OpenBIOS November 2007

openbios@openbios.org

5 participants
46 discussions

by svn＠openbios.org

Author: wmb Date: 2007-11-30 20:38:32 +0100 (Fri, 30 Nov 2007) New Revision: 742 Modified: cpu/x86/pc/olpc/usb8388.bth Log: OLPC build process - check Libertas MD5 sum from individual checksum file instead of from MD5SUMS, which is less prone to accidental corruption of the checksum. Modified: cpu/x86/pc/olpc/usb8388.bth =================================================================== --- cpu/x86/pc/olpc/usb8388.bth 2007-11-28 03:21:05 UTC (rev 741) +++ cpu/x86/pc/olpc/usb8388.bth 2007-11-30 19:38:32 UTC (rev 742) @@ -9,11 +9,12 @@ " wget -q http://dev.laptop.org/pub/firmware/libertas/usb8388-${WLAN_VERSION}.bin" expand$ $sh -" wget -q -O - http://dev.laptop.org/pub/firmware/libertas/MD5SUMS | grep ${WLAN_VERSION} >usb8388.md5" expand$ $sh +" wget -q http://dev.laptop.org/pub/firmware/libertas/usb8388-${WLAN_VERSION}.bin.md5" expand$ $sh -" md5sum usb8388-${WLAN_VERSION}.bin | cmp - usb8388.md5" expand$ $sh +" md5sum -b usb8388-${WLAN_VERSION}.bin | cmp - usb8388-${WLAN_VERSION}.bin.md5" expand$ $sh " mv usb8388-${WLAN_VERSION}.bin usb8388.bin" expand$ $sh +" rm usb8388-${WLAN_VERSION}.bin.md5" expand$ $sh \ This forces the creation of a .log file, so we don't re-fetch writing usb8388.version

13 years, 2 months

r741 - cpu/x86/pc/olpc

by svn＠openbios.org

Author: wmb Date: 2007-11-28 04:21:05 +0100 (Wed, 28 Nov 2007) New Revision: 741 Modified: cpu/x86/pc/olpc/versions.fth Log: Q2D05 Modified: cpu/x86/pc/olpc/versions.fth =================================================================== --- cpu/x86/pc/olpc/versions.fth 2007-11-28 00:08:05 UTC (rev 740) +++ cpu/x86/pc/olpc/versions.fth 2007-11-28 03:21:05 UTC (rev 741) @@ -2,10 +2,10 @@ \ The overall firmware revision macro: FW_MAJOR D -macro: FW_MINOR 04n +macro: FW_MINOR 05 \ The EC microcode -macro: EC_VERSION test4c +macro: EC_VERSION test4e \ Alternate command for getting EC microcode, for testing new versions. \ Temporarily uncomment the line and modify the path as necessary @@ -16,7 +16,7 @@ \ The wireless LAN module firmware macro: WLAN_RPM ${WLAN_VERSION}-1.olpc1 -macro: WLAN_VERSION 5.110.20.p4 +macro: WLAN_VERSION 5.110.20.p1 \ The bios_verify image macro: CRYPTO_VERSION 0.2

13 years, 3 months

r740 - cpu/x86/pc/olpc

by svn＠openbios.org

Author: wmb Date: 2007-11-28 01:08:05 +0100 (Wed, 28 Nov 2007) New Revision: 740 Modified: cpu/x86/pc/olpc/versions.fth Log: Changed EC and Libertas firmware versions for testing. Modified: cpu/x86/pc/olpc/versions.fth =================================================================== --- cpu/x86/pc/olpc/versions.fth 2007-11-27 00:23:05 UTC (rev 739) +++ cpu/x86/pc/olpc/versions.fth 2007-11-28 00:08:05 UTC (rev 740) @@ -2,21 +2,21 @@ \ The overall firmware revision macro: FW_MAJOR D -macro: FW_MINOR 04 +macro: FW_MINOR 04n \ The EC microcode -macro: EC_VERSION d02 +macro: EC_VERSION test4c \ Alternate command for getting EC microcode, for testing new versions. \ Temporarily uncomment the line and modify the path as necessary -\ macro: GET_EC wget http://dev.laptop.org/~rsmith/shiny6.bin -O ec.img +\ macro: GET_EC cp shiny4c.bin ec.img macro: KEYS mpkeys \ macro: KEYS testkeys \ The wireless LAN module firmware macro: WLAN_RPM ${WLAN_VERSION}-1.olpc1 -macro: WLAN_VERSION 5.110.20.p0 +macro: WLAN_VERSION 5.110.20.p4 \ The bios_verify image macro: CRYPTO_VERSION 0.2

13 years, 3 months

Determining whether BARs point to (non)relocatable memory/io space.

by Asif Haswarey

Hi everyone! I've run into what seems to be a contradiction with reference to the "relocatability" of a BAR's mem/io space. In the following example the spec. of the device says that BARs 0x10, 0x14 and 0x30 are relocatable. assigned-addresses 81001810 00000000 00000300 00000000 00000100 82001814 00000000 00100000 00000000 00002000 82001830 00000000 00180000 00000000 00080000 reg 00001800 00000000 00000000 00000000 00000000 01001810 00000000 00000000 00000000 00000100 02001814 00000000 00000000 00000000 00002000 02001830 00000000 00000000 00000000 00080000 So when I create and assign values to the reg property, I set the "n" bit to 0. Example BAR 0x10: 01001810 As defined in the OpenFirmware IEEE-1275 PCI Bus Binding 2.1, the "n" bit indicates (non)relocatability of the corresponding BAR's mem/io address space. The assigned-addresses property, however, is created by the system OpenFirmware (Sun SPARC Blade 1500 in this example). And as you can see the "n" bit in the phy-hi values is set for all the BARs (0x10, 0x14, 0x30), indicating that the corresponding BAR's space is non-relocatable. Example BAR 0x10: 81001810 I know, I'm being lazy! I can look for the piece of code in Sun's OBP source. But I'm trying to understand the rule for setting this "n" bit as far as the PCI spec. is concerned. I could not find anything indicating the attribute of relocatibility in the PCI spec. (including the PCI Architecture 2.1 book from Mindshare.) The only place I was able to make any sense of relocatability was the type bit field in the BAR[2:1]: For BAR[0] = 0, this BAR is associated with a memory address space: Type field BAR[2:1] = 00 (locate anywhere in lower 4GB) 01 (locate below 1MB (reserved in PCI 2.2 spec)) 10 (locate anywhere over 4GB (implies that register is 64-bits wide)) 11 (reserved) For BAR[0] = 1, this BAR is associated with a IO address space. There is no type field in this case. All this leads me to conclude: - all memory-address-space BARs are relocatable; - all io-address-space BARs are non-relocatable. Is this conclusion correct? _____________ Äsif Haswaréy QLogic Corporation FCode Driver Development Phone: 949.389.7607 Fax: 949.389.6308 asif.haswarey(a)qlogic.com

13 years, 3 months

r739 - cpu/x86/pc/olpc

by svn＠openbios.org

Author: wmb Date: 2007-11-27 01:23:05 +0100 (Tue, 27 Nov 2007) New Revision: 739 Modified: cpu/x86/pc/olpc/gxmsrs.fth cpu/x86/pc/olpc/lxmsrs.fth cpu/x86/pc/olpc/rtcwake.fth Log: OLPC Trac 5128 - Set RTC alarm day, alarm month, and century offset MSRs, and set century byte to 20. Modified: cpu/x86/pc/olpc/gxmsrs.fth =================================================================== --- cpu/x86/pc/olpc/gxmsrs.fth 2007-11-20 07:40:33 UTC (rev 738) +++ cpu/x86/pc/olpc/gxmsrs.fth 2007-11-27 00:23:05 UTC (rev 739) @@ -291,6 +291,9 @@ \ msr: 5140.004e 00000000.ef2500c0. \ LPC_SIRQ \ msr: 5140.004e 00000000.effd0080. \ LPC_SIRQ msr: 5140.004e 00000000.effd00c0. \ LPC_SIRQ - Active high bit 12 (mouse) and 2 (kbd), Enable, Quiet, 17 frames +msr: 5140.0055 00000000.0000003d. \ RTC alarm day (offset into CMOS RAM) +msr: 5140.0056 00000000.0000003e. \ RTC alarm month (offset into CMOS RAM) +msr: 5140.0057 00000000.00000032. \ RTC century (offset into CMOS RAM) \ USB host controller msr: 5120.0001 0000000b.00000000. \ USB_GLD_MSR_CONFIG - 5536 page 262 Modified: cpu/x86/pc/olpc/lxmsrs.fth =================================================================== --- cpu/x86/pc/olpc/lxmsrs.fth 2007-11-20 07:40:33 UTC (rev 738) +++ cpu/x86/pc/olpc/lxmsrs.fth 2007-11-27 00:23:05 UTC (rev 739) @@ -241,6 +241,9 @@ \ msr: 5140.004e 00000000.ef2500c0. \ LPC_SIRQ \ msr: 5140.004e 00000000.effd0080. \ LPC_SIRQ msr: 5140.004e 00000000.effd00c0. \ LPC_SIRQ - Active high bit 12 (mouse) and 2 (kbd), Enable, Quiet, 17 frames + msr: 5140.0055 00000000.0000003d. \ RTC alarm day (offset into CMOS RAM) + msr: 5140.0056 00000000.0000003e. \ RTC alarm month (offset into CMOS RAM) + msr: 5140.0057 00000000.00000032. \ RTC century (offset into CMOS RAM) \ USB host controller msr: 5120.0001 0000000b.00000000. \ USB_GLD_MSR_CONFIG - 5536 page 262 Modified: cpu/x86/pc/olpc/rtcwake.fth =================================================================== --- cpu/x86/pc/olpc/rtcwake.fth 2007-11-20 07:40:33 UTC (rev 738) +++ cpu/x86/pc/olpc/rtcwake.fth 2007-11-27 00:23:05 UTC (rev 739) @@ -4,17 +4,17 @@ : enable-rtc-irq ( -- ) h# a1 pc@ h# fe and h# a1 pc! ; : disable-rtc-irq ( -- ) h# a1 pc@ 1 or h# a1 pc! ; -0 value cmos-alarm-day \ Offset of day alarm in CMOS -0 value cmos-alarm-month \ Offset of month alarm in CMOS +\ Must agree with lxmsrs.fth +h# 3d constant cmos-alarm-day \ Offset of day alarm in CMOS +h# 3e constant cmos-alarm-month \ Offset of month alarm in CMOS +h# 32 constant cmos-century \ Offset of century byte in CMOS -: set-alarm-offsets ( -- ) - h# 10. h# 5140.0055 wrmsr - h# 11. h# 5140.0056 wrmsr +dev /rtc + cmos-alarm-day " alarm_day" integer-property + cmos-alarm-month " alarm_month" integer-property + cmos-century " century" integer-property +dend - h# 5140.0055 rdmsr drop to cmos-alarm-day - h# 5140.0056 rdmsr drop to cmos-alarm-month -; - false value enable-rtc-irq? : rtc-handler ( -- ) h# c cmos@ drop ." R" ; \ Clear RTC interrupt flags @@ -51,12 +51,15 @@ 0 0 acpi-l! \ Disable RTC SCI ; : rtc-wackup - set-alarm-offsets 0 begin pm-sleep-rtc space dup . (cr 1+ key? until key drop ; +stand-init: Century + d# 20 cmos-century cmos! +; + \ LICENSE_BEGIN \ Copyright (c) 2007 FirmWorks \

13 years, 3 months

Strange request when attempting to build the latest release

by Gregg Levine

Hello! It seems that with the later releases of OpenBIOS from the SVN place I get a strange request: root@jimkirk:/usr/src/openbios/openbios-devel# make Please install the OpenBIOS fcode utilities. Download with : $ svn co svn://openbios.org/openbios/fcode-utils make: *** [requirements] Error 1 root@jimkirk:/usr/src/openbios/openbios-devel# Well I did that. I then repeated the make command, it said the same thing. Any suggestions for configuring the fcode-utils, and where they should be placed? I stuck my copy inside the same tree as the OpenBIOS SVN release. -- Gregg C Levine gregg.drwho8(a)gmail.com "This signature was once found posting rude messages in English in the Moscow subway."

13 years, 3 months

r738 - ofw/fs/fatfs

by svn＠openbios.org

Author: wmb Date: 2007-11-20 08:40:33 +0100 (Tue, 20 Nov 2007) New Revision: 738 Modified: ofw/fs/fatfs/write.fth Log: OLPC trac 5029 - Fixed a problem that caused files written from OFW to be 0x400 bytes too short in certain cases. Modified: ofw/fs/fatfs/write.fth =================================================================== --- ofw/fs/fatfs/write.fth 2007-11-19 23:48:42 UTC (rev 737) +++ ofw/fs/fatfs/write.fth 2007-11-20 07:40:33 UTC (rev 738) @@ -80,7 +80,7 @@ \ If we need more cluster(s) and \ we're at or beyond the end of the file, add new cluster(s) dup 0> - current-position fh_length l@ > last-cluster? or and if + current-position fh_length l@ > last-cluster? or and if extend-file ?dup if ?flush-fat-cache nip exit then extended? on else \ same cluster? @@ -94,17 +94,19 @@ ?flush-fat-cache true exit then - \ If this is a partial cluster without extension, - \ then adjust the length field - remaining @ /cluster < extended? @ invert and if + \ Adjust the length field unless extend-file already did it + extended? @ 0= if current-position remaining @ + fh_length l! then \ Continue with the rest of the transfer #cont-cls @ /cluster * dup bufadr +! negate remaining +! - next-cl# @ ?dup if 1st-cl# ! 1 #cont-cls ! next-cl# off - else 1st-cl# off #cont-cls off then + next-cl# @ ?dup if + 1st-cl# ! 1 #cont-cls ! next-cl# off + else + 1st-cl# off #cont-cls off + then last-cluster? 0= if to-next-cluster then repeat drop ( )

13 years, 3 months

r737 - ofw/fs/fatfs

by svn＠openbios.org

Author: wmb Date: 2007-11-20 00:48:42 +0100 (Tue, 20 Nov 2007) New Revision: 737 Modified: ofw/fs/fatfs/device.fth Log: OLPC trac 5022 - Fixed FAT size determination code in the face of crazy media type values, using Microsoft's recommended algorithm for determining whether to use FAT-12, FAT-16, or FAT-32. Modified: ofw/fs/fatfs/device.fth =================================================================== --- ofw/fs/fatfs/device.fth 2007-11-18 03:54:48 UTC (rev 736) +++ ofw/fs/fatfs/device.fth 2007-11-19 23:48:42 UTC (rev 737) @@ -121,6 +121,9 @@ then ; +\ This version of read-bpb implements Microsoft's recommended +\ method for determining whether a filesystem is FAT-16 or FAT-32. +variable nsects : ?read-bpb ( -- ) /sector 0= if \ Read bpb if necessary init-sector-size @@ -132,25 +135,14 @@ then bp_bps lew@ bps w! bp_spc c@ spc c! - bp_spf lew@ ?dup if - media c@ h# f8 = if - dup d# 12 < if fat12 else fat16 then - else - fat12 - then 0 0 - 0 fsinfo ! 0 fssector ! false fsinfos-dirty c! - else - bp_bspf lel@ fat32 - bp_rdirclus lel@ - bp_fsinfos lew@ dup read-fsinfo - then - fsinfos w! - rdirclus l! - fat-type c! - spf l! + bp_spf lew@ ?dup 0= if bp_bspf lel@ then spf l! + bp_nsects lew@ ?dup 0= if bp_xnsects lel@ then nsects l! - init-fat-cache + bp_ndirs lew@ /dirent * /sector 1- + /sector / #dir-sectors w! + \ nsects is TotSec + \ #dir-sectors is RootDirSectors + \ Sector number where the FAT starts. \ bp_nhid is the number of sectors before the BPB sector. \ bp_res is the number of sectors from the BPT sector to the @@ -158,17 +150,24 @@ \ If the underlying disk driver handles partition offsets, \ we don't need to handle bp_nhid here. - bp_res lew@ ( bp_nhid lel@ + ) fat-sector0 w! + bp_res lew@ ( bp_nhid lel@ + ) + dup fat-sector0 w! ( #resv-sectors ) + spf l@ bp_nfats c@ * + ( #early-sectors ) + dup dir-sector0 w! ( #early-sectors ) + #dir-sectors w@ + cl-sector0 l! ( ) - rdirclus @ ?dup if - dv_cwd-cl l! + \ cl-sector0 is FirstDataSector + + nsects l@ #dir-sectors w@ - spc c@ / ( #clusters ) + dup d# 65525 >= if ( #clusters ) + drop fat32 fat-type c! 0 dir-sector0 w! - 0 #dir-sectors w! - spf l@ bp_nfats c@ * fat-sector0 w@ + cl-sector0 l! + bp_rdirclus lel@ rdirclus l! + bp_fsinfos lew@ dup read-fsinfo fsinfos w! else - spf l@ bp_nfats c@ * fat-sector0 w@ + dir-sector0 w! - bp_ndirs lew@ /dirent * /sector / #dir-sectors w! - dir-sector0 w@ #dir-sectors w@ + cl-sector0 l! + d# 4085 >= if fat16 else fat12 then fat-type c! + 0 fsinfo ! 0 fssector ! false fsinfos-dirty c! + 0 fsinfos w! 0 rdirclus l! then \ The number of clusters is limited both by space for clusters numbers @@ -180,16 +179,22 @@ spf l@ /sector * bytes>cl-entries ( #clusters ) \ Compare it with the number of clusters for which there is disk space - bp_nsects lew@ ?dup 0= if bp_xnsects lel@ then + nsects l@ cl-sector0 l@ - spc c@ / 2 + min 1- max-cl# l! free-bpb + + \ Start with the root directory as the current working directory + rdirclus @ dv_cwd-cl l! + + init-fat-cache then ; : set-device ( device# -- ) (set-device) ?read-bpb ; \ : stand-init ( -- ) stand-init clear-device-records ; + \ LICENSE_BEGIN \ Copyright (c) 2006 FirmWorks \

13 years, 3 months

r179 - openbios-devel/drivers

by svn＠openbios.org

Author: blueswirl Date: 2007-11-19 20:11:38 +0100 (Mon, 19 Nov 2007) New Revision: 179 Modified: openbios-devel/drivers/obio.c Log: Improved Module ID generation Modified: openbios-devel/drivers/obio.c =================================================================== --- openbios-devel/drivers/obio.c 2007-11-17 19:00:58 UTC (rev 178) +++ openbios-devel/drivers/obio.c 2007-11-19 19:11:38 UTC (rev 179) @@ -891,7 +891,15 @@ push_str("cache-coherence?"); fword("property"); - PUSH(i); + switch (machine_id) { + case 0x71: + case 0x72: + PUSH(i + 8); + break; + case 0x80: + PUSH(i); + break; + } fword("encode-int"); push_str("mid"); fword("property"); @@ -1110,6 +1118,7 @@ sparc_header->smp_ctx = context; sparc_header->valid = 1; + cpu &= 7; intregs->cpu_intregs[cpu].set = SUN4M_SOFT_INT(14); return 0;

13 years, 3 months

r736 - cpu/x86/pc/olpc

by svn＠openbios.org

Author: wmb Date: 2007-11-18 04:54:48 +0100 (Sun, 18 Nov 2007) New Revision: 736 Modified: cpu/x86/pc/olpc/fw.bth cpu/x86/pc/olpc/loaddropins.fth cpu/x86/pc/olpc/security.fth Log: OLPC - Cosmetic improvements in secure boot error messages and some stuff to help the factory repair systems after FQA. Modified: cpu/x86/pc/olpc/fw.bth =================================================================== --- cpu/x86/pc/olpc/fw.bth 2007-11-16 22:29:19 UTC (rev 735) +++ cpu/x86/pc/olpc/fw.bth 2007-11-18 03:54:48 UTC (rev 736) @@ -332,9 +332,9 @@ fload ${BP}/cpu/x86/pc/olpc/lzip.fth \ Access zip images from memory fload ${BP}/cpu/x86/pc/olpc/gamekeys.fth fload ${BP}/cpu/x86/pc/olpc/linux.fth +fload ${BP}/cpu/x86/pc/olpc/setwp.fth fload ${BP}/cpu/x86/pc/olpc/security.fth fload ${BP}/cpu/x86/pc/olpc/fsupdate.fth -fload ${BP}/cpu/x86/pc/olpc/setwp.fth fload ${BP}/ofw/gui/ofpong.fth fload ${BP}/cpu/x86/pc/olpc/life.fth fload ${BP}/cpu/x86/pc/olpc/sound.fth @@ -520,12 +520,13 @@ ." Tests complete - powering off" cr d# 5000 ms power-off then ; + : ?scan-nand ( -- ) rocker-up game-key? if text-on scan-nand then ; : ?fs-update ( -- ) button-check button-x or button-o or button-square or ( mask ) - game-key-mask over and = if + game-key-mask = if try-fs-update then ; @@ -547,6 +548,8 @@ no-page + ?factory-mode + console-start ?resume-botch Modified: cpu/x86/pc/olpc/loaddropins.fth =================================================================== --- cpu/x86/pc/olpc/loaddropins.fth 2007-11-16 22:29:19 UTC (rev 735) +++ cpu/x86/pc/olpc/loaddropins.fth 2007-11-18 03:54:48 UTC (rev 736) @@ -27,7 +27,7 @@ " ${BP}/ofw/inet/telnetd.fth" " telnetd" $add-deflated-dropin - " ${BP}/cpu/x86/pc/olpc/images/warnings.565" " warnings.565" $add-deflated-dropin +\ " ${BP}/cpu/x86/pc/olpc/images/warnings.565" " warnings.565" $add-deflated-dropin " ${BP}/cpu/x86/pc/olpc/images/lightdot.565" " lightdot.565" $add-deflated-dropin " ${BP}/cpu/x86/pc/olpc/images/yellowdot.565" " yellowdot.565" $add-deflated-dropin " ${BP}/cpu/x86/pc/olpc/images/darkdot.565" " darkdot.565" $add-deflated-dropin Modified: cpu/x86/pc/olpc/security.fth =================================================================== --- cpu/x86/pc/olpc/security.fth 2007-11-16 22:29:19 UTC (rev 735) +++ cpu/x86/pc/olpc/security.fth 2007-11-18 03:54:48 UTC (rev 736) @@ -13,8 +13,8 @@ 0 0 2value base-xy 0 0 2value next-xy -d# 410 d# 540 2constant progress-xy -d# 557 d# 283 2constant sad-xy +d# 463 d# 540 2constant progress-xy +d# 552 d# 283 2constant sad-xy true value debug-security? : ?lease-debug ( msg$ -- ) @@ -23,6 +23,9 @@ : ?lease-debug-cr ( msg$2 -- ) debug-security? if type cr else 2drop then ; +: ?lease-error-cr ( msg$2 -- ) + debug-security? if red-letters type black-letters cr else 2drop then +; : fail-load ( -- ) text-on @@ -31,15 +34,21 @@ begin again ; +code halt hlt c; \ To save power + 0 value security-off? : security-failure ( -- ) visible - ." Stopping" cr - security-off? if quit then + security-off? if ." Stopping" cr quit then - d# 10000 ms - power-off + button-check game-key? if + ." Use power button to power off" cr + begin halt again + else + ." Powering off in 10 seconds" cr + d# 10000 ms power-off + then ; : +icon-xy ( delta-x,y -- ) icon-xy d+ to icon-xy ; @@ -56,6 +65,12 @@ sad-xy to icon-xy " sad" show-icon to icon-xy ; +: .security-failure ( error$ -- ) + visible red-letters type black-letters cr + show-sad + security-failure +; + : show-lock ( -- ) " lock" show-icon ; : show-unlock ( -- ) " unlock" show-icon ; : show-child ( -- ) @@ -63,14 +78,16 @@ d# 552 d# 383 to icon-xy " rom:xogray.565" $show-opaque progress-xy to icon-xy \ For boot progress reports ; + +0 [if] : show-warnings ( -- ) " erase-screen" $call-screen d# 48 d# 32 to icon-xy " rom:warnings.565" $show-opaque dcon-freeze ; +[then] - h# 20 buffer: cn-buf \ filename prefix - either "act" or "run" h# 20 buffer: fn-buf \ filename tail - either "os" or "rd" h# 100 buffer: pn-buf \ pathname - either "\boot" or "\boot-alt" @@ -211,7 +228,7 @@ pubkey$ 2swap signature-bad? ( error? ) dup if - " Signature invalid" ?lease-debug-cr + " Signature invalid" ?lease-error-cr else " Signature valid" ?lease-debug-cr then @@ -233,7 +250,7 @@ 2dup our-pubkey? if false exit then ( rem$ line$ ) 2drop ( rem$ ) repeat ( rem$ ) - " No signature for our key" ?lease-debug-cr + " No signature for our key" ?lease-error-cr 2drop true ; @@ -266,94 +283,64 @@ over r@ - r> ( tail$ head$ ) ; -0. 2value exp-seconds \ Accumulator for parsing data/time strings - \ numfield is a factor used for parsing 2-digit fields from date/time strings. -\ Radix is the number to scale the result by, i.e. one more than the maximum -\ value of the field. Adjust is 0 for fields whose first valid value is 0 -\ (hours, minutes, seconds) or 1 for fields that start at 1 (month,day). - -: numfield ( exp$ adjust radix -- exp$' ) - >r >r ( exp$ r: radix adjust ) - 2 break$ $number throw ( exp$' num r: radix adjust ) - r> - ( exp$ num' r: radix ) - dup r@ u>= throw ( exp$ num r: radix ) - - \ No need to multiply the top half because it can only become nonzero - \ on the last call to scale-time - exp-seconds drop r> um* ( exp$ num d.seconds ) - rot 0 d+ to exp-seconds ( exp$ ) +: numfield ( exp$ min max -- exp$' ) + >r >r ( exp$ r: max min ) + 2 break$ $number throw ( exp$' num r: max min ) + dup r> < throw ( exp$ num r: max ) + dup r> > throw ( exp$ num ) ; \ expiration-to-seconds parses an expiration date string like \ "20070820T130401Z", converting it to (double precision) seconds \ according to the simplified calculation described above for "get-date" -: (expiration-to-seconds) ( expiration$ -- true | d.seconds false ) - 4 break$ $number throw ( exp$' year ) - dup d# 2999 u> throw ( exp$' year ) - 0 to exp-seconds ( exp$' ) - - 1 d# 12 numfield ( exp$' ) \ Month - 1 d# 31 numfield ( exp$' ) \ Day - - 1 break$ " T" $= 0= throw ( exp$' ) - - 0 d# 24 numfield ( exp$' ) \ Hour - 0 d# 60 numfield ( exp$' ) \ Minute - 0 d# 60 numfield ( exp$' ) \ Second - - " Z" $= 0= throw ( ) - exp-seconds +: (expiration-to-seconds) ( expiration$ -- d.seconds ) + 4 break$ $number throw >r ( exp$' r: y ) + 1 d# 12 numfield >r ( exp$' r: y m ) + 1 d# 31 numfield >r ( exp$' r: y m d ) + 1 break$ " T" $= 0= throw ( exp$' r: y m d ) + 0 d# 23 numfield >r ( exp$' r: y m d h ) + 0 d# 59 numfield >r ( exp$' r: y m d h m ) + 0 d# 59 numfield >r ( exp$' r: y m d h m s ) + " Z" $= 0= throw ( r: y m d h m s ) + r> r> r> r> r> r> ( s m h m d y ) + >unix-seconds ; -: expiration-to-seconds ( expiration$ -- true | d.seconds false ) +: expiration-to-seconds ( expiration$ -- true | seconds false ) push-decimal - ['] (expiration-to-seconds) catch ( x x true | d.seconds false ) + ['] (expiration-to-seconds) catch ( x x true | seconds false ) pop-base dup if nip nip then ; -\ earliest is the earliest acceptable date value (in seconds). -\ It is the date that the first test version of this code was -\ deployed. If a laptop has any earlier date that than, that -\ date is presumed bogus. +0 value current-seconds -" 20070101T000000Z" expiration-to-seconds drop 2constant earliest +: date-bad? ( -- flag ) + current-seconds 0= if + time&date >unix-seconds to current-seconds + then -0. 2value current-seconds + \ earliest is the earliest acceptable date value (in seconds). + \ It is the date that the first test version of this code was + \ deployed. If a laptop has any earlier date that than, that + \ date is presumed bogus. -\ get-date reads the date and time from the real time clock -\ and converts it to seconds. - -\ The seconds conversion uses a simplified approach that ignores -\ leap years and the like - it assumes that all months are 31 days. -\ This is sufficient for comparison purposes so long as we use the -\ same calculation in all cases. It is not good for doing -\ arithmetic on dates. -: get-date ( -- ) - time&date ( s m h d m y ) - d# 12 * swap 1- + ( s m h d m' ) \ Months start at 1 - d# 31 * swap 1- + ( s m h d' ) \ Days start at 1 - d# 24 * + ( s m h' ) - d# 60 * + ( s m' ) \ Can't overflow so far - d# 60 um* ( s d.s' ) - swap 0 d+ to current-seconds + current-seconds [ " 20070101T000000Z" expiration-to-seconds drop ] literal - 0< ; + \ expired? determines whether or not the expiration time string is \ earlier than this machine's current time (from the real time clock). : expired? ( expiration$ -- bad? ) - expiration-to-seconds if true exit then ( d.seconds ) + expiration-to-seconds if true exit then ( seconds ) - current-seconds 0. d= if ( d.seconds ) - then - \ If the date is bad, leases are deemed to have expired - current-seconds earliest d< if 2drop true exit then + date-bad? if drop true exit then ( seconds ) - current-seconds d< + current-seconds - 0< ; d# 1024 constant /sec-line-max @@ -380,11 +367,11 @@ : get-my-sn ( -- error? ) " SN" find-tag 0= if - " No serial number in mfg data" ?lease-debug-cr + " No serial number in mfg data" ?lease-error-cr true exit then ( adr len ) ?-null dup d# 11 <> if - " Invalid serial number" ?lease-debug-cr + " Invalid serial number" ?lease-error-cr 2drop true exit then ( adr len ) machine-id-buf swap move @@ -392,11 +379,11 @@ [char] : machine-id-buf d# 11 + c! " U#" find-tag 0= if - " No UUID in mfg data" ?lease-debug-cr + " No UUID in mfg data" ?lease-error-cr true exit then ( adr len ) ?-null dup d# 36 <> if - " Invalid UUID" ?lease-debug-cr + " Invalid UUID" ?lease-error-cr 2drop true exit then ( adr len ) machine-id-buf d# 12 + swap move @@ -419,12 +406,12 @@ 2dup " 00000000T000000Z" $= if 0 exit then dup d# 16 <> if ( expiration$ ) - " has bad expiration format" ?lease-debug-cr + " has bad expiration format" ?lease-error-cr -1 exit then ( expiration$ ) 2dup expired? if - " expired" ?lease-debug-cr + " expired" ?lease-error-cr -1 exit then ( expiration$ ) 0 @@ -454,7 +441,10 @@ my-sn$ $= 0= if 2drop 0 exit then ( rem$ ) \ Disposition code - bl left-parse-string 1 <> if 3drop -1 exit then ( rem$ disp-adr ) + bl left-parse-string 1 <> if + " No disposition code" ?lease-error-cr + 3drop -1 exit + then ( rem$ disp-adr ) set-disposition ( rem$ ) bl left-parse-string check-expiry if 4drop -1 exit then ( sig$ exp$ ) @@ -470,7 +460,7 @@ : check-lease ( act01-lease$ -- -1|0|1 ) bl left-parse-string " act01:" $= 0= if - " Not act01:" ?lease-debug-cr + " Not act01:" ?lease-error-cr 2drop -1 exit then ( rem$ ) check-timed-signature ( -1|0|1 ) @@ -498,7 +488,7 @@ -1 of r> close-file drop " lock" show-icon false exit endof endcase repeat - " No matching records" ?lease-debug-cr + " No matching records" ?lease-error-cr r> close-file drop false ; @@ -575,6 +565,7 @@ ; false value secure? +false value in-factory? stand-init: wp " wp" find-tag if 2drop true to secure? then @@ -612,7 +603,7 @@ -1 of r> close-file drop false exit endof endcase repeat - " No matching records" ?lease-debug-cr + " No matching records" ?lease-error-cr r> close-file drop false ; @@ -625,13 +616,13 @@ base @ >r d# 36 base ! fw#buf 5 $number if show-x - visible ." Invalid firmware version number" security-failure + " Invalid firmware version number" .security-failure then pop-base ; : firmware-up-to-date? ( img$ -- ) - /flash <> if show-x visible ." Invalid Firmware image" security-failure then ( adr ) + /flash <> if show-x " Invalid Firmware image" .security-failure then ( adr ) (fw-version) ( file-version# ) rom-pa (fw-version) ( file-version# rom-version# ) u<= @@ -680,7 +671,7 @@ ['] ?enough-power catch ?dup if visible - .error + red-letters .error black-letters security-failure then @@ -689,9 +680,7 @@ reflash \ Should power-off and reboot show-x - visible - ." Reflash returned, unexpectedly" cr - security-failure + " Reflash returned, unexpectedly" .security-failure then show-lock then @@ -757,7 +746,6 @@ next-xy to icon-xy ( list$ ) repeat ( list$ ) - " sad" show-icon ( list$ ) 2drop false ( ) ; @@ -765,15 +753,19 @@ : all-devices$ ( -- list$ ) " disk: sd: nand:" ; +: secure-startup ( -- ) + in-factory? if + button-check button-x or button-o or button-square or button-rotate or ( mask ) + game-key-mask = if exit then + then -: secure-startup ( -- ) ['] noop to ?show-device ['] noop to load-done ['] noop to load-started set-alternate - button-rotate game-key? if show-warnings then +\ button-rotate game-key? if show-warnings then show-child ?force-secure @@ -788,18 +780,58 @@ persistent-devkey? if true to security-off? visible exit then - get-my-sn if visible ." No serial number" cr show-sad security-failure then + get-my-sn if " No serial number" .security-failure then - get-date current-seconds earliest d< if + date-bad? if \ This is not fatal, because we don't want a brick if the RTC battery fails - visible ." Invalid system date" cr show-sad + visible red-letters ." Invalid system date" black-letters cr show-sad then - load-crypto if visible ." Crytpo load failed" cr show-sad security-failure then ( ) + load-crypto if " Crytpo load failed" .security-failure then ( ) alternate? if " \boot-alt" else " \boot" then pn-buf place all-devices$ load-from-list if exit then \ Returns only if no images found - visible ." Boot failed" cr show-sad security-failure + " Boot failed" .security-failure ; + +: efface-md ( -- ) + " md" find-tag 0= if exit then ( data$ ) + + 2 + flash-base - ( flash-offset ) + spi-start spi-identify ( flash-offset ) + " MD" rot write-spi-flash ( ) + spi-reprogrammed ( ) +; + +: days>seconds ( n -- seconds ) [ d# 60 d# 60 * d# 24 * ] literal * ; +: ?factory-mode ( -- ) + date-bad? if efface-md exit then + " md" find-tag if ( data$ ) + 0 left-parse-string 2nip ( time$ ) + \ Erase the tag if it is invalid + expiration-to-seconds if efface-md exit then ( begin-seconds ) + dup 3 days>seconds + ( begin-seconds end-seconds ) + \ Erase the tag if its time is up + current-seconds -rot within 0= if efface-md exit then ( ) + true to in-factory? + then +; + +\ iso8601 date construction for activation key +: .2digits ( .. roll# -- .. ) roll u# u# drop ; +: >iso8601$ ( s m h d m y -- adr len ) + push-decimal + <# + [char] Z hold 5 .2digits 4 .2digits 3 .2digits + [char] T hold 2 .2digits 1 .2digits u# u# u# u# + u#> + pop-base +; + +: factory-mode ( -- ) + " md" find-tag if ." md tag already exists" cr 2drop exit then + " MD" find-tag if ." MD tag already exists" cr 2drop exit then + date-bad? if ." The RTC is not set correctly" cr exit then + time&date >iso8601$ " md" $add-tag +;

13 years, 3 months