cr1 cr2 cr3 cr4 cr5 cr6 cr7 cr8 cr9 cr10 cr11 cr12 cr13 cr14 cr15
regs: r0 r1 r2 r3 r4 r5 r6 r7 r8 r9 r10 r11 r12 r13 r14 r15
+dregs: d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10 d11 d12 d13 d14 d15 d16 d17 d18 d19 d20 d21 d22 d23 d24 d25 d26 d27 d28 d29 d30 d31
+sregs: s0 s1 s2 s3 s4 s5 s6 s7 s8 s9 s10 s11 s12 s13 s14 s15 s16 s17 s18 s19 s20 s21 s22 s23 s24 s25 s26 s27 s28 s29 s30 s31
previous definitions
also arm-assembler definitions
-: and 0000.0000 {cond/s} amode-rrop2 ;
-: eor 0020.0000 {cond/s} amode-rrop2 ;
-: sub 0040.0000 {cond/s} amode-rrop2 ;
-: rsb 0060.0000 {cond/s} amode-rrop2 ;
-: add 0080.0000 {cond/s} amode-rrop2 ;
-: adc 00a0.0000 {cond/s} amode-rrop2 ;
-: sbc 00c0.0000 {cond/s} amode-rrop2 ;
-: rsc 00e0.0000 {cond/s} amode-rrop2 ;
-: orr 0180.0000 {cond/s} amode-rrop2 ;
-: bic 01c0.0000 {cond/s} amode-rrop2 ;
-
-: clz 016f.0f10 {cond/s} amode-rdop2 ;
-: mov 01a0.0000 {cond/s} amode-rdop2 ;
-: mvn 01e0.0000 {cond/s} amode-rdop2 ;
-: movw 0300.0000 {cond} amode-movw ;
-
-: mul 0000.0090 {cond/s} amode-mul ;
-: mla 0020.0090 {cond/s} amode-mla ;
-: umull 0080.0090 {cond/s} amode-lmul ;
-: umlal 00a0.0090 {cond/s} amode-lmul ;
-: smull 00c0.0090 {cond/s} amode-lmul ;
-: smlal 00e0.0090 {cond/s} amode-lmul ;
-
-: tst 0110.0000 {cond} amode-rnop2 ;
-: teq 0130.0000 {cond} amode-rnop2 ;
-: cmp 0150.0000 {cond} amode-rnop2 ;
-: cmn 0170.0000 {cond} amode-rnop2 ;
-
-: mrs 010f.0000 {cond} amode-mrs ;
-: msr 0120.f000 {cond} amode-msr ;
-
-: ldc 0c10.0000 {cond} amode-lsc ;
-: stc 0c00.0000 {cond} amode-lsc ;
-: cdp 0e00.0000 {cond} amode-cdp ;
-: mcr 0e00.0010 {cond} amode-copr ;
-: mrc 0e10.0010 {cond} amode-copr ;
-
-: swi 0f00.0000 {cond} amode-imed24 ;
-
-: b 0a00.0000 {cond} amode-bbl ;
-: bl 0b00.0000 {cond} amode-bbl ;
-
-: bx 012f.ff10 {cond} amode-bx ;
-: blx 012f.ff30 {cond} amode-bx ;
-
-: swp 0100.0090 {cond} {b} amode-swp ;
-: strex 0180.0f90 {cond} {bh} amode-swp ;
-: ldrex 0190.0f9f {cond} {bh} amode-ldrex ;
-
-: ldm 0810.0000 {cond} 1 parse-inc 1 amode-lsm ;
-: popm 08bd.0000 {cond} 0 amode-lsm ;
-: stm 0800.0000 {cond} 0 parse-inc 1 amode-lsm ;
-: pushm 092d.0000 {cond} 0 amode-lsm ;
-
-: ldr ( -- ) 0410.0000 {cond} {shbt} ;
-: str ( -- ) 0400.0000 {cond} {hbt} ;
-
-: rev ( -- ) 06bf0f30 {cond} amode-rev ;
-: rev16 ( -- ) 06bf0fb0 {cond} amode-rev ;
-: revsh ( -- ) 06ff0f30 {cond} amode-rev ;
+: and 0000.0000 {cond/s} amode-rrop2 ;
+: eor 0020.0000 {cond/s} amode-rrop2 ;
+: sub 0040.0000 {cond/s} amode-rrop2 ;
+: rsb 0060.0000 {cond/s} amode-rrop2 ;
+: add 0080.0000 {cond/s} amode-rrop2 ;
+: adc 00a0.0000 {cond/s} amode-rrop2 ;
+: sbc 00c0.0000 {cond/s} amode-rrop2 ;
+: rsc 00e0.0000 {cond/s} amode-rrop2 ;
+: orr 0180.0000 {cond/s} amode-rrop2 ;
+: bic 01c0.0000 {cond/s} amode-rrop2 ;
+: uadd16 0650.0f10 {cond} amode-rrop2 ;
+: uasx 0650.0f30 {cond} amode-rrop2 ;
+: uadd8 0650.0f90 {cond} amode-rrop2 ;
+: uxtab 06e0.0070 {cond} amode-rrop2 ;
+
+: clz 016f.0f10 {cond/s} amode-rdop2 ;
+: mov 01a0.0000 {cond/s} amode-rdop2 ;
+: mvn 01e0.0000 {cond/s} amode-rdop2 ;
+: mvn 01e0.0000 {cond/s} amode-rdop2 ;
+
+: movw 0300.0000 {cond} amode-movw ;
+
+: mul 0000.0090 {cond/s} amode-mul ;
+: mla 0020.0090 {cond/s} amode-mla ;
+: umull 0080.0090 {cond/s} amode-lmul ;
+: umlal 00a0.0090 {cond/s} amode-lmul ;
+: smull 00c0.0090 {cond/s} amode-lmul ;
+: smlal 00e0.0090 {cond/s} amode-lmul ;
+
+: tst 0110.0000 {cond} amode-rnop2 ;
+: teq 0130.0000 {cond} amode-rnop2 ;
+: cmp 0150.0000 {cond} amode-rnop2 ;
+: cmn 0170.0000 {cond} amode-rnop2 ;
+
+: mrs 010f.0000 {cond} amode-mrs ;
+: msr 0120.f000 {cond} amode-msr ;
+
+: ldc 0c10.0000 {cond} amode-lsc ;
+: stc 0c00.0000 {cond} amode-lsc ;
+: cdp 0e00.0000 {cond} amode-cdp ;
+: mcr 0e00.0010 {cond} amode-copr ;
+: mrc 0e10.0010 {cond} amode-copr ;
+
+: swi 0f00.0000 {cond} amode-imed24 ;
+
+: b 0a00.0000 {cond} amode-bbl ;
+: bl 0b00.0000 {cond} amode-bbl ;
+
+: bx 012f.ff10 {cond} amode-bx ;
+: blx 012f.ff30 {cond} amode-bx ;
+
+: rev 06bf.0f30 {cond} amode-rev ;
+: rev16 06bf.0fb0 {cond} amode-rev ;
+: revsh 06ff.0f30 {cond} amode-rev ;
+
+: swp 0100.0090 {cond} {b} amode-swp ;
+: strex 0180.0f90 {cond} {bh} amode-swp ;
+: ldrex 0190.0f9f {cond} {bh} amode-ldrex ;
+
+: ldm 0810.0000 {cond} " " 1 parse-inc 1 amode-lsm ;
+: popm 08bd.0000 {cond} 0 amode-lsm ;
+: stm 0800.0000 {cond} " " 0 parse-inc 1 amode-lsm ;
+: pushm 092d.0000 {cond} 0 amode-lsm ;
+
+: ldr 0410.0000 {cond} {shbt} ;
+: str 0400.0000 {cond} {hbt} ;
+
+: vldr 0d10.0a00 {cond} amode-vldst ;
+: vstr 0d00.0a00 {cond} amode-vldst ;
+: vldm 0c10.0b00 {cond} " ia" 1 parse-inc 1 amode-vlsm ;
+: vstm 0c00.0b00 {cond} " ia" 0 parse-inc 1 amode-vlsm ;
+: vmsr 0ee0.0a10 {cond} amode-vmsr ;
+: vmrs 0ef0.0a10 {cond} amode-vmrs ;
: rd-field ( reg# -- ) d# 12 set-field ;
: rb-field ( reg# -- ) d# 16 set-field ;
@@ -1027,11 +1162,15 @@
true asm-const ( reg# op )
then
else ( reg# imm imm )
+[ifdef] armv7
0 1.0000 within if ( reg# imm )
set-imm16 0300.0000 ( reg# op ) \ movw rN,#<imm16>
else ( reg# imm )
false asm-const ( reg# op )
then
+[else]
+ drop false asm-const ( reg# op )
+[then]
then ( reg# op )
then ( reg# op )
iop rd-field !op
Modified: cpu/arm/code.fth
==============================================================================
--- cpu/arm/code.fth Mon Feb 28 18:36:42 2011 (r2175)
+++ cpu/arm/code.fth Thu Mar 3 07:14:41 2011 (r2176)
@@ -4,6 +4,10 @@
\ These words are specific to the virtual machine implementation
: assembler ( -- ) arm-assembler ;
+variable pre-asm-base
+: stash-base base @ pre-asm-base ! ;
+: restore-base pre-asm-base @ base ! ;
+
only forth also arm-assembler also helpers also arm-assembler also definitions
\ Forth Virtual Machine registers
@@ -54,6 +58,7 @@
: exitcode ( -- )
['] $interpret-do-undefined is $do-undefined
previous
+ restore-base
;
' exitcode is do-exitcode
headers
@@ -119,6 +124,8 @@
also forth definitions
headerless
: entercode ( -- )
+ stash-base
+ decimal
also assembler
\ false is disassembling?
[ also helpers ]
Modified: cpu/arm/disassem.fth
==============================================================================
--- cpu/arm/disassem.fth Mon Feb 28 18:36:42 2011 (r2175)
+++ cpu/arm/disassem.fth Thu Mar 3 07:14:41 2011 (r2176)
@@ -37,6 +37,21 @@
: forth-regs ['] (forth-regs is regs ;
forth-regs
+string-array dregs
+," d0" ," d1" ," d2" ," d3" ," d4" ," d5" ," d6" ," d7"
+," d8" ," d9" ," d10" ," d11" ," d12" ," d13" ," d14" ," d15"
+," d16" ," d17" ," d18" ," d19" ," d20" ," d21" ," d22" ," d23"
+," d24" ," d25" ," d26" ," d27" ," d28" ," d29" ," d30" ," d31"
+end-string-array
+
+string-array sregs
+," s0" ," s1" ," s2" ," s3" ," s4" ," s5" ," s6" ," s7"
+," s8" ," s9" ," s10" ," s11" ," s12" ," s13" ," s14" ," s15"
+," s16" ," s17" ," s18" ," s19" ," s20" ," s21" ," s22" ," s23"
+," s24" ," s25" ," s26" ," s27" ," s28" ," s29" ," s30" ," s31"
+end-string-array
+
+
: udis. ( n -- )
push-hex
<#
@@ -77,24 +92,40 @@
: ., ( -- ) ." , " ;
: .[ ( -- ) ." [" ;
: .] ( -- ) ." ]" ;
+: .{ ( -- ) ." {" ;
+: .} ( -- ) ." }" ;
: .rm ( -- ) 0 .reg ;
: .rs ( -- ) 8 .reg ;
-: .rd, ( -- ) d# 12 .reg ., ;
+: .rd ( -- ) d# 12 .reg ;
+: .rd, ( -- ) .rd ., ;
: op.rd, ( -- ) op-col .rd, ;
: .rb ( -- ) d# 16 .reg ;
alias .rn .rb
: rn ( -- rn ) d# 16 4bits ;
-: .rm,shift ( -- )
+: .rm,shift ( rsr? -- )
.rm
d# 4 8bits if \ LSL #0 is no-shift; this isn't it
.,
- 4 8bits 6 = if ." rrx" exit then
+ 4 8bits 6 = if ." rrx" drop exit then
5 2 " lsllsrasrror" 3 .fld ." "
- 4 bit? if .rs else ." #" 7 5 bits .d then
+ ( rsr? ) 4 bit? and if .rs else ." #" 7 5 bits .d then
+ else
+ drop
+ then
+;
+
+: get-vd ( dbit-lo? -- reg# )
+ if
+ d# 12 4bits 1 << d# 22 bit? if 1+ then
+ else
+ d# 12 4bits d# 22 bit? if h# 10 + then
then
;
+: .dreg ( reg# -- ) dregs ". ;
+: .sreg ( reg# -- ) sregs ". ;
+: op.vd, ( dbit-lo? -- ) op-col get-vd 8 bit? if .dreg else .sreg then ., ;
: u.h ( n -- ) dup d# 9 u> if ." 0x" then (u.) type ;
: ror ( n cnt -- ) 2dup d# 32 swap - lshift -rot rshift or ;
@@ -116,7 +147,7 @@
: +/- ( -- ) d#23 bit? 0= if ." -" then ;
: .r/imm ( -- )
- d#25 bit? if ." #" .imm else .rm,shift then
+ d#25 bit? if ." #" .imm else 1 .rm,shift then
;
\ Indicates the form of the instruction that affects both PC and CPSR/SPSR
: {p} ( -- )
@@ -278,7 +309,7 @@
\ : byte ( -- ) 00400000 op-or ; \ ldr str operate bytewide
: .mregs ( -- )
- ." {" ( )
+ .{ ( )
0 d# 16 bits false ( n need,? )
d# 16 0 do ( n need,? )
over 1 and if ( n need,? )
@@ -288,7 +319,7 @@
swap 2/ swap ( n need,?' )
loop ( n need,?' )
2drop ( )
- ." }" ( )
+ .} ( )
;
: .inc ( -- ) d#23 2 " daiadbib" 2 .fld ;
: .ldm/stm ( -- ) \ d# 25 3 bits 4 =
@@ -300,18 +331,25 @@
: imm12 ( -- n ) 0 d# 12 bits ;
: ,.addr-mode ( -- )
d#25 bit? if
- ., +/- .rm,shift
+ ., +/- 1 .rm,shift
else
imm12 if ., ." #" +/- imm12 u.h then
then
;
: .rev ( -- ) {<cond>} op.rd, .rm ;
+: .uadd ( -- ) {<cond>} op.rd, .rn ., .rm ;
+: .uxtab ( -- ) {<cond>} op.rd, .rn ., 0 .rm,shift ;
: .stuff ( -- )
- 0 d# 28 bits h# 0fff.0ff0 and
+ 0 d# 28 bits h# 0ff0.00f0 and
case
- h# 06bf0f30 of ." rev" .rev endof
- h# 06bf0fb0 of ." rev16" .rev endof
- h# 06ff0f30 of ." revsh" .rev endof
+ h# 0650.0090 of ." uadd8" .uadd endof
+ h# 0650.0010 of ." uadd16" .uadd endof
+ h# 0650.0030 of ." uasx" .uadd endof
+ h# 06b0.0030 of ." rev" .rev endof
+ h# 06b0.00b0 of ." rev16" .rev endof
+ h# 06e0.0070 of ." uxtab" .uxtab endof
+ h# 06f0.0030 of ." revsh" .rev endof
+ h# 06f0.0030 of ." revsh" .rev endof
( default )
." undefined" {<cond>}
endcase
@@ -348,10 +386,31 @@
10 8 fops rmf sin fml cos fdv tan frd asn
18 4 fops pol acs ??? atn
[then]
+: .fpspec ( -- )
+ d# 16 4bits case
+ 0 of ." fpsid" endof
+ 1 of ." fpscr" endof
+ 8 of ." fpexc" endof
+ ." fpxxx" endcase
+;
+
: p# ( -- n ) 8 4bits ;
: .p#, ( n -- ) ." p" p# n.d ., ;
: .offset8 ( -- ) ." #" +/- 0 8bits 4 * u.h ;
+: .vldst ( -- )
+ 24 bit? 21 bit? not and if
+ ." v" .ld/st ." r" {<cond>} true op.vd, .[ .rb
+ 0 8bits if ., .offset8 then .]
+ else
+ ." v" .ld/st ." m" {<cond>} .inc
+ op-col .rb {!} ., .{
+ false get-vd 0 8bits 2/ bounds over -rot do
+ i .dreg i 1+ over <> if ., then
+ loop drop .}
+ then
+;
: .ldc/stc ( -- )
+ 9 3 bits 5 = if .vldst exit then
.ld/st ." c" {<cond>} " l" d#22 ?.bit
op-col .p#, d# 12 .creg ., .[ .rn
p-bit if ., .offset8 .] {!} else .] ., .offset8 then
@@ -415,9 +474,19 @@
dup 7 8 between if drop ., .flushes exit then
d# 15 = if .clocks then \ SA-110
;
+: .vfp ( -- ) \ Decode VFP ops
+ d# 20 bit? if
+ ." vmrs" op.rd, .fpspec
+ else
+ ." vmsr" op-col .fpspec ., .rd
+ then
+;
+
: .coproc ( -- )
p-bit if .swi exit then
+
d# 4 bit? if \ MRC and MCR
+ p# 1 >> 5 = if .vfp exit then
d# 20 1 " mcrmrc" 3 .fld {<cond>}
op-col
p# d# 15 = if \ System Control Coprocessor
Modified: cpu/arm/kerncode.fth
==============================================================================
--- cpu/arm/kerncode.fth Mon Feb 28 18:36:42 2011 (r2175)
+++ cpu/arm/kerncode.fth Thu Mar 3 07:14:41 2011 (r2176)
@@ -753,6 +753,7 @@
: unaligned-l! ( l adr -- ) unaligned-! ;
: unaligned-d! ( d adr -- ) tuck na1+ unaligned-! unaligned-! ;
: d@ ( adr -- d ) dup @ swap na1+ @ ;
+: d! ( d adr -- ) tuck na1+ ! ! ;
code c! ( char adr -- ) pop r0,sp strb r0,[tos] pop tos,sp c;
code 2@ ( adr -- n-high n-low )
Modified: cpu/arm/kernel.bth
==============================================================================
--- cpu/arm/kernel.bth Mon Feb 28 18:36:42 2011 (r2175)
+++ cpu/arm/kernel.bth Thu Mar 3 07:14:41 2011 (r2176)
@@ -92,6 +92,7 @@
fload ${BP}/forth/kernel/uservars.fth
fload ${BP}/cpu/arm/muldiv.fth \ Uses "defer"; must be after uservars
+fload ${BP}/cpu/arm/sqroot.fth
fload ${BP}/cpu/arm/dodoesad.fth
fload ${BP}/cpu/arm/version.fth
Modified: cpu/arm/sqroot.fth
==============================================================================
--- cpu/arm/sqroot.fth Mon Feb 28 18:36:42 2011 (r2175)
+++ cpu/arm/sqroot.fth Thu Mar 3 07:14:41 2011 (r2176)
@@ -1,6 +1,7 @@
purpose: Integer square-root for ARM processors
\ See license at end of file
+0 [if]
\ u1 -- 32-bit unsigned
\ n -- significant digits
\ 16 -> sqrt-integer
@@ -26,6 +27,24 @@
top r1 mov c;
: sqrt ( u1 -- u2 ) td 16 (sqrt ;
+[then]
+
+\ 32bit -> 16bit fixed point square root
+\ see http://www.finesse.demon.co.uk/steven/sqrt.html
+code sqrt ( n -- root )
+ mov r0, tos \ n
+ mov tos, `1 d# 30 <<` \ root
+ mov r1, `3 d# 30 <<` \ offset
+ mov r2, 0 \ loop count
+ begin
+ cmp r0, tos, ror r2
+ subhs r0, r0, tos, ror r2
+ adc tos, r1, tos, lsl #1
+ inc r2, #2
+ cmp r2, #32
+ = until
+ bic tos, tos, `3 d# 30 <<`
+c;
\ LICENSE_BEGIN
\ Copyright (c) 2008 FirmWorks
Modified: forth/lib/loclabel.fth
==============================================================================
--- forth/lib/loclabel.fth Mon Feb 28 18:36:42 2011 (r2175)
+++ forth/lib/loclabel.fth Thu Mar 3 07:14:41 2011 (r2176)
@@ -17,7 +17,7 @@
headerless
20 constant #references-max
-10 constant #labels-max
+20 constant #labels-max
#labels-max #references-max * /n* buffer: references
#labels-max /n* buffer: local-labels
@@ -59,11 +59,12 @@
;
headerless
-: init-labels ( -- )
+: (init-labels) ( -- )
#labels-max 0 do i clear-label loop
;
-init-labels
+defer init-labels
+' (init-labels) is init-labels
also forth definitions
headers
8 years, 4 months