October 2011 - openfirmware - mail.coreboot.org

[commit] r2658 - cpu/arm/olpc/1.75
by repository service Oct. 31, 2011

Oct. 31, 2011

Author: quozl Date: Mon Oct 31 10:10:14 2011 New Revision: 2658 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2658 Log: OLPC XO-1.75 C1 - accelerometer, use 100 kHz SCL rate now that SDA and SCL pullups are present, continue to use 25 kHz SCL rate for older devices. #11367. Modified: cpu/arm/olpc/1.75/accelerometer.fth Modified: cpu/arm/olpc/1.75/accelerometer.fth ============================================================================== --- cpu/arm/olpc/1.75/accelerometer.fth Fri Oct 28 04:46:45 2011 (r2657) +++ cpu/arm/olpc/1.75/accelerometer.fth Mon Oct 31 10:10:14 2011 (r2658) @@ -298,9 +298,10 @@ ; : delay ( -- ) d# 30 ms ; +d# 25,000 value bus-speed : open ( -- flag ) my-unit " set-address" $call-parent - d# 25,000 " set-bus-speed" $call-parent + bus-speed " set-bus-speed" $call-parent ['] accelerometer-on catch 0= ; : close ( -- ) @@ -389,10 +390,11 @@ : probe ( -- ) h# 3a 6 " set-address" $call-parent - d# 25,000 " set-bus-speed" $call-parent + d# 25,000 " set-bus-speed" $call-parent \ XO-1.75 B1 lacks pullups SCL SDA ['] accelerometer-on catch if \ The attempt to talk at the old address failed, so we assume the new chip \ Support for new LIS3DHTR chip + d# 100,000 to bus-speed d# 50 to min-x d# 50 to min-y d# 50 to min-z d# 150 to max-x d# 150 to max-y d# 450 to max-z h# 32 6 encode-phys " reg" property @@ -401,6 +403,7 @@ accelerometer-off \ Something responded to the old address, so we assume it's the old chip \ Support for old LIS33DE chip + d# 25,000 to bus-speed d# 20 to min-x d# 20 to min-y d# 20 to min-z d# 400 to max-x d# 400 to max-y d# 400 to max-z h# 3a 6 encode-phys " reg" property

1 0

[commit] r2657 - in cpu: arm/olpc/1.75 x86/pc/olpc/via
by repository service Oct. 27, 2011

Oct. 27, 2011

Author: quozl Date: Fri Oct 28 04:46:45 2011 New Revision: 2657 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2657 Log: OLPC XO-1.75 XO-1.5 - minor file comment fix Modified: cpu/arm/olpc/1.75/usb.fth cpu/x86/pc/olpc/via/usb.fth Modified: cpu/arm/olpc/1.75/usb.fth ============================================================================== --- cpu/arm/olpc/1.75/usb.fth Thu Oct 27 03:53:58 2011 (r2656) +++ cpu/arm/olpc/1.75/usb.fth Fri Oct 28 04:46:45 2011 (r2657) @@ -1,4 +1,4 @@ -purpose: USB elaborations for the BIOS loaded OFW +purpose: Platform-specific USB elaborations \ See license at end of file 0 0 " d4208000" " /" begin-package \ USB Host Controller Modified: cpu/x86/pc/olpc/via/usb.fth ============================================================================== --- cpu/x86/pc/olpc/via/usb.fth Thu Oct 27 03:53:58 2011 (r2656) +++ cpu/x86/pc/olpc/via/usb.fth Fri Oct 28 04:46:45 2011 (r2657) @@ -1,4 +1,4 @@ -purpose: USB elaborations for the BIOS loaded OFW +purpose: Platform-specific USB elaborations \ See license at end of file 0 config-int usb-delay \ Milliseconds to wait before set-address

1 0

[commit] r2656 - cpu/arm/olpc/1.75
by repository service Oct. 26, 2011

Oct. 26, 2011

Author: quozl Date: Thu Oct 27 03:53:58 2011 New Revision: 2656 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2656 Log: OLPC XO-1.75 - accelerometer ADC test words Modified: cpu/arm/olpc/1.75/accelerometer.fth Modified: cpu/arm/olpc/1.75/accelerometer.fth ============================================================================== --- cpu/arm/olpc/1.75/accelerometer.fth Wed Oct 26 08:40:16 2011 (r2655) +++ cpu/arm/olpc/1.75/accelerometer.fth Thu Oct 27 03:53:58 2011 (r2656) @@ -16,15 +16,24 @@ 0 [if] +select /accelerometer + : temperature-on ( -- ) h# c0 h# 1f acc-reg! ; +: bdu ( -- ) h# 23 acc-reg@ h# 80 or h# 23 acc-reg! ; : adc1@ ( -- ) h# 08 acc-reg@ h# 09 acc-reg@ bwjoin wext 5 >>a ; : adc2@ ( -- ) h# 0a acc-reg@ h# 0b acc-reg@ bwjoin wext 5 >>a ; -: adc3@ ( -- ) h# 0c acc-reg@ h# 0d acc-reg@ bwjoin wext 5 >>a ; + +: wait-adc3 begin h# 07 acc-reg@ h# 04 and until ; +: adc3@ ( -- l h ) + h# 0c acc-reg@ h# 0d acc-reg@ bwjoin wext 5 >>a +; +: 1hz ( -- ) h# 20 acc-reg@ h# 0f and h# 10 or h# 20 acc-reg! ; : .xr ( n -- ) push-hex 2 .r pop-base space ; : .dr ( n -- ) push-decimal 3 .r pop-base space ; : .br ( n -- ) push-binary 0 <# # # # # [char] . hold # # # # #> type pop-base space ; +: .status-aux h# 07 acc-reg@ .br ; : .reg ( register -- value ) space

1 0

[commit] r2655 - cpu/arm/olpc/1.75
by repository service Oct. 25, 2011

Oct. 25, 2011

Author: quozl Date: Wed Oct 26 08:40:16 2011 New Revision: 2655 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2655 Log: Q4C02 Modified: cpu/arm/olpc/1.75/ec-version.fth cpu/arm/olpc/1.75/fw-version.fth Modified: cpu/arm/olpc/1.75/ec-version.fth ============================================================================== --- cpu/arm/olpc/1.75/ec-version.fth Tue Oct 25 08:16:34 2011 (r2654) +++ cpu/arm/olpc/1.75/ec-version.fth Wed Oct 26 08:40:16 2011 (r2655) @@ -1,5 +1,5 @@ \ The EC microcode -macro: EC_VERSION 4_0_3_01 +macro: EC_VERSION 4_0_3_02 \ Alternate command for getting EC microcode, for testing new versions. \ Temporarily uncomment the line and modify the path as necessary Modified: cpu/arm/olpc/1.75/fw-version.fth ============================================================================== --- cpu/arm/olpc/1.75/fw-version.fth Tue Oct 25 08:16:34 2011 (r2654) +++ cpu/arm/olpc/1.75/fw-version.fth Wed Oct 26 08:40:16 2011 (r2655) @@ -1,3 +1,3 @@ \ The overall firmware revision macro: FW_MAJOR C -macro: FW_MINOR 01 +macro: FW_MINOR 02

1 0

[commit] r2654 - dev/olpc/kb3700
by repository service Oct. 24, 2011

Oct. 24, 2011

Author: quozl Date: Tue Oct 25 08:16:34 2011 New Revision: 2654 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2654 Log: OLPC XO-1.75, XO-1.5, XO-1 - battery state of charge did not properly show 100%. The cause was that a previous definition of 2.d was being used, which used (.2), truncating any higher digits. This wasn't a problem until the XO-1.75 EC started using 100% instead of 97% in commit ea52c0c. Batteries moved from XO-1.75 to XO-1.5 and XO-1 may show 00% in watch-battery or .bat until this fix is released. Modified: dev/olpc/kb3700/batstat.fth Modified: dev/olpc/kb3700/batstat.fth ============================================================================== --- dev/olpc/kb3700/batstat.fth Tue Oct 25 00:57:39 2011 (r2653) +++ dev/olpc/kb3700/batstat.fth Tue Oct 25 08:16:34 2011 (r2654) @@ -79,8 +79,7 @@ if ." MPPT" then ; -\needs 2.d : 2.d ( n -- ) push-decimal <# u# u#s u#> type pop-base ; -: .% ( n -- ) 2.d ." %" ; +: .% ( n -- ) push-decimal <# [char] % hold u# u#s u#> type pop-base ; : .bat ( -- ) bat-status@ ( stat ) ." AC:" dup h# 10 and if ." on " else ." off " then ( stat )

1 0

[commit] r2653 - cpu/x86/pc/alex
by repository service Oct. 24, 2011

Oct. 24, 2011

Author: lwalter Date: Tue Oct 25 00:57:39 2011 New Revision: 2653 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2653 Log: Include /supplicant package for wifi. Modified: cpu/x86/pc/alex/fw.bth Modified: cpu/x86/pc/alex/fw.bth ============================================================================== --- cpu/x86/pc/alex/fw.bth Tue Oct 25 00:56:15 2011 (r2652) +++ cpu/x86/pc/alex/fw.bth Tue Oct 25 00:57:39 2011 (r2653) @@ -174,10 +174,22 @@ true ' local-mac-address? set-config-int-default fload ${BP}/ofw/inet/loadtcp.fth +[ifdef] resident-packages +support-package: nfs + fload ${BP}/ofw/fs/nfs/loadpkg.fth +end-support-package +[then] +devalias nfs net//obp-tftp:last//nfs + support-package: http fload ${BP}/ofw/inet/http.fth \ HTTP client end-support-package +fload ${BP}/ofw/wifi/wifi-cfg.fth +support-package: supplicant +fload ${BP}/ofw/wifi/loadpkg.fth +end-support-package + fload ${BP}/cpu/x86/pc/boot.fth fload ${BP}/cpu/x86/pc/linux.fth @@ -251,7 +263,7 @@ " probe-" do-drop-in probe-pci report-pci-fb -\ silent-probe-usb + silent-probe-usb ; \ This reduces processor use when waiting for a key. It helps

1 0

[commit] r2652 - ofw/wifi
by repository service Oct. 24, 2011

Oct. 24, 2011

Author: lwalter Date: Tue Oct 25 00:56:15 2011 New Revision: 2652 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2652 Log: Make eapol key handshake more robust and fix WPA/WPA2 auto cipher bug Modified: ofw/wifi/data.fth ofw/wifi/eapol.fth Modified: ofw/wifi/data.fth ============================================================================== --- ofw/wifi/data.fth Mon Oct 24 08:53:41 2011 (r2651) +++ ofw/wifi/data.fth Tue Oct 25 00:56:15 2011 (r2652) @@ -27,6 +27,8 @@ : set-key-type ( ctp ctg ktype -- ) " set-key-type" $call-parent ; : set-country-info ( adr len -- ) " set-country-info" $call-parent ; : set-atim-window ( n -- ) " set-atim-window" $call-parent ; +: set-gtk-idx ( n -- ) " set-gtk-idx" ['] $call-parent catch if 3drop then ; +: disconnected? ( -- flag ) " disconnected?" $call-parent ; \ ======================================================================= \ Global data Modified: ofw/wifi/eapol.fth ============================================================================== --- ofw/wifi/eapol.fth Mon Oct 24 08:53:41 2011 (r2651) +++ ofw/wifi/eapol.fth Tue Oct 25 00:56:15 2011 (r2652) @@ -9,9 +9,38 @@ : vtype ( adr len -- ) debug? if ??cr type cr else 2drop then ; \ ======================================================================= -\ EAPOL data definitions +\ Handshake variables and helper words + +false value send-error? +2 constant send-retry-cnt + +d# 30,000 constant timeout-eapol-limit +d# 5,000 constant timeout-msg-limit + +0 value timeout-eapol \ Global timeout value in ms +0 value timeout-msg \ Timeout waiting for msg from AP in ms + +: set-eapol-timeout ( -- ) get-msecs timeout-eapol-limit + to timeout-eapol ; +: set-msg-timeout ( -- ) get-msecs timeout-msg-limit + to timeout-msg ; +: eapol-timeout? ( -- timeout? ) get-msecs timeout-eapol >= ; +: msg-timeout? ( -- timeout? ) get-msecs timeout-msg >= ; +: timeout? ( -- timeout? ) eapol-timeout? msg-timeout? or ; + +false value done-pairwise-key? +false value done-group-key? -d# 10,000 constant eapol-wait +: done? ( -- done? ) done-group-key? done-pairwise-key? and ; + +: send-eapol-msg ( adr len -- ok? ) + send-retry-cnt 0 do + 2dup tuck write-force = if 2drop true unloop exit then + loop 2drop + false " Fail to send message" vtype + true to send-error? +; + +\ ======================================================================= +\ EAPOL data definitions 1 value eapol-ver 3 constant eapol-key @@ -68,6 +97,13 @@ 1000 constant ki-encr-key \ >kdata encrypted (WPA2) 2000 constant ki-smk-msg \ STSL master key +\ Vendor IE's OUI +h# 0050f2.01 constant oui-wpa \ Microsoft OUI's +h# 0050f2.02 constant oui-wmm +h# 0050f2.04 constant oui-wps +h# 00904c.33 constant oui-ht \ Broadcom OUI's +h# 00904c.34 constant oui-ht+ + \ ======================================================================= \ Nonce @@ -179,14 +215,13 @@ \ ======================================================================= \ Group temporal keys (GTK) -false value done-group-key? - d# 32 dup constant /gtk buffer: gtk 0 value gtk-idx \ Active GTK idx d# 32 buffer: ek \ Temporary rc4 key : install-group-key ( -- ) gtk ctype-g ct>klen set-gtk ; +: install-gtk-idx ( -- ) ktype kt-wpa2 = if gtk-idx set-gtk-idx then ; : set-gtk-supplicant ( adr -- ) dup gtk d# 16 move @@ -252,12 +287,54 @@ then ; +\ ======================================================================= +\ WPA/WPA2 handshake states +\ State Comment Process +\ s1/4 initial state Reset wait for 1/4 timeout +\ authenticate done pairwise key received? = false +\ associate done group key received? = false +\ Wait for 1/4 msg from AP +\ 1/4 timeout -> exit fail +\ s3/4 got 1/4 msg Send 2/4 msg to AP +\ Reset 3/4 timeout +\ Wait for 3/4 msg from AP +\ 3/4 timeout -> s2 +\ got 3/4 msg -> send 4/4 msg to AP +\ If we got group key, exit ok +\ Reset 1/2 timeout +\ s1/2 pairwise key done Wait for 1/2 msg from AP +\ 1/2 timeout -> s2 +\ got 1/2 msg -> send 2/2 msg to AP +\ exit ok +\ exit ok Install keys +\ +\ Anytime a 1/4 msg is received, goto s3/4 +\ Anytime deauthenticate is received, exit fail +\ Global timeout, exit fail + +0 value eapol-state +0 constant s0/4 \ Initial state; also invalid msg +1 constant s1/4 \ Waiting for 1/4 from AP; also msg# +2 constant s3/4 \ Waiting for 3/4 from AP; also msg# +3 constant s1/2 \ Waiting for 1/2 from AP; also msg# + +: set-s/4 ( -- ) + false to done-group-key? + false to done-pairwise-key? + set-msg-timeout +; +: set-s0/4 ( -- ) s0/4 to eapol-state set-s/4 ; +: set-s1/4 ( -- ) s1/4 to eapol-state set-s/4 ; +: set-s3/4 ( -- ) s3/4 to eapol-state set-s/4 ; +: set-s1/2 ( -- ) + false to done-group-key? + s1/2 to eapol-state + set-msg-timeout +; \ ======================================================================= \ WPA/WPA2 pairwise keys four-way handshake - -false value pairwise-rekey? -false value group-rekey? +\ WPA group key 2-way handshake /buf buffer: data @@ -269,35 +346,9 @@ ktype kt-wpa2 = if ki-encr-key or ki-secure or then ; : pct>kinfo4 ( -- kinfo ) pct>kinfo2 ktype kt-wpa2 = if ki-secure or then ; -: wait-for-eapol-key ( -- flag ) - false eapol-wait 0 do - data /buf read-force 0> if - data >etype be-w@ eapol-type = \ EAPOL frame - data >ptype c@ eapol-key = and \ EAPOL-key - data >rcnt last-rcnt /rcnt comp and if \ A new eapol-key record - data >rcnt last-rcnt! \ Update last replay counter - drop true - leave - then - then - 1 ms - loop -; - -: process-1-of-4 ( -- ok? ) - " Process EAPOL-key message 1 of 4" vtype - data >ver c@ to eapol-ver - data >dtype c@ kt>dtype <> if false exit then \ Descriptor type mismatch - data >kinfo be-w@ pct>kinfo1 <> if false exit then \ Bad key info - data >klen be-w@ ctype-p ct>klen <> if false exit then \ Bad key len - data >knonce anonce$ move \ Save authenticator nonce - true -; - -: wait-for-1-of-4 ( -- ok? ) - " Waiting for EAPOL-key message 1 of 4..." vtype - wait-for-eapol-key if process-1-of-4 else false then -; +: gct>kinfo ( -- kinfo ) ctype-p ct-aes = if ki-sha1-aes else ki-md5-rc4 then ; +: gct>kinfo1 ( -- kinfo ) gct>kinfo ki-mic or ki-secure or ki-ack or ; +: gct>kinfo2 ( -- kinfo ) gct>kinfo ki-mic or ki-secure or ; : set-eapolbuf-common ( -- ) eapolbuf /eapol-key erase @@ -310,132 +361,168 @@ ctype-p ct>klen eapolbuf >klen be-w! \ Key length last-rcnt eapolbuf >rcnt /rcnt move \ Replay counter ; -: send-2-of-4 ( -- error? ) + +: resync-gtk ( -- ) + last-rcnt++ + set-eapolbuf-common + 0 eapolbuf >klen be-w! \ Key length + /eapol-body eapolbuf >plen be-w! \ Packet length + ctype-g ct>klen eapolbuf >klen be-w! \ Key length + gct>kinfo ki-request or eapolbuf >kinfo be-w! \ Key info + eapolbuf /eapol-key send-eapol-msg +; + +: send-2/4 ( -- ok? ) " Send EAPOL-key message 2 of 4" vtype compute-snonce compute-ptk-supplicant set-eapolbuf-common - /eapol-body wpa-ie$ nip + eapolbuf >plen be-w! \ Packet length - pct>kinfo2 eapolbuf >kinfo be-w! \ Key info - snonce eapolbuf >knonce /nonce move \ Supplicant nonce - wpa-ie$ nip eapolbuf >kdlen be-w! \ Key data length - wpa-ie$ eapolbuf >kdata swap move \ Key data is WPA IE (or RSN IE) + /eapol-body wpa-ie$ nip + eapolbuf >plen be-w! \ Packet length + pct>kinfo2 eapolbuf >kinfo be-w! \ Key info + snonce eapolbuf >knonce /nonce move \ Supplicant nonce + wpa-ie$ nip eapolbuf >kdlen be-w! \ Key data length + wpa-ie$ eapolbuf >kdata swap move \ Key data is WPA IE (or RSN IE) eapolbuf >ver eapolbuf >plen be-w@ 4 + ctype-p compute-mic - mic eapolbuf >kmic /mic move \ Key mic - eapolbuf /eapol-key wpa-ie$ nip + tuck write-force <> + mic eapolbuf >kmic /mic move \ Key mic + eapolbuf /eapol-key wpa-ie$ nip + send-eapol-msg ; - -: process-3-of-4 ( -- ok? ) - " Process EAPOL-key message 3 of 4" vtype - data >dtype c@ kt>dtype <> if false exit then \ Descriptor type mismatch - data >kinfo be-w@ pct>kinfo3 <> if false exit then \ Bad key info - data >klen be-w@ ctype-p ct>klen <> if false exit then \ Bad key len - data >knonce anonce /nonce comp if false exit then \ Nonce differ - data ctype-p mic-ok? 0= if true to pairwise-rekey? false exit then \ Bad mic - data >kinfo be-w@ ki-encr-key and if data decrypt-parse-ie else true then -; -: wait-for-3-of-4 ( -- ok? ) - " Waiting for EAPOL-key message 3 of 4..." vtype - wait-for-eapol-key if process-3-of-4 else false then -; - -: send-4-of-4 ( -- error? ) +: send-4/4 ( -- ok? ) " Send EAPOL-key message 4 of 4" vtype set-eapolbuf-common - /eapol-body eapolbuf >plen be-w! \ Packet length - pct>kinfo4 eapolbuf >kinfo be-w! \ Key info + /eapol-body eapolbuf >plen be-w! \ Packet length + pct>kinfo4 eapolbuf >kinfo be-w! \ Key info eapolbuf >ver eapolbuf >plen be-w@ 4 + ctype-p compute-mic - mic eapolbuf >kmic /mic move \ Key mic - eapolbuf /eapol-key tuck write-force <> + mic eapolbuf >kmic /mic move \ Key mic + eapolbuf /eapol-key send-eapol-msg ; - -: do-pairwise-key-handshake ( -- ok? ) - false - wait-for-1-of-4 0= if ." Failed to get the first pairwise key" cr exit then - send-2-of-4 if ." Failed to send the second pairwise key" cr exit then - wait-for-3-of-4 0= if ." Failed to get the third pairwise key" cr exit then - send-4-of-4 if ." Failed to send the fourth pairwise key" cr exit then - install-pairwise-key - done-group-key? if install-group-key then - drop true +: send-2/2 ( -- ok? ) + " Send EAPOL-key message 2 of 2" vtype + set-eapolbuf-common + /eapol-body eapolbuf >plen be-w! \ Packet length + gct>kinfo2 eapolbuf >kinfo be-w! \ Key info + ctype-g ct>klen eapolbuf >klen be-w! \ Key length + eapolbuf >ver eapolbuf >plen be-w@ 4 + ctype-p compute-mic + mic eapolbuf >kmic /mic move \ Key mic + eapolbuf /eapol-key send-eapol-msg +; +: 1/4-valid? ( -- ok? ) + data >dtype c@ kt>dtype <> if false exit then \ Descriptor type mismatch + data >klen be-w@ ctype-p ct>klen <> if false exit then \ Bad key len + true ; - -\ ======================================================================= -\ WPA group keys 2-way handshake - -: gct>kinfo ( -- kinfo ) ctype-g ct-aes = if ki-sha1-aes else ki-md5-rc4 then ; -: gct>kinfo1 ( -- kinfo ) gct>kinfo ki-mic or ki-secure or ki-ack or ; -: gct>kinfo2 ( -- kinfo ) gct>kinfo ki-mic or ki-secure or ; - -: process-1-of-2 ( -- ok? ) - " Process EAPOL-key message 1 of 2" vtype - data >dtype c@ kt>dtype <> if false exit then \ Descriptor type mismatch data >kinfo be-w@ ki-idx-mask invert and gct>kinfo1 <> if false exit then \ Bad key info - data >klen be-w@ ctype-g ct>klen <> if false exit then \ Bad key len - data ctype-p mic-ok? 0= if true to group-rekey? false exit then \ Bad mic - data decrypt-gtk +: process-1/4 ( -- ) + " Process EAPOL-key message 1 of 4" vtype + eapol-state s1/4 <> if " Unexpected pairwise key message 1 of 4" vtype then + 1/4-valid? if + data >ver c@ to eapol-ver + data >knonce anonce$ move + send-2/4 if set-s3/4 then + then ; -: wait-for-1-of-2 ( -- ok? ) - " Waiting for EAPOL-key message 1 of 2..." vtype - wait-for-eapol-key if process-1-of-2 else false then +: 3/4-valid? ( -- ok? ) + data >dtype c@ kt>dtype <> if false exit then \ Descriptor type mismatch + data >klen be-w@ ctype-p ct>klen <> if false exit then \ Bad key len + data >knonce anonce /nonce comp if false exit then \ Nonce differ + data ctype-p mic-ok? 0= if false exit then \ Bad mic + data >kinfo be-w@ ki-encr-key and if data decrypt-parse-ie else true then ; - -: send-2-of-2 ( -- error? ) - " Send EAPOL-key message 2 of 2" vtype - set-eapolbuf-common - /eapol-body eapolbuf >plen be-w! \ Packet length - gct>kinfo2 eapolbuf >kinfo be-w! \ Key info - eapolbuf >ver eapolbuf >plen be-w@ 4 + ctype-g compute-mic - mic eapolbuf >kmic /mic move \ Key mic - eapolbuf /eapol-key tuck write-force <> +: process-3/4 ( -- ) + " Process EAPOL-key message 3 of 4" vtype + eapol-state s3/4 = if + 3/4-valid? if + send-4/4 if + install-pairwise-key + true to done-pairwise-key? + " Install pairwise key" vtype + done-group-key? if + \ WPA2 3/4 contains the group key + " Install group key" vtype + install-gtk-idx + install-group-key + else + " Wait for group key message 1 of 2" vtype + set-s1/2 + then + then + then + else + " Unexpected pairwise key message 3 of 4" vtype + then ; - -: do-group-key-handshake ( -- ) - wait-for-1-of-2 0= if ." Failed to get the first group key" cr exit then - send-2-of-2 if ." Failed to send the second group key" cr exit then - install-group-key +: 1/2-valid? ( -- ok? ) + data >dtype c@ kt>dtype <> if false exit then \ Descriptor type mismatch + data >klen be-w@ ctype-g ct>klen <> if false exit then \ Bad key len + data ctype-p mic-ok? 0= if false exit then \ Bad MIC + data decrypt-gtk ; - -\ ======================================================================= -\ WPA/WPA2 keys handshake - -: request-rekey ( ki-pairwise -- ) - " Send rekey message" vtype - last-rcnt++ - set-eapolbuf-common - 0 eapolbuf >klen be-w! \ Key length - /eapol-body eapolbuf >plen be-w! \ Packet length - ( ki-pairwise ) dup gct>kinfo or ki-request or ki-error or ki-mic or - eapolbuf >kinfo be-w! \ Key info - eapolbuf >ver eapolbuf >plen be-w@ 4 + - rot ( ki-pairwise ) if ctype-p else ctype-g then compute-mic - mic eapolbuf >kmic /mic move \ Key mic - eapolbuf /eapol-key tuck write-force <> +: process-1/2 ( -- ) + eapol-state s1/2 = if + 1/2-valid? if + send-2/2 if + " Install group key" vtype + install-gtk-idx + install-group-key + true to done-group-key? + then + then + else + " Unexpected group key message 1 of 2" vtype + then ; - -: resync-gtk ( -- ) 0 request-rekey ; - -: (do-key-handshakes) ( -- ) - false to done-group-key? - false to pairwise-rekey? - do-pairwise-key-handshake if - done-group-key? 0= if - 2 0 do - false to group-rekey? - do-group-key-handshake - group-rekey? if resync-gtk else leave then - loop +: get-eapol-msg# ( -- state ) + data >kinfo be-w@ + dup ki-pairwise and if \ Pairwise key info + dup pct>kinfo1 = if drop s1/4 exit then + pct>kinfo3 = if s3/4 exit then + else + ki-idx-mask invert and gct>kinfo1 = if s1/2 exit then + then + s0/4 +; +: process-eapol-key ( -- ) + get-eapol-msg# case + s1/4 of process-1/4 endof + s3/4 of process-3/4 endof + s1/2 of process-1/2 endof + endcase +; +: (process-eapol) ( -- ) + data >etype be-w@ eapol-type = \ EAPOL frame + data >ptype c@ eapol-key = and if \ EAPOL-key + data >ver c@ to eapol-ver + data >rcnt last-rcnt /rcnt comp 0> if \ A new eapol-key record + data >rcnt /rcnt cdump cr + data >rcnt last-rcnt! \ Update last replay counter + process-eapol-key + else + " Same replay count; packet discarded" vtype then + else + " Non EAPOL frame received" vtype then ; - +: (do-key-handshakes) ( -- ) + begin + data /buf read-force 0> if (process-eapol) then + timeout? disconnected? or done? or send-error? or until +; : do-key-handshakes ( -- ) - 2 0 do + false to send-error? + set-eapol-timeout + set-s1/4 + (do-key-handshakes) + done? disconnected? or send-error? or if exit then + done-pairwise-key? if + " Timeout waiting for group key; request group key" vtype + resync-gtk + set-eapol-timeout + set-s1/2 (do-key-handshakes) - pairwise-rekey? if ki-pairwise request-rekey else leave then - loop + else + " Timeout waiting for 4-way handshake" vtype + then ; - \ ======================================================================= \ Scan dump @@ -447,6 +534,21 @@ loop cr ; +: .vendor ( adr -- ) + dup 2 + swap 1+ c@ ( adr' len ) + ?dup 0= if drop exit then + dup 4 < if ." Vendor Specific: " cdump cr exit then + over be-l@ case ( adr len ) + oui-wpa of ." WPA: " endof + oui-wmm of ." WMM: " endof + oui-wps of ." WPS: " endof + oui-ht of ." HT: " endof + oui-ht+ of ." More HT: " endof + ( otherwise ) ." Vendor Specific: " + endcase + cdump cr +; + : .ie ( adr -- ) dup c@ ." " case @@ -482,7 +584,7 @@ cr endof d# 48 of ." Robust security network: " dup 2 + swap 1+ c@ cdump cr endof d# 50 of ." Extended supported rates (Mbit/s): " .rates endof - d# 221 of ." Wi-Fi Protected Access: " dup 2 + swap 1+ c@ cdump cr endof + d# 221 of .vendor endof ( default ) ." Unknown IE type: " swap dup 1+ c@ 2 + cdump cr endcase ; @@ -573,14 +675,6 @@ dup le-w@ swap 2 + swap d# 19 /string ( adr' len' ) r> (find-ie) ( adr' len' true | false ) ; -: find-ie2 ( ie-adr,len adr ie-type -- adr len true | false ) - >r ( ie-adr,len adr ) ( R: ie-type ) - dup le-w@ swap 2 + swap d# 19 /string - ( ie-adr,len adr len ) ( R: ie-type ) - 3 pick rot - - ( ie-adr,len len' ) ( R: ie-type ) - swap /string ( adr' len' ) ( R: ie-type ) - r> (find-ie) ( adr' len' true | false ) -; : find-ssid ( ssid$ scanbuf-adr -- ap-adr' true | false ) false swap ( ssid$ flag adr ) dup 3 + ( ssid$ flag 'ap ) @@ -728,14 +822,19 @@ dup 0= if ." Keys in wifi-cfg are not valid - " then ; -h# 0050.f201 constant wpa-tag -: process-wpa-ie ( ie-adr,len adr -- ) - \ Some AP such as Linksys has a bogus WPA IE in addition to the real thing. - \ Skip over the bad one and see if there's another WPA IE - 2 pick be-l@ wpa-tag = if drop set-wpa-ktype exit then - d# 221 find-ie2 if - over be-l@ wpa-tag = if set-wpa-ktype else 2drop then - then +: (process-vendor-ie) ( adr -- ) + dup 2 + swap 1+ c@ ( adr' len ) + dup 4 > if + over be-l@ oui-wpa = if set-wpa-ktype exit then + then 2drop +; + +: process-vendor-ie ( adr -- ) + dup le-w@ swap 2 + swap d# 19 /string ( adr' len ) + begin dup 0> while ( adr len ) + over c@ h# dd = if over (process-vendor-ie) then + over 1+ c@ 2 + /string ( adr' len' ) + repeat 2drop ; : ssid-valid? ( adr -- flag ) @@ -754,7 +853,7 @@ dup 6 find-ie if drop 2 + le-w@ set-atim-window then \ ATIM window dup 7 find-ie 0= if null$ then do-set-country-info \ Country channel/power info dup d# 48 find-ie if set-wpa2-ktype drop key-ok? exit then \ Favor RSN(WPA2) over WPA - dup d# 221 find-ie if 2 pick process-wpa-ie then + dup process-vendor-ie \ E.g. WPA drop key-ok? ; @@ -764,7 +863,7 @@ cr ktype=wpa? if do-key-handshakes - done-group-key? + done? else true then @@ -831,15 +930,8 @@ " Process EAPOL message" vtype 2dup vdump data swap move - data >ptype c@ eapol-key = and if \ EAPOL-key - data >rcnt last-rcnt /rcnt comp if \ A new eapol-key record - " Got EAPOL-key message 1 of 2" vtype - data >rcnt last-rcnt! \ Update last replay counter - process-1-of-2 if - send-2-of-2 0= if install-group-key then - then - then - then + set-s1/2 + (process-eapol) ;

1 0

[commit] r2651 - cpu/x86/pc/olpc/via
by repository service Oct. 23, 2011

Oct. 23, 2011

Author: quozl Date: Mon Oct 24 08:53:41 2011 New Revision: 2651 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2651 Log: Q3B22 (note that this version was not svn head at this point, but rather svn 2478 [q3b19] with 2589 and 2597 [q3b20]) (that is, q3b21 with another EC version) Modified: cpu/x86/pc/olpc/via/ec-version.fth cpu/x86/pc/olpc/via/fw-version.fth Modified: cpu/x86/pc/olpc/via/ec-version.fth ============================================================================== --- cpu/x86/pc/olpc/via/ec-version.fth Mon Oct 24 04:51:51 2011 (r2650) +++ cpu/x86/pc/olpc/via/ec-version.fth Mon Oct 24 08:53:41 2011 (r2651) @@ -1,5 +1,5 @@ \ The EC microcode -macro: EC_VERSION 2_2_9 +macro: EC_VERSION 2_2_10 \ Alternate command for getting EC microcode, for testing new versions. \ Temporarily uncomment the line and modify the path as necessary Modified: cpu/x86/pc/olpc/via/fw-version.fth ============================================================================== --- cpu/x86/pc/olpc/via/fw-version.fth Mon Oct 24 04:51:51 2011 (r2650) +++ cpu/x86/pc/olpc/via/fw-version.fth Mon Oct 24 08:53:41 2011 (r2651) @@ -1,3 +1,3 @@ \ The overall firmware revision macro: FW_MAJOR B -macro: FW_MINOR 21 +macro: FW_MINOR 22

1 0

[commit] r2650 - cpu/arm/olpc/1.75
by repository service Oct. 23, 2011

Oct. 23, 2011

Author: wmb Date: Mon Oct 24 04:51:51 2011 New Revision: 2650 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2650 Log: Q4C01 Modified: cpu/arm/olpc/1.75/fw-version.fth Modified: cpu/arm/olpc/1.75/fw-version.fth ============================================================================== --- cpu/arm/olpc/1.75/fw-version.fth Mon Oct 24 04:51:41 2011 (r2649) +++ cpu/arm/olpc/1.75/fw-version.fth Mon Oct 24 04:51:51 2011 (r2650) @@ -1,3 +1,3 @@ \ The overall firmware revision -macro: FW_MAJOR B -macro: FW_MINOR 12mk +macro: FW_MAJOR C +macro: FW_MINOR 01

1 0

[commit] r2649 - cpu/arm/olpc/1.75
by repository service Oct. 23, 2011

Oct. 23, 2011

Author: wmb Date: Mon Oct 24 04:51:41 2011 New Revision: 2649 URL: http://tracker.coreboot.org/trac/openfirmware/changeset/2649 Log: OLPC XO-1.75 - Include EC code version 4_0_3_01 Modified: cpu/arm/olpc/1.75/ec-version.fth Modified: cpu/arm/olpc/1.75/ec-version.fth ============================================================================== --- cpu/arm/olpc/1.75/ec-version.fth Mon Oct 24 04:51:29 2011 (r2648) +++ cpu/arm/olpc/1.75/ec-version.fth Mon Oct 24 04:51:41 2011 (r2649) @@ -1,5 +1,5 @@ \ The EC microcode -macro: EC_VERSION 4_0_3_00 +macro: EC_VERSION 4_0_3_01 \ Alternate command for getting EC microcode, for testing new versions. \ Temporarily uncomment the line and modify the path as necessary

1 0