February 2009 - openfirmware - mail.coreboot.org

r1115 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 22:50:02 +0100 (Fri, 20 Feb 2009) New Revision: 1115 Modified: cpu/x86/pc/olpc/versions.fth Log: Q2E33 Modified: cpu/x86/pc/olpc/versions.fth =================================================================== --- cpu/x86/pc/olpc/versions.fth 2009-02-20 21:44:09 UTC (rev 1114) +++ cpu/x86/pc/olpc/versions.fth 2009-02-20 21:50:02 UTC (rev 1115) @@ -2,7 +2,7 @@ \ The overall firmware revision macro: FW_MAJOR E -macro: FW_MINOR 32 +macro: FW_MINOR 33 \ The EC microcode macro: EC_VERSION e32

1 0

r1114 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 22:44:09 +0100 (Fri, 20 Feb 2009) New Revision: 1114 Modified: cpu/x86/pc/olpc/mfgdata.fth Log: OLPC - improved the pagination for manufacturing data display in the presence of long key override tags. Modified: cpu/x86/pc/olpc/mfgdata.fth =================================================================== --- cpu/x86/pc/olpc/mfgdata.fth 2009-02-20 21:43:04 UTC (rev 1113) +++ cpu/x86/pc/olpc/mfgdata.fth 2009-02-20 21:44:09 UTC (rev 1114) @@ -90,6 +90,7 @@ dup >r blue-letters bounds ?do + #out @ 3 + rmargin @ > if exit? ?leave then 3 .tab i c@ <# u# u# u#> type loop black-letters

1 0

r1113 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 22:43:04 +0100 (Fri, 20 Feb 2009) New Revision: 1113 Modified: cpu/x86/pc/olpc/smbios.fth Log: OLPC SMBIOS - cleaned up a few things that were misreported. Modified: cpu/x86/pc/olpc/smbios.fth =================================================================== --- cpu/x86/pc/olpc/smbios.fth 2009-02-20 19:01:27 UTC (rev 1112) +++ cpu/x86/pc/olpc/smbios.fth 2009-02-20 21:43:04 UTC (rev 1113) @@ -77,7 +77,7 @@ ( 06 ) h# f000 w, \ BIOS starting address segment ( 08 ) 3 c, \ Release date index ( 09 ) h# 0f c, \ BIOS ROM size in 64K chunks, minus 1 -( 0a ) h# 91880 l, \ BIOS characteristics - PCI, Reflash, selectable boot, EDD +( 0a ) h# 99880 l, \ BIOS characteristics - PCI, Reflash, shadowing, CD boot, selectable boot, EDD ( 0e ) 0 l, \ Vendor and system specific BIOS characteristics ( 12 ) 1 c, \ ACPI is supported \ " OLPC" c$, \ Vendor string @@ -92,12 +92,13 @@ ( 04 ) 1 c, \ Manufacturer string index ( 05 ) 2 c, \ Product name string index ( 06 ) 3 c, \ Version string index -( 07 ) 0 c, \ Serial number string index +( 07 ) 4 c, \ Serial number string index ( 08 ) here to uuid-adr h# 10 allot \ SETME ( 18 ) 6 c, \ Wake up type \ " OLPC" c$, \ 1: Manufacturer \ " XO" c$, \ 2: Product Name \ " 1" c$, \ 3: Version +\ " <sn>" c$, \ 4: Serial number \ 0 c, create base-board-info @@ -108,7 +109,7 @@ ( 05 ) 2 c, \ Product string index ( 06 ) 3 c, \ Version string index ( 07 ) 4 c, \ Serial number string index -\ " OLPC" c$, \ 1: Manufacturer +\ " QUANTA" c$, \ 1: Manufacturer \ " XO" c$, \ 2: Product Name \ " 1" c$, \ 3: Version @@ -157,7 +158,7 @@ ( 01 ) h# 13 c, \ Length ( 02 ) h# 0701 w, \ Handle ( 04 ) 0 c, \ Socket string index -( 05 ) h# 181 w, \ Writeback, enabled, internal, not socketed, L1 +( 05 ) h# 180 w, \ Writeback, enabled, internal, not socketed, L1 ( 07 ) h# 8002 w, \ Max size - 128K ( 09 ) h# 8002 w, \ Installed size - 128K ( 0b ) 1 w, \ Supported SRAM type - unknown @@ -174,7 +175,7 @@ ( 01 ) h# 13 c, \ Length ( 02 ) h# 0702 w, \ Handle ( 04 ) 0 c, \ Socket string index -( 05 ) h# 181 w, \ Writeback, enabled, internal, not socketed, L1 +( 05 ) h# 180 w, \ Writeback, enabled, internal, not socketed, L1 ( 07 ) h# 8001 w, \ Max size - 64K ( 09 ) h# 8001 w, \ Installed size - 64K ( 0b ) 1 w, \ Supported SRAM type - unknown @@ -191,7 +192,7 @@ ( 01 ) h# 13 c, \ Length ( 02 ) h# 0703 w, \ Handle ( 04 ) 0 c, \ Socket string index -( 05 ) h# 182 w, \ Writeback, enabled, internal, not socketed, L2 +( 05 ) h# 181 w, \ Writeback, enabled, internal, not socketed, L2 ( 07 ) h# 8002 w, \ Max size - 128K ( 09 ) h# 8002 w, \ Installed size - 128K ( 0b ) 1 w, \ Supported SRAM type - unknown @@ -396,10 +397,11 @@ +OLPC " XO" +smbios$ " 1" +smbios$ \ Version + " SN" get-tag$ +smbios$ end-smbios-table base-board-info copy-smbios-table ( adr ) - +OLPC + " QUANTA" +smbios$ " XO" +smbios$ " 1" +smbios$ \ Version " SN" get-tag$ +smbios$

1 0

r1112 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 20:01:27 +0100 (Fri, 20 Feb 2009) New Revision: 1112 Modified: cpu/x86/pc/olpc/fw.bth Log: OLPC - turned off "stand-init-debug" to reduce the verbosity of the startup messages on the serial port. The messages that it eliminates are not associated with any common failure modes, so that particular verbosity is pointless. Modified: cpu/x86/pc/olpc/fw.bth =================================================================== --- cpu/x86/pc/olpc/fw.bth 2009-02-20 18:59:25 UTC (rev 1111) +++ cpu/x86/pc/olpc/fw.bth 2009-02-20 19:01:27 UTC (rev 1112) @@ -233,8 +233,8 @@ ; fload ${BP}/dev/geode/usb.fth -\ false to stand-init-debug? -true to stand-init-debug? +false to stand-init-debug? +\ true to stand-init-debug? hex : i-key-wait ( ms -- pressed? )

1 0

r1111 - dev/olpc/spiflash
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 19:59:25 +0100 (Fri, 20 Feb 2009) New Revision: 1111 Modified: dev/olpc/spiflash/spiui.fth Log: OLPC reflasher - use verify-firmware instead of verify after reflashing. It makes no difference for the normal OLPC build, but it prevents spurious verify failures in the custom build case where you enable config variable storage in SPI FLASH. Modified: dev/olpc/spiflash/spiui.fth =================================================================== --- dev/olpc/spiflash/spiui.fth 2009-02-20 18:56:33 UTC (rev 1110) +++ dev/olpc/spiflash/spiui.fth 2009-02-20 18:59:25 UTC (rev 1111) @@ -265,11 +265,11 @@ write-firmware spi-us d# 20 < if - ['] verify catch if + ['] verify-firmware catch if ." Verify failed. Retrying once" cr spi-identify write-firmware - verify + verify-firmware then flash-write-disable else

1 0

r1110 - cpu/x86/pc
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 19:56:33 +0100 (Fri, 20 Feb 2009) New Revision: 1110 Modified: cpu/x86/pc/biosints.fth Log: Biosints - turned off the "unimplemented INT" messages for the two functions that we intentionally don't implement. Modified: cpu/x86/pc/biosints.fth =================================================================== --- cpu/x86/pc/biosints.fth 2009-02-20 18:16:26 UTC (rev 1109) +++ cpu/x86/pc/biosints.fth 2009-02-20 18:56:33 UTC (rev 1110) @@ -839,8 +839,9 @@ h# c1 of 'ebda 4 rshift rm-es! endof \ Segment address of extended BIOS data area h# c2 of handle-mouse endof h# e8 of bigmem-int endof -\ h# e9 of endof \ Don't know what this is. Ralf Brown's interrupt list says -\ \ PhysTechSoft PTS ROM-DOS, but I doubt that is right + h# e9 of rm-set-cf endof \ Don't know what this is. Ralf Brown's interrupt list says + \ PhysTechSoft PTS ROM-DOS, but I doubt that is right + \ Windows invokes it but seems to be okay with it failing ( default ) rm-set-cf ." Unsupported INT 15 AH=" dup . cr endcase @@ -927,7 +928,10 @@ h# 0b of rm-cl@ cfgadr config-b! endof h# 0c of rm-cx@ cfgadr config-w! endof h# 0d of rm-ecx@ cfgadr config-l! endof + + h# 0e of h# 81 rm-ah! rm-set-cf endof \ Fail PCI interrupt routing for now \ h# 0e of pci-int-rout endof + \ h# 0f of set-pci-int endof ( default ) h# 81 rm-ah! rm-set-cf

1 0

r1109 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 19:16:26 +0100 (Fri, 20 Feb 2009) New Revision: 1109 Modified: cpu/x86/pc/olpc/smbios.fth Log: SMBIOS - fixed the UUID routine. Modified: cpu/x86/pc/olpc/smbios.fth =================================================================== --- cpu/x86/pc/olpc/smbios.fth 2009-02-20 09:17:56 UTC (rev 1108) +++ cpu/x86/pc/olpc/smbios.fth 2009-02-20 18:16:26 UTC (rev 1109) @@ -337,23 +337,26 @@ ; : get-tag$ ( tag$ -- value$ ) find-tag 0= if " Not Available" then ?-null ; -: too-long? ( dst-adr -- dst-adr flag ) dup pad - d# 10 >= ; +: too-long? ( dst-adr -- dst-adr flag ) dup pad - h# 10 >= ; : (uuid) ( -- true | adr len false ) " U#" find-tag if ( adr len ) + ?-null ( adr len' ) pad -rot ( dst-adr adr len ) bounds ?do ( dst-adr ) too-long? if drop true unloop exit then i c@ h# 10 digit if ( dst-adr digith ) i 1+ c@ h# 10 digit if ( dst-adr digith digitl ) - swap bwjoin over c! ( dst-adr ) - 1+ ( dst-adr' ) + swap 4 lshift or ( dst-adr byte ) + over c! 1+ ( dst-adr' ) + 2 ( dst-adr advance ) else ( dst-adr digith char ) 3drop true unloop exit then ( dst-adr ) else ( dst-adr char ) [char] - <> if drop true unloop exit then - then ( dst-adr ) - loop ( dst-adr ) + 1 ( dst-adr advance ) + then ( dst-adr advance ) + +loop ( dst-adr ) pad tuck - ( adr len ) dup h# 10 = if ( adr len ) false \ Good UUID ( adr len false )

1 0

r1108 - in cpu/x86/pc: biosload olpc
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 10:17:56 +0100 (Fri, 20 Feb 2009) New Revision: 1108 Modified: cpu/x86/pc/biosload/c32vesa.fth cpu/x86/pc/biosload/callvbe.fth cpu/x86/pc/biosload/fw.bth cpu/x86/pc/olpc/iflash.fth Log: Biosload version - improvements to VESA BIOS handling under Syslinux. Modified: cpu/x86/pc/biosload/c32vesa.fth =================================================================== --- cpu/x86/pc/biosload/c32vesa.fth 2009-02-20 09:14:16 UTC (rev 1107) +++ cpu/x86/pc/biosload/c32vesa.fth 2009-02-20 09:17:56 UTC (rev 1108) @@ -1,8 +1,8 @@ purpose: Call VESA BIOS functions via syslinux c32 callback -[ifdef] notdef-get-com32-ptr -\ sp 0 #) mov \ Save COM32 stack pointer to get parameters -[then] +\ [ifdef] notdef-get-com32-ptr + sp 0 #) mov \ Save COM32 stack pointer for possible later use +\ [then] [ifdef] notdef-getvbe \ There is an extra return address on the stack from the e9 call above d# 16 [sp] bp mov \ COM32 intcall helper function Modified: cpu/x86/pc/biosload/callvbe.fth =================================================================== --- cpu/x86/pc/biosload/callvbe.fth 2009-02-20 09:14:16 UTC (rev 1107) +++ cpu/x86/pc/biosload/callvbe.fth 2009-02-20 09:17:56 UTC (rev 1108) @@ -9,12 +9,12 @@ d# 16 [ax] bx mov \ COM32 intcall helper function d# 20 [ax] dx mov \ bounce buffer address - 4f02 # d# 36 [dx] mov \ AX + h# 4f02 # d# 36 [dx] mov \ AX cx d# 24 [dx] mov \ BX - \ dx push - 0 # push dx push +\ 0 # push + dx push h# 10 # push bx call @@ -22,3 +22,118 @@ bp pop di pop si pop c; + +: +c32-regs ( offset -- adr ) 0 @ d# 20 + @ + ; +: 'c32-gs ( -- adr ) 0 +c32-regs ; +: 'c32-fs ( -- adr ) 2 +c32-regs ; +: 'c32-es ( -- adr ) 4 +c32-regs ; +: 'c32-ds ( -- adr ) 6 +c32-regs ; + +: 'c32-di ( -- adr ) d# 8 +c32-regs ; +: 'c32-si ( -- adr ) d# 12 +c32-regs ; +: 'c32-bp ( -- adr ) d# 16 +c32-regs ; +: 'c32-bx ( -- adr ) d# 24 +c32-regs ; +: 'c32-dx ( -- adr ) d# 28 +c32-regs ; +: 'c32-cx ( -- adr ) d# 32 +c32-regs ; +: 'c32-ax ( -- adr ) d# 36 +c32-regs ; +: 'c32-eflags ( -- adr ) d# 40 +c32-regs ; + +code c32-intcall ( int# -- ) + cx pop + + si push di push bp push + 0 #) ax mov \ Pointer to COM32 args + d# 16 [ax] bx mov \ COM32 intcall helper function + d# 20 [ax] dx mov \ bounce buffer address + + dx push +\ 0 # push + dx push + cx push \ Int# + + bx call + ax pop ax pop ax pop + + bp pop di pop si pop +c; + +: vbe-call ( function# -- ) + 'c32-ax l! h# 10 c32-intcall + 'c32-ax l@ h# 4f <> abort" VESA BIOS call failed" +; + +: c32-es:di! ( adr -- ) >seg:off 'c32-es w! 'c32-di w! ; +: vbe-info ( -- adr ) + h# 200 +c32-regs ( adr ) + h# 32454256 over l! ( adr ) + c32-es:di! + h# 4f00 vbe-call + h# 200 +c32-regs ( adr ) +; +: .vesa-modes ( -- ) + vbe-info ( adr ) + dup l@ h# 41534556 <> if ( adr ) + ." VBE info call failed" cr + drop exit + then ( adr ) + dup 6 + seg:off@ .cstr cr ( adr ) + d# 14 + seg:off@ ( 'mode-list ) + begin dup w@ dup h# ffff <> while . wa1+ repeat cr 2drop +; +: vesa-mode-info ( mode# -- adr ) + 'c32-cx l! ( ) + h# 200 +c32-regs c32-es:di! ( ) + h# 4f01 vbe-call ( ) + h# 200 +c32-regs ( adr ) +; + +[ifdef] Commentary +00.w mode attributes +02.b wina attributes +03.b winb attributes +04.w winGranularity +06.w WinSize +08.w WinASegment +0a.w WinBSegment +0c.l WinFuncPtr +10.w BytesPerScanLine +12.w XResolution +14.w YResolution +16.b XCharSize +17.b YCharSize +18.b NumberOfPlanes +19.b BitsPerPixel +1a.b NumberOfBanks +1b.b MemoryModel +1c.b BankSize +1d.b NumberOfImagePlanes +1e.b Reserved +1f.b RedMaskSize +20.b RedFieldPosition +21.b GreenMaskSize +22.b GreenFieldPosition +23.b BlueMaskSize +24.b BlueFieldPosition +25.b RsvdMaskSize +26.b RsvdFieldPosition +27.b DirectColorModeInfo +\ Following are VBE 2.0 and above +28.l PhysBasePtr +2c.l Reserved +30.w Reserved +\ Following are VBE 3.0 and above +32.w LinBytesPerScanLine +34.b BnkNumberOfImagePages +35.b LinNumberOfImagePages +36.b LinRedMaskSize +37.b LinRedFieldPosition +38.b LinGreenMaskSize +39.b LinGreenFieldPosition +3a.b LinBlueMaskSize +3b.b LinBlueFieldPosition +3c.b LinRsvdMaskSize +3d.b LinRsvdFieldPosition +3e.l MaxPixelClock +[then] + +: current-vesa-mode ( -- mode# ) h# 4f03 vbe-call 'c32-bx l@ ; Modified: cpu/x86/pc/biosload/fw.bth =================================================================== --- cpu/x86/pc/biosload/fw.bth 2009-02-20 09:14:16 UTC (rev 1107) +++ cpu/x86/pc/biosload/fw.bth 2009-02-20 09:17:56 UTC (rev 1108) @@ -226,6 +226,10 @@ fload ${BP}/cpu/x86/pc/biosload/callbios.fth \ fload ${BP}/cpu/x86/pc/biosload/rmenter.fth +[ifdef] syslinux-loaded +fload ${BP}/cpu/x86/pc/biosload/callvbe.fth +[then] + \ false to stand-init-debug? true to stand-init-debug? Modified: cpu/x86/pc/olpc/iflash.fth =================================================================== --- cpu/x86/pc/olpc/iflash.fth 2009-02-20 09:14:16 UTC (rev 1107) +++ cpu/x86/pc/olpc/iflash.fth 2009-02-20 09:17:56 UTC (rev 1108) @@ -22,6 +22,7 @@ /mfg-data-merge move ; +[ifdef] get-mfg-data \ unpollute-mfg-data clears the Insyde BIOS residue from the beginning \ of the manufacturing data area, restoring it to erased state : unpollute-mfg-data ( -- ) @@ -30,6 +31,7 @@ mfg-data-buf mfg-data-merge-offset h# ff fill put-mfg-data ; +[then] : get-insyde ( ["filename"] -- ) parse-word ( adr len )

1 0

r1107 - cpu/x86/pc cpu/x86/pc/olpc dev/geode/display
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 10:14:16 +0100 (Fri, 20 Feb 2009) New Revision: 1107 Modified: cpu/x86/pc/biosints.fth cpu/x86/pc/olpc/dsdt.dsl dev/geode/display/gxvga.fth Log: OLPC - Fixed the VESA BIOS emulation for Windows, supporting mode 112 (640x480x32), 115 (800x600x32), and 118 (1024x768x32) Modified: cpu/x86/pc/biosints.fth =================================================================== --- cpu/x86/pc/biosints.fth 2009-02-20 08:23:30 UTC (rev 1106) +++ cpu/x86/pc/biosints.fth 2009-02-20 09:14:16 UTC (rev 1107) @@ -200,18 +200,52 @@ create mode12-info \ w b b w w w w d w w w b b b b b b b b b h# db w, \ Linear, (VGA), Graphics, Color, no TTY Output, D1=1, hardware-supported +create mode-112-info +\ h# bf w, \ Linear, VGA, Graphics, Color, TTY Output, D1=1, hardware-supported +\ 7 c, 0 c, d# 64 w, d# 64 w, h# a000 w, 0 w, 0 l, \ Not windowed + h# fb w, \ Linear, noVGA, Graphics, Color, noTTY Output, D1=1, hardware-supported + 0 c, 0 c, d# 0 w, d# 0 w, 0 w, 0 w, 0 l, \ Not windowed + d# 640 /l* w, \ BytesPerScanLine + d# 640 w, d# 480 w, \ X, Y res + d# 8 c, d# 16 c, \ Char width, height + 1 c, \ NumPlanes + d# 32 c, \ Bits/pixel + 1 c, \ NumBanks + 6 c, \ MemModel - DirectColor + 0 c, \ Bank size (not banked) + 2 c, \ NumImagePages + 1 c, \ Reserved + + \ Banked Color info + 8 c, d# 16 c, 8 c, 8 c, 8 c, 0 c, 0 c, 0 c, \ {RGBX}{Bits,Pos} + 0 c, \ Gamma fixed (change if we implement function 9) + + fb-pci-base l, 0 l, 0 w, \ Framebuffer address + + \ Linear info + d# 640 /l* w, \ Bytes per scan line + 0 c, \ No banks + +\ fbsize d# 640 / d# 480 / 4 / c, ( need to account for cmd buffer ) + d# 11 c, \ Number of images that will fit in framebuffer + + 8 c, d# 16 c, 8 c, 8 c, 8 c, 0 c, 0 c, 0 c, \ {RGBX}{Bits,Pos} + + d# 56,200,000 l, \ Max pixel clock +here mode-112-info - constant /mode-112-info + create mode-115-info - h# fb w, \ Linear, NotVGA, Graphics, Color, no TTY Output, D1=1, hardware-supported - 0 c, 0 c, 0 w, d# 64 w, 0 w, 0 w, 0 l, \ Not windowed - 0 w, \ BytesPerScanLine irrelevant in linear mode + h# fb w, \ Linear, VGA, Graphics, Color, TTY Output, D1=1, hardware-supported + 0 c, 0 c, d# 0 w, d# 0 w, h# 0 w, 0 w, 0 l, \ Not windowed + d# 800 /l* w, \ BytesPerScanLine d# 800 w, d# 600 w, \ X, Y res - d# 15 c, d# 18 w, \ Char width, height (sort of irrelevant) - 1 c, \ NumPlanes (irrelevant for linear?) - d# 24 c, \ Bits/pixel + d# 8 c, d# 16 c, \ Char width, height + 1 c, \ NumPlanes + d# 32 c, \ Bits/pixel 1 c, \ NumBanks - 6 c, \ MemModel - DirectColor (?) + 6 c, \ MemModel - DirectColor 0 c, \ Bank size (not banked) - 0 c, \ NumImagePages + 2 c, \ NumImagePages 1 c, \ Reserved \ Banked Color info @@ -221,11 +255,11 @@ fb-pci-base l, 0 l, 0 w, \ Framebuffer address \ Linear info - d# 4096 /w* w, \ Bytes per scan line + d# 800 /l* w, \ Bytes per scan line 0 c, \ No banks -\ fbsize d# 1200 / d# 900 / 2/ c, - 3 c, \ Number of images that will fit in framebuffer +\ fbsize d# 800 / d# 600 / 4/ c, + 7 c, \ Number of images that will fit in framebuffer 8 c, d# 16 c, 8 c, 8 c, 8 c, 0 c, 0 c, 0 c, \ {RGBX}{Bits,Pos} @@ -233,17 +267,18 @@ here mode-115-info - constant /mode-115-info create mode-118-info - h# fb w, \ Linear, NotVGA, Graphics, Color, no TTY Output, D1=1, hardware-supported - 0 c, 0 c, 0 w, d# 64 w, 0 w, 0 w, 0 l, \ Not windowed - 0 w, \ BytesPerScanLine irrelevant in linear mode + h# fb w, \ Linear, VGA, Graphics, Color, TTY Output, D1=1, hardware-supported + 0 c, 0 c, d# 0 w, d# 0 w, h# 0 w, 0 w, 0 l, \ Not windowed + d# 1024 /l* w, \ BytesPerScanLine d# 1024 w, d# 768 w, \ X, Y res - d# 15 c, d# 18 w, \ Char width, height (sort of irrelevant) - 1 c, \ NumPlanes (irrelevant for linear?) - d# 24 c, \ Bits/pixel +\ d# 12 c, d# 15 c, \ Char width, height + d# 8 c, d# 16 c, \ Char width, height (sort of irrelevant) + 1 c, \ NumPlanes + d# 32 c, \ Bits/pixel 1 c, \ NumBanks - 6 c, \ MemModel - DirectColor (?) + 6 c, \ MemModel - DirectColor 0 c, \ Bank size (not banked) - 0 c, \ NumImagePages + 2 c, \ NumImagePages 1 c, \ Reserved \ Banked Color info @@ -253,11 +288,11 @@ fb-pci-base l, 0 l, 0 w, \ Framebuffer address \ Linear info - d# 4096 /w* w, \ Bytes per scan line + d# 1024 /l* w, \ Bytes per scan line 0 c, \ No banks -\ fbsize d# 1200 / d# 900 / 2/ c, - 3 c, \ Number of images that will fit in framebuffer +\ fbsize d# 1024 / d# 768 / 4 / c, + 4 c, \ Number of images that will fit in framebuffer 8 c, d# 16 c, 8 c, 8 c, 8 c, 0 c, 0 c, 0 c, \ {RGBX}{Bits,Pos} @@ -266,11 +301,11 @@ create mode-120-info h# fb w, \ Linear, NotVGA, Graphics, Color, no TTY Output, D1=1, hardware-supported - 0 c, 0 c, 0 w, d# 64 w, 0 w, 0 w, 0 l, \ Not windowed - 0 w, \ BytesPerScanLine irrelevant in linear mode + 0 c, 0 c, d# 0 w, d# 0 w, h# 0 w, 0 w, 0 l, \ Not windowed + d# 1200 /w* w, \ BytesPerScanLine d# 1200 w, d# 900 w, \ X, Y res - d# 15 c, d# 18 w, \ Char width, height (sort of irrelevant) - 1 c, \ NumPlanes (irrelevant for linear?) + d# 15 c, d# 18 c, \ Char width, height (sort of irrelevant) + 1 c, \ NumPlanes d# 16 c, \ Bits/pixel 1 c, \ NumBanks 6 c, \ MemModel - DirectColor (?) @@ -289,7 +324,7 @@ 0 c, \ No banks \ fbsize d# 1200 / d# 900 / 2/ c, - 7 c, \ Number of images that will fit in framebuffer + 6 c, \ Number of images that will fit in framebuffer 5 c, d# 11 c, 6 c, 5 c, 5 c, 0 c, 0 c, 0 c, \ {RGBX}{Bits,Pos} @@ -309,9 +344,11 @@ : vbe-ok ( -- ) h# 4f rm-ax! ; : vbe-modes ( -- ) +\ ." vbe-modes" cr ?vbe2 h# 41534556 0 >vbe-pa l! \ VbeSignature - h# 0300 4 >vbe-pa w! \ VbeVersion +\ h# 0300 4 >vbe-pa w! \ VbeVersion + h# 0200 4 >vbe-pa w! \ VbeVersion 1 d# 10 >vbe-pa l! \ Capabilities - 8-bit DACs vbebuf d# 34 + d# 14 >vbe-pa seg:off! \ VbeFarPtr to mode list fbsize d# 16 rshift d# 18 >vbe-pa w! \ TotalMemory @@ -328,33 +365,56 @@ d# 34 >vbe-pa \ 3 w!++ \ Text mode 3 \ h# 12 w!++ \ Graphics mode 12 - h# 115 w!++ \ 800x600x24 - h# 118 w!++ \ 1024x768x24 - h# 120 w!++ \ OLPC native mode + h# 112 w!++ \ 640x480x32 + h# 115 w!++ \ 800x600x32 + h# 118 w!++ \ 1024x768x32 +\ h# 120 w!++ \ OLPC native mode -1 swap w! \ End of list vbe-ok ; : vbe-get-mode ( -- ) - rm-cx@ case - h# 115 of mode-115-info /mode-115-info endof - h# 118 of mode-118-info /mode-118-info endof - h# 120 of mode-120-info /mode-120-info endof + + rm-cx@ +\ ." vbe-get-mode " dup . cr +h# 1ff and case + h# 112 of mode-112-info /mode-112-info endof + h# 115 of mode-115-info /mode-115-info endof + h# 118 of mode-118-info /mode-118-info endof +\ h# 120 of mode-120-info /mode-120-info endof ( default ) ." Bad VBE mode number " dup . cr 0 0 rot endcase ( adr len ) 0 >vbe-pa h# 100 erase 0 >vbe-pa swap move + vbe-ok ; +: vbe-error ( -- ) h# 014f rm-ax! ; +0 value vbe-this-mode : vbe-set-mode ( -- ) - ." VBE set mode " rm-cx@ . cr - debug-me + rm-bx@ case + h# 4112 of " 640x480x32" endof + h# 4115 of " 800x600x32" endof + h# 4118 of " 1024x768x32" endof + ( default ) drop vbe-error exit + endcase + + " screen-ih" eval ['] $call-method catch if + 3drop vbe-error exit + then + rm-bx@ to vbe-this-mode + vbe-ok ; +: vbe-current-mode ( -- ) + vbe-this-mode rm-bx! + vbe-ok +; : vesa-bios ( -- ) rm-al@ case h# 00 of vbe-modes endof h# 01 of vbe-get-mode endof h# 02 of vbe-set-mode endof + h# 03 of vbe-current-mode endof ( default ) ." Unsupported VBE function" dup .x cr endcase ; Modified: cpu/x86/pc/olpc/dsdt.dsl =================================================================== --- cpu/x86/pc/olpc/dsdt.dsl 2009-02-20 08:23:30 UTC (rev 1106) +++ cpu/x86/pc/olpc/dsdt.dsl 2009-02-20 09:14:16 UTC (rev 1107) @@ -771,7 +771,6 @@ Name (_UID, One) Method (_CRS, 0, NotSerialized) { Name (MBRB, ResourceTemplate () { - Memory32Fixed (ReadWrite, 0x00000000, 0x000A0000, ) Memory32Fixed (ReadOnly, 0x000E0000, 0x00020000, ) Memory32Fixed (ReadWrite, 0x00100000, 0x00000000, _Y06) // Edited below // Assumes that the SMM memory is at 8040.0000, I think @@ -781,7 +780,6 @@ IO (Decode16, 0x0092, 0x0092, 0x00, 0x01, ) IO (Decode16, 0x0030, 0x0030, 0x00, 0x10, ) // I/O ports to fake out SMI interrupts - IO (Decode16, 0x03c0, 0x03c0, 0x00, 0x20, ) // Claim VGA I/O ports // We claim the EC ports here instead of in an EC node // because we don't want Windows to load an EC driver // nor to complain about not being able to do so. @@ -839,7 +837,7 @@ Name (_HID, EisaId ("PNP0501")) Name (_UID, One) Name (_DDN, "COM1") - Name (_STA, 3) // Present and decoding resources, but shouldn't be shown + Name (_STA, 0xf) // You can use the serial port if you open the box Name (_CRS, ResourceTemplate () { IO (Decode16, 0x03F8, 0x03F8, 0x00, 0x08, ) IRQNoFlags () {4} Modified: dev/geode/display/gxvga.fth =================================================================== --- dev/geode/display/gxvga.fth 2009-02-20 08:23:30 UTC (rev 1106) +++ dev/geode/display/gxvga.fth 2009-02-20 09:14:16 UTC (rev 1107) @@ -554,26 +554,38 @@ 1 ms \ Let memory accesses finish 4 dc@ 1 invert and 4 dc! \ Turn off FIFO load ; -: 8bpp ( -- ) 8 dc@ h# 300 invert and 8 dc! ; \ drop down to 8bpp mode +: 8bpp ( -- ) 8 dc@ h# 300 invert and 8 dc! ; +: 16bpp ( -- ) 8 dc@ h# 300 invert and h# 100 or 8 dc! ; +: 24bpp ( -- ) 8 dc@ h# 300 invert and h# 200 or 8 dc! ; +: 32bpp ( -- ) 8 dc@ h# 300 or 8 dc! ; : vga-start ( -- ) 8 dc@ 1 or 8 dc! \ Timing generator \ Fixed timing must be off to use the upscaler 4 dc@ h# 40000 invert and h# 80 or 4 dc! \ Fixed Timing (40000) off, VGA (80) on ; -: display-lfb ( -- ) 5601 4 dc! ; \ Enable linear framebuffer +: display-lfb ( -- ) 6501 4 dc! ; \ Enable linear framebuffer -: setup-filter-mode3 ( -- ) -\ d# 400 1- d# 640 1- wljoin h# 5c dc! +: setup-filter-720x400 ( -- ) d# 400 1- d# 720 1- wljoin h# 5c dc! h# 1c802666 h# 90 dc! filter-taps ; -: setup-filter-mode12 ( -- ) +: setup-filter-640x480 ( -- ) d# 480 1- d# 640 1- wljoin h# 5c dc! h# 22222222 h# 90 dc! filter-taps ; +: setup-filter-800x600 ( -- ) + d# 600 1- d# 800 1- wljoin h# 5c dc! + h# 2ab42ab4 h# 90 dc! + filter-taps +; +: setup-filter-1024x768 ( -- ) + d# 768 1- d# 1024 1- wljoin h# 5c dc! + h# 369d369d h# 90 dc! + filter-taps +; : filteron ( -- ) h# 1000 h# 94 dc! ; : filteroff ( -- ) h# 40004000 h# 90 dc! @@ -629,7 +641,7 @@ \ otherwise the filter and the VGA don't synchronize to the \ frame and the sequencer shuts off after one frame. \ That can be fixed with all-off vga-start - setup-filter-mode3 filteron + setup-filter-720x400 filteron all-off 8bpp vga-start use-vga-dac @@ -642,7 +654,7 @@ \ otherwise the filter and the VGA don't synchronize to the \ frame and the sequencer shuts off after one frame. \ That can be fixed with all-off vga-start - setup-filter-mode12 filteron + setup-filter-640x480 filteron all-off 8bpp vga-start use-vga-dac @@ -657,6 +669,36 @@ : graphics-mode12 ( -- ) switch-to-vga-mode12 graphics-mode12 ; +: 640x480x32 ( -- ) \ vesa-mode112 + set-mode \ Get back into native resolution linear mode + unlock + 32bpp + d# 640 d# 32 * 6 rshift dup h# 30 dc! h# 34 dc! + all-off setup-filter-640x480 filteron + display-on + lock +; + +: 800x600x32 ( -- ) \ vesa-mode115 + set-mode \ Get back into native resolution linear mode + unlock + 32bpp + d# 800 d# 32 * 6 rshift dup h# 30 dc! h# 34 dc! + all-off setup-filter-800x600 filteron + display-on + lock +; + +: 1024x768x32 ( -- ) \ vesa-mode118 + set-mode \ Get back into native resolution linear mode + unlock + 32bpp + d# 1024 d# 32 * 6 rshift dup h# 30 dc! h# 34 dc! + all-off setup-filter-1024x768 filteron + display-on + lock +; + warning ! \ LICENSE_BEGIN

1 0

r1106 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 20, 2009

Feb. 20, 2009

Author: wmb Date: 2009-02-20 09:23:30 +0100 (Fri, 20 Feb 2009) New Revision: 1106 Modified: cpu/x86/pc/olpc/vsapci.fth Log: OLPC - Fixed the longstanding problem whereby the Windows XO Display Driver wouldn't claim the legacy resources. The problem was that the "IO Space Enable" bit in the virtual config register for the display was off. That made the Windows VideoPort driver refuse to claim the resources. Modified: cpu/x86/pc/olpc/vsapci.fth =================================================================== --- cpu/x86/pc/olpc/vsapci.fth 2009-02-12 08:26:53 UTC (rev 1105) +++ cpu/x86/pc/olpc/vsapci.fth 2009-02-20 08:23:30 UTC (rev 1106) @@ -157,7 +157,7 @@ ff800000 , ffffc000 , ffffc000 , ffffc000 , 0 , 0 , 0 , 0 , - 30100b , 2200002 , 3000000 , 8 , + 30100b , 2200003 , 3000000 , 8 , fb-pci-base , gp-pci-base , dc-pci-base , vp-pci-base , \ FB, GP, DC, VP 0 , 0 , 0 , 30100b , \ VIP (LX only) 0 , 0 , 0 , 0 ,

1 0

Towers of Hanoi in OF
by C.W. Betts Feb. 19, 2009

Feb. 19, 2009

Has anyone tried to get the Towers of Hanoi to run in Open Firmware? http://www.kernelthread.com/mac/apme/hanoi/ Qemu has started bundling Open Firmware with its PowerPC implementation, and this would be a good test to see if it could use Towers of Hanoi like the Mac's Firmware.

2 1

r1105 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 12, 2009

Feb. 12, 2009

Author: rsmith Date: 2009-02-12 09:26:53 +0100 (Thu, 12 Feb 2009) New Revision: 1105 Modified: cpu/x86/pc/olpc/versions.fth Log: OLPC: q2e32 Modified: cpu/x86/pc/olpc/versions.fth =================================================================== --- cpu/x86/pc/olpc/versions.fth 2009-02-12 06:38:23 UTC (rev 1104) +++ cpu/x86/pc/olpc/versions.fth 2009-02-12 08:26:53 UTC (rev 1105) @@ -2,10 +2,10 @@ \ The overall firmware revision macro: FW_MAJOR E -macro: FW_MINOR 31 +macro: FW_MINOR 32 \ The EC microcode -macro: EC_VERSION e21 +macro: EC_VERSION e32 \ Alternate command for getting EC microcode, for testing new versions. \ Temporarily uncomment the line and modify the path as necessary

1 0

r1104 - dev/olpc/kb3700
by svn＠openfirmware.info Feb. 11, 2009

Feb. 11, 2009

Author: rsmith Date: 2009-02-12 07:38:23 +0100 (Thu, 12 Feb 2009) New Revision: 1104 Modified: dev/olpc/kb3700/battery.fth dev/olpc/kb3700/ecio.fth Log: OLPC: Integrate a sizeable chunk of the tools used in the extern batman.fth diagnostic script into the firmware directly. Modified: dev/olpc/kb3700/battery.fth =================================================================== --- dev/olpc/kb3700/battery.fth 2009-02-12 05:40:08 UTC (rev 1103) +++ dev/olpc/kb3700/battery.fth 2009-02-12 06:38:23 UTC (rev 1104) @@ -5,6 +5,7 @@ : ec@ wbsplit h# 381 pc! h# 382 pc! h# 383 pc@ ; : ec! wbsplit h# 381 pc! h# 382 pc! h# 383 pc! ; [then] + : wextend ( w -- n ) dup h# 8000 and if h# ffff.0000 or then ; : rr ec@ . ; : ww ec! ; @@ -115,8 +116,348 @@ begin (cr .bat kill-line d# 1000 ms key? until key drop ; -\ send questions to andrew at gold peak +\ A few commands for looking at what the EC is doing in +\ the battery state machine. +\ Unless you are on a serial console with stdout turned off +\ (stdout off) this will miss some state changes. +\ +: next-bstate + begin + f780 ec@ tuck <> + until +; + +: see-bstate + 0 begin + next-bstate dup . + dup 4 = if + cr + then key? + until +; + +: next-e1-state + begin + fae1 ec@ tuck <> + until +; + +: see-e1 + 0 begin + next-e1-state dup . + key? + until +; + +\ Turn on the charging mosfet +: bat-enable-charge ( -- ) fc21 ec@ 40 or fc21 ec! ; + +\ Turn off the charging mosfet +: bat-disable-charge ( -- ) fc21 ec@ 40 invert and fc21 ec! 03 fc24 ec! ; + +\ Turn on the trickle charge with max system voltage +: bat-enable-trickle ( -- ) fc23 ec@ 1 or fc23 ec! 01 fc24 ec! ; + +\ Turn off the trickle charger +: bat-disable-trickle ( -- ) fc23 ec@ 1 invert and fc23 ec! ; + +\ Access the 1-wire data line via the EC GPIO ports + +h# 383 constant dataport +h# 382 constant lowadr +4 constant dq-bit +0 value high +0 value low +false value 1w-initialized +0 value fileih + +: disable-ec-charging + 1 h# fa07 ec! +; + +: disable-ec-1w + 1 h# fa08 ec! +; + +: enable-ec-charging + 0 h# fa07 ec! +; + +: enable-ec-1w + 0 h# fa08 ec! +; + +: 1w-init ( -- ) + disable-interrupts + h# fc24 ec@ dq-bit invert and fc24 ec! + h# fc14 ec@ dup dq-bit or to low dq-bit invert and to high + high fc14 ec! + true to 1w-initialized + 1 ms +; + + +\ New ec revs can turn off the battery subsystem without +\ putting the ec into reset allowing the user to +\ use the XO keyboard. +: init-ec-live + disable-ec-charging + disable-ec-1w + 1w-init +; + +: init-ec-dead + kbc-off + 1w-init +; + +\ EC code at the time these commands were added can stop +\ the battery subsystem with out putting the EC into reset. +\ so we always do a live init +: batman-init init-ec-live 500 ms bat-disable-charge bat-disable-trickle ; +: batman-init? 1w-initialized if else batman-init then ; + + +: batman-start batman-init? ; + +: batman-stop + enable-ec-1w + enable-ec-charging + enable-interrupts + false to 1w-initialized +; + +: bit? ( us -- flag ) \ Test the data after a delay + us h# 34 lowadr pc! dataport pc@ dq-bit and 0<> +; + +: 1w-pulse ( us -- ) \ Pulse the wire low for some duration + h# 14 lowadr pc! low dataport pc! us high dataport pc! +; + +\ Generic 1-wire primitives + +: 1w-reset ( -- ) + d# 480 1w-pulse + d# 67 bit? abort" No response from battery" + begin 1 bit? until +; + +: 1w-write-byte ( byte -- ) + 8 0 do ( byte ) + dup 1 and if ( byte ) + 1 1w-pulse d# 60 us ( byte ) + else ( byte ) + d# 60 1w-pulse ( byte ) + then ( byte ) + 2/ ( byte' ) + loop ( byte ) + drop ( ) +; + +: 1w-read-byte ( -- ) + 0 8 0 do + 1 1w-pulse + \ Shift bits in from the left, little endian + d# 10 bit? h# 100 and or 2/ d# 50 us + loop +; + +: 1w-skip-address ( -- ) h# cc 1w-write-byte ; + +: 1w-cmd ( arg cmd -- ) 1w-reset 1w-skip-address 1w-write-byte 1w-write-byte ; + +\ Basic commands for the DS2756 chip + +: 1w-read ( adr len start -- ) + h# 69 1w-cmd ( adr len ) + bounds ?do 1w-read-byte i c! loop ( ) + 1w-reset +; + +: 1w-write ( adr len start -- ) + h# 6c 1w-cmd ( adr len ) + bounds ?do i c@ 1w-write-byte loop ( ) + 1w-reset +; + +: 1w-write-start ( start -- ) + h# 6c 1w-cmd +; + +: 1w-copy ( start -- ) h# 48 1w-cmd d# 10 ms ; + +: 1w-recall ( start -- ) h# b8 1w-cmd ; + +\ : 1w-lock ( start -- ) h# 6a 1w-cmd ; + +: 1w-sync ( -- ) 0 h# d2 1w-cmd ; + +\ Some higher-level commands for accessing battery data + +\ buffer for reading bank data +h# 20 constant /ds-bank \ Bytes per bank in the battery sensor chip +h# 60 constant /ds-eeprom \ Bytes in the eeprom +h# 01 constant ni-mh +h# 02 constant li-fe +h# 00 constant ds-regs +h# 10 constant ds-acr +h# 20 constant ds-bank0 +h# 25 constant ds-bat-misc-flag +h# 2d constant ds-last-dis-soc +h# 2e constant ds-last-dis-acr-msb +h# 2f constant ds-last-dis-acr-lsb +h# 40 constant ds-bank1 +h# 60 constant ds-bank2 +h# 68 constant ds-remain-acr-msb +h# 69 constant ds-remain-acr-lsb +h# 5f constant ds-batid + +h# 01 constant ds-bat-low-volt +h# 20 constant ds-bat-full + +h# 20 buffer: ds-bank-buf + +: ds-bank$ ( -- adr len ) ds-bank-buf /ds-bank ; + +: bat-id ( -- id ) ds-bank-buf 1 ds-batid 1w-read ds-bank-buf c@ h# 0f and ; + +: read-bank ( offset -- ) ds-bank$ rot 1w-read ; + +: bat-read-eeprom ( -- ) ds-bank-buf /ds-eeprom ds-bank0 1w-read ; + +: bat-ds-regs@ ( -- ) ds-regs read-bank ; +: bat-bank0@ ( -- ) ds-bank0 read-bank ; +: bat-bank1@ ( -- ) ds-bank1 read-bank ; +: bat-bank2@ ( -- ) ds-bank2 read-bank ; + +: bat-dump-bank ( -- ) ds-bank$ dump ; + +: bat-dump-banks ( -- ) + batman-init? + cr ." Regs" + bat-ds-regs@ bat-dump-bank + cr cr ." Bank 0" + bat-bank0@ bat-dump-bank + cr cr ." Bank 1" + bat-bank1@ bat-dump-bank + cr cr ." Bank 2" + bat-bank2@ bat-dump-bank +; + +: bat-dump-regs ( -- ) batman-init? bat-ds-regs@ bat-dump-bank ; + +: w16>d32 ( 16bit -- 32bit_sign-extended ) + d# 16 << d# 16 >>a +; + +: bat-save ( -- ) + " disk:\battery.dmp" + 2dup ['] $delete catch if 2drop then ( name$ ) + $create-file to fileih + + 1w-init + h# 80 0 do + ds-bank$ i 1w-read + ds-bank$ " write" fileih $call-method + /ds-bank +loop + + fileih close-dev +; + +\ bg-* words access the gauge directly via 1w rather than +\ read the value from the ec cache +: bg-acr@ ( -- acr ) + batman-init? + ds-bank-buf 2 ds-acr 1w-read ( ) + ds-bank-buf c@ 8 << ( msb ) + ds-bank-buf 1 + c@ or w16>d32 ( acr ) +; + +: bg-acr! ( acr -- ) + batman-init? + wbsplit + ds-acr 1w-write-start + 1w-write-byte + 1w-write-byte + 1w-reset +; + +: bg-last-dis-soc@ ( -- last-dis-soc ) + batman-init? + ds-bank-buf 1 ds-last-dis-soc 1w-read + ds-bank-buf c@ +; + +: bg-last-dis-soc! ( soc -- ) + batman-init? + ds-bank0 1w-recall + ds-last-dis-soc 1w-write-start + 1w-write-byte + ds-last-dis-soc 1w-copy + 1w-reset +; + +: bg-last-dis-acr@ ( -- last-dis-acr ) + batman-init? + ds-bank-buf 2 ds-last-dis-acr-msb 1w-read ( ) + ds-bank-buf c@ 8 << ( last-dis-acr-msb ) + ds-bank-buf 1 + c@ or w16>d32 ( last-dis-acr ) +; + +: bg-last-dis-acr! ( last-dis-acr --) + batman-init? + wbsplit + ds-bank0 1w-recall + ds-last-dis-acr-msb 1w-write-start + 1w-write-byte + 1w-write-byte + ds-last-dis-acr-msb 1w-copy + 1w-reset +; + +: bg-misc@ ( -- misc-flag ) + batman-init? + ds-bank-buf 1 ds-bat-misc-flag 1w-read + ds-bank-buf c@ +; + +: bg-set-full-flag ( -- ) + bg-misc@ + ds-bat-full or + ds-bank0 1w-recall + ds-bat-misc-flag 1w-write-start + 1w-write-byte + ds-bat-misc-flag 1w-copy + 1w-reset +; + +: bg-clear-full-flag ( -- ) + bg-misc@ + ds-bat-full not and + ds-bank0 1w-recall + ds-bat-misc-flag 1w-write-start + 1w-write-byte + ds-bat-misc-flag 1w-copy + 1w-reset +; + +: .bg-eeprom + base @ >r + decimal + ." acr: " bg-acr@ . cr + ." Last dis soc: " bg-last-dis-soc@ . cr + ." Last dis acr: " bg-last-dis-acr@ . cr + hex + ." Misc flags: " bg-misc@ dup . ." : " + dup ds-bat-full and if ." fully charged " then + dup ds-bat-low-volt and if ." low voltage " then + drop + cr + r> base ! +; + dev / new-device " battery" device-name Modified: dev/olpc/kb3700/ecio.fth =================================================================== --- dev/olpc/kb3700/ecio.fth 2009-02-12 05:40:08 UTC (rev 1103) +++ dev/olpc/kb3700/ecio.fth 2009-02-12 06:38:23 UTC (rev 1104) @@ -8,6 +8,18 @@ : ec@ ( index -- b ) wbsplit iobase 1+ pc! iobase 2+ pc! iobase 3 + pc@ ; : ec! ( b index -- ) wbsplit iobase 1+ pc! iobase 2+ pc! iobase 3 + pc! ; +\ Write a word to an EC index. +: ecw! ( wdata index ) + wbsplit ( wdata index_lsb index_msb ) + iobase 1+ pc! ( wdata index_lsb ) + dup iobase 2+ pc! ( wdata index_lsb ) + swap wbsplit ( index_lsb wdata_lsb wdata_msb ) + iobase 3 + pc! ( index_lsb wdata_lsb ) + swap 1 + ( wdata_lsb index_lsb+1 ) + iobase 2+ pc! ( wdata_lsb ) + iobase 3 + pc! +; + : autowak! ( value -- ) f64f ec! ; : autowak-on ( -- ) 1 autowak! ; : autowak-off ( -- ) 0 autowak! ; @@ -145,6 +157,8 @@ too-many-retries ; +: ec-wakeup! ( wakeup -- ) lbsplit h# 36 ec-cmd-out ec-wb ec-wb ec-wb ec-wb ; + : ec-abnormal@ ( -- b ) h# 1f ec-cmd-b@ ; : bat-init-nimh-gp ( -- ) h# 20 ec-cmd ;

1 0

r1103 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 11, 2009

Feb. 11, 2009

Author: wmb Date: 2009-02-12 06:40:08 +0100 (Thu, 12 Feb 2009) New Revision: 1103 Modified: cpu/x86/pc/olpc/keyject.fth Log: Removed keyjector warnings. Modified: cpu/x86/pc/olpc/keyject.fth =================================================================== --- cpu/x86/pc/olpc/keyject.fth 2009-02-12 02:37:52 UTC (rev 1102) +++ cpu/x86/pc/olpc/keyject.fth 2009-02-12 05:40:08 UTC (rev 1103) @@ -5,13 +5,6 @@ \ See HowItWorks near end of file for a description of the overall procedure -create test-me -[ifdef] test-me -.( WARNING WARNING - keyjector test code installed) cr -[else] -.( Warning - keyjector version) cr -[then] - \ !!! Re-implement this for each different deployment : wrong-sku? ( -- flag ) " P#" find-tag 0= if true exit then ( pn$ )

1 0

r1102 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 11, 2009

Feb. 11, 2009

Author: wmb Date: 2009-02-12 03:37:52 +0100 (Thu, 12 Feb 2009) New Revision: 1102 Modified: cpu/x86/pc/olpc/kjversions.fth cpu/x86/pc/olpc/versions.fth Log: Q2E31 Modified: cpu/x86/pc/olpc/kjversions.fth =================================================================== --- cpu/x86/pc/olpc/kjversions.fth 2009-02-10 23:59:41 UTC (rev 1101) +++ cpu/x86/pc/olpc/kjversions.fth 2009-02-12 02:37:52 UTC (rev 1102) @@ -2,7 +2,7 @@ \ The overall firmware revision macro: FW_MAJOR E -macro: FW_MINOR 30b +macro: FW_MINOR 30x \ The EC microcode macro: EC_VERSION e21 Modified: cpu/x86/pc/olpc/versions.fth =================================================================== --- cpu/x86/pc/olpc/versions.fth 2009-02-10 23:59:41 UTC (rev 1101) +++ cpu/x86/pc/olpc/versions.fth 2009-02-12 02:37:52 UTC (rev 1102) @@ -2,7 +2,7 @@ \ The overall firmware revision macro: FW_MAJOR E -macro: FW_MINOR 30 +macro: FW_MINOR 31 \ The EC microcode macro: EC_VERSION e21

1 0

r1101 - cpu/x86/pc/olpc dev/olpc/spiflash
by svn＠openfirmware.info Feb. 10, 2009

Feb. 10, 2009

Author: wmb Date: 2009-02-11 00:59:41 +0100 (Wed, 11 Feb 2009) New Revision: 1101 Added: cpu/x86/pc/olpc/keyject.bth cpu/x86/pc/olpc/keyject.fth cpu/x86/pc/olpc/kjversions.fth Modified: cpu/x86/pc/olpc/fw.bth cpu/x86/pc/olpc/mfgdata.fth cpu/x86/pc/olpc/security.fth cpu/x86/pc/olpc/setwp.fth dev/olpc/spiflash/spiui.fth Log: OLPC - keyjector support. Modified: cpu/x86/pc/olpc/fw.bth =================================================================== --- cpu/x86/pc/olpc/fw.bth 2009-02-02 08:47:00 UTC (rev 1100) +++ cpu/x86/pc/olpc/fw.bth 2009-02-10 23:59:41 UTC (rev 1101) @@ -529,7 +529,6 @@ install-dual-console ?gui-banner - " probe-" do-drop-in \ ." nvramrc" cr \ use-nvramrc? if nvramrc safe-evaluate then ; @@ -603,12 +602,17 @@ read-game-keys stdout off \ + + " probe-" do-drop-in + probe-pci sound open-keyboard ?games ['] false to interrupt-auto-boot? probe-usb + " probe+" do-drop-in + interpreter-init ?scan-nand ?diags Added: cpu/x86/pc/olpc/keyject.bth =================================================================== --- cpu/x86/pc/olpc/keyject.bth (rev 0) +++ cpu/x86/pc/olpc/keyject.bth 2009-02-10 23:59:41 UTC (rev 1101) @@ -0,0 +1,121 @@ +purpose: Construct the Open Firmware module collection + +command: &builder &this +in: ${BP}/cpu/x86/pc/olpc/build/ec.img +in: ${BP}/cpu/x86/pc/olpc/build/romreset.di +\ in: ${BP}/cpu/x86/pc/olpc/build/romstart.di +in: ${BP}/cpu/x86/pc/olpc/build/resume.di +in: ${BP}/cpu/x86/pc/olpc/build/rmstart.img +in: ${BP}/cpu/x86/pc/olpc/build/paging.di +in: ${BP}/cpu/x86/pc/olpc/build/fw.dic +in: ${BP}/dev/mmc/sdhci/build/sdhci.fc +in: ${BP}/dev/mmc/sdhci/build/sdmmc.fc +in: ${BP}/dev/geode/ac97/build/ac97.fc +in: ${BP}/dev/usb2/hcd/ohci/build/ohci.fc +in: ${BP}/dev/usb2/hcd/ehci/build/ehci.fc +in: ${BP}/dev/usb2/device/hub/build/hub.fc +in: ${BP}/dev/usb2/device/net/build/usbnet.fc +in: ${BP}/dev/usb2/device/serial/build/usbserial.fc +in: ${BP}/dev/usb2/device/storage/build/usbstorage.fc +in: ${BP}/dev/usb2/device/keyboard/build/usbkbd.fc +in: ${BP}/dev/usb2/device/wlan/build/usb8388.fc +in: ${BP}/dev/olpc/cafenand/build/cafenand.fc +in: ${BP}/dev/olpc/cafecamera/build/cafecamera.fc +in: ${BP}/clients/memtest86/memtest +in: ${BP}/clients/emacs/x86/emacs +in: ${BP}/cpu/x86/pc/olpc/build/verify.img +in: ${BP}/cpu/x86/pc/olpc/build/usb8388.bin +in: ${BP}/dev/pci/build/pcibridg.fc +in: ${BP}/cpu/x86/pc/olpc/build/dsdt.aml +in: ${BP}/cpu/x86/pc/olpc/build/mcastnand.bin + +build-now + +fload ${BP}/cpu/x86/pc/olpc/kjversions.fth + +" macro: FW_VERSION Q2${FW_MAJOR}${FW_MINOR}" expand$ eval + + +fload ${BP}/cpu/x86/pc/olpc/config.fth + +\ Always re-create the builton.fth file when we make a new rom.img +fload ${BP}/cpu/x86/pc/builton.bth +fload ${BP}/cpu/x86/pc/olpc/sourceurl.fth + +fload ${BP}/forth/lib/crc32.fth + +hex + +: pad-file ( location -- ) + ofd @ fsize + 2dup u< abort" The ROM image is too large" + ?do h# ff ofd @ fputc loop +; + +.( --- Saving as ) +" ${FW_VERSION}.rom" expand$ 2dup lower ( adr len ) +2dup type cr ( adr len ) +$new-file + " ec.img" $add-file + + dropin-base rom-pa - pad-file + + " romreset.di" $add-file +\ " romstart.di" $add-file + " resume.di" $add-file + + fload ${BP}/cpu/x86/pc/olpc/loaddropins.fth + " sourceurl" " sourceurl" $add-dropin + + " ${BP}/clients/memtest86/memtest" " memtest" $add-deflated-dropin +\ " ${BP}/cpu/x86/pc/olpc/images/olpc16EQima.wav" " splash" $add-deflated-dropin +\ " ${BP}/cpu/x86/pc/olpc/images/Edge1-16k-EQ-Comp-Amp-ima.wav" " splash" $add-deflated-dropin + " ${BP}/cpu/x86/pc/olpc/images/Edge1-8k-EQ-Comp-Amp-Short.wav" " splash" $add-deflated-dropin + " ${BP}/cpu/x86/pc/olpc/build/dsdt.aml" " dsdt" $add-deflated-dropin +\ " ${BP}/cpu/x86/pc/olpc/build/ssdt.aml" " ssdt" $add-deflated-dropin + + " ${BP}/cpu/x86/pc/olpc/build/nandblaster_rx.bin" " nb_rx" $add-deflated-dropin + " ${BP}/cpu/x86/pc/olpc/build/nandblaster_tx.bin" " nb_tx" $add-deflated-dropin +\ " ${BP}/cpu/x86/pc/olpc/build/multicast-nand/mcastnand.elf" " mcastnand" $add-deflated-dropin +\ " ${BP}/cpu/x86/pc/olpc/build/blaster.bin" " blaster" $add-deflated-dropin +\ " ${BP}/cpu/x86/pc/olpc/build/cloner.bin" " cloner" $add-deflated-dropin + + " ${BP}/cpu/x86/pc/olpc/keyject.fth" " probe+" $add-deflated-dropin + " /space/bios-crypto/build/k2.public" " o1" $add-dropin \ Incompressible + " /space/bios-crypto/build/k2.public" " s1" $add-dropin + " /space/bios-crypto/build/k2.public" " d1" $add-dropin + " /space/bios-crypto/build/k2.public" " w1" $add-dropin + " /space/bios-crypto/build/k2.public" " a1" $add-dropin + + /rom h# 400 - pad-file \ rmstart image must start 0x400 from end + " rmstart.img" $add-file + +\ Insert the revision signature +/rom h# 40 - ofd @ fseek +h# 10 buffer: signature +signature h# 10 blank +" CL1" signature swap move +" ${FW_VERSION}" expand$ signature 6 + swap move +" Q2${FW_MAJOR}" expand$ signature d# 13 + swap move +signature h# 10 ofd @ fputs + +/l buffer: crcbuf +/rom buffer: filebuf +0 ofd @ fseek + +\ Read the entire image, compute the CRC, and store it h# 30 from the end +filebuf /rom ofd @ fgets /rom <> abort" Can't read back image" +0 crctab filebuf /rom ($crc) crcbuf ! + +/rom h# 30 - ofd @ fseek +crcbuf /l ofd @ fputs + +ofd @ fclose + +\ Creating olpc.version serves two purposes: +\ a) It reports the firmware revision for use by external scripts +\ b) It provides an olpc.* artifact for the builder dependency management. + +writing keyject.version +" ${FW_VERSION}" expand$ 2dup lower ofd @ fputs +ofd @ fclose Added: cpu/x86/pc/olpc/keyject.fth =================================================================== --- cpu/x86/pc/olpc/keyject.fth (rev 0) +++ cpu/x86/pc/olpc/keyject.fth 2009-02-10 23:59:41 UTC (rev 1101) @@ -0,0 +1,243 @@ +purpose: Inject additional keys into manufacturing data +\ See license at end of file + +\ Search for !!! for things that may need to change for different deployments + +\ See HowItWorks near end of file for a description of the overall procedure + +create test-me +[ifdef] test-me +.( WARNING WARNING - keyjector test code installed) cr +[else] +.( Warning - keyjector version) cr +[then] + +\ !!! Re-implement this for each different deployment +: wrong-sku? ( -- flag ) + " P#" find-tag 0= if true exit then ( pn$ ) + + -null ( pn$' ) + 2dup " 1CL11ZP0KD6" $= if 2drop false exit then ( pn$ ) \ UY BYD LiFePO4 + 2dup " 1CL11ZP0KD7" $= if 2drop false exit then ( pn$ ) \ UY GP NiMH + 2dup " 1CL11ZP0KD9" $= if 2drop false exit then ( pn$ ) \ UY GP LiFePO4 +[ifdef] test-me + 2dup " 1CL11ZU0KDB" $= if 2drop false exit then ( pn$ ) \ US for testing +[then] + 2drop + + true +; + +\ !!! Change the date for each different deployment +: keyject-expired? ( -- flag ) " 20090401T000000Z" expired? ; + +\ !!! Change the key list for each different deployment +: new-key-list$ ( -- ) " o1 s1 d1 w1 a1" ; + +\ True if the all the requested tags are already present. +\ This prevents endless looping. +: already-injected? ( -- flag ) + new-key-list$ begin dup while ( $ ) + bl left-parse-string ( $' name$ ) + find-tag if ( $ value$ ) + 2drop ( $ ) + else ( $ ) + 2drop false exit + then ( $ ) + repeat ( $ ) + 2drop true +; + +: inject-key ( keyname$ -- ) + 2dup find-drop-in if ( keyname$ value$ ) + 2over ram-find-tag if ( keyname$ value$ oldvalue$ ) + 2 pick <> if ( keyname$ value$ oldvalue$ ) + 3drop ( keyname$ ) + ." Warning: inconsistent old tag length for " type cr ( ) + exit + then ( keyname$ value$ oldvalue-adr ) + >r 2tuck r> swap move ( valu$ keyname$ ) + green-letters + ." Replaced " type cr ( value$ ) + black-letters + else ( keyname$ value$ ) + 2swap ( value$ keyname$ ) + 2over 2over ( value$ keyname$ value$ keyname$ ) + ($add-tag) ( value$ keyname$ ) + green-letters + ." Added " type cr ( value$ ) + black-letters + then ( value$ ) + free-mem ( ) + else ( keyname$ ) + ." Warning: Can't find a dropin module for " type cr ( ) + then ( ) +; + +: inject-keys ( -- ) + get-mfg-data + new-key-list$ begin dup while ( $ ) + bl left-parse-string ( $' name$ ) + inject-key ( $ ) + repeat ( $ ) + 2drop ( ) + (put-mfg-data) ( ) +; + +: keyject-error ( msg$ -- ) + cr + red-letters ." Not injecting because: " type cr black-letters + cr + ." Will update firmware in 20 seconds" cr + d# 20,000 ms +; + +: do-keyject? ( -- flag ) + wrong-sku? if + " Wrong SKU" keyject-error + false exit + then + keyject-expired? if + " Date Expired" keyject-error + false exit + then + already-injected? if + " Keys Already Present" keyject-error + false exit + then + true +; + +false value new-firmware? +: got-firmware? ( dev$ -- flag ) + 2dup ." Looking for new bootfw2.zip on " type cr ( dev$ ) + dn-buf place ( ) + " \boot" pn-buf place ( ) + filesystem-present? 0= if false exit then ( ) + null$ cn-buf place ( ) + " bootfw2" bundle-present? 0= if false exit then ( ) + ." Found" cr ( ) + secure? if ( ) + load-crypto if ( ) + ." Crypto load failed" cr false exit ( ) + then ( ) + fwkey$ to pubkey$ ( ) + img$ sig$ fw-valid? 0= if ( ) + ." Bad signature" cr ( ) + false exit + then ( ) + then ( ) + img$ tuck flash-buf swap /flash min move ( len ) + ['] ?image-valid catch if ( x ) + ." Bad firmware image" cr ( x ) + drop false exit + then ( ) + ." Good image" cr ( ) + + true to new-firmware? ( ) + true ( true ) +; + +: get-new-firmware ( -- ) + all-devices$ begin dup while ( $ ) + bl left-parse-string ( $' dev$ ) + got-firmware? if 2drop exit then ( $ ) + repeat ( $ ) + 2drop +; + +\ Firmware is in flash-buf +: update-firmware ( -- ) + write-firmware + + ['] verify-firmware catch if + ." Verify failed. Retrying once" cr + spi-identify + write-firmware + verify-firmware + then +; + +: ?keyject ( -- ) + visible + green-letters cr ." Security Key Injector" cr cr black-letters + \ Get the new firmware first, so any security checks use the old keys + get-new-firmware + do-keyject? if + flash-write-enable + inject-keys + new-firmware? if update-firmware then + flash-write-disable \ Should reboot + else + \ If we can't update the firmware, don't touch the SPI FLASH, lest + \ we get into an infinite reboot cycle. + new-firmware? if + ." Updating firmware ..." cr + flash-write-enable + update-firmware + flash-write-disable \ Should reboot + then + then +; + +?keyject + +[ifdef] HowItWorks +OLPC signs bootfw.zip containing OFW image A and bootfw2.zip containing OFW image B. +* A is an OFW with additional keyjector functionality +* B is an ordinary OFW +Version number B > version number A. + +bootfw.zip and bootfw2.zip are presented to a deployment machine in the usual manner, +either on a USB key or as part of a signed OS image. + +On a deployment machine with firmware X (version X < version A): + +The deployment machine (with old firmware version X < A) +auto-reflashes itself with firmware A via the existing secure +reflash mechanism. + +1) Firmware A starts and chacks that: + + The SKU is for the intended deployment + + the date is before the keyjector expiration date + + The override keys are not already present +so it + ! Injects the new keys +then it + ! Reads bootfw2.zip, checks its signature, and reflashes with firmware B (version > A) + ! Reboots + +2) Firmware B starts, performs the normal fw update attempt step, +noticing that data.img (firmware A) is downrev, and proceeds to +boot normally. + +In step (1), on a non-UY SKU, or after the expiration date, +firmware A skips the keyjection step and instead goes straight +to the "reflash firmware B" step. + +In either case, the machine ends up with (normal) firmware B. +[then] + +\ LICENSE_BEGIN +\ Copyright (c) 2007 FirmWorks +\ +\ Permission is hereby granted, free of charge, to any person obtaining +\ a copy of this software and associated documentation files (the +\ "Software"), to deal in the Software without restriction, including +\ without limitation the rights to use, copy, modify, merge, publish, +\ distribute, sublicense, and/or sell copies of the Software, and to +\ permit persons to whom the Software is furnished to do so, subject to +\ the following conditions: +\ +\ The above copyright notice and this permission notice shall be +\ included in all copies or substantial portions of the Software. +\ +\ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, +\ EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +\ MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND +\ NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE +\ LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION +\ OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION +\ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +\ +\ LICENSE_END Added: cpu/x86/pc/olpc/kjversions.fth =================================================================== --- cpu/x86/pc/olpc/kjversions.fth (rev 0) +++ cpu/x86/pc/olpc/kjversions.fth 2009-02-10 23:59:41 UTC (rev 1101) @@ -0,0 +1,31 @@ +\ Version numbers of items included in the OLPC firmware image + +\ The overall firmware revision +macro: FW_MAJOR E +macro: FW_MINOR 30b + +\ The EC microcode +macro: EC_VERSION e21 + +\ Alternate command for getting EC microcode, for testing new versions. +\ Temporarily uncomment the line and modify the path as necessary +\ macro: GET_EC cp pq2e18c.img ec.img + +macro: KEYS mpkeys +\ macro: KEYS testkeys + +\ The wireless LAN module firmware +macro: WLAN_RPM ${WLAN_VERSION}.olpc1 +macro: WLAN_VERSION 5.110.22.p23 + +\ The bios_verify image +macro: CRYPTO_VERSION 0.2 + +\ The multicast NAND updater code version +\ Use a specific git commit ID for a formal release or "test" for development. +\ With a specific ID, mcastnand.bth will download a tarball without .git stuff. +\ With "test", mcastnand.bth will clone the git head if build/multicast-nand/ +\ is not already present, then you can modify the git subtree as needed. +macro: MCNAND_VERSION 0c73b4a084a27f0687b152dd0395c67fdf54b10f +\ macro: MCNAND_VERSION test +\ macro: MCNAND_VERSION HEAD Modified: cpu/x86/pc/olpc/mfgdata.fth =================================================================== --- cpu/x86/pc/olpc/mfgdata.fth 2009-02-02 08:47:00 UTC (rev 1100) +++ cpu/x86/pc/olpc/mfgdata.fth 2009-02-10 23:59:41 UTC (rev 1101) @@ -51,8 +51,8 @@ nip if drop flash-base h# f.0000 + then ; -: find-tag ( name$ -- false | data$ true ) - drop >r mfg-data-top ( adr r: name-adr ) +: (find-tag) ( name$ top-adr -- false | data$ true ) + -rot drop >r ( adr r: name-adr ) begin another-tag? while ( adr' data$ tname-adr r: name-adr ) r@ 2 comp 0= if ( adr' data$ r: name-adr ) r> drop rot drop ( data$ ) @@ -62,6 +62,9 @@ repeat ( adr' r: name-adr ) r> 2drop false ; +: find-tag ( name$ -- false | data$ true ) + mfg-data-top (find-tag) +; \ Remove bogus null characters from the end of mfg data tags (old machines \ have malformed tags) Modified: cpu/x86/pc/olpc/security.fth =================================================================== --- cpu/x86/pc/olpc/security.fth 2009-02-02 08:47:00 UTC (rev 1100) +++ cpu/x86/pc/olpc/security.fth 2009-02-10 23:59:41 UTC (rev 1101) @@ -20,15 +20,15 @@ debug-security? if red-letters type black-letters cr else 2drop then ; +code halt hlt c; \ To save power + : fail-load ( -- ) + show-sad text-on ." OS Load Failed" cr - quit - begin again + begin halt again ; -code halt hlt c; \ To save power - 0 value security-off? : security-failure ( -- ) @@ -943,7 +943,7 @@ banner then - load-crypto if " Crytpo load failed" .security-failure then ( ) + load-crypto if " Crypto load failed" .security-failure then ( ) alternate? if " \boot-alt" else " \boot" then pn-buf place Modified: cpu/x86/pc/olpc/setwp.fth =================================================================== --- cpu/x86/pc/olpc/setwp.fth 2009-02-02 08:47:00 UTC (rev 1100) +++ cpu/x86/pc/olpc/setwp.fth 2009-02-10 23:59:41 UTC (rev 1101) @@ -7,9 +7,16 @@ \ require a full erase. That is faster and safer than copying out the \ data, erasing the block and rewriting it. -: set-wp ( -- ) +: enable-security ( "serialnumber" -- ) + board-revision h# b48 < abort" Only supported on B4 and later" h# fffefffe 2 " wp" $= if ." wp is already set" cr exit then - " SN" find-tag 0= abort" No serial number (SN tag); enabling security would brick me." 2drop + " SN" find-tag 0= abort" No serial number (SN tag); enabling security would brick me." ( sn$ ) + safe-parse-word ( sn$ confirmation$ ) + $= 0= abort" Confirmation code mismatch" + + " Enabling security is dangerous. Are you sure you want to do it?" confirmedn? + 0= abort" Canceled" + " U#" find-tag 0= abort" No U# tag; enabling security would brick me." 2drop h# fffefffe 2 " ww" $= 0= abort" No ww tag" spi-start spi-identify @@ -25,10 +32,18 @@ rom-pa mfg-data-offset + mfg-data-buf /flash-block lmove ; +: (put-mfg-data) + mfg-data-buf mfg-data-end-offset mfg-data-offset write-flash-range +; + +: ram-find-tag ( name$ -- false | data$ true ) + mfg-data-buf /flash-block + (find-tag) +; + \ Write mfg data from RAM to FLASH : put-mfg-data ( -- ) spi-start spi-identify - mfg-data-buf mfg-data-end-offset mfg-data-offset write-flash-range + (put-mfg-data) spi-reprogrammed ; @@ -52,12 +67,14 @@ r> ; +[ifdef] notdef \ Change the "ww" tag to "wp" : hard-set-wp ( -- ) " ww" mfg-data-setup ( ram-adr ) [char] p swap 1+ c! ( ) put-mfg-data ; +[then] \ Change the "wp" tag to "ww" : clear-wp ( -- ) @@ -68,14 +85,12 @@ alias disable-security clear-wp -: enable-security ( -- ) - board-revision h# b48 < abort" Only supported on B4 and later" - set-wp -; - : ?tagname-valid ( tagname$ -- tagname$ ) dup 2 <> abort" Tag name must be 2 characters long" ; +: ?wp-tag ( tagname$ -- tagname$ ) + 2dup " wp" $= abort" Use enable-security for the wp tag" +; : tag-setup ( tagname$ -- ram-value$ ) ?tagname-valid get-mfg-data @@ -100,6 +115,7 @@ ; : $change-tag ( value$ tagname$ -- ) + ?wp-tag tag-setup ( new-value$ old-value$ ) 2over 2over value-mismatch? abort" New value and old value have different lengths" drop swap move ( ) @@ -116,11 +132,7 @@ mfg-data-buf /flash-block + last-mfg-data ; -: $add-tag ( value$ name$ -- ) - ?tagname-valid ( value$ name$ ) - 2dup find-tag abort" Tagname already exists" ( value$ name$ ) - - get-mfg-data +: ($add-tag) ( value$ name$ -- ) ram-last-mfg-data >r ( value$ name$ r: adr ) \ Check for enough space for the new tag @@ -152,7 +164,14 @@ \ Copy the value data over - swap move ( ) +; +: $add-tag ( value$ name$ -- ) + ?wp-tag + ?tagname-valid ( value$ name$ ) + 2dup find-tag abort" Tagname already exists" ( value$ name$ ) + get-mfg-data ( value$ name$ ) + ($add-tag) ( ) put-mfg-data ( ) ; Modified: dev/olpc/spiflash/spiui.fth =================================================================== --- dev/olpc/spiflash/spiui.fth 2009-02-02 08:47:00 UTC (rev 1100) +++ dev/olpc/spiflash/spiui.fth 2009-02-10 23:59:41 UTC (rev 1101) @@ -40,8 +40,9 @@ ." Verifying" cr ?do ( adr ) i .x (cr - dup i + /flash-block i flash-verify - /flash-block +loop ( adr ) + dup /flash-block i flash-verify + /flash-block + ( adr' ) + /flash-block +loop ( adr ) cr drop ( ) ; @@ -226,8 +227,19 @@ then ; -: write-firmware ( -- ) +: verify-firmware ( -- ) [ifdef] use-flash-nvram + flash-buf nvram-offset 0 verify-flash-range \ Verify first part +[else] + flash-buf mfg-data-offset 0 verify-flash-range \ Verify first part +[then] + + \ Don't verify the block containing the manufacturing data + + flash-buf mfg-data-end-offset + /flash mfg-data-end-offset verify-flash-range \ Verify last part +; +: write-firmware ( -- ) +[ifdef] use-flash-nvram flash-buf nvram-offset 0 write-flash-range \ Write first part [else] flash-buf mfg-data-offset 0 write-flash-range \ Write first part @@ -235,8 +247,7 @@ \ Don't write the block containing the manufacturing data - flash-buf mfg-data-end-offset + ( adr ) - /flash mfg-data-end-offset write-flash-range \ Write last part + flash-buf mfg-data-end-offset + /flash mfg-data-end-offset write-flash-range \ Write last part ; : .verify-msg ( -- ) @@ -334,11 +345,11 @@ ?file \ Don't overwrite the EC code - flash-buf /flash /ec write-flash-range + flash-buf /ec + /flash /ec write-flash-range .verify-msg ; -: verify-bios ( -- ) flash-buf /flash /ec verify-flash-range ; +: verify-bios ( -- ) flash-buf /ec + /flash /ec verify-flash-range ; : flash-ec ( -- ) ?file

1 0

r1100 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 2, 2009

Feb. 2, 2009

Author: wmb Date: 2009-02-02 09:47:00 +0100 (Mon, 02 Feb 2009) New Revision: 1100 Modified: cpu/x86/pc/olpc/biosresume.fth Log: OLPC trac 9229 - pretty boot was hiding the Windows logo screen. Modified: cpu/x86/pc/olpc/biosresume.fth =================================================================== --- cpu/x86/pc/olpc/biosresume.fth 2009-02-02 08:42:58 UTC (rev 1099) +++ cpu/x86/pc/olpc/biosresume.fth 2009-02-02 08:47:00 UTC (rev 1100) @@ -71,7 +71,7 @@ ; : xo-platform-fixup ( -- ) disable-uoc - dcon-unfreeze unfreeze + visible wlan-reset ; ' xo-platform-fixup to more-platform-fixup

1 0

r1099 - cpu/x86/pc/olpc
by svn＠openfirmware.info Feb. 2, 2009

Feb. 2, 2009

Author: wmb Date: 2009-02-02 09:42:58 +0100 (Mon, 02 Feb 2009) New Revision: 1099 Modified: cpu/x86/pc/olpc/dsdt.dsl Log: OLPC trac 9228 - Windows shows Embedded Controller device with non-functional driver. Fixed by eliminating the EC0 node in the ACPI tables, moving the EC's registers under the MEM node. Modified: cpu/x86/pc/olpc/dsdt.dsl =================================================================== --- cpu/x86/pc/olpc/dsdt.dsl 2009-02-02 07:01:43 UTC (rev 1098) +++ cpu/x86/pc/olpc/dsdt.dsl 2009-02-02 08:42:58 UTC (rev 1099) @@ -782,6 +782,15 @@ IO (Decode16, 0x0030, 0x0030, 0x00, 0x10, ) // I/O ports to fake out SMI interrupts IO (Decode16, 0x03c0, 0x03c0, 0x00, 0x20, ) // Claim VGA I/O ports + // We claim the EC ports here instead of in an EC node + // because we don't want Windows to load an EC driver + // nor to complain about not being able to do so. + IO (Decode16, 0x62, 0x62, 0, 1) // Claim EC I/O ports + IO (Decode16, 0x66, 0x66, 0, 1) + IO (Decode16, 0x61, 0x61, 0, 1) // Extra EC I/O ports + IO (Decode16, 0x68, 0x68, 0, 1) + IO (Decode16, 0x6c, 0x6c, 0, 1) + IO (Decode16, 0x380, 0x380, 0, 4) }) CreateDWordField (MBRB, \_SB.PCI0.SBF0.MEM._CRS._Y06._LEN, EM1L) Store (MEMK, Local0) // Memory size in Kbytes (from EBDA) @@ -837,27 +846,27 @@ }) } - Device(EC0) { - Name (_HID, EISAID("PNP0C09")) - Name (_CRS, ResourceTemplate() { - IO (Decode16, 0x62, 0x62, 0, 1) - IO (Decode16, 0x66, 0x66, 0, 1) - IO (Decode16, 0x61, 0x61, 0, 1) - IO (Decode16, 0x68, 0x68, 0, 1) - IO (Decode16, 0x6c, 0x6c, 0, 1) - IO (Decode16, 0x380, 0x380, 0, 4) - }) - -// // Define that the EC SCI is bit 0 of the GP_STS register -// Name(_GPE, 0) { -// OperationRegion(ECOR, EmbeddedControl, 0, 0xFF) -// Field(ECOR, ByteAcc, Lock, Preserve) { -// // Field definitions go here -// } -// } +// Device(EC0) { +// Name (_HID, EISAID("PNP0C09")) +// Name (_CRS, ResourceTemplate() { +// IO (Decode16, 0x62, 0x62, 0, 1) +// IO (Decode16, 0x66, 0x66, 0, 1) +//// IO (Decode16, 0x61, 0x61, 0, 1) +//// IO (Decode16, 0x68, 0x68, 0, 1) +//// IO (Decode16, 0x6c, 0x6c, 0, 1) +//// IO (Decode16, 0x380, 0x380, 0, 4) +// }) +// +// // Define that the EC SCI is bit 0 of the GP_STS register +// Name(_GPE, 0) // Possibly wrong +// +//// OperationRegion(ECOR, EmbeddedControl, 0, 0xFF) { +//// Field(ECOR, ByteAcc, Lock, Preserve) { +//// // Field definitions go here +//// } +//// } +// } - } - // Elided superio // XXX ??? Maybe this should be moved out a level so it's not under SB F0

1 0

r1098 - ofw/fs/cdfs
by svn＠openfirmware.info Feb. 1, 2009

Feb. 1, 2009

Author: wmb Date: 2009-02-02 08:01:43 +0100 (Mon, 02 Feb 2009) New Revision: 1098 Modified: ofw/fs/cdfs/cdfs.fth Log: Fixed the ISO9660 filesystem reader so it doesn't convert names to upper case, because lower case and mixed case names can appear on media. Modified: ofw/fs/cdfs/cdfs.fth =================================================================== --- ofw/fs/cdfs/cdfs.fth 2009-02-02 06:58:07 UTC (rev 1097) +++ ofw/fs/cdfs/cdfs.fth 2009-02-02 07:01:43 UTC (rev 1098) @@ -137,7 +137,6 @@ max-name min ( adr len' ) tuck sought-name swap move ( len' ) sought-name swap ( adr' len' ) - 2dup upper ( adr' len' ) -period ( adr' len'' ) ; : froot ( -- ) root-dirent select-file set-dir ;

1 0

r1097 - cpu/x86/pc cpu/x86/pc/olpc dev/usb2/device/wlan
by svn＠openfirmware.info Feb. 1, 2009

Feb. 1, 2009

Author: wmb Date: 2009-02-02 07:58:07 +0100 (Mon, 02 Feb 2009) New Revision: 1097 Modified: cpu/x86/pc/biosints.fth cpu/x86/pc/olpc/biosresume.fth dev/usb2/device/wlan/wlan.fth Log: OLPC trac 9227 - reset wlan when starting Windows because the Windows driver fails if the wlan firmware has already been loaded. Modified: cpu/x86/pc/biosints.fth =================================================================== --- cpu/x86/pc/biosints.fth 2009-02-02 06:47:25 UTC (rev 1096) +++ cpu/x86/pc/biosints.fth 2009-02-02 06:58:07 UTC (rev 1097) @@ -1078,6 +1078,7 @@ ; : get-one-sector ( dev$ -- error? ) open-dev to bios-ih + bios-ih 0= if true exit then load-base 0 1 (bios-read-sectors) ( #read ) close-bios-ih ( #read ) 1 <> ( error? ) @@ -1104,7 +1105,6 @@ : open " sd:1" ntfs? if " sd:0" set-hd-boot - dcon-unfreeze unfreeze true exit then false Modified: cpu/x86/pc/olpc/biosresume.fth =================================================================== --- cpu/x86/pc/olpc/biosresume.fth 2009-02-02 06:47:25 UTC (rev 1096) +++ cpu/x86/pc/olpc/biosresume.fth 2009-02-02 06:58:07 UTC (rev 1097) @@ -69,7 +69,13 @@ h# 5120.000b msr@ 2 invert and h# 5120.000b msr! h# 5101.0020 p2d-bm-off ; -' disable-uoc to more-platform-fixup +: xo-platform-fixup ( -- ) + disable-uoc + dcon-unfreeze unfreeze + wlan-reset +; +' xo-platform-fixup to more-platform-fixup + : video-refresh-off ( -- ) h# 1000.002a msr@ drop d# 12 lshift ( dc-base ) h# 4758 over l! 0 swap 4 + l! Modified: dev/usb2/device/wlan/wlan.fth =================================================================== --- dev/usb2/device/wlan/wlan.fth 2009-02-02 06:47:25 UTC (rev 1096) +++ dev/usb2/device/wlan/wlan.fth 2009-02-02 06:58:07 UTC (rev 1097) @@ -84,17 +84,20 @@ my-args " supplicant" $open-package to supplicant-ih supplicant-ih 0= if end-bulk-in end-out-ring free-buf false exit then nonce-cmd - force-open? if true exit then - link-up? 0= if - ['] 2drop to ?process-eapol - do-associate 0= if free-buf false exit then - ds-disconnected reset-driver-state - ds-associated set-driver-state - ['] do-process-eapol to ?process-eapol + force-open? 0= if + link-up? 0= if + ['] 2drop to ?process-eapol + do-associate 0= if free-buf false exit then + ds-disconnected reset-driver-state + ds-associated set-driver-state + ['] do-process-eapol to ?process-eapol + then + start-nic then - start-nic then - use-promiscuous? if enable-promiscuous else disable-promiscuous then + force-open? 0= if + use-promiscuous? if enable-promiscuous else disable-promiscuous then + then opencount @ 1+ opencount ! true ;

1 0

r1096 - cpu/x86/pc/olpc dev/geode
by svn＠openfirmware.info Feb. 1, 2009

Feb. 1, 2009

Author: wmb Date: 2009-02-02 07:47:25 +0100 (Mon, 02 Feb 2009) New Revision: 1096 Modified: cpu/x86/pc/olpc/vsapci.fth dev/geode/msr.fth Log: OLPC trac 9226 - virtualize AC97 PCI config base address so Windows XO Audio driver works whether or not the audio device is enabled at Windows boot. Modified: cpu/x86/pc/olpc/vsapci.fth =================================================================== --- cpu/x86/pc/olpc/vsapci.fth 2009-01-29 17:03:19 UTC (rev 1095) +++ cpu/x86/pc/olpc/vsapci.fth 2009-02-02 06:47:25 UTC (rev 1096) @@ -242,11 +242,18 @@ h# 51010081 2 pick 4 and if msr-set else msr-clr then ; : ac97-cmd-reg ( object value -- ) - set-cmd-reg ( adr value ) + set-cmd-reg + h# 300 ?bus-master ( adr value ) + 1 and if ( adr ) + >bar-info ( base-adr size ) + 2dup 1 h# 5100.0026 set-io-rconf ( base-adr size ) + h# a h# 5101.00e1 set-iod-bm ( ) + else ( adr ) + drop ( ) + h# 5100.0026 rconf-off ( ) + h# 5101.00e1 iod-bm-off ( ) + then - h# 300 ?bus-master ( adr value ) - - 2drop \ The MSRs are: \ 5101.00e1 a0000001.480fff80. IOD_BM \ 5100.0026 014f0001.01480001. io-rconf Modified: dev/geode/msr.fth =================================================================== --- dev/geode/msr.fth 2009-01-29 17:03:19 UTC (rev 1095) +++ dev/geode/msr.fth 2009-02-02 06:47:25 UTC (rev 1096) @@ -36,7 +36,7 @@ : p2d-range-off ( msr# -- ) p2d-range-disabled rot msr! ; -: >p2d-bm ( ( base size type -- d.msrval ) +: >p2d-bm ( base size type -- d.msrval ) >r ( base size r: type ) negate page# ( base mask-page r: type ) swap page# ( base-page mask-page r: type ) @@ -59,6 +59,15 @@ : p2d-bm-off ( msr# -- ) p2d-bm-disabled rot msr! ; +: >iod-bm ( base size type -- d.msrval ) + >r ( base size r: type ) + negate h# fffff and ( base mask r: type ) + swap ( mask base r: type ) + 0 r> p2d-format ( msr.low msr.hi' ) +; +: set-iod-bm ( base size type msr# -- ) >r >iod-bm r> msr! ; +alias iod-bm-off p2d-bm-off + : >rconf ( base-adr size mode -- d.msrval ) >r ( base-adr size r: mode ) bounds ( end-adr base-adr r: mode ) @@ -69,6 +78,15 @@ : set-rconf ( base-adr size mode msr# -- ) >r >rconf r> msr! ; +: >io-rconf ( base-adr size mode -- d.msrval ) + >r ( base-adr size r: mode ) + bounds ( end-adr base-adr r: mode ) + d# 12 lshift r> or ( end-adr msr.low ) + swap 4 - d# 12 lshift 1 or ( msr.low msr.high ) +; + +: set-io-rconf ( base-adr size mode msr# -- ) >r >io-rconf r> msr! ; + : rconf-off ( msr# -- ) rconf-disabled rot msr! ;

1 0