openfirmware August 2010

openfirmware@openfirmware.info

1 participants
48 discussions

[commit] r1906 - cpu/x86/pc/olpc
by repository service 03 Aug '10

03 Aug '10

Author: wmb Date: Tue Aug 3 21:52:21 2010 New Revision: 1906 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/1906 Log: svn ignore *.public Added: cpu/x86/pc/olpc/keyinsert.fth Modified: cpu/x86/pc/olpc/ (props changed) Added: cpu/x86/pc/olpc/keyinsert.fth ============================================================================== --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ cpu/x86/pc/olpc/keyinsert.fth Tue Aug 3 21:52:21 2010 (r1906) @@ -0,0 +1,172 @@ +purpose: Inject additional keys into manufacturing data on non-secured machine +\ See license at end of file + +\ This key insertion program is for use in manufacturing before the machines +\ have been secured, i.e. before the wp tag is created. + +\ Search for !!! for things that may need to change for different deployments + +\ !!! Change the list of SKUs according to which ones should receive the new keys +: wrong-sku? ( -- flag ) + " P#" find-tag 0= if true exit then ( pn$ ) + + -null ( pn$' ) + 2dup " 1CL1BZP0KD6" $= if 2drop false exit then ( pn$ ) \ 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 + 2dup " 1CL1B000003" $= if 2drop false exit then ( pn$ ) \ Preproduction XO-1.5 +[ifdef] test-me + 2dup " 1CL11ZU0KDB" $= if 2drop false exit then ( pn$ ) \ US for testing +[then] + 2drop + + true +; + +\ !!! Change the key list according to which keys the deployment wants +: new-key-list$ ( -- ) " a1 d1 o1 s1 t1 w1 " ; + +\ !!! Change the pattern according to the location of key data files +: get-key-file ( keyname$ -- false | value$ true ) +\ " ext:\\%s.public" \ For an SD card formatted with EXT2 (capable of handling long names) + " u:\\%s.pub" \ For a USB driver formatted with FAT (8.3 filename restriction) + sprintf $read-file 0= +; + +: keyject-expired? ( -- flag ) false ; + +\ True if the all the requested tags are already present. +\ This prevents endless looping. +: already-injected? ( -- flag ) + new-key-list$ 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 get-key-file 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 +; + +: 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 +; + +: ac-connected? ( -- flag ) bat-status@ h# 10 and 0<> ; + +\ Empirically, a weak-but-present battery can present the "trickle charge" (80) +\ but not present the "present" bit (01). +: battery-present? ( -- flag ) bat-status@ h# 81 and 0<> ; + +\ Similarly, a weak-but-present battery can present the "trickle charge" (80) +\ but not present the "battery low" bit (04). +: battery-strong? ( -- flag ) bat-status@ h# 84 and 0= ; + +: wait-enough-power ( -- ) + ac-connected? 0= if + ." Please connect the AC adapter to continue..." + begin d# 100 ms ac-connected? until + cr + then + battery-present? 0= if + ." Please insert a well-charged battery to continue..." + begin d# 100 ms battery-present? until + cr + then + battery-strong? 0= if + ." The battery is low. Please insert a charged one to continue..." + begin d# 100 ms battery-present? battery-strong? and until + then +; + +: ?keyject ( -- ) + visible + green-letters cr ." Security Key Injector" cr cr black-letters + do-keyject? if + wait-enough-power + flash-write-enable + inject-keys + flash-write-disable \ Should reboot + then +; + +?keyject + +\ LICENSE_BEGIN +\ Copyright (c) 2010 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

1 0

[commit] r1905 - cpu/x86/pc/olpc/via
by repository service 03 Aug '10

03 Aug '10

Author: wmb Date: Tue Aug 3 21:50:49 2010 New Revision: 1905 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/1905 Log: Q3A48 Modified: cpu/x86/pc/olpc/via/fw-version.fth Modified: cpu/x86/pc/olpc/via/fw-version.fth ============================================================================== --- cpu/x86/pc/olpc/via/fw-version.fth Tue Aug 3 21:43:50 2010 (r1904) +++ cpu/x86/pc/olpc/via/fw-version.fth Tue Aug 3 21:50:49 2010 (r1905) @@ -1,3 +1,3 @@ \ The overall firmware revision macro: FW_MAJOR A -macro: FW_MINOR 47 +macro: FW_MINOR 48

1 0

[commit] r1904 - cpu/x86/pc/olpc/via
by repository service 03 Aug '10

03 Aug '10

Author: wmb Date: Tue Aug 3 21:43:50 2010 New Revision: 1904 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/1904 Log: Checked-in auxiliary code to draw waveforms on "Tektronix vector displays" (as emulated by TeraTERM and xterm). Added: cpu/x86/pc/olpc/via/tekwaveform.fth Added: cpu/x86/pc/olpc/via/tekwaveform.fth ============================================================================== --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ cpu/x86/pc/olpc/via/tekwaveform.fth Tue Aug 3 21:43:50 2010 (r1904) @@ -0,0 +1,137 @@ +\ TeraTERM must be in AutoSwitch mode - Setup>Terminal - for to-tek and to-vt +\ to work. Otherwise you have to invoke the Tek emulator manually (Control>Open TEK) +\ and switch back and forth with the mouse +: to-tek ( -- ) h# 1f emit ; +: to-vt ( -- ) h# 1b emit [char] 2 emit ; + +: tek-page ( -- ) h# 1b emit h# c emit ; \ Clear screen + +: tek-alpha ( -- ) h# 1f emit ; +: tek-plot ( -- ) h# 1d emit ; + +: component ( n tag -- ) swap h# 1f and or emit ; + +d# 1024 constant screen-width +d# 780 constant screen-height + +: tek-move ( x y -- ) + \ Clip the xy components so they can be represented in the + \ 10 bit width of the encoding protocol. + d# 1023 min ( x y' ) + dup 5 rshift h# 20 component ( x y ) \ hi-y + h# 60 component ( x ) \ lo-y + d# 1023 min ( x' ) + dup 5 rshift h# 20 component ( x ) \ hi-x + h# 40 component ( ) \ lo-x +; +\ Here's how line drawing works. Starting in alphanumeric mode: +\ Execute tek-plot to get into line drawing mode. The pen is +\ now up, so the first line is invisible, i.e. it's a move. +\ The pen automatically goes down after the first move, and +\ subsequent moves draw lines. tek-alpha gets you out of +\ drawing mode. + +: tek-box ( -- ) + to-tek + tek-plot + 10 10 tek-move \ initial non-drawn move + 40 10 tek-move + 40 60 tek-move + 10 60 tek-move + 10 10 tek-move + tek-alpha + to-vt +; + +\ PostScript-style line drawing commands + +variable cur-x variable cur-y +: moveto ( x y -- ) cur-y ! cur-x ! ; +: tek-at ( x y -- ) + moveto tek-plot cur-x @ cur-y @ tek-move tek-alpha +; +: lineto ( x y -- ) +\ to-tek + tek-plot + cur-x @ cur-y @ tek-move ( x y ) + 2dup moveto tek-move ( ) + tek-alpha +\ to-vt +; +: rline ( dx dy -- ) swap cur-x @ + swap cur-y @ + lineto ; + +: box2 ( -- ) + tek-page + 10 10 moveto + 40 10 lineto + 40 60 lineto + 10 60 lineto + 10 10 lineto + to-vt +; + +alias clear-plot tek-page +0 constant white +1 constant black +2 constant red +3 constant green +4 constant blue +5 constant cyan +6 constant magenta +7 constant yellow +: set-fg ( n -- ) h# 1b emit ." ML" h# 30 + emit ; + +0 [if] +\ Incremental plotting mode; not very useful. Exit with tek-alpha +: tek-turtle ( -- ) h# 1e emit ; +: unpen ( -- ) bl emit ; +: pen ( -- ) [char] P emit ; +: east ( -- ) [char] A emit ; +: west ( -- ) [char] B emit ; +: north ( -- ) [char] D emit ; +: south ( -- ) [char] H emit ; +: nw ( -- ) [char] F emit ; +: ne ( -- ) [char] E emit ; +: sw ( -- ) [char] J emit ; +: se ( -- ) [char] I emit ; +[then] + +\ End of tek.fth + + +7 constant wave-scale +: wave-height 1 d# 15 wave-scale - << 2+ ; + +: wave0 ( -- ) screen-height wave-height - ; + +: plot0 ( -- x y ) 0 d# 10 ; + +variable ylim variable ymin +1 value stretch +: clip-y ( value -- value' ) ymin @ - 0 max ylim @ min ; +: plot ( xt xmin xmax xscale ymin ymax -- ) + over - ylim ! ymin ! to stretch ( xt xmin xmax ) + over 3 pick execute clip-y ( xt xmin xmax y-at-xmin ) + plot0 2 pick - moveto ( xt xmin xmax y-at-xmin ) + -rot swap 1+ ?do ( xt last ) + i 2 pick execute clip-y ( xt last value ) + tuck - ( xt value delta ) + stretch swap rline ( xt value ) + loop ( xt last ) + 2drop ( ) +; + +: clear-waveform ( -- ) clear-plot ; +: waveform-start ( -- ) 0 wave0 moveto ; +: draw-wave ( adr ) + 0 swap ( last adr ) + screen-width 0 do ( last adr ) + tuck <w@ ( adr last this-unscaled ) + wave-scale >>a ( adr last this ) + tuck swap - ( adr this distance ) + 1 swap rline ( adr this ) + swap wa1+ ( this adr ) + loop ( last adr ) + 2drop +; +: waveform ( adr -- ) clear-waveform waveform-start draw-wave to-vt ;

1 0

[commit] r1903 - dev/video/common
by repository service 03 Aug '10

03 Aug '10

Author: wmb Date: Tue Aug 3 21:42:23 2010 New Revision: 1903 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/1903 Log: Checked-in experimental screenshot code. Added: dev/video/common/screenshot.fth Added: dev/video/common/screenshot.fth ============================================================================== --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ dev/video/common/screenshot.fth Tue Aug 3 21:42:23 2010 (r1903) @@ -0,0 +1,75 @@ +\ Screenshot for XO-1.5 (requires 32-bit display mode) +\ +\ Copy this file to a USB stick - preferably FAT format +\ ok fload u:\scrnshot.fth +\ When you want to take a screenshot, hit the frame key. +\ The text on the screen will flash to show you it has happened. +\ To save the screenshot to file: +\ ok save-screenshot u:\shot.bmp +\ The format is .BMP at 32 bits per pixel + +: screen-bounds ( -- x y w h ) 0 0 screen-wh ; +: screen-depth ( -- n ) " depth" $call-screen ; +: total-pixels ( -- n ) screen-wh * ; +: screenshot ( -- ) + load-base screen-bounds " native-read-rectangle" $call-screen + screen-ih package( blink-screen )package +; +4 buffer: bitbuf +: fput-le32 ( l -- ) bitbuf le-l! bitbuf 4 ofd @ fputs ; +: fput-le16 ( w -- ) bitbuf le-w! bitbuf 2 ofd @ fputs ; + + +: save-screenshot8 ( -- ) + ." 8 bit depth not yet implemented for save-screenshot" cr +; +: 16>32-screenshot + ." 16 bit screenshots not yet implemented" cr +; +h# 4000 buffer: temp-line +: reorder-lines ( -- ) + screen-wh swap /l* >r ( height r: line-width ) + r@ * ( total-bytes ) + load-base + load-base ( end-line start-line r: line-width ) + begin 2dup u> while ( end-line start-line r: line-width ) + dup temp-line r@ lmove ( end-line start-line r: line-width ) + 2dup r@ lmove ( end-line start-line r: line-width ) + temp-line third r@ lmove ( end-line start-line r: line-width ) + swap r@ - swap r@ + ( end-line' start-line' r: line-width ) + repeat ( end-line start-line r: line-width ) + r> 3drop +; +: save-screenshot32 ( "filename" -- ) + writing + " BM" ofd @ fputs + total-pixels /l* d# 54 + fput-le32 \ File size + 0 fput-le32 \ File creator 1 and 2 + d# 54 fput-le32 \ Offset to bits + d# 40 fput-le32 \ DIB header size + screen-wh swap fput-le32 fput-le32 \ Width and height + 1 fput-le16 \ Number of color planes + d# 32 fput-le16 \ Depth + 0 fput-le32 \ No compression + total-pixels /l* fput-le32 \ Bitmap size + d# 7874 dup fput-le32 fput-le32 \ H and V resolution in pixels/meter + 0 fput-le32 \ Colors in palette + 0 fput-le32 \ All colors are important + + reorder-lines + + load-base total-pixels /l* ofd @ fputs + ofd @ fclose +; + +: save-screenshot ( -- ) + screen-depth case + 8 of save-screenshot8 endof + d# 16 of 16>32-screenshot save-screenshot32 endof + d# 24 of save-screenshot32 endof + d# 32 of save-screenshot32 endof + ( default ) ." Unsupported depth " dup .d cr + endcase +; + +: do-screenshot ( -- ) screenshot 0 ; +patch do-screenshot allow-user-aborts? user-abort

1 0

[commit] r1902 - dev/mmc/sdhci
by repository service 03 Aug '10

03 Aug '10

Author: wmb Date: Tue Aug 3 21:31:37 2010 New Revision: 1902 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/1902 Log: SD driver - fixed the ADATA SD card problems without a delay, by using READ/WRITE_SINGLE (instead of _MULTIPLE) for one-block transfers. Modified: dev/mmc/sdhci/sdhci.fth Modified: dev/mmc/sdhci/sdhci.fth ============================================================================== --- dev/mmc/sdhci/sdhci.fth Tue Aug 3 11:53:23 2010 (r1901) +++ dev/mmc/sdhci/sdhci.fth Tue Aug 3 21:31:37 2010 (r1902) @@ -432,6 +432,9 @@ : write-single ( address -- ) h# 183a h# 03 cmd ; \ CMD24 R1 WRITE_SINGLE_BLOCK : write-multiple ( address -- ) h# 193a h# 27 cmd ; \ CMD25 R1 WRITE_MULTIPLE +: issue-write ( address #blocks -- ) 1 = if write-single else write-multiple then ; +: issue-read ( address #blocks -- ) 1 = if read-single else read-multiple then ; + : program-csd ( -- ) 0 h# 1b1a 0 cmd ; \ CMD27 R1 UNTESTED : protect ( group# -- ) h# 1c1b 0 cmd ; \ CMD28 R1b UNTESTED : unprotect ( group# -- ) h# 1d1b 0 cmd ; \ CMD29 R1b UNTESTED @@ -711,7 +714,7 @@ \ This is the correct way to wait for programming complete. : wait-write-done ( -- error? ) - writing? 0= if false exit then ( limit ) + writing? 0= if false exit then get-msecs d# 4000 + ( limit ) \ Wait for return to "tran" state (4) @@ -899,14 +902,12 @@ rot dma-setup ( block# r: #blocks fresh? in? ) wait-write-done if drop r> r> r> 3drop true exit then ( block# ) address-shift lshift r> if ( block# r: #blocks fresh? ) - r> r> 2drop ( block# ) - read-multiple ( ) + r> drop ( block# r: #blocks ) + r> issue-read ( ) else ( block# r: #blocks fresh? ) - \ Ugly ugly workaround for a problem with AData Class 6 SD cards - \ They hang (data timeout) if you start a write too soon after a read. - " d# 1200 us" evaluate - r> if r> pre-write-erase else r> drop then ( block# ) - write-multiple true to writing? ( ) + r> if r@ pre-write-erase then ( block# r: #blocks ) + r> issue-write ( ) + true to writing? ( ) then ( ) true to dma? ( ) false ( error? )

1 0

[commit] r1901 - cpu/x86/pc/olpc/via
by repository service 03 Aug '10

03 Aug '10

Author: wmb Date: Tue Aug 3 11:53:23 2010 New Revision: 1901 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/1901 Log: OLPC XO-1.5 - Reduce the SD clock to 33 MHz for Windows; Windows doesn't work reliably otherwise. Modified: cpu/x86/pc/olpc/via/acpi.fth Modified: cpu/x86/pc/olpc/via/acpi.fth ============================================================================== --- cpu/x86/pc/olpc/via/acpi.fth Tue Aug 3 10:19:53 2010 (r1900) +++ cpu/x86/pc/olpc/via/acpi.fth Tue Aug 3 11:53:23 2010 (r1901) @@ -319,6 +319,9 @@ : rm-platform-fixup ( -- ) \ Disable the internal SD to prevent Windows from making it C: h# f9 h# 6099 config-b! + h# 80 h# 6088 config-b! \ Set timeout clock 0:33Mhz + h# 00 h# 6089 config-b! \ Set max clock to 33Mhz + true windows-mode-adr ! \ Tell the resume code to do it too more-platform-fixup

1 0

[commit] r1900 - cpu/x86/pc/olpc/via
by repository service 03 Aug '10

03 Aug '10

Author: wmb Date: Tue Aug 3 10:19:53 2010 New Revision: 1900 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/1900 Log: OLPC XO-1.5 - Add 10 mS delay after turning on SD card power in Windows resume-from-S3 path, and a few other such tweaks to make Windows resume work reliably. Modified: cpu/x86/pc/olpc/via/startpcirestore.fth Modified: cpu/x86/pc/olpc/via/startpcirestore.fth ============================================================================== --- cpu/x86/pc/olpc/via/startpcirestore.fth Mon Aug 2 07:43:02 2010 (r1899) +++ cpu/x86/pc/olpc/via/startpcirestore.fth Tue Aug 3 10:19:53 2010 (r1900) @@ -7,7 +7,7 @@ 0a 603c config-wb \ Interrupt line \ This is a workaround for an odd problem with the Via Vx855 chip. - \ You have to tell it to use 1.8 V, otherwise when you tell it + \ You have to tell it once to use 1.8 V, otherwise when you tell it \ it to use 3.3V, it will use 1.8 V instead! You only have to \ do this 1.8V thing once after power-up to fix it until the \ next power cycle. The "fix" survives resets; it takes a power @@ -28,12 +28,10 @@ al 8000.1029 #) mov al 8000.2029 #) mov -\ 0e # al mov al 8000.0029 #) mov -\ al 8000.1029 #) mov -\ al 8000.2029 #) mov - f9 6099 config-wb \ Two SD slots (correct for XP, wrong for Linux) - 03 6084 config-wb \ D3 power state + d# 10000 wait-us \ 10 ms delay to let power come on + 80 6088 config-wb \ Set timeout clock 0:33Mhz + 00 6089 config-wb \ Set max clock to 33Mhz \ \ Enable System Management Mode, assuming that the in-memory data structures are already set up \ 21 383 config-wb \ Enable A/Bxxxx range as memory instead of frame buffer (with fxxxx region R/O)

1 0

[commit] r1899 - dev/olpc/keyboard
by repository service 02 Aug '10

02 Aug '10

Author: wmb Date: Mon Aug 2 07:43:02 2010 New Revision: 1899 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/1899 Log: OLPC trac 10262 - In keyboard selftest, distinguish between a key that has been pressed and one that is currently down. Modified: dev/olpc/keyboard/selftest.fth Modified: dev/olpc/keyboard/selftest.fth ============================================================================== --- dev/olpc/keyboard/selftest.fth Thu Jul 29 08:41:26 2010 (r1898) +++ dev/olpc/keyboard/selftest.fth Mon Aug 2 07:43:02 2010 (r1899) @@ -489,7 +489,8 @@ raw-scancode value cur-sc-table [then] -h# 07ff constant pressed-key-color +h# f81f constant down-key-color +h# 07ff constant tested-key-color h# 001f constant idle-key-color h# ffff constant kbd-bc @@ -511,8 +512,9 @@ " fill-rectangle" $call-screen then ; -: key-down ( key# -- ) pressed-key-color draw-key ; -: key-up ( key# -- ) idle-key-color draw-key ; +: key-tested ( key# -- ) tested-key-color draw-key ; +: key-down ( key# -- ) down-key-color draw-key ; +: key-up ( key# -- ) idle-key-color draw-key ; : fill-screen ( color -- ) 0 0 " dimensions" $call-screen " fill-rectangle" $call-screen @@ -536,6 +538,7 @@ else ( scan key# ) swap h# 80 and if \ Up ( key# ) final-test? smt-test? or if ( key# ) + dup key-tested ( key# ) dup 0= last-1 0= and last-2 0= and if ( key# ) drop true ( exit? ) else ( key# )

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

openfirmware August 2010