sys.sysctl.extra_free_kbytes property controls the value set to the
/proc/sys/vm/extra_free_kbytes out-of-tree kernel knob. Replace its
use with execution of extra_free_kbytes.sh script which calculates
and sets corresponding value to /proc/sys/vm/watermark_scale_factor
upstream-supported kernel knob.
Bug: 109664768
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Iaece28eb858a20c8c39578a28dc6bbc6adc13c40
Android uses an out-of-tree /proc/sys/vm/extra_free_kbytes kernel knob
to increase low watermark. A similar /proc/sys/vm/watermark_scale_factor
knob supported in upstream kernel can be used for the same purpose.
The script implements conversion from desired extra_free_kbytes value
to watermark_scale_factor value and sets watermark_scale_factor according
to the value for extra_free_kbytes provided as a parameter.
Note that watermark_scale_factor affects high watermark differently
from extra_free_kbytes. While extra_free_kbytes increases low and high
watermarks by the same amount, watermark_scale_factor affects high
watermark 2x times. Therefore it is expected that replacing the usage
extra_free_kbytes with watermark_scale_factor will keep the same low
watermark but high watermark will be higher than before.
Usage:
extra_free_kbytes.sh <extra_free_kbytes value>
Bug: 109664768
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Id8949be731da77be884cdeec3a765ea291dbc2f0
socket.cpp fails to compile on musl with a missing definition of
fd_set. Include sys/select.h for the definition.
Bug: 190084016
Test: m USE_HOST_MUSL=true fastboot
Change-Id: I095a56381a85981f42444f4c39ae8e7786d8640f
This change is to update the doc to reflect the current state and it
doesn't affect the runtime behavior.
The max packet size was updated to 256 by aosp/1226715.
Bug: 196676392
Test: Presubmit; doc update doesn't affect build
Change-Id: I73b0a416730d09fdb24ea81bffe52ddd89534132
* changes:
Reduce the queue depth of loop devices used by the zram driver
libdm: Export ExtractBlockDeviceName()
libdm: Make ExtractBlockDeviceName() return its result
This allows us to mount with such fstab:
/dev/block/by-name/system /system ext4 ro slotselect
/dev/block/by-name/system /system erofs ro slotselect
This will be used when mount partition through ensure_path_mounted
in recovery.
Bug: 195726623
Test: mount with different image on the same fstab
Change-Id: I7761489a3e2bf46331563e872b6dfaa8b685a082
Signed-off-by: Huang Jianan <huangjianan@oppo.com>
Make the queue depth of loop devices identical to that of the underlying
storage device. This patch reduces latency by lowering the queue depth.
With this patch applied I see the following:
# cat /sys/block/loop30/queue/nr_requests
32
Bug: 194450129
Test: Built Android images, installed these and verified that the queue depth of loop devices is 32 instead of 256.
Change-Id: Ifa16084c7df3a54d9559c2388abc4a8392ff88c6
Signed-off-by: Bart Van Assche <bvanassche@google.com>
Make this function available to libdm users. A caller outside libdm will
be added by a later patch.
Bug: 194450129
Test: mm libfs_mgr libdm_test
Change-Id: I3e3560f3cdef8978eac644d5b53cf3851209c0c2
Signed-off-by: Bart Van Assche <bvanassche@google.com>
From https://engdoc.corp.google.com/eng/doc/devguide/cpp/styleguide.md:
"Prefer using return values over output parameters: they improve
readability, and often provide the same or better performance (see the C++
Primer)." Implement this advice for ExtractBlockDeviceName(). This patch
does not change any functionality.
Bug: 194450129
Test: mm libfs_mgr libdm_test
Change-Id: I6363781163eba08e6128507b403200f472f68a59
Signed-off-by: Bart Van Assche <bvanassche@google.com>
The KeyMint TA may send responses that are longer thant the 8K buffer
that the KeyMint HAL holds. This patch introduces
trusty_keymaster_call_2 which can grow the receive buffer on demand.
Bug: 195622501
Test: VTS and CTS test for regression testing.
Change-Id: Ia06e590e547e649ca81cda9a71851f334970788f
Merged-In: Ia06e590e547e649ca81cda9a71851f334970788f
After compiler update, infinite side effect free loops are replaced with trap
instructions. So use -fno-finite-loop to disable this behavior.
Bug: 196162833
Test: run debuggerd_test.
Change-Id: I057263360a5df64af18c17a025fab48887d0b470
There is an unclear error message if we run
`fastboot --disable-verification flash boot boot.img` on a
AVB-disabled device without a /vbmeta partition.
Error message:
terminating with uncaught exception of type std::out_of_range: basic_string
Aborted
This is because the buf->fd of the loaded boot.img is not reset,
before returning from function copy_boot_avb_footer().
Also reset some changes in copy_boot_avb_footer() from commit
If83f785e235569ee8ef0de2b37f11dbd2a9a71f4, to prevent the
unnecessary read of the boot vbmeta if no need to copy the
boot image avb footer.
Bug: 191903922
Test: `fastboot --disable-verification flash boot boot.img` on a device
without /vbmeta and without setting BOARD_AVB_ENABLE to true.
Change-Id: If84f08f1b2e12c4c6ded5cafda2bd1d30e75c662
With VAB we encourage a smaller super partition, so make sure we delete
"other" slot partitions during flashing. Otherwise, we may not have
enough space in super.
Bug: 195930130
Test: set_active, flashall, ensure other slot is deleted
Change-Id: Ic6a4f60e8f4c7abb8af7b38228753ad8ed85eedf
F2FS requires a larger minimal size than the current value of 16MiB.
Bug: 171368484
Fixes: 171368484
Test: Presubmit
Change-Id: I085d91634ada673388b07a0f5c6b1847e261b7f8
Also the check that linker configuration should not be updated if APEX
is in updatable state, is removed as now with refactoring of code and its
movement to apex project, that is also needed.
Resolves: https://b.corp.google.com/issues/192838499
Test: Run the AVD Device and check that it boots and run atest
ApexTestCases
Change-Id: I80a94fa0a4145d30546bb34a6d478a20a2881368
The main users of this instance are KFENCE and MTE-aided KASAN, which are only supported on arm64.
Skip creation of this tracing instance on 32-bit systems to save ~6Mb memory on low-end devices.
Bug: 195089948
Bug: 194719088
Bug: 194156700
Change-Id: Icaf762715fed7a282b1ad738c10bcb45dc848f4d
The user interface of Android devices becomes very slow under memory
pressure. This is because Android uses the zram driver on top of the loop
driver for swapping, because under memory pressure the swap code alternates
reads and writes quickly, because mq-deadline is the default scheduler for
loop devices and because mq-deadline delays writes by five seconds for such
a workload with default settings. Fix this by selecting I/O scheduler 'none'
for loop devices.
Bug: 194450129
Test: Built Android images, installed these and verified that the I/O scheduler of all loop devices is 'none' instead of 'mq-deadline'.
Change-Id: Ia5f606504b663948ab56955cad5a71885a356430
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Improve code readability by renaming 'device' into 'loop_device' and
'device_fd' into 'loop_fd'.
Bug: 194894000
Test: Built Android images and installed these on an Android device.
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Change-Id: Ia9c2d7a525e727f8706e66631b97fc4678c6a4d9
When the device is rebooted after OTA is done,
daemon will read through all the dynamic partitions
to bring the blocks to memory. When update-verifier
runs, all the blocks would be in page-cache thereby
cutting down the boot time.
Boot time improvements on Pixel:
Full OTA:
===========
VABC (Without this patch): 37.308 seconds
VABC (With this patch): 28.604 seconds
Incremental OTA:
=================
VABC (Without this patch): 39.072 seconds
VABC (With this patch): 27.523 seconds
We read the blocks only during second stage transition. Thus,
it shouldn't impact when snapuserd is spin up during
post-install phase or during first-stage init.
Bug: 193863442
Test: Full and Incremental OTA on pixel
Signed-off-by: Akilesh Kailash <akailash@google.com>
Change-Id: Id654449238e22125f7d6288c7100fde512cc2ced
* changes:
libsnapshot: Don't PrepMergeOps on resume
snapuserd: Add support for Xor ops in snapuserd
snapuserd: Rename Read Ahead Iterator
snapuserd: Add XorSink
libsnapshot: Clone worker readers from snapuserd
libsnapshot: Add support for Xor ops in Cow Format
When creating uid/pid hierarchy, cgroup.subtree_control should be set at
every level of that hierarchy except for the leaf level.
Bug: 195149205
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Iedc2e859715b31db62158c85016708f252db2b70
