December 2012 - openfirmware - mail.coreboot.org

On Raspberry Pi
by Mark Morgan Lloyd Jan. 5, 2015

Jan. 5, 2015

Is it possible to put OpenFirmware on an SD card and use it to boot off e.g. a USB-attached mass storage device or using TFTP over a LAN? -- Mark Morgan Lloyd markMLl .AT. telemetry.co .DOT. uk [Opinions above are the author's, not those of his employers or colleagues]

4 10

[commit] r3504 - in cpu/arm/olpc: . 4.0
by repository service Dec. 31, 2012

Dec. 31, 2012

Author: quozl Date: Mon Dec 31 11:43:11 2012 New Revision: 3504 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3504 Log: OLPC XO-4 - touchscreen, change to version 0.0.0.9, which caused a few more timing changes, requiring some fixes to the driver. Modified: cpu/arm/olpc/4.0/nn-version.fth cpu/arm/olpc/nn-touchscreen.fth Modified: cpu/arm/olpc/4.0/nn-version.fth ============================================================================== --- cpu/arm/olpc/4.0/nn-version.fth Mon Dec 31 07:36:36 2012 (r3503) +++ cpu/arm/olpc/4.0/nn-version.fth Mon Dec 31 11:43:11 2012 (r3504) @@ -1,2 +1,2 @@ \ The Neonode touchscreen firmware -macro: NN_VERSION 0_0_0_8 +macro: NN_VERSION 0_0_0_9 Modified: cpu/arm/olpc/nn-touchscreen.fth ============================================================================== --- cpu/arm/olpc/nn-touchscreen.fth Mon Dec 31 07:36:36 2012 (r3503) +++ cpu/arm/olpc/nn-touchscreen.fth Mon Dec 31 11:43:11 2012 (r3504) @@ -111,7 +111,21 @@ then ( id limit ) dup get-msecs - 0< ( id limit timeout? ) until ( id limit ) - 2drop ( ) + drop ( id ) + ." timeout waiting for " 2u.x cr ( ) + r> rstrace >r + pbuf /pbuf erase +; + +: ?id-abort ( id $message -- ) + rot pbuf 2+ c@ <> if + pbuf plen cdump + r> rstrace >r + cr + $abort + else + 2drop + then ; : ?missing @@ -120,22 +134,24 @@ : flush-input ( -- ) begin - 0 0 anticipate - pbuf 2+ c@ h# 07 = if ?missing then + in? if + pbuf 2+ c@ h# 07 = if ?missing then + then no-data? until ; : read-boot-complete ( -- ) - 0 pbuf 2+ c! - h# 07 d# 0 anticipate - pbuf 2+ c@ h# 07 <> abort" response other than boot complete" - ?missing + in? if + h# 07 " response other than boot complete" ?id-abort + ?missing + then ; + : read-version h# 1e h# 01 h# ee 3 bytes-out h# 1e d# 100 anticipate - pbuf 2+ c@ h# 1e <> abort" response other than status" + h# 1e " response other than status" ?id-abort pbuf 9 + le-w@ pbuf 7 + le-w@ wljoin pbuf 5 + le-w@ pbuf 3 + le-w@ wljoin to version# ; @@ -151,12 +167,14 @@ : start ( -- ) h# 04 h# 01 h# ee 3 bytes-out ; : deactivate ( -- ) - h# 00 h# 01 h# ee 3 bytes-out h# 00 d# 60 anticipate + h# 00 h# 01 h# ee 3 bytes-out h# 00 d# 100 anticipate ; : deconfigure ( -- ) - deactivate - true to configure? + configure? 0= if + deactivate + true to configure? + then ; : configure ( -- ) @@ -216,7 +234,6 @@ ; : stream-poll? ( -- false | x y buttons true ) - 0 pbuf 2+ c! in? if \ FIXME: only handles one subpacket pbuf 2+ c@ h# 04 = if @@ -245,7 +262,6 @@ : test-response ( -- ) 6 0 do h# 40 h# 0 h# 0f 3 h# ee 5 bytes-out \ fss full level - 0 pbuf 2+ c! h# 0f d# 130 anticipate pbuf 2+ c@ dup 0= abort" missing response after fixed signal strength" h# 0f <> abort" response other than fixed signal strength" @@ -405,8 +421,8 @@ : test-os-axis ( axis -- ) h# 21 h# 02 h# ee 4 bytes-out - h# 21 d# 60 anticipate - pbuf 2+ c@ h# 21 <> abort" response other than open short" + h# 21 d# 100 anticipate + h# 21 " response other than open short" ?id-abort pbuf d# 5 + ( addr ) @@ -466,7 +482,7 @@ : test-fss-axis ( axis -- ) d# 64 swap h# 0f h# 03 h# ee 5 bytes-out h# 0f d# 60 anticipate - pbuf 2+ c@ h# 0f <> abort" response other than fixed signal strength" + h# 0f " response other than fixed signal strength" ?id-abort 8sp push-decimal pbuf 4 + c@ 0 do ( ) @@ -601,7 +617,7 @@ : test-ls-axis ( axis -- ) h# 0d h# 02 h# ee 4 bytes-out h# 0d d# 200 anticipate - pbuf 2+ c@ h# 0d <> abort" response other than low signals" + h# 0d " response other than low signals" ?id-abort 8sp pbuf 5 + ( addr ) @@ -1010,15 +1026,25 @@ ; +: (lg-tooling) test-os test-fll ; + : lg-tooling ( -- error? ) - open if test-os test-fll else fault then + open if + ['] (lg-tooling) catch ?dup if .error fault then + else + fault + then faults close ; : ir-pcb-smt ( -- error? ) hold-reset connect - open if test-os else fault then + open if + ['] test-os catch ?dup if .error fault then + else + fault + then close hold-reset disconnect faults @@ -1027,8 +1053,10 @@ : ir-pcb-assy ( -- error? ) hold-reset connect open if - test-fll - faults 0= if test-adjacent-axes then + ['] test-fll catch ?dup if .error fault then + faults 0= if + ['] test-adjacent-axes catch ?dup if .error fault then + then else fault then @@ -1039,14 +1067,14 @@ : mb-smt ( -- error? ) open 0= if true exit then - show-version + ['] show-version catch ?dup if .error fault then close - false + faults ; : mb-assy ( -- error? ) open 0= if true exit then - test-adjacent-axes + ['] test-adjacent-axes catch ?dup if .error fault then close faults ;

1 0

[commit] r3503 - forth/lib
by repository service Dec. 31, 2012

Dec. 31, 2012

Author: quozl Date: Mon Dec 31 07:36:36 2012 New Revision: 3503 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3503 Log: .caller - minor stack comment fix, no execution effect Modified: forth/lib/caller.fth Modified: forth/lib/caller.fth ============================================================================== --- forth/lib/caller.fth Fri Dec 28 09:14:36 2012 (r3502) +++ forth/lib/caller.fth Mon Dec 31 07:36:36 2012 (r3503) @@ -5,7 +5,7 @@ \ .current-word displays the name of that \ definition. -\ .caller ( -- ) Displays the colon definition name as in +\ .caller ( ip -- ) Displays the colon definition name as in \ .current-word, and also the address "ip" decimal

1 0

[commit] r3502 - cpu/arm/olpc
by repository service Dec. 28, 2012

Dec. 28, 2012

Author: quozl Date: Fri Dec 28 09:14:36 2012 New Revision: 3502 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3502 Log: OLPC XO-4 - fix support for non-touchscreen SKUs, compatible with touchscreen SKUs, treats TOUCH_HD# signal as active high, since the touchscreen controller is pulling the signal low in normal operation, momentarily enables a pull-down in conflict against the permanent pull-up, in order to sample the signal properly. Requires a BOM change. #12391 Modified: cpu/arm/olpc/nn-touchscreen.fth Modified: cpu/arm/olpc/nn-touchscreen.fth ============================================================================== --- cpu/arm/olpc/nn-touchscreen.fth Tue Dec 18 06:58:54 2012 (r3501) +++ cpu/arm/olpc/nn-touchscreen.fth Fri Dec 28 09:14:36 2012 (r3502) @@ -70,7 +70,12 @@ : reset ( -- ) touch-rst-gpio# dup gpio-clr gpio-set d# 250 ms ; : hold-reset ( -- ) touch-rst-gpio# gpio-clr ; : no-data? ( -- no-data? ) touch-int-gpio# gpio-pin@ ; -: absent? ( -- flag ) touch-hd-gpio# gpio-pin@ 0= ; +: absent? ( -- flag ) + touch-hd-gpio# af@ +pull-dn touch-hd-gpio# af! + 1 ms + touch-hd-gpio# gpio-pin@ + touch-hd-gpio# af@ 0 +pull-dn invert and touch-hd-gpio# af! +; d# 250 constant /pbuf 0 value pbuf

1 0

[commit] r3501 - cpu/arm/olpc/4.0
by repository service Dec. 18, 2012

Dec. 18, 2012

Author: rsmith Date: Tue Dec 18 06:58:54 2012 New Revision: 3501 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3501 Log: OLPC XO-4 - EC Code 0.3.06 Modified: cpu/arm/olpc/4.0/ec-version.fth Modified: cpu/arm/olpc/4.0/ec-version.fth ============================================================================== --- cpu/arm/olpc/4.0/ec-version.fth Sat Dec 15 06:14:47 2012 (r3500) +++ cpu/arm/olpc/4.0/ec-version.fth Tue Dec 18 06:58:54 2012 (r3501) @@ -1,6 +1,6 @@ \ The EC microcode macro: EC_PLATFORM cl4 -macro: EC_VERSION 7_0_3_05 +macro: EC_VERSION 7_0_3_06 \ Alternate command for getting EC microcode, for testing new versions. \ Temporarily uncomment the line and modify the path as necessary

1 0

[commit] r3500 - cpu/x86/pc/olpc/via
by repository service Dec. 15, 2012

Dec. 15, 2012

Author: quozl Date: Sat Dec 15 06:14:47 2012 New Revision: 3500 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3500 Log: OLPC - allow a touchscreen, if present, to prevent automatic test execution after left rocker key, #11817 Modified: cpu/x86/pc/olpc/via/mfgtest.fth Modified: cpu/x86/pc/olpc/via/mfgtest.fth ============================================================================== --- cpu/x86/pc/olpc/via/mfgtest.fth Fri Dec 14 19:27:33 2012 (r3499) +++ cpu/x86/pc/olpc/via/mfgtest.fth Sat Dec 15 06:14:47 2012 (r3500) @@ -202,7 +202,8 @@ (cr kill-line 0 d# 30 do i d# 10 mod 0= if (cr i d# 10 / .d then - mouse-event? dup if nip nip nip then ( moused? ) + touchscreen-event? dup if nip nip nip then ( touched? ) + mouse-event? dup if nip nip nip then or ( touched-or-moused? ) key? or if ( ) drop ( ) menu-interact ( )

1 0

[commit] r3499 - cpu/arm/olpc/4.0
by repository service Dec. 14, 2012

Dec. 14, 2012

Author: wmb Date: Fri Dec 14 19:27:33 2012 New Revision: 3499 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3499 Log: Q7B09 Modified: cpu/arm/olpc/4.0/fw-version.fth Modified: cpu/arm/olpc/4.0/fw-version.fth ============================================================================== --- cpu/arm/olpc/4.0/fw-version.fth Fri Dec 14 19:23:22 2012 (r3498) +++ cpu/arm/olpc/4.0/fw-version.fth Fri Dec 14 19:27:33 2012 (r3499) @@ -1,7 +1,7 @@ \ The overall firmware revision macro: FW_PREFIX Q7 macro: FW_MAJOR B -macro: FW_MINOR 08 +macro: FW_MINOR 09 \ Create a 2-character build/fw-suffix file to personalize your test builds " fw-suffix" $file-exists? [if]

1 0

[commit] r3498 - dev
by repository service Dec. 14, 2012

Dec. 14, 2012

Author: wmb Date: Fri Dec 14 19:23:22 2012 New Revision: 3498 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3498 Log: EDID parser - fixed a bug in the parsing of audio information - it got the format wrong and ignored the number of channels. Modified: dev/edid.fth Modified: dev/edid.fth ============================================================================== --- dev/edid.fth Fri Dec 14 15:35:10 2012 (r3497) +++ dev/edid.fth Fri Dec 14 19:23:22 2012 (r3498) @@ -159,7 +159,7 @@ ; string-array format-names -( 0 ) ," " +( 0 ) ," Reserved0" ( 1 ) ," LPCM" ( 2 ) ," AC-3" ( 3 ) ," MPEG1" @@ -174,12 +174,14 @@ ( 12 ) ," TrueHD" ( 13 ) ," DST" ( 14 ) ," WMA" +( 15 ) ," Format15" end-string-array : .cea-audio ( offset size -- ) ." Audio: " bounds ?do - i cea@ format-names count type ." ," + i cea@ dup 3 rshift h# f and format-names count type ." ," ( byte ) + 7 and 1+ .d ." channels," i 1+ cea@ .cea-freqs ." @" i 2+ cea@ .cea-bits space 3 +loop

1 0

[commit] r3497 - cpu/arm/olpc
by repository service Dec. 14, 2012

Dec. 14, 2012

Author: wmb Date: Fri Dec 14 15:35:10 2012 New Revision: 3497 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3497 Log: OLPC ARM - The SP command protocol had redundant "unlock"s that may have caused conflicts between foreground and interrupt queue polling. Modified: cpu/arm/olpc/spcmd.fth Modified: cpu/arm/olpc/spcmd.fth ============================================================================== --- cpu/arm/olpc/spcmd.fth Fri Dec 14 08:32:27 2012 (r3496) +++ cpu/arm/olpc/spcmd.fth Fri Dec 14 15:35:10 2012 (r3497) @@ -80,14 +80,12 @@ ; : deque? ( channel# -- false | entry true ) - lock - select-queue if unlock false exit then + select-queue if false exit then head @ tail @ <> if q head @ ca+ c@ head inc-q-ptr true else false then - unlock ; 0 value reg-base

1 0

[commit] r3496 - dev/olpc/dcon
by repository service Dec. 14, 2012

Dec. 14, 2012

Author: quozl Date: Fri Dec 14 08:32:27 2012 New Revision: 3496 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/3496 Log: OLPC XO-1.75 XO-4 - fix DCON unfreeze glitch in "test /display", #12394, but needs review; calling screen wake and sleep from /dcon seems ugly. Modified: dev/olpc/dcon/mmp2dcon.fth Modified: dev/olpc/dcon/mmp2dcon.fth ============================================================================== --- dev/olpc/dcon/mmp2dcon.fth Fri Dec 14 08:25:26 2012 (r3495) +++ dev/olpc/dcon/mmp2dcon.fth Fri Dec 14 08:32:27 2012 (r3496) @@ -59,6 +59,20 @@ : dcon@ ( reg# -- word ) " reg-w@" $call-parent ; : dcon! ( word reg# -- ) " reg-w!" $call-parent ; +: mode@ ( -- mode ) 1 dcon@ ; +: mode! ( mode -- ) 1 dcon! ; +: hres! ( hres -- ) 2 dcon! ; \ def: h# 458 d# 1200 +: htotal! ( htotal -- ) 3 dcon! ; \ def: h# 4e8 d# 1256 +: hsync! ( sync -- ) 4 dcon! ; \ def: h# 1808 d# 24,8 +: vres! ( vres -- ) 5 dcon! ; \ def: h# 340 d# 900 +: vtotal! ( htotal -- ) 6 dcon! ; \ def: h# 390 d# 912 +: vsync! ( sync -- ) 7 dcon! ; \ def: h# 403 d# 4,3 +: timeout! ( to -- ) 8 dcon! ; \ def: h# ffff +: scanint! ( si -- ) 9 dcon! ; \ def: h# 0000 + +: scanint-on ( -- ) mode@ h# 100 or mode! ; +: scanint-off ( -- ) mode@ h# 100 invert and mode! ; + : dcon-load ( -- ) dcon-load-gpio# gpio-set ; : dcon-unload ( -- ) dcon-load-gpio# gpio-clr ; @@ -72,12 +86,30 @@ 1 value vga? \ VGA 0 value color? \ COLOUR -d# 905 value resumeline \ Configurable; should be set from args - -: wait-output ( -- ) d# 40 ms ; - +d# 850 value resumeline : mark-time ( -- start-time ) get-msecs ; : delta-ms ( start-time -- elapsed-ms ) mark-time swap - ; + +: wait-output ( -- ) + mark-time ( start-time ) + resumeline scanint! setup-dcon-irq scanint-on ( ) + begin ( start-time ) + dcon-irq? if ( start-time ) + setup-dcon-irq ( start-time ) + begin ( start-time ) + dcon-irq? if ( start-time ) + drop scanint-off exit ( ) + then ( start-time ) + dup delta-ms d# 100 > ( start-time ) + until ( start-time ) + then ( start-time ) + dup delta-ms d# 100 > ( start-time reached? ) + until ( start-time ) + drop ( ) + ." Timeout leaving DCON mode" cr ( ) + scanint-off ( ) +; + : wait-dcon-mode ( -- retry? ) mark-time ( start-time ) begin ( start-time ) @@ -103,23 +135,22 @@ dup vga? = if drop exit then ( source ) dup to vga? ( source ) if -\ unblank-display - d# 50 ms - wait-output - dcon-load \ Put the DCON in VGA-refreshed mode - d# 25 ms \ Ensure that that DCON sees the DCONLOAD high -\ display-on + wait-output \ Wait for the DCON to reach the scan line + " wake" $call-screen \ Enable video signal from SoC + d# 5 ms + dcon-load \ Put the DCON in VGA-refreshed mode + d# 25 ms \ Ensure that that DCON sees the DCONLOAD high else has-dcon-ram? if begin ( ) setup-dcon-irq dcon-unload \ Put the DCON in self-refresh mode wait-dcon-mode ( retry? ) - \ display-off ( retry? ) while ( ) \ We got a false ack from the DCON so start over from LOAD state dcon-load d# 25 ms ( ) repeat ( ) + " sleep" $call-screen then then ; @@ -132,16 +163,6 @@ ['] dcon! catch if 2drop bus-reset then ; -: mode@ ( -- mode ) 1 dcon@ ; -: mode! ( mode -- ) 1 dcon! ; -: hres! ( hres -- ) 2 dcon! ; \ def: h# 458 d# 1200 -: htotal! ( htotal -- ) 3 dcon! ; \ def: h# 4e8 d# 1256 -: hsync! ( sync -- ) 4 dcon! ; \ def: h# 1808 d# 24,8 -: vres! ( vres -- ) 5 dcon! ; \ def: h# 340 d# 900 -: vtotal! ( htotal -- ) 6 dcon! ; \ def: h# 390 d# 912 -: vsync! ( sync -- ) 7 dcon! ; \ def: h# 403 d# 4,3 -: timeout! ( to -- ) 8 dcon! ; \ def: h# ffff -: scanint! ( si -- ) 9 dcon! ; \ def: h# 0000 [ifdef] old-way : dcon-bright! ( level -- ) d# 10 dcon! ; \ def: h# xxxF ' dcon-bright! to bright! @@ -261,6 +282,16 @@ then ; +[ifdef] notdef +: test-dcon-freeze-glitch + screen-ih remove-output + " gvsr" $call-screen + begin dcon-freeze dcon-unfreeze key? until key drop + screen-ih add-output + page +; +[then] + \ LICENSE_BEGIN \ Copyright (c) 2010 FirmWorks \

1 0