We are changing the default emerge options to use --oneshot, so make sure
our scripts select the packages we care about for @world.
BUG=None
TEST=setup_board+build_packages for tegra2 adds pkgs to world
Change-Id: I5fc68c538a1a1c846bd9724f27ef717029d11d42
Reviewed-on: https://gerrit.chromium.org/gerrit/16838
Reviewed-by: David James <davidjames@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
BUG=None
TEST=`cros_sdk --delete && cros_sdk` still works
Change-Id: I9a5d5e5eedb950dcfaf9108c22087a5a7fa7cd47
Reviewed-on: https://gerrit.chromium.org/gerrit/16825
Reviewed-by: Brian Harring <ferringb@chromium.org>
Reviewed-by: David James <davidjames@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
BUG=chromium-os:26824
TEST=repro'd chromium-os:26790 on a test system with this config,
demonstrating that it no longer masks the bug. Confirmed
dbus-monitoring still works while doing so. (Tested with
help from jorgelo.) Confirmed the 26790 fix is not regressed by this
either.
Change-Id: I60ce833ffb14a9a709ca6bfecf40221940845aa0
Reviewed-on: https://gerrit.chromium.org/gerrit/16739
Reviewed-by: Scott James Remnant <keybuk@chromium.org>
Commit-Ready: Jim Hebert <jimhebert@chromium.org>
Tested-by: Jim Hebert <jimhebert@chromium.org>
The logic of IGNORE_PREFLIGHT_PREBUILTS is currently inverted in make_chroot,
causing preflight prebuilts to be ignored by default, but then used in the one
situation where they should be ignored. This CL fixes that.
BUG=chromium-os:26834
TEST=Verify preflight prebuilts are now used by default in make_chroot.
Change-Id: I626319d6de79b6f201f993a05e34610e944b161e
Reviewed-on: https://gerrit.chromium.org/gerrit/16614
Reviewed-by: Richard Barnette <jrbarnette@chromium.org>
Commit-Ready: David James <davidjames@chromium.org>
Tested-by: David James <davidjames@chromium.org>
The --jobs=<n> option wasn't actually being passed to parallel_emerge --
which meant it didn't do anything.
BUG=chromium-os:26827
TEST=Ran with --jobs=N and made sure only N parallel_emerge processes were
started.
Change-Id: I581fc5588b54e246acaefd0c7e528e55adf9ba8a
Reviewed-on: https://gerrit.chromium.org/gerrit/16570
Reviewed-by: Michael Krebs <mkrebs@chromium.org>
Commit-Ready: David James <davidjames@chromium.org>
Tested-by: David James <davidjames@chromium.org>
Also, blacklist /usr/share/locale. Most recently this unused directory crept
back into the image after some "sudo" package changes/upgrades.
BUG=chromium-os:11820
TEST=build_image dev for x86-alex, tegra2, amd64-generic
Change-Id: I0a209a5030a3da3674d3a38faf2367032c6e3423
Reviewed-on: https://gerrit.chromium.org/gerrit/16352
Tested-by: Darin Petkov <petkov@chromium.org>
Reviewed-by: Chris Sosa <sosa@chromium.org>
Commit-Ready: Darin Petkov <petkov@chromium.org>
Previously this script never honored the user --port option.
BUG=None
TEST=Manual: Run with or without --port option
Change-Id: I971ee69818a37e6706ce70327b1d243fd78d214d
Reviewed-on: https://gerrit.chromium.org/gerrit/16429
Reviewed-by: Ryan Cairns <rtc@chromium.org>
Commit-Ready: Paul Stewart <pstew@chromium.org>
Tested-by: Paul Stewart <pstew@chromium.org>
If we try to setup multiple boards that use the same toolchain, we end up
deleting the package.{mask,keywords} files for the existing toolchain so
that we can detect if there are upgrades available. If there aren't, the
files don't get recreated (since we were relying on crossdev for it), and
the installed toolchain packages may no longer be directly reinstalled if
they aren't incidentally marked stable.
Normally this isn't an issue since people build for target boards only
rather than installing toolchains, but this breaks the sdk bot since it
sets up a bunch of bots and then tries to emerge all the toolchain pkgs
directly. With packages we've pinned that are unstable, portage fails:
The following keyword changes are necessary to proceed:
#required by =cross-arm-none-eabi/newlib-1.18.0 (argument)
>=cross-arm-none-eabi/newlib-1.18.0 **
So backup & restore the files in the update path rather than just blowing
them away. If we end up installing things anyways, crossdev will do the
right thing.
BUG=None
TEST=`cbuildbot chromiumos-sdk` passes
Change-Id: I6efc71179d6f897af91deaefe1ce6c62df10ebf6
Reviewed-on: https://gerrit.chromium.org/gerrit/16220
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
The arm-none-eabi toolchain doesn't use glibc, so trying to copy
the glibc binpkg won't work.
BUG=None
TEST=`cbuildbot chromiumos-sdk` passes
Change-Id: I47d420fa18466128fa319f15702d4864e5f7dce0
Reviewed-on: https://gerrit.chromium.org/gerrit/16225
Reviewed-by: Zdenek Behan <zbehan@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
We want to have some 64bit boards also install the 32bit toolchain, so
extend the toolchain.conf file to allow specifying of multiple targets.
BUG=chromium-os:25969
TEST=build_packages for x86-mario installed both i686-pc-linux-gnu and x86_64-cros-linux-gnu
Change-Id: Ic2ca1aeeefa5ea2207d3c963ac856909cb40cae5
CQ-DEPEND=Iea6f682e258d59c0b3b3ca2cad7e15ae521dcc89
CQ-DEPEND=Ib71e7503c3ebc98549d09e35cb0bb9687b8fafb8
Reviewed-on: https://gerrit.chromium.org/gerrit/15902
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
When we create a chroot for the first time, we see:
...
Exiting: Nothing left to do; exiting. :)
Elapsed time: 5m10s
cros_sdk: All set up. To enter the chroot, run:"
cros_sdk: $ cros_sdk --enter "
CAUTION: Do *NOT* rm -rf the chroot directory; if there are stale bind
mounts you may end up deleting your source tree too. To unmount and
delete the chroot cleanly, use:
$ cros_sdk --delete
(cr) (v) vapier@vapier ~/trunk/src/scripts $
Kill the spurious double quotes, add new lines to make the text easier
to pick out from the previous command/new prompt, and only show the
sdk prefix on the first line.
BUG=None
TEST=`cros_sdk --delete && cros_sdk` looks nice
Change-Id: Iba0ee68035fffe6fc946e9b3476cc0d0b55abfc5
Reviewed-on: https://gerrit.chromium.org/gerrit/16075
Reviewed-by: David James <davidjames@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
Having introduced autodetection and autoselection of images, some users
may be surprised to find that the script will happily choose to copy an
image other than chromiumos_image.bin when only one image is present.
Since this is substantially different from the previous behavior, we now
present a menu (with a single choice) in this case as well.
Also fixing the list of images to be detected and properly prioritizing
a default image over others. The code should already work with the
future image naming scheme, where chromiumos_image.bin will be a symlink
to one of six concrete images (with unique names).
BUG=None
TEST=Ran script with different image configurations and flags.
Change-Id: I5d448c5425754496d256258281694cd7ec2554d2
Reviewed-on: https://gerrit.chromium.org/gerrit/15982
Commit-Ready: Gilad Arnold <garnold@chromium.org>
Reviewed-by: Gilad Arnold <garnold@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
I use these wrappers all the time on my own system.
BUG=None
TEST=`./setup_board --board=x86-mario && qlist-x86-mario -Iv` works
Change-Id: I609ea07337b2fd14257fae9a0c89b544a40beab9
Reviewed-on: https://gerrit.chromium.org/gerrit/16071
Reviewed-by: David James <davidjames@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
The -d option to umount is unnecessary in cases where the original
mount was created with '-o loop' (it says so in the man page).
Moreover, there's a demonstrable bug when using 'umount -d' in this
case: umount complains that it's unable to delete the loop device
in this case. This change silences those messages.
The demonstrated bug in umount is a plausible explanation for build
failures characterized by unexpected failures of 'mount -o loop';
see the BUG, below.
BUG=chromium-os:24975
TEST=run the changed commands, see no loopback devices leaked
Change-Id: I1df9f60b40eabd279164b79ee66994ff9f644ee9
Reviewed-on: https://gerrit.chromium.org/gerrit/15998
Tested-by: Richard Barnette <jrbarnette@chromium.org>
Reviewed-by: David James <davidjames@chromium.org>
Commit-Ready: Richard Barnette <jrbarnette@chromium.org>
Without this fix, when the script is invoked with (say) --to=/dev/sdk
but /dev/sdk is not a valid block device, we will interpret it as if the
user intended to copy the image to a file of the same name. However, it
is more likely that the user specified the wrong device name. Hence,
this fix catches this type of error and will fail the operation with
a corresponding error message.
Also includes minor changes to printed messages.
BUG=None
TEST=Ran the script with different values of --to
Change-Id: Ie060ff2be77b8c065ba95f635455049e034d4bbc
Reviewed-on: https://gerrit.chromium.org/gerrit/15822
Reviewed-by: Gilad Arnold <garnold@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
Commit-Ready: Gilad Arnold <garnold@chromium.org>
The recent changes to image_to_usb.sh assumed that chromiumos_image.bin
is always used for generating a test image. This wasn't the case before
these changes were applied, where a preexisting test image would have
sufficed. This fix eliminates image autodetection entirely when
--test_image it used and falls back to the old behavior: if a test image
exists, it will use it; otherwise, it will generate it from the image
specified by --image_name, or from the default chromiumos_image.bin.
This is not to say that I like this behavior: it is subtle, may lead to
errors (e.g. using a stale test image whereas the user assumes a new one
should be generated), and inconsistent with other behaviors (e.g.
installing a recovery image, where no image generation takes place at
all). I believe that long term we should eliminate the generation of
test images from image_to_usb.sh and have users invoke the necessary
scripts (like mod_image_for_test.sh) explicitly.
BUG=chromium-os:26239
TEST=Ran script with different combinations of --test_image,
--image_name and image files present.
Change-Id: Ib2491a7e45e23eb51ced6ef31d3f129ec7863a25
Reviewed-on: https://gerrit.chromium.org/gerrit/15764
Reviewed-by: Richard Barnette <jrbarnette@chromium.org>
Commit-Ready: Gilad Arnold <garnold@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
This both fixes an incorrect behavior where --to_product would unsafely
lead to using a target device not intended to by the user; and reuses
the existing logic for device autodetection to work with provided
product strings. As a result, it is easier to find devices, and the
script's behavior is more consistent as far as device detection and
selection. This preserves the pattern matching on product string (now
properly highlighted in usage text and inline comments).
BUG=None
TEST=Ran script with various flag combinations (--to, --to_product),
product strings, and devices connected.
Change-Id: I80bfca086029555853a680e7c97251b11adc912e
Reviewed-on: https://gerrit.chromium.org/gerrit/15757
Reviewed-by: Richard Barnette <jrbarnette@chromium.org>
Commit-Ready: Gilad Arnold <garnold@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
The gcc-config file name could be different from the gcc atom that creates it.
Changed the logic to first find out the binary path of gcc and then to query
portage to see what package owns that path.
BUG=none
TEST=cros_sdk --bootstrap --replace
cros_sdk -- ./setup_board --board=x86-zgb --nousepkg
Change-Id: I3ccfa9948aaba6ba28107f150523c8a5d7cf1260
Reviewed-on: https://gerrit.chromium.org/gerrit/15556
Reviewed-by: David James <davidjames@chromium.org>
Reviewed-by: Zdenek Behan <zbehan@chromium.org>
Tested-by: asharif <asharif@chromium.org>
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Commit-Ready: asharif <asharif@chromium.org>
Every consumer that I can find of get_latest_image.sh expects this
script to output something useful. Many don't check the result and
end up spitting out unhelpful messages when it doesn't. So make a
missing images dir fatal for this script so everyone on top of us
gets this checking for free.
If, in the future, someone actually wants this to output nothing
when there's nothing found, we can add a dedicated flag. That mode
of operation is a lot less common than having it die by default.
BUG=None
TEST=delete image tree, run ./get_latest_image.sh --board=${BOARD}, see it fail nicely
TEST=delete image tree, run ./image_to_vm.sh --board=${BOARD}, see it fail nicely
Change-Id: I0a2ffd0b9084297b6d2346f02dbdb7bd0ef667bf
Reviewed-on: https://gerrit.chromium.org/gerrit/15589
Reviewed-by: David James <davidjames@chromium.org>
Reviewed-by: Chris Sosa <sosa@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
This is needed as part of transition to cros-workon.
Crossdev has a -S option to install the current stable, which is
the preferred future option, but would greatly complicate the logic.
BUG=chromium-os:25338
TEST=setup_board --board=any --force, see it build the right versions
TEST=trybot chromiumos-sdk
Change-Id: I60dfffcfa38e482499f909787ead26fd27ef36e1
Reviewed-on: https://gerrit.chromium.org/gerrit/15702
Commit-Ready: Zdenek Behan <zbehan@chromium.org>
Reviewed-by: Zdenek Behan <zbehan@chromium.org>
Tested-by: Zdenek Behan <zbehan@chromium.org>
Now that aura is default, we don't need to special-case it anymore.
BUG=chromium-os:25818
TEST=Verify that default prebuilts uploaded by the x86-generic builder
are used now instead of the aura-specific prebuilts. These prebuilts
have aura enabled now, since it is the default.
Change-Id: I8b78f53f5ceab6e5907390e03554120f20ea96fe
Reviewed-on: https://gerrit.chromium.org/gerrit/15726
Tested-by: Ryan Cui <rcui@chromium.org>
Reviewed-by: Ryan Cui <rcui@chromium.org>
Commit-Ready: David James <davidjames@chromium.org>
Tested-by: David James <davidjames@chromium.org>
This reverts commit 19754342048a44304150981b2938f604751605e7
The tree is on fire: the x86 canaries and x86 generic full
bots are dying with internal gold link errors.
http://crosbug.com/26168
Only happens on new sdks, so we're shotgunning a few
suspicious revs and hoping for the best.
Change-Id: Iae05c876ff2a84f4a9f549f74b2ee7e0903f2b5a
Reviewed-on: https://gerrit.chromium.org/gerrit/15669
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
This is needed as part of transition to cros-workon.
Crossdev has a -S option to install the current stable, which is
the preferred future option, but would greatly complicate the logic.
BUG=chromium-os:25338
TEST=setup_board --board=any --force, see it build the right versions
TEST=trybot chromiumos-sdk
Change-Id: I07c1ade5731e9391fb22a572f7fe8517011e6239
Reviewed-on: https://gerrit.chromium.org/gerrit/15176
Reviewed-by: David James <davidjames@chromium.org>
Tested-by: Zdenek Behan <zbehan@chromium.org>
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Commit-Ready: Zdenek Behan <zbehan@chromium.org>
The refresh-packages builder needs to start using auth tokens for logging into
Google Docs, rather than email/password each time, so this copies ~/.gdata_token
into the chroot in the same way as ~/gdata_cred.txt.
BUG=chromium-os:23819
TEST=Put a file at ~/.gdata_token outside chroot, then run:
`cros_sdk true ; cmp {chroot/home/$USER/,~/}.gdata_token`, which passes.
Change-Id: Ib688a01aa88e0fd0f0211236bb0354813ef1fbb3
Reviewed-on: https://gerrit.chromium.org/gerrit/15550
Commit-Ready: Matt Tennant <mtennant@chromium.org>
Reviewed-by: Matt Tennant <mtennant@chromium.org>
Tested-by: Matt Tennant <mtennant@chromium.org>
The script does not assume a single source image (chromiumos_image.bin)
but detects which images of a list of candidate images are present, and
lets the user select one. The default choice is the first image in the
list to be detected. If only one image was detected, it will be
automatically selected. The list contains the aforementioned standard
image, as well as the default names for recovery, test, factory and
factory install images. If the script is invoked with --test, --factory
or --factory_install flags, it will only seek for the standard image
(and attempt to generate the desired image from it, as was previously
done).
Also fixed some log messages; option strings; and improved the logic for
unmounting the target device, eliminating an unnecessary message and
a 3 second delay.
BUG=chromium-os:26010
TEST=Tested image_to_usb.sh with different images and flags.
CQ-DEPEND=I53a42a46a3c90fd486fead578bfbae248f64cfc2
Change-Id: I0d2f20dc8d62ce5fa18c10d9f8b51a46b2ddca5d
Reviewed-on: https://gerrit.chromium.org/gerrit/15528
Reviewed-by: Richard Barnette <jrbarnette@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
Commit-Ready: Gilad Arnold <garnold@chromium.org>
Defined a bash function choose(), which allows to present a menu and
prompting for a selection, with support for default choice. Usage is
commented appropriately.
Originally reviewed in CL I0d2f20dc8d62ce5fa18c10d9f8b51a46b2ddca5d.
BUG=chromium-os:26010
TEST=Tested in conjunction with a use in image_to_usb.sh, works fine.
Change-Id: I53a42a46a3c90fd486fead578bfbae248f64cfc2
Reviewed-on: https://gerrit.chromium.org/gerrit/15586
Reviewed-by: Gilad Arnold <garnold@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
Commit-Ready: Gilad Arnold <garnold@chromium.org>
First, add a enable_strict_sudo helpers that scripts that are
sudo strict can invoke. This does a sanity check bailing immediately
if invoked from a non sudo-keep-alive context.
Second, update enter_chroot.sh and make_chroot.sh to be strict.
While this is strict, that's the point. The aim of this is to block
the previous sudo interactive mess for spreading, let alone reappearing
in scripts/code that has been cleansed.
BUG=chromium-os:18393
TEST=cros_sdk --replace; in the midst of it, do sudo -k.
cros_sdk should thus bail out w/ an appropriate error.
CQ-DEPEND=I01bb1466cf027401fa387af7fad15e42fd33aea4
Change-Id: I76c5b87a812cc78c30a2eb1a0c56b9e438f4a98f
Reviewed-on: https://gerrit.chromium.org/gerrit/15294
Commit-Ready: Brian Harring <ferringb@chromium.org>
Reviewed-by: Brian Harring <ferringb@chromium.org>
Tested-by: Brian Harring <ferringb@chromium.org>
Motivated by the need to share the default recovery image name across
scripts (image_to_usb.sh in this case) and have it setup in common.sh.
Includes some general cleanup steps such flags and usage strings,
migrating function definitions to the beginning of the script, etc.
BUG=chromium-os:26010
TEST=Tested mod_image_for_recovery.sh on local host.
Change-Id: Ife3731915302a2e7be8beb750501d0ae9a0c43a4
Reviewed-on: https://gerrit.chromium.org/gerrit/15523
Reviewed-by: Richard Barnette <jrbarnette@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
Commit-Ready: Gilad Arnold <garnold@chromium.org>
This is a preparatory step for changes to image_to_usb.sh and
mod_image_for_recovery.sh. In general, by migrating the default
recovery image name to common.sh we can improve the interoperability
between the different scripts.
BUG=chromium-os:26010
TEST=Obvious change; tested to see the mod_image_for_recovery.sh and
image_to_usb.sh still work.
Change-Id: I765d665b656d5d25b2d9b0e40bf5c05048c3f255
Reviewed-on: https://gerrit.chromium.org/gerrit/15507
Commit-Ready: Gilad Arnold <garnold@chromium.org>
Reviewed-by: Gilad Arnold <garnold@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
When we source the local copy of shflags, we don't check for errors like
we do when sourcing the external one. Add a `die` call for that too.
BUG=None
TEST=`rm lib/shflags/shflags && (. ./common.sh)` now exits properly
Change-Id: I62c9b33dd96f2934691c43d53dea012097116ca1
Reviewed-on: https://gerrit.chromium.org/gerrit/15443
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
The shflags sourcing logic tries to use `die` before we've defined it, so
if there's an issue, we end up with:
bash: die: command not found
and the shflags code doesn't actually get loaded.
So relocate these small helpers to the top so we don't have to worry about
when it's safe to use these things.
BUG=None
TEST=`rm lib/shflags/shflags && (. ./common.sh)` now exits properly
Change-Id: Ibdc268e6c081aa07679dc9fce76e5603b7217b20
Reviewed-on: https://gerrit.chromium.org/gerrit/15442
Reviewed-by: Brian Harring <ferringb@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
Migrated function definition to the beginning of the code; some minor
cosmetics. This is a preliminary fix to the actual feature mentioned in
the cited issue.
BUG=chromium-os:26010
TEST=Executed in chroot environment with various combinations of flags
and connected devices
Change-Id: Ib73328e738ebecc38e6faafbd4feb33ced8804ad
Reviewed-on: https://gerrit.chromium.org/gerrit/15438
Reviewed-by: Chris Sosa <sosa@chromium.org>
Commit-Ready: Gilad Arnold <garnold@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
If no target device is provided, image_to_usb.sh will let the user
select one out of a list of autodetected devices. If only one device is
detected, it will be automatically selected. Also improves the
descriptors shown for candidate/chosen target device(s), and prints
a noticeable warning when the target device does not appear to be
a USB/MMC one.
Also changed all fail/warning messages to use 'die' and 'warn',
respectively, for compliance with other scripts. Slightly massaged
error/warning strings to be more compact and to-the-point.
BUG=chromium-os:25878
TEST=Executed in chroot environment with various combinations of flags
and connected devices
Change-Id: If248993b8e6f3bc8654c2c8f25f1e54e7899330d
Reviewed-on: https://gerrit.chromium.org/gerrit/15270
Reviewed-by: Richard Barnette <jrbarnette@chromium.org>
Tested-by: Gilad Arnold <garnold@chromium.org>
Commit-Ready: Gilad Arnold <garnold@chromium.org>
Via distfiles bind mounting, chrome build output was being cached
across chroot replacements. This potentially is desirable, but
wasn't explicitly planned for and violates an assumption of the
canaries.
As such move the content to it's new location.
BUG=None
TEST=emerge-<board> chromeos-chrome
CQ-DEPEND=Ibfc5d42e74861ff498dd2cb6cc2d7be6ec0ded60
Change-Id: I18e384f0fc1cedb3c70ab85178103765322370eb
Reviewed-on: https://gerrit.chromium.org/gerrit/15383
Reviewed-by: David James <davidjames@chromium.org>
Tested-by: Brian Harring <ferringb@chromium.org>
Purpose of this is to allow us to avoid re-downloading everything
every time we rebuild the chroot.
This maintains two directories; host and target. Future enhancement
involves collapsing this into one- this requires some host work however,
and has some potential gotchas in doing so.
Meanwhile, we now store distfiles in repo/distfiles/{host,target},
and mount bind repo/distfiles into /var/cache/distfiles.
An upgrade script in turn optimistically tries to move the content
into the new location; if it can't complete the move, it wipes the
content and the user has to redownload it (acceptable, if annoying).
BUG=chromium-os:13115
TEST=cbuildbot x86-generic-full
Change-Id: Iea96429df0e1fdc4ac0860fbce0daabc90c4c2a3
Reviewed-on: https://gerrit.chromium.org/gerrit/15189
Reviewed-by: Brian Harring <ferringb@chromium.org>
Tested-by: Brian Harring <ferringb@chromium.org>
Commit-Ready: Brian Harring <ferringb@chromium.org>
Now that make_chroot.sh lives alongside enter_chroot, we're able
to invoke it directly and use internal options not exposed through
cros_sdk. Thus enter_chroot grows a --early_make_chroot flag.
This flag bypasses the normal sudo usage, and runs the command
as root. It is needed and used by make_chroot for when sudo
may not yet exist (--bootstrap via stage3 lacks sudo).
For where sudo exists and we need access to the source tree,
distfiles, or profile sourcing, enter_chroot is directly used.
For all other invocations we use a single sudo chroot call to
bypass the overhead of enter_chroot.
This is a complete fix to chromium-os:25697 while addressing
all refactoring requests that came up in I8a6c5a26.
Finally, convert all comments into complete sentences, upper
case the leading word, etc to keep reviewers happy.
BUG=chromium-os:25697
TEST=Within a trybot, cros_sdk --bootstrap --replace
TEST=cros_sdk --replace
TEST=cros_sdk --delete
TEST=cros_sdk
Change-Id: I270ee7fc325ef0bea74c61505d25cdbb49a9a333
Reviewed-on: https://gerrit.chromium.org/gerrit/15322
Commit-Ready: Brian Harring <ferringb@chromium.org>
Reviewed-by: Brian Harring <ferringb@chromium.org>
Tested-by: Brian Harring <ferringb@chromium.org>
Imported as of 09d3117c483c60f55f552694b84b0964dfc17477.
Only modification is so that it can find common.sh.
BUG=chromium-os:25697
TEST=None
Change-Id: Idb43f1ecff3dff10a61396629a344c8bd10f505e
Reviewed-on: https://gerrit.chromium.org/gerrit/15317
Tested-by: Brian Harring <ferringb@chromium.org>
Reviewed-by: David James <davidjames@chromium.org>
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Commit-Ready: Brian Harring <ferringb@chromium.org>
With Aura, we won't have a need for the WM any more.
TEST=built an image without chromeos-wm, verified that the build system didn't complain.
Change-Id: Id6e5006770f72ea45f440dbe56ff7f53262864c1
Reviewed-on: https://gerrit.chromium.org/gerrit/15212
Tested-by: Stéphane Marchesin <marcheu@chromium.org>
Reviewed-by: Daniel Erat <derat@chromium.org>
Commit-Ready: Stéphane Marchesin <marcheu@chromium.org>
Add a clearer error message if the source image was not found.
This can occur if the --from directory exists, but the selected
image within it does not. An easy case is --from'ing a directory
from a chromeos-images zip.
Previously image_to_usb would copy zero bytes and return success.
BUG=None
TEST=Ran image_to_usb.sh for default, directory and .bin --froms.
Change-Id: I826f2bbc8effd9554f558e517b51cc03b4f832d5
Reviewed-on: https://gerrit.chromium.org/gerrit/15120
Tested-by: Chris Wolfe <cwolfe@chromium.org>
Reviewed-by: Jon Kliegman <kliegs@chromium.org>
Commit-Ready: Chris Wolfe <cwolfe@chromium.org>
This is needed as part of transition to cros-workon.
Crossdev has a -S option to install the current stable, which is
the preferred future option, but would greatly complicate the logic.
BUG=chromium-os:25338
TEST=setup_board --board=any --force, see it build the right versions
Change-Id: I9139b8a38c9156ea27dfb428cff8c7deffb0e6de
Reviewed-on: https://gerrit.chromium.org/gerrit/15011
Tested-by: Zdenek Behan <zbehan@chromium.org>
Reviewed-by: Mike Frysinger <vapier@chromium.org>
Commit-Ready: Zdenek Behan <zbehan@chromium.org>
BUG=chromium-os:23229
TEST=ran this against a ToT alex img, image_to_usb, boots, installs, works
Change-Id: Ia1e589a8b920332c1a247986e6bbc3cad17ed751
Reviewed-on: https://gerrit.chromium.org/gerrit/14650
Commit-Ready: Jim Hebert <jimhebert@chromium.org>
Tested-by: Jim Hebert <jimhebert@chromium.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Chris Sosa <sosa@chromium.org>
This is needed as part of transition to cros-workon.
Crossdev has a -S option to install the current stable, which is
the preferred future option, but would greatly complicate the logic.
BUG=chromium-os:25338
TEST=setup_board --board=any --force, see it build the right one
Change-Id: Ibb4fd18d122f69835eb471b2fa1a258c571dfdd1
Reviewed-on: https://gerrit.chromium.org/gerrit/14878
Tested-by: Zdenek Behan <zbehan@chromium.org>
Commit-Ready: Zdenek Behan <zbehan@chromium.org>
Reviewed-by: Zdenek Behan <zbehan@chromium.org>
This adds repo --references awareness (which maps down to git shared object
pools; see git clone -s); specifically, it sets up the appropriate indirection
bindings back to the actual shared source for use from within the chroot.
Note that the approach is a bit complex, but it's required complexity- via
this approach the references are accessible both from within, and without
the chroot.
BUG=chromium-os:19939
TEST=cbuildbot x86-generic-full --clobber --notests --nouprev
TEST=cbuildbot x86-generic-full --notests --nouprev
CQ-DEPEND=I986f17503dc154234ecadd90f6975d7164117cdb
Change-Id: I8498863010cefe0bf5b8f20350fa45a5f2a093d3
Reviewed-on: https://gerrit.chromium.org/gerrit/13467
Tested-by: Brian Harring <ferringb@chromium.org>
Reviewed-by: David James <davidjames@chromium.org>
Commit-Ready: Brian Harring <ferringb@chromium.org>
dump_syms can only dump the symbols of an executable with the same ELF
format as itself. Some recent build configurations now have binaries with
differing ELF formats. For example, issue 25468 has a 64-bit kernel where
everything else is 32-bit, and issue 25466 indicates there's a test that
contains a 32-bit executable.
BUG=chromium-os:25496, chromium-os:25468, chromium-os:25466
TEST=Ran cros_generate_breakpad_symbols on 32/64-bit executables
Change-Id: I15f5115585b3ed54ca7ae7b631216285baef8580
Reviewed-on: https://gerrit.chromium.org/gerrit/14835
Tested-by: Michael Krebs <mkrebs@chromium.org>
Reviewed-by: David James <davidjames@chromium.org>
Commit-Ready: Michael Krebs <mkrebs@chromium.org>
We have helper funcs already for displaying messages, so convert
some raw `echo` calls to them.
BUG=None
TEST=build_image still works and boots
Change-Id: Ie66cc59d0362ef6aa19011fa6cb0bc64a6a4fce8
Reviewed-on: https://gerrit.chromium.org/gerrit/14478
Reviewed-by: Richard Barnette <jrbarnette@chromium.org>
Commit-Ready: Mike Frysinger <vapier@chromium.org>
Tested-by: Mike Frysinger <vapier@chromium.org>
We have updated the toolchain in the chroot to use a newer version of
libstdc++ than what is installed on our workstations. This means
this additional library is required to generate deltas outside of the
chroot.
BUG=chromium-os:25277
TEST=Generated zip file, and ran contents outside of chroot on my workstation.
Change-Id: I5a90ee355aabd4849a9186a9a66e3dc9b1c51d52
Reviewed-on: https://gerrit.chromium.org/gerrit/14597
Reviewed-by: Eric M. Blake <eblake@google.com>
Reviewed-by: Eric Blake <eblake@chromium.org>
Tested-by: Don Garrett <dgarrett@chromium.org>
Commit-Ready: Don Garrett <dgarrett@chromium.org>
