December 2009 - openfirmware - mail.coreboot.org

r1631 - dev
by svn＠openfirmware.info Dec. 18, 2009

Dec. 18, 2009

Author: wmb Date: 2009-12-18 04:44:13 +0100 (Fri, 18 Dec 2009) New Revision: 1631 Modified: dev/libertas.fth Log: Marvel WLAN driver - inability to read the MAC adress from the WLAN module is now a hard failure. Modified: dev/libertas.fth =================================================================== --- dev/libertas.fth 2009-12-17 23:54:58 UTC (rev 1630) +++ dev/libertas.fth 2009-12-18 03:44:13 UTC (rev 1631) @@ -590,11 +590,12 @@ respbuf >fw-data 2 + mac-adr$ move false ; -: marvel-get-mac-address ( -- ) +: marvel-get-mac-address ( -- error? ) 4 0 do - (marvel-get-mac-address) 0= if unloop exit then + (marvel-get-mac-address) 0= if false unloop exit then loop ." marvel-get-mac-address failed" cr + true ; : marvel-set-mac-address ( -- ) @@ -1523,14 +1524,15 @@ headers -: ?make-mac-address-property ( -- ) - driver-state ds-ready < if exit then - " mac-address" get-my-property if - marvel-get-mac-address +: ?make-mac-address-property ( -- error? ) + driver-state ds-ready < if false exit then + " mac-address" get-my-property if ( ) + marvel-get-mac-address if true exit then mac-adr$ encode-bytes " local-mac-address" property mac-address encode-bytes " mac-address" property - else - 2drop + false + else ( adr len ) + 2drop false then ; : set-frame-size ( -- ) @@ -1542,7 +1544,7 @@ ; : init-net ( -- ) - ?make-mac-address-property + ?make-mac-address-property drop ; : ?load-fw ( -- error? ) @@ -1553,8 +1555,7 @@ then ds-ready to driver-state then - init-net - false + ?make-mac-address-property ; false instance value use-promiscuous?

1 0

r1630 - cpu/x86/pc/olpc/via
by svn＠openfirmware.info Dec. 18, 2009

Dec. 18, 2009

Author: pgf Date: 2009-12-18 00:54:58 +0100 (Fri, 18 Dec 2009) New Revision: 1630 Modified: cpu/x86/pc/olpc/via/runin.fth Log: bypass $rename on int:, fix stack/parameter problems in show-tag and handle-tag, set final TS tag to SHIP, and don't close CIFS connection while handling tags -- the factory server seems to not respond on a quick reconnect. Modified: cpu/x86/pc/olpc/via/runin.fth =================================================================== --- cpu/x86/pc/olpc/via/runin.fth 2009-12-17 07:42:02 UTC (rev 1629) +++ cpu/x86/pc/olpc/via/runin.fth 2009-12-17 23:54:58 UTC (rev 1630) @@ -110,10 +110,10 @@ put-ascii-tag ; : show-tag ( value$ -- ) - $tag-printable? if ?-null type else wrapped-cdump then + 2dup $tag-printable? if ?-null type else wrapped-cdump then ; : do-tag-error ( -- ) - \ Don't know what to do here + ." Problem with tag processing. Halting." cr begin halt again ; : handle-tag ( value$ key$ -- ) @@ -122,11 +122,10 @@ 2>r 2over 2r@ $= if ( value$ key$ r: old-value$' ) 2drop 2drop 2r> 2drop ( ) else ( value$ key$ r: old-value$' ) - 2drop 2drop 2r> 2drop ( ) - \ type ." tag changed!" cr ( value$ r: old-value$' ) - \ ." Old: " r> show-tag cr ( value$ ) - \ ." New: " show-tag cr ( ) - \ do-tag-error + type ." tag changed!" cr ( value$ r: old-value$' ) + ." Old: " 2r> show-tag cr ( value$ ) + ." New: " show-tag cr ( ) + do-tag-error then else ( value$ key$ ) \ New tag, add it put-tag @@ -172,11 +171,9 @@ " BD" ($delete-tag) " MD" ($delete-tag) make-md-tag - " FINAL" " TS" ($add-tag) response$ parse-tags - \ per request from wei-heng " TS" ($delete-tag) " SHIP" " TS" put-ascii-tag @@ -276,6 +273,11 @@ cifs-ih 0= abort" Cannot open SMB share" ; +\ $rename gives "Unimplemented package interface procedure" on ext2 +: do-rename + 2>r 2dup 2r> $copy $delete +; + : finish-final-test ( -- ) wait-connections @@ -284,19 +286,19 @@ \ verify-rtc-date ." Getting final tags .. " - cifs-connect final-tag-exchange cifs-disconnect + cifs-connect final-tag-exchange \ Note: no disconnect... ." Done" cr - \ save a copy of the factory server string, before we destroy MS tag - factory-server$ dup alloc-mem $save - inject-tags - my-cifs-connect final-result cifs-disconnect + ." Submitting results .. " + final-result cifs-disconnect + ." Done" cr + " int:\runin\olpc.fth" $delete-all \ Ultimately this should just be delete of runin\olpc.fth - \ " int:\runin\olpc.fth" " int:\runin\runin.sav" $rename + " int:\runin\olpc.fth" " int:\runin\runin.sav" do-rename ; \ Make the "wait for SD insertion" step highly visible @@ -317,7 +319,7 @@ \ the Linux-based runin tests again. : rerunin ( -- ) " int:\runin\olpc.fth" $delete-all - fail-log-file$ fail-backup-file$ $rename + fail-log-file$ fail-backup-file$ do-rename ; : after-runin ( -- )

1 0

r1629 - forth/lib
by svn＠openfirmware.info Dec. 17, 2009

Dec. 17, 2009

Author: wmb Date: 2009-12-17 08:42:02 +0100 (Thu, 17 Dec 2009) New Revision: 1629 Modified: forth/lib/fileed.fth Log: OLPC trac 9909 - Page Fault on ESC h - (find-previous-word was backing up before the beginning of the buffer. Modified: forth/lib/fileed.fth =================================================================== --- forth/lib/fileed.fth 2009-12-17 05:17:15 UTC (rev 1628) +++ forth/lib/fileed.fth 2009-12-17 07:42:02 UTC (rev 1629) @@ -228,6 +228,7 @@ \ Doesn't move the cursor or change the display. Internal. : find-previous-word ( -- adr ) + cursor-adr line-start-adr = if line-start-adr exit then line-start-adr dup cursor-adr 1- ?do ( linestart ) i c@ bl <> if drop i leave then -1 +loop

1 0

r1628 - cpu/x86/pc/olpc/via
by svn＠openfirmware.info Dec. 17, 2009

Dec. 17, 2009

Author: wmb Date: 2009-12-17 06:17:15 +0100 (Thu, 17 Dec 2009) New Revision: 1628 Modified: cpu/x86/pc/olpc/via/assy.fth cpu/x86/pc/olpc/via/runin.fth Log: Enabled revision keyword for assy.fth and runin.fth Property changes on: cpu/x86/pc/olpc/via/assy.fth ___________________________________________________________________ Added: svn:keywords + Revision Property changes on: cpu/x86/pc/olpc/via/runin.fth ___________________________________________________________________ Added: svn:keywords + Revision

1 0

r1627 - cpu/x86/pc/olpc/via
by svn＠openfirmware.info Dec. 16, 2009

Dec. 16, 2009

Author: wmb Date: 2009-12-16 09:45:45 +0100 (Wed, 16 Dec 2009) New Revision: 1627 Modified: cpu/x86/pc/olpc/via/versions.fth Log: Q3A25 Modified: cpu/x86/pc/olpc/via/versions.fth =================================================================== --- cpu/x86/pc/olpc/via/versions.fth 2009-12-16 08:37:06 UTC (rev 1626) +++ cpu/x86/pc/olpc/via/versions.fth 2009-12-16 08:45:45 UTC (rev 1627) @@ -2,7 +2,7 @@ \ The overall firmware revision macro: FW_MAJOR A -macro: FW_MINOR 24 +macro: FW_MINOR 25 \ The EC microcode macro: EC_VERSION 1_9_18

1 0

r1626 - in cpu/x86/pc/olpc: . via
by svn＠openfirmware.info Dec. 16, 2009

Dec. 16, 2009

Author: wmb Date: 2009-12-16 09:37:06 +0100 (Wed, 16 Dec 2009) New Revision: 1626 Modified: cpu/x86/pc/olpc/rtcwake.fth cpu/x86/pc/olpc/via/probemem.fth cpu/x86/pc/olpc/via/rtcwake.fth Log: OLPC trac 9887 - added suspend/resume test to memory selftest. Modified: cpu/x86/pc/olpc/rtcwake.fth =================================================================== --- cpu/x86/pc/olpc/rtcwake.fth 2009-12-16 08:20:30 UTC (rev 1625) +++ cpu/x86/pc/olpc/rtcwake.fth 2009-12-16 08:37:06 UTC (rev 1626) @@ -51,7 +51,7 @@ : show-rtc-wake ." R" ; -d# 1 constant rtc-alarm-delay +d# 1 value rtc-alarm-delay : pm-sleep-rtc ( -- ) 0 acpi-l@ h# 400.0000 or 0 acpi-l! \ Enable RTC SCI ['] show-rtc-wake rtc-alarm-delay " set-alarm" clock-node @ $call-method @@ -103,6 +103,22 @@ autowack-off ; +: s3-selftest ( -- error? ) + \ The general failure mode here is that it won't wake up, so + \ it's hard to return a real error code. We just have to rely + \ on the operator. + ." Testing S3 suspend/resume" cr + rtc-alarm-delay >r + ." Sleeping for 1 second .. " 1 to rtc-alarm-delay pm-sleep-rtc cr + ." Sleeping for 4 seconds .. " 4 to rtc-alarm-delay pm-sleep-rtc cr + ." Sleeping for 8 seconds .. " 8 to rtc-alarm-delay pm-sleep-rtc cr + r> to rtc-alarm-delay + false +; +dev /memory +[ifdef] test-s3 ' s3-selftest to test-s3 [then] +dend + stand-init: Century h# 20 cmos-century cmos! \ The century is in BCD, hence h# ; Modified: cpu/x86/pc/olpc/via/probemem.fth =================================================================== --- cpu/x86/pc/olpc/via/probemem.fth 2009-12-16 08:20:30 UTC (rev 1625) +++ cpu/x86/pc/olpc/via/probemem.fth 2009-12-16 08:37:06 UTC (rev 1626) @@ -79,6 +79,7 @@ : 8u.h ( n -- ) push-hex (.8) type pop-base ; [then] : .chunk ( adr len -- ) ." Testing memory at: " swap 8u.h ." size " 8u.h cr ; +defer test-s3 ( -- error? ) ' false is test-s3 : selftest ( -- error? ) " available" get-my-property if ." No available property" cr true exit then ( adr len ) @@ -88,7 +89,8 @@ \ We maintain a 1-1 convenience mapping so explicit mapping is unnecessary memory-test-suite if 2drop true exit then ( rem$ ) repeat drop - false + + test-s3 ; device-end Modified: cpu/x86/pc/olpc/via/rtcwake.fth =================================================================== --- cpu/x86/pc/olpc/via/rtcwake.fth 2009-12-16 08:20:30 UTC (rev 1625) +++ cpu/x86/pc/olpc/via/rtcwake.fth 2009-12-16 08:37:06 UTC (rev 1626) @@ -51,7 +51,7 @@ : show-rtc-wake ." R" ; -d# 1 constant rtc-alarm-delay +d# 1 value rtc-alarm-delay : pm-sleep-rtc ( -- ) 0 acpi-l@ h# 400.0000 or 0 acpi-l! \ Enable RTC SCI ['] show-rtc-wake rtc-alarm-delay " set-alarm" clock-node @ $call-method @@ -63,6 +63,19 @@ begin pm-sleep-rtc space dup . (cr 1+ key? until key drop ; +: s3-selftest ( -- error? ) + \ The general failure mode here is that it won't wake up, so + \ it's hard to return a real error code. We just have to rely + \ on the operator. + ." Testing S3 suspend/resume" cr + rtc-alarm-delay >r + ." Sleeping for 1 second .. " 1 to rtc-alarm-delay pm-sleep-rtc cr + ." Sleeping for 4 seconds .. " 4 to rtc-alarm-delay pm-sleep-rtc cr + ." Sleeping for 8 seconds .. " 8 to rtc-alarm-delay pm-sleep-rtc cr + r> to rtc-alarm-delay + false +; +dev /memory ' s3-selftest to test-s3 dend stand-init: Century h# 20 cmos-century cmos! \ The century is in BCD, hence h#

1 0

r1625 - cpu/x86/pc/olpc/via
by svn＠openfirmware.info Dec. 16, 2009

Dec. 16, 2009

Author: wmb Date: 2009-12-16 09:20:30 +0100 (Wed, 16 Dec 2009) New Revision: 1625 Modified: cpu/x86/pc/olpc/via/runin.fth Log: Via manufacturing scripts - extra messages after restarting a failed runin. Modified: cpu/x86/pc/olpc/via/runin.fth =================================================================== --- cpu/x86/pc/olpc/via/runin.fth 2009-12-16 08:16:08 UTC (rev 1624) +++ cpu/x86/pc/olpc/via/runin.fth 2009-12-16 08:20:30 UTC (rev 1625) @@ -328,15 +328,21 @@ key drop cr cr list ." Type R to restart runin, any other key to power off " - key dup emit upc [char] R = if rerunin then + key dup emit cr upc [char] R = if + ." Resetting state to restart runin." cr + ." The old failure log is in " fail-backup-file$ type cr + rerunin + else + power-off + then else autorun-mfg-tests pass? if finish-final-test then show-result-screen - ." Type a key to power off" - key cr then + ." Type a key to power off" + key cr power-off ;

1 0

r1624 - cpu/x86/pc/olpc/via
by svn＠openfirmware.info Dec. 16, 2009

Dec. 16, 2009

Author: wmb Date: 2009-12-16 09:16:08 +0100 (Wed, 16 Dec 2009) New Revision: 1624 Modified: cpu/x86/pc/olpc/via/runin.fth Log: Via manufacturing scripts - added option to restart a failed runin. Modified: cpu/x86/pc/olpc/via/runin.fth =================================================================== --- cpu/x86/pc/olpc/via/runin.fth 2009-12-16 08:05:00 UTC (rev 1623) +++ cpu/x86/pc/olpc/via/runin.fth 2009-12-16 08:16:08 UTC (rev 1624) @@ -243,7 +243,7 @@ exit then ( data$ tag$ ) - 2swap ?-null 2swap + 2swap ?-null 2swap ( data$' tag$ ) put-key:value ( ) ; @@ -308,14 +308,16 @@ warning ! dend - +: fail-backup-file$ ( -- name$ ) + time&date format-date " int:\runin\fail-%s.log" sprintf +; : fail-log-file$ ( -- name$ ) " int:\runin\fail.log" ; \ The operator can type this to reset the state to run \ the Linux-based runin tests again. : rerunin ( -- ) " int:\runin\olpc.fth" $delete-all - " int:\runin\fail.log" $delete-all + fail-log-file$ fail-backup-file$ $rename ; : after-runin ( -- ) @@ -325,14 +327,16 @@ ." Type a key to see the failure log" key drop cr cr list + ." Type R to restart runin, any other key to power off " + key dup emit upc [char] R = if rerunin then else autorun-mfg-tests pass? if finish-final-test then show-result-screen + ." Type a key to power off" + key cr then - ." Type a key to power off" - key cr power-off ;

1 0

r1623 - cpu/x86/pc/olpc/via
by svn＠openfirmware.info Dec. 16, 2009

Dec. 16, 2009

Author: wmb Date: 2009-12-16 09:05:00 +0100 (Wed, 16 Dec 2009) New Revision: 1623 Modified: cpu/x86/pc/olpc/via/assy.fth Log: Via manufacturing scripts - Added auto-firmware-update to the assy script. Modified: cpu/x86/pc/olpc/via/assy.fth =================================================================== --- cpu/x86/pc/olpc/via/assy.fth 2009-12-16 07:56:51 UTC (rev 1622) +++ cpu/x86/pc/olpc/via/assy.fth 2009-12-16 08:05:00 UTC (rev 1623) @@ -2,11 +2,41 @@ visible +\ If the firmware file is on a CIFS share on the factory server, it +\ should be read-only so multiple clients can read it simultaneously. + +false value update-firmware? \ Make this true to update firmware +: wanted-fw$ ( -- $ ) " q3a25" ; \ Set this value to the firmware version + : swid$ ( -- adr len ) " OFW ASSY test $Revision$" ; \ Location of the files containing KA tag data : ka-dir$ ( -- adr len ) " http:\\10.1.0.1\ka\" ; +: find-firmware-file ( -- name$ ) + wanted-fw$ " u:\\boot\\%s.rom" sprintf ( name$ ) + ." Trying " 2dup type cr ( name$ ) + 2dup $file-exists? if exit then ( name$ ) + 2drop ( ) + + wanted-fw$ factory-server$ " %s\\%s.rom" sprintf ( name$ ) + ." Trying " 2dup type cr ( name$ ) + 2dup $file-exists? if exit then ( name$ ) + 2drop + + true abort" Can't find new firmware file" +; + +: ?update-firmware ( -- ) + update-firmware? 0= if exit then + \ Exit if the existing firmware and the wanted firmware are the same + fw-version$ wanted-fw$ nocase-$= if exit then + ." Updating firmware to version " fw-version$ type cr + d# 2000 ms + ?enough-power + find-firmware-file $get-file reflash +; + : put-ka-tag ( value$ key$ -- ) 2over 8 min ka-dir$ " %s%s" sprintf ( value$ key$ filename$ ) $read-file if ( value$ key$ ) @@ -201,6 +231,8 @@ ; : start-assy-test ( -- ) + ?update-firmware + wait-lan get-info

1 0

r1622 - dev/olpc/viacamera
by svn＠openfirmware.info Dec. 16, 2009

Dec. 16, 2009

Author: wmb Date: 2009-12-16 08:56:51 +0100 (Wed, 16 Dec 2009) New Revision: 1622 Modified: dev/olpc/viacamera/camera.fth Log: Via diagnostics - don't fail the camera test if serial is preventing it from running. Modified: dev/olpc/viacamera/camera.fth =================================================================== --- dev/olpc/viacamera/camera.fth 2009-12-16 07:24:34 UTC (rev 1621) +++ dev/olpc/viacamera/camera.fth 2009-12-16 07:56:51 UTC (rev 1622) @@ -507,7 +507,7 @@ : selftest ( -- error? ) serial-enabled? if ." The serial port is in use so the camera cannot be used" cr - true exit + false exit then open 0= if true exit then d# 300 ms

1 0