Devices launching with P VINTF requirements
(including HALs and kernels) will
not be able to update to Android U.
If a device wish to do so, it must update the
vendor / odm partitions to match newer VINTF
requirements.
Test: TH
Bug: 205175891
Change-Id: I0b6c0bf0955a1e9b41e84352cc706f7b73404980
No longer keeping a compatibility_matrix.current.xml. The matrices will
always have the level in the name.
Test: na
Bug: 266392619
Change-Id: I59f35a1d47d594644140318ff0ed1115224550f6
Remove L2TP, PPP and PPTP related kernel configs from
the android-base.config as these are converted to
modules.
Bug: 232431151
Test: TH
Change-Id: I4c6e6f9f14663e756a2f825469f975fe232e2559
Signed-off-by: Ramji Jiyani <ramjiyani@google.com>
Technically this is not a valid change, as it restricts valid
configurations of the kernel, but... in practice... this is required
already by at least Android 11 (R) in order for traffic accounting
to work correctly.
When using xt_qtaguid Android R CTS fails:
testTrafficStatsForLocalhost fails "java.lang.IllegalArgumentException: lower must be less than or equal to upper"
due to an extremely subtle difference in the implementation of
'-m owner --uid-owner XXX' (which xt_qtaguid breaks, by
effectively making the condition always true).
This was reported by Nvidia back in August 2021 in bug 196522401.
It takes a lot of painful debugging before we get to comment 36
which tries to make sense of things... and then comment 38 gets
to the root of the problem.
If someone does fail this (ie. VINTF) they'll need to flip:
CONFIG_NETFILTER_XT_MATCH_QTAGUID -> n
CONFIG_NETFILTER_XT_MATCH_OWNER -> y
CONFIG_NET_CLS_BPF -> y
If someone actually *relies* on xt_qtaguid (and thus can't simply
flip the kernel config options) they'll need to request
a waiver for VINTF... However, it's not at all clear how that could even happen considering the platform (userspace) side of the qtaguid implementation code was deleted in Android S (and was mainlined in T).
Additionally they'll then fail CTS, and we don't grant a waiver there.
So they simply won't be able to update to U (but I'm certain they
wouldn't have been able to run S or T either, so they'd need to be
upgrading straight from R).
Bug: 196522401
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Change-Id: I883e0c521e615a99f08361328dde2cd0d9a3d861
U will not support 4.9
(not to mention that, per https://www.kernel.org/category/releases.html,
4.9 LTS goes EOL in January 2023 [ie. in 2 months])
Test: TreeHugger
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Change-Id: Iafae8ea1e70984f84b76db9854aa074a460bce1f
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: I99c476221a177158facadfaffe0821fea70b3473
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: I66d72e8bcf69ee1f3b3dfab48c6bbc61b77ff087
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: I882f06e86ffcafb36a7eb80e9877375614ea64d0
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: I931f2671f688ab9907169754d3fd10e981be4a9e
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: I4d1fab878de42df92900ddbc306f74f49ed5039c
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: I9993666bedc9e077a1e5a3553de0bb4e8d0b3116
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: Ib3a48adaa5996bea244f6fe6a736bbeef223a537
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: Icf22688caa251689f552dd8cc9924076039989bf
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: I58ede037ed7cf9a5cea169fc0513787b387a0529
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: Ia59182ee7308966bdad6a54aaceef86e87a7d5aa
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://android-review.googlesource.com/c/kernel/configs/+/2186824
Change-Id: Ifcf7d8cb4a66f6c3b6134dd94171f9f1b0cdb9a1
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/kernel/configs/+/19094454
Change-Id: Ie5cc038bc7145d2bdafd59679b56316759a29d96
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/kernel/configs/+/19094454
Change-Id: I4f1b7e8e33c5fb7c2f091a5ace108160c6738784
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
am skip reason: Merged-In I6c9f71288d46c310c96bbfadfcf18a35a30e8511 with SHA-1 659aee1b68 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/kernel/configs/+/19094454
Change-Id: I887a5929cab9e92a3c8f51aea2d9b28a1f95b4eb
Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
These two drivers support every >1 gigabit usb ethernet dongle
I've managed to get my hands on.
In practice already enabled on GKI 5.10+ since S.
$ for i in AND510{S,T} AND515T ANDM; do cd /git/$i; echo; echo $i; git grep '^CONFIG_USB_NET_AQC111=y$' | egrep gki_defconfig; done
AND510S
arch/arm64/configs/gki_defconfig:342:CONFIG_USB_NET_AQC111=y
arch/x86/configs/gki_defconfig:314:CONFIG_USB_NET_AQC111=y
AND510T
arch/arm64/configs/gki_defconfig:354:CONFIG_USB_NET_AQC111=y
arch/x86/configs/gki_defconfig:325:CONFIG_USB_NET_AQC111=y
AND515T
arch/arm64/configs/gki_defconfig:350:CONFIG_USB_NET_AQC111=y
arch/x86/configs/gki_defconfig:320:CONFIG_USB_NET_AQC111=y
ANDM
arch/arm64/configs/gki_defconfig:342:CONFIG_USB_NET_AQC111=y
arch/x86/configs/gki_defconfig:315:CONFIG_USB_NET_AQC111=y
(while CONFIG_USB_NET_CDC_NCM defaults to 'y')
Test: TreeHugger
Bug: 183564444
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Change-Id: I6c9f71288d46c310c96bbfadfcf18a35a30e8511
Can be used for implementing better ingress rate limiting.
In practice already enabled on GKI 5.10+ since S.
$ for i in AND510{S,T} AND515T ANDM; do cd /git/$i; echo; echo $i; git grep '^CONFIG_IFB=y$' | egrep gki_defconfig; done
AND510S
arch/arm64/configs/gki_defconfig:324:CONFIG_IFB=y
arch/x86/configs/gki_defconfig:296:CONFIG_IFB=y
AND510T
arch/arm64/configs/gki_defconfig:337:CONFIG_IFB=y
arch/x86/configs/gki_defconfig:308:CONFIG_IFB=y
AND515T
arch/arm64/configs/gki_defconfig:333:CONFIG_IFB=y
arch/x86/configs/gki_defconfig:303:CONFIG_IFB=y
ANDM
arch/arm64/configs/gki_defconfig:325:CONFIG_IFB=y
arch/x86/configs/gki_defconfig:298:CONFIG_IFB=y
Test: TreeHugger
Bug: 179454839
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Change-Id: I04cf0ca5dae83d73f91b75b99fd6263c7e686aef
