summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* futility: updater: Add more legacy optionsfactory-nocturne-11066.BHung-Te Lin2019-04-221-0/+12
| | | | | | | | | | | | | | | | | | | To simplify the migration plan, we want to support the legacy arguments used by FAFT and factory test "UpdateFirmware": --noupdate_ec => --host_only --noupdate_pd => --host_only --nocheck_keys => --force --update_{main,ec,pd} => ignore BUG=chromium:882445,b:118509893,b:130401936,chromium:953993,b:130115628 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I31652806085937fe5ca2f2facc7321021977cbb7 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1310253 Reviewed-by: Randall Spangler <rspangler@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/vboot_reference/+/1577344
* futility: updater: Correct HWID digest when preserving HWIDHung-Te Lin2019-01-164-13/+41
| | | | | | | | | | | | | | | | | | | | | | | | | | Starting from GBB 1.2, a digest is stored in GBB and must be updated whenever the HWID string is changed. In shell script version of updater, the digest is automatically updated when we do "futility gbb -s --hwid=XXX", but in native updater implementation we only updated the HWID string and left digest unchanged, this leaves devices generating wrong PCR1 values. `cmd_gbb_utility` updates the digest by calling `update_hwid_digest` using vboot1 structure, so we should introduce a new vboot2 friendly function, `vb2_change_hwid`, which changes both HWID string and digest at same time. Note this has no impact for end user's devices with write protection enabled. Only changes dogfood units AU results. BUG=b:122248649 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=none Change-Id: I6ad2754e6df3c9dd66d71c560a2afc26d14eae33 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/c/1411941
* futility: Add firmware updater as 'futility update'Hung-Te Lin2018-11-0511-1871/+4060
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The factory branches may need to run firmware updater extracted from newer release images which relies on 'futility update' to run. This change collects all related changes: cabd2f8c futility: update: Fix update failure in factory mode due to wrong WP logic ac8cb8dd futility: update: Add legacy option '--factory' 79bdb0cc futility: Remove FUTIL_STATIC build rules d6f7c16b futility: update: Skip TPM check if tpm_fwver is not valid. f6198074 futility: update: Preserve new sections: RO_PRESERVE and RW_PRESERVE f67b991a futility: update: Preserve RO_FSG ee96e741 futility: update: Preserve RW_ELOG b07165b5 futility: update: Allow tpm_fwver=0 and allow --force to waive TPM check 5a5be0c7 futility: update: Revise error messages ba73f35e futility: update: Refactor 'emulation' (--emulate). eb166257 futility: update: Add --programmer to override flashrom programmer for servo 5fa5e3b9 futility: update: Refactor: move updater logic to 'updater.c' 243e2115 futility: updater: Refactor: move command line processing to cmd_update.c e7bcda54 futility: updater: Refactor: localize temp files management 8810b089 futility: updater: Refactor: move quirks to 'updater_quirks.c' a1bfb650 futility: updater: Add default quirks a2b9e356 futility: updater: Add quirk 'daisy_snow_dual_model' for daisy_snow ba30c93b futility: updater: Preserve SMMSTORE and add quirk 'eve_smm_store' 38d06701 futility: updater: Support reading main image from stdin f1609b8c futility: updater: Revise verbosity and error messages fabf760f futility: updater: Add ASPRINTF macro 7aa34302 futility: updater: Refactor function names 077026fd futility: updater: Add '--archive' to read from an archive or directory f7c44a32 futility: updater: Refactor how arguments were configured 1847ba4f futility: updater: Add --manifest to scan and print archive info a71e0ccd futility: updater: Allow patching rootkey and vblock files 2ae23df7 futility: updater: Change default model name to FWID platform name 3d615c91 futility: updater: Do not preserve GBB flags in --factory mode. 10b3d4ed futility: updater: Add '--model' and select images by system model 622c35aa futility: updater: Support white label from VPD and --signature_id c6781206 futility: updater: Support --mode=output and --output_dir fb947ae0 futility: update: Add `--host_only` argument 67d66f18 futility: updater: Support --repack and --unpack 1f489393 futility: updater: Fix model detection error BUG=chromium:882445 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Signed-off-by: Hung-Te Lin <hungte@chromium.org> Change-Id: I6c14133f59561a422a190b17fa8ff7e79c45d953 Reviewed-on: https://chromium-review.googlesource.com/c/1317057
* crossystem: fix field alignmentMike Frysinger2018-09-131-3/+6
| | | | | | | | | | | | | | clear_tpm_owner_request is 23 chars now. BUG=None TEST=`crossystem` is aligned BRANCH=None Change-Id: I6d077b7311c74c51fd608281ad48b29fc6219937 Reviewed-on: https://chromium-review.googlesource.com/1218502 Commit-Ready: Mike Frysinger <vapier@chromium.org> Tested-by: Mike Frysinger <vapier@chromium.org> Reviewed-by: Randall Spangler <rspangler@chromium.org>
* crossystem: resort param listMike Frysinger2018-09-131-13/+13
| | | | | | | | | | | | | | A bunch of these fields are slightly missorted. BUG=None TEST=`crossystem` is sorted BRANCH=None Change-Id: I9e90343f5034e7a8a2d81c9b8eeb4b1d7286f157 Reviewed-on: https://chromium-review.googlesource.com/1218503 Commit-Ready: Mike Frysinger <vapier@chromium.org> Tested-by: Mike Frysinger <vapier@chromium.org> Reviewed-by: Randall Spangler <rspangler@chromium.org>
* accessory: script to generically generate accessory keysstabilize-jetstream-11056.BNick Sanders2018-09-106-81/+1
| | | | | | | | | | | | | | | | | | | All accessories leverage the key format of Hammer therefore this script calls Hammer's one to generate a key pair and renames them. The key name isn't referenced by the signer anymore, so we will name them all "hammerlike". BUG=chromium:859269 TEST=Run this script in the chroot. BRANCH=None Change-Id: Iba35b03e59216e96a99f8aa471b660f3805c1f23 Reviewed-on: https://chromium-review.googlesource.com/1205636 Commit-Ready: Nick Sanders <nsanders@chromium.org> Tested-by: Nick Sanders <nsanders@chromium.org> Reviewed-by: Marco Chen <marcochen@chromium.org> Reviewed-by: Nicolas Boichat <drinkcat@chromium.org> Reviewed-by: Mike Frysinger <vapier@chromium.org>
* futility: cmd_update: Add quirk 'min_platform_version'Hung-Te Lin2018-09-102-1/+60
| | | | | | | | | | | | | | | | | | | | Many device may have some minor difference in early builds, for example (board id) rev 0 and rev 1 may have GPIO pins connected to different components. Usually the firmware should read board identifier and do the right mapping, but sometimes the firmware may be totally incompatible and no way to workaround (for example even the CPU may be different). The min_platform_version is introduced so we can prevent updating to incompatible systems, by reading $(mosys platform version). BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I418fee1aad884551b38ac25c340b2797b8503596 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1198815 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Add quirk 'unlock_me_for_update'Hung-Te Lin2018-09-102-13/+65
| | | | | | | | | | | | | | | | | | | | On recent Intel platforms, flashing to SI_ME may get corrupted due to ME execution in parallel. If we lock SI_ME immediately (by writing the new SI_DESC), the device may fail to boot due to ME execution failure. As a result, a quirk is added so the firmware updater will never lock SI_ME. The Flash Master values are always unlocked when updating SI_ME, and after system reboot, a board-postinst script should check ME status and reflash SI_DESC only if SI_ME looks all good. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I584aa373797e2b4c2608f07aac21c16cdb34a5c4 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1198807 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Add quirk 'enlarge_image'Hung-Te Lin2018-09-102-7/+86
| | | | | | | | | | | | | | | | | Some devices may have shipped with a smaller image that the real flash may be larger, especially if the device's original flash has been EOL'ed. The quirk 'enlarge_image' allows changing image size according to current_image size by padding 0xFF (flash default value). BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I84373cfa9bcbd98a2cd96a7dd4bed27a6f724cf3 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1198806 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Add --quirks for device-specific hacksHung-Te Lin2018-09-102-4/+125
| | | | | | | | | | | | | | | | | | | | | The firmware updater usually needs to apply many special rules for particular device, previously done by the 'updater_custom.sh' script. In futility updater, we want to support that by a 'quirks' system, that the updater package can declare a list of needed quirks and send to updater as `futility update --quirks LIST`. Currently only a dummy "test" quirks is defined. The real quirks will be added in follow up changes. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: Ic935d69a54473f2347964e7c161ffcdc0af43ec6 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1198804 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Create and remove temporary files properlyHung-Te Lin2018-09-101-5/+64
| | | | | | | | | | | | | | | | | When running on DUT we should create temporary files using system calls instead of using hard-coded path and file name. The new create_temp_file() will collect all temporary files and remove them all when the remove_temp_files() is invoked (usually at end of program). BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I866dd8dfe7acbf8c5a586249ea2d19f33891672d Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1203334 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Add new 'legacy' modeHung-Te Lin2018-09-102-2/+33
| | | | | | | | | | | | | | | | | | | | | For devices that do not have update tag provisioned in legacy CBFS, we need a way to push and enforce the updater to complete first migration. The '--mode=legacy' provides a short cut to do "flashrom -p host -w image -i RW_LEGACY" Devices that need newer (or latest) legacy firmware should invoke firmware updater in their initialization or setup process, to enforce updating RW_LEGACY. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I87db067ad134e82bbbdc937bd2880c6731ec892b Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1198808 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Correctly handle error counterHung-Te Lin2018-09-071-4/+4
| | | | | | | | | | | | | | | | The errorcnt in do_update should not add function return values directly because the function may return negative values, which would lead to wrong results. Instead we can process using '!!' so the returned value will always be zero or positive integers. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I9f450b2ee8d86035288f06bdb314dfb1dce3ca64 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1209023 Reviewed-by: Joel Kitching <kitching@chromium.org>
* validate_rec_mrc: Check size before reading metadataFurquan Shaikh2018-09-071-0/+6
| | | | | | | | | | | | | | | | This change ensures that the slot_len is enough to hold the metadata before attempting to actually read it. BUG=None BRANCH=None TEST=None Change-Id: Ief50edccc7a73c2c3002f772c616e4ed5557ecbd Signed-off-by: Furquan Shaikh <furquan@google.com> Reviewed-on: https://chromium-review.googlesource.com/1212528 Commit-Ready: Furquan Shaikh <furquan@chromium.org> Tested-by: Furquan Shaikh <furquan@chromium.org> Reviewed-by: Hung-Te Lin <hungte@chromium.org>
* futility: drop static buildMike Frysinger2018-09-071-10/+3
| | | | | | | | | | | | | | We shouldn't need this anywhere anymore, so drop it. BUG=chromium:765499 TEST=precq passes BRANCH=none Change-Id: I0f1adb2bf120e1a20c79d2641a0d2fe96d8e6908 Reviewed-on: https://chromium-review.googlesource.com/1210342 Commit-Ready: Mike Frysinger <vapier@chromium.org> Tested-by: Mike Frysinger <vapier@chromium.org> Reviewed-by: Hung-Te Lin <hungte@chromium.org>
* futility: cmd_update: Revise debug and error message macroHung-Te Lin2018-09-051-75/+63
| | | | | | | | | | | | | | | | | To simplify debugging with logs, we want to print "current function" in almost every debug and error messages. To prevent typing __FUNCTION__ everywhere, two new macros ERROR and DEBUG are introduced, to print messages in ERROR: <function>: <message>\n and DEBUG: <function>: <message>\n . BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: If680f436042d58a32e2b4f534d4c22958f68d021 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1197023 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Preserve sections blocked by management engineHung-Te Lin2018-09-051-1/+44
| | | | | | | | | | | | | | | | | | | | | On Intel platforms, when management engine is not in manufacturing mode the SI_DESC will be read only and SI_ME can't be accessed by CPU. For RW-FULL mode when we want to reflash whole firmware, flashrom will skip SI_ME but it'll still try to update SI_DESC if the content looks different, which would make all firmware update to fail when a new image is pushed with different SI_DESC contents. As a result, we have to preserve and skip SI_DESC changes when ME is locked. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I834405ad519dcb7ccd44073addfd63e844b74168 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1197022 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* firmware/lib/vboot_ui: Add a VbExSleepMs to the developer mode UIsRaul E Rangel2018-09-052-0/+9
| | | | | | | | | | | | | | | This won't have any real power savings until the APIC delay is merged and enabled. BUG=b:109749762 BRANCH=none TEST=Booted grunt and made sure the developer screens still worked. Change-Id: I7d75198771946415fa6a8fa69dff024d87ba5ef0 Signed-off-by: Raul E Rangel <rrangel@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1182190 Commit-Ready: ChromeOS CL Exonerator Bot <chromiumos-cl-exonerator@appspot.gserviceaccount.com> Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Correct updating logic for RW_LEGACYHung-Te Lin2018-09-051-3/+60
| | | | | | | | | | | | | | | The RW_LEGACY logic has been changed recently and need cbfstool to help identifying if update can be performed silently. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: Ida38bb8886b17c2f7bbb2c14d072508d4b9c5809 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1194821 Reviewed-by: Randall Spangler <rspangler@chromium.org> Reviewed-by: Joel Kitching <kitching@chromium.org>
* futility: cmd_update: Add more legacy updater optionsHung-Te Lin2018-09-051-1/+31
| | | | | | | | | | | | | | | | | | To make it easier for integration with legacy firmware updater, we want to add few options: -m, --mode MODE -d, --debug -v, --verbose BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I6f045db0a8e9b5c73c1f0be2b52a71a7ee2a495e Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1193043 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Check RO and RW signing compatibility before updatingHung-Te Lin2018-09-052-0/+145
| | | | | | | | | | | | | | | | If the RW is not signed by RO root key then verified boot will fail and enter recovery mode. This may happen when user is trying to flash a DEV (or PreMP) signed firmware on a MP-signed device, with write protection enabled. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I1cee0b5f42f1f403d9baa5f9b2659f75511fbcb8 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183659 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Check TPM key versions before updatingHung-Te Lin2018-09-052-11/+150
| | | | | | | | | | | | | | | | | | | In verified boot, the key versions stored in TPM will be checked before being able to load and run a signed RW firmware. This is also known as anti-rollback check. To prevent user installing an incompatible RO (even RW) and then being not able to boot, we should check TPM key versions (by `tpm_fwver` system property) before starting to update. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I6d50a6e475001d76fbcbe680a3f8b10f62354096 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1189249 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Check platform compatibility before updatingHung-Te Lin2018-09-052-2/+70
| | | | | | | | | | | | | | | | | | A safety check so people won't accidentally flashed wrong firmware image and then being not able to boot. The platform is decided by extracting the first component (delimited by dot '.') of firmware ID. For example, platform for "Google_Link.123" is "Google_Link". BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I90a1631f6b3e9a675fe1990cf9c204d763faf54c Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1189248 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Add vboot1 updater logicHung-Te Lin2018-09-052-12/+47
| | | | | | | | | | | | | | | | | | There are still many devices running vboot1 and we need to support them as well. There is no way to determine if a firmware is vboot2 or not, so we can only rely on the system property "fw_vboot2". If fw_vboot2 is 0, then we should always update section B and compare content with section A. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: Iefdcb81099914c2183c627a33eb73678d1269bc1 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1184952 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Use real system write protection statusHung-Te Lin2018-09-051-7/+97
| | | | | | | | | | | | | | | | | | | | | | | The updater logic is heavily based on write protection status. The write protection must be decided by two sources: hardware ("write protection switch", known as `wpsw` in crossystem) and software (on most SPI, this is controlled by SRP0 register using flashrom). When debugging firmware updating issues, it is very important to have complete logs for status of all WP sources (hw and sw, and the final decision by updater itself - maybe overridden by --wp). This change tries to handle WP properly and also leaving enough information of how the WP logic was decided. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I15dc2dbcefc421c1194aa623e15f00d793653e93 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183658 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Check contents before starting to updateHung-Te Lin2018-09-052-8/+86
| | | | | | | | | | | | | | | | | | In try-boot updating flow, we have to first check if RO content needs to be changed or not, and do full (ro+ro) update if WP is disabled. Also, before starting to update RW-A or RW-B, we should also check if the active system already has same firmware contents. An --force is also added to allow skipping the check. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: Ie2f75f9aab4696c75aedafbf45e418ee98a2a4b4 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183654 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* add extern C markers to installed headersMike Frysinger2018-09-0415-0/+121
| | | | | | | | | | | | | | | Some of the headers have extern C markings already, so add to the rest of the installed files so users don't have to. BUG=chromium:878440 TEST=build passes BRANCH=none Change-Id: I3edf56ca2235269803049207806a9f7eb4c664f2 Reviewed-on: https://chromium-review.googlesource.com/1201042 Commit-Ready: Mike Frysinger <vapier@chromium.org> Tested-by: Mike Frysinger <vapier@chromium.org> Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Preserve image sections before updateHung-Te Lin2018-09-012-0/+115
| | | | | | | | | | | | | | | | | When updating RO (or going to compare with contents), we need to preserve (copy) section data from system active firmware image. The `preserve_images` will try to preserve a list of known sections in full update (`--wp=0`) mode, so we VPD data and HWID won't be lost. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh BRANCH=None Change-Id: I85c4ba972853dbc0fc101bee269c0effe70988b1 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183653 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Implement updater logic "TRY-RW" (--try)Hung-Te Lin2018-08-312-3/+129
| | | | | | | | | | | | | | | | | | | | In vboot2, to try one RW (unused) section on next boot, we have to: - Find mainfw_act - Select and update to the "other" slot - Set system property fw_try_{next,count} values to try in next boot. The new '--try' (-t) option can trigger the mode if available. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility futility update --emulation FILE -i IMAGE -t --sys_prop 0; # Updates to B. futility update --emulation FILE -i IMAGE -t --sys_prop 1; # Updates to A. BRANCH=None Change-Id: I4b4662616a7181d2f37307238b7b80ae82369768 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1188017 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Add 'mainfw_act' system propertyHung-Te Lin2018-08-312-2/+36
| | | | | | | | | | | | | | | | | | | Add the system property 'mainfw_act'. In both vboot1 and vboot2, the try-rw update process will need to figure out what is current (active) firmware slot, which is the "mainfw_act" system property. BUG=chromium:875551 TEST=make futil; tests/futility/run_test_scripts.sh $(pwd)/build/futility futility --debug update -i IMAGE --sys_prop 0; futility --debug update -i IMAGE --sys_prop 1; BRANCH=None Change-Id: Ie745726107bff416549ba095a3defdd4cc98d32d Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183652 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Implement updater logic "RW UPDATE".Hung-Te Lin2018-08-312-6/+109
| | | | | | | | | | | | | | | | | | The logic is same as --mode=recovery,--wp=1 in legacy firmware updater. An debugging option '--wp' is introduced so user can easily switch between FULL UPDATE (--wp=0) or RW UPDATE (--wp=1). BUG=chromium:875551 TEST=make futil; futility update -i IMAGE --wp=0; futility update -i IMAGE --wp=1; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: Ic7d8aa8b327296988ebf80a8e737e8893b7870ea Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1188016 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Add "system property" and '--sys_props' to overrideHung-Te Lin2018-08-312-0/+146
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When updating firmware, we may need to get some system environment status, like which firmware slot was selected and boot (active), or setting cookies so the next boot will be using right (updated) slot. In verified boot, these status are manipulated by "system property" using API Vb{Get,Set}SystemProperty{String,Int}. The user land tool is `crossystem`. In order to run the firmware updater for testing and debugging, we need an easy way to toggle getting real system status, or fetch from predefined values. A new 'system_property' structure is introduced and included as part of `updater_config`. Each property can be access by `get_system_property(property_type)` function. If the value was not fetched yet, the function will call corresponding 'getter' function defined in property and then cache it. A new parameter '--sys_props` is also introduced so we can easily override them from command line so the updater will not get status from running system. The --sys_props takes a list of integers, eliminated by space or comma. For example, "1,2,3" => overrides [0]=1, [1]=2, [2]=3. "1 2,3" => overrides [0]=1, [1]=2, [2]=3. "1, ,3" => overrides [0]=1, [2]=3. BUG=chromium:875551 TEST=make futil; futility update -i IMAGE tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: Ia2e06a953da1480da9a94f7f397802caa7468efa Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1188015
* futility: cmd_update: Implement updater logic "FULL UPDATE".stabilize-atlas-11022.BHung-Te Lin2018-08-312-14/+179
| | | | | | | | | | | | | | | The logic is same as --mode=factory or --mode=recovery,--wp=0 in legacy firmware updater. BUG=chromium:875551 TEST=make futil; futility update -i IMAGE tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: Ifbfc4fb76f954483e779c8b508377d07561b67da Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183651 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Add '--emulate' optionHung-Te Lin2018-08-312-7/+62
| | | | | | | | | | | | | | | | | | To help debugging and testing, we may want to run updater against an image file instead of modifying real system firmware. The --emulate allows running with all checks and reading, and outputs to given file. BUG=chromium:875551 TEST=make futil; futility update -i IMAGE --emulate IMAGE2 tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: Ic52fe60a1468f29245cade70f859513d8d117c9c Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1184953 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Access system firmware using external flashromHung-Te Lin2018-08-312-4/+100
| | | | | | | | | | | | | | | | | | | | | To manipulate the firmware contents on device, we need to access the flash chipset (usually via SPI) on system. The `host_flashrom` provides a way to call external program "flashrom" for reading and writing firmware. So the `update_firmware` can now load "system current firmware" using flashrom. Note in the future we may want to statically link the flashrom as library so there won't be external dependency. BUG=chromium:875551 TEST=make futil; futility update -i IMAGE tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I52f2d4fe4fe4dd660f762a5a75e3367820717e19 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183650 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Process FMAP and parse image versionHung-Te Lin2018-08-312-1/+105
| | | | | | | | | | | | | | | | | Add `find_firmware_section` and `firmware_section_exists` utility functions to manipulate FMAP based sections easily. It is used by parsing of image version strings, which helps indicating the firmware versions going to be updated. BUG=chromium:875551 TEST=make futil; futility update -i /build/eve/firmware/image.bin tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I458fa8c31c45dbbd29614c3d6ccd586e46a3ed0b Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183649 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: cmd_update: Support loading image files by -i, -e, --pd_imageHung-Te Lin2018-08-312-2/+92
| | | | | | | | | | | | | | | | | | | To specify images, we want to read them from files specified from command line: -i: AP (host) firmware image. -e: EC firmware image. --pd_image: PD firmware image (deprecated). BUG=chromium:875551 TEST=make futil; futility update -i /build/eve/firmware/image.bin tests/futility/run_test_script.sh $(pwd)/build/futility BRANCH=None Change-Id: I3c2dbe3d3ce4619aa7e044a154be3aba7ab9181c Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183648 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: Add 'update' command for updating firmwareHung-Te Lin2018-08-314-0/+119
| | | | | | | | | | | | | | | | A reference firmware updater for all systems running vboot using FMAP for layout. The updater is currently a dummy implementation and will be completed with incoming changes. BUG=chromium:875551 TEST=make futil; build/futility/futility update; tests/futility/run_test_scripts.sh $(pwd)/build/futility BRANCH=None Change-Id: I57bec91c178749b79a19789f9599f5f9048fced8 Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1182701 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* futility: Add 'Error' utility macroHung-Te Lin2018-08-311-0/+3
| | | | | | | | | | | | | An 'Error' is easier than writing fprintf(stderr, "ERROR: %s", ...). BUG=chromium:875551 TEST=make futil; BRANCH=None Change-Id: Id4a849014dc202319dc1932289c68f43b2430c7d Signed-off-by: Hung-Te Lin <hungte@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1183647 Reviewed-by: Randall Spangler <rspangler@chromium.org>
* cgpt: add edit commandstabilize-nocturne-10986.Bstabilize-11021.84.Bstabilize-11020.Bstabilize-11005.Bstabilize-10985.Brelease-R70-11021.Bfirmware-servo-11011.Bfirmware-nocturne-10984.Bfactory-nocturne-10984.BMatt Delco2018-08-158-0/+161
| | | | | | | | | | | | | | | | This change adds a command to cgpt to change the GUID of the drive. BRANCH=none BUG=None TEST=Compiled and ran utility to verify that GUID changes. Also verified that the new and existing tests completed successfully. Change-Id: Ia8a815447509626312e2b06c6f293901290c73c3 Signed-off-by: Matt Delco <delco@google.com> Reviewed-on: https://chromium-review.googlesource.com/1171834 Reviewed-by: Julius Werner <jwerner@chromium.org>
* cgpt: show verbose details when primary ignoredMatt Delco2018-08-141-4/+5
| | | | | | | | | | | | | | | | | | A later change I authored has the tests check the GUID of the drive. When the primary table is ignored the GUID from the secondary wasn't being displayed either. This change has the details of the secondary table get displayed when the primary table is ignored. BRANCH=none BUG=None TEST=Compiled. The change was runtime tested as part of a larger change though I didn't unit test this particular change after it was split out into a separate commit. Change-Id: I300511cf65c67f4888e08ab49cd72c7acf234507 Signed-off-by: Matt Delco <delco@google.com> Reviewed-on: https://chromium-review.googlesource.com/1173410 Reviewed-by: Julius Werner <jwerner@chromium.org>
* vboot_reference: sync flags with eve-campfire branchTing Shen2018-08-143-1/+11
| | | | | | | | | | | | | BUG=b:112520234 TEST=make runtests BRANCH=master Change-Id: I84ea07f948fec9aa2945c10831f434e77b0e435d Signed-off-by: Ting Shen <phoenixshen@google.com> Reviewed-on: https://chromium-review.googlesource.com/1172305 Commit-Ready: Hung-Te Lin <hungte@chromium.org> Tested-by: Ting Shen <phoenixshen@chromium.org> Reviewed-by: Hung-Te Lin <hungte@chromium.org>
* vboot: remove line length check from lsb_syntaxcheck()Bob Moragues2018-08-131-8/+0
| | | | | | | | | | | | | | | | | | Unibuilds, such as Octopus, generate long lines listing the individual board names. Removing a check which is restricting the unibuilds ability to add additional boards to the list. BUG=chromium:873552 BRANCH=none TEST=none Change-Id: I080f4f251935eb19ee3377556500a5bd98117a2f Reviewed-on: https://chromium-review.googlesource.com/1173256 Commit-Ready: Bob Moragues <moragues@chromium.org> Tested-by: Bob Moragues <moragues@chromium.org> Reviewed-by: Julius Werner <jwerner@chromium.org>
* vboot_reference: Change OOBE autoconfig TPM space to storing 32-byte hashTudor Brindus2018-08-081-4/+4
| | | | | | | | | | | | | | | | | This commit switches from claiming 33 bytes of NVRAM for the compressed representation of a prime256v1 compressed public key to 32 bytes for its SHA256 hash. This makes it easier to process with the standard OpenSSL binary. BUG=chromium:845589 TEST=make runtests BRANCH=none Change-Id: Ic641b800bcbf2158d52ffbebbf143c47061e8cc3 Reviewed-on: https://chromium-review.googlesource.com/1161496 Commit-Ready: Tudor Brindus <tbrindus@chromium.org> Tested-by: Tudor Brindus <tbrindus@chromium.org> Reviewed-by: Randall Spangler <rspangler@chromium.org>
* Add AltOS NVRAM flagsTing Shen2018-08-076-1/+31
| | | | | | | | | | | | | | | | Port CL:1009444 to ToT. Adds (enable|disable)_alt_os_request flag for AltOS boot flow. BRANCH=none BUG=b:70804764 TEST=1. make runtests 2. Manually, set and get new flags via crossystem Change-Id: Ie7fe2620f736335f11c39cbfe37b3fdf400ff926 Reviewed-on: https://chromium-review.googlesource.com/1014840 Commit-Ready: ChromeOS CL Exonerator Bot <chromiumos-cl-exonerator@appspot.gserviceaccount.com> Tested-by: Ting Shen <phoenixshen@chromium.org> Reviewed-by: Randall Spangler <rspangler@chromium.org>
* vboot: bulk fix tabs and spacing inconsistencyJoel Kitching2018-08-0373-1054/+1081
| | | | | | | | | | | | | | | | | | | Problem files were found with: find . -name '*.c' -o -name '*.h' | xargs grep '^ [^*]' and edited manually. Ignores utility/ and cgpt/, since they seem to globally adhere to a two-space tab convention. BUG=None TEST=make clean runtests TEST=emerge vboot_reference depthcharge Change-Id: I5a678484a119c8f1911f717e1968bdb4f1a0810f Reviewed-on: https://chromium-review.googlesource.com/1160131 Commit-Ready: Joel Kitching <kitching@chromium.org> Tested-by: Joel Kitching <kitching@chromium.org> Reviewed-by: Randall Spangler <rspangler@chromium.org>
* vboot_reference: Add recoverysw_is_virtual flag to crossystemTudor Brindus2018-08-022-0/+8
| | | | | | | | | | | | | | | | | | This commit adds a flag recoverysw_is_virtual for determining whether a device's recovery switch status (as given by recoverysw_cur) is from a physical button or a line connected to Servo, without a physical button (e.g. veyron_minnie). BRANCH=none BUG=chromium:845589 TEST=manually tested on cave and veyron_minnie; make runtests Change-Id: If8e54e1df78b25a52dbf359ce641bea75533d705 Reviewed-on: https://chromium-review.googlesource.com/1157537 Commit-Ready: Tudor Brindus <tbrindus@chromium.org> Tested-by: Tudor Brindus <tbrindus@chromium.org> Reviewed-by: Randall Spangler <rspangler@chromium.org> Reviewed-by: Julius Werner <jwerner@chromium.org>
* vboot: Add .clang-format fileRaul E Rangel2018-08-021-0/+10
| | | | | | | | | | | | | Copied from depthcharge. BUG=none BRANCH=none TEST=formatted some code Change-Id: I9b9916df7da6195c753f2ce9ddbf37baf8a3e747 Signed-off-by: Raul E Rangel <rrangel@chromium.org> Reviewed-on: https://chromium-review.googlesource.com/1160930 Reviewed-by: Martin Roth <martinroth@chromium.org>
* vbutil_what_keys: support 16 MB firmware imagesJoel Kitching2018-07-311-1/+1
| | | | | | | | | | | | | | | | Previously, vbutil_what_keys assumed any input files of size 8 MB or lower are firmware images. Push that size up to 16 MB to support larger firmware images, such as that of eve's. BUG=None TEST=vbutil_what_keys on eve image.bin Change-Id: Iaf07ad3f419f5e79584391a2b846100e3fae61dc Reviewed-on: https://chromium-review.googlesource.com/1156326 Commit-Ready: Joel Kitching <kitching@chromium.org> Tested-by: Joel Kitching <kitching@chromium.org> Reviewed-by: Bill Richardson <wfrichar@chromium.org>
* image_signing: strip_boot: support device node inputsMike Frysinger2018-07-272-10/+15
| | | | | | | | | | | | | | | | To speed things up, support stripping /boot from a rootfs block device. This way we can mount an image via loopback and pass that in directly. BRANCH=None BUG=chromium:714598 TEST=strip_boot_from_image.sh on image files works, and on loopback partitions Change-Id: Ie74d3f239ac29533f4325d0c1f75e3cce5fab7a5 Reviewed-on: https://chromium-review.googlesource.com/1152075 Commit-Ready: ChromeOS CL Exonerator Bot <chromiumos-cl-exonerator@appspot.gserviceaccount.com> Tested-by: Mike Frysinger <vapier@chromium.org> Reviewed-by: David Riley <davidriley@chromium.org> Reviewed-by: Chris Ching <chingcodes@chromium.org>
View Lorry Controller Status