Try to mitigate potential future driver core api changes by adding a
padding to a bunch of filesystem structures.
Based on a change made to the RHEL/CENTOS 8 kernel.
Bug: 151154716
Change-Id: Ida6d98d30f292c980ab07e0250fec5268c4c87ed
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Try to mitigate potential future driver core api changes by adding a
padding to struct dentry and struct dentry_operations.
Based on a change made to the RHEL/CENTOS 8 kernel.
Bug: 151154716
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Idde3c6e99bd4af3a91ba115b8ec148e3e1cdd4a9
Try to mitigate potential future driver core api changes by adding a
padding to struct bio_integrity_payload and struct bio_set.
Based on a change made to the RHEL/CENTOS 8 kernel.
Bug: 151154716
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I0397ede2e11560ad9422cd7765434fcd4f7a6dd8
This reverts commit ee5fed1cef.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: If0ec2bbe712e7dfd5bf43f0f266ddc255a2d28b3
This reverts commit 34c16f1a20.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I2654f99101dc95994cb5cc3328f288401dfe389f
This reverts commit 93d8bbeafb.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I965b9e7225127b5b0e58082e05d80bcebdb1e547
This reverts commit 990e8bd6d9.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Id685a5c9bca0cd0aa8bcc0d2da6288b9220456c8
This reverts commit ab02bc73ec.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I5a655c83ddbb6b70841f91e74924504ea6c320ec
This reverts commit 240526f655.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ic76a90ab14c2e03b2e40a1e7385f3cf5cb562ec8
This reverts commit 5cc70dfa47.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I8727d672ee887e1598d1661bbcd99061dcdae4d2
This reverts commit 7be3754601.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I94e86dbe2ba7f2f860854cf1f7f825a9a8215853
This reverts commit a717ddfe1a.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ia97655771b9a074f4d0724dc5ca543f63d139af4
This reverts commit 7f7da7f236.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ife780bae48b0b10e7502aa90a701f631f9b5a32f
This reverts commit 0c5b51622c.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I5b2c6e89737a2ac647f19719a1ccf256c2794a02
This reverts commit c31598eb0b.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ia44bf70cb550e4840e3be1be7c8b2b0bea0a330e
This reverts commit ff1561ac7f.
It is no longer needed as we are able to update the abi at this point in
time.
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I20acdca0aa5b59436e890887c03f125f027a9d45
The list_for_each_entry_safe() macro saves the current item (n) and
the item after (n+1), so that n can be safely removed without
corrupting the list. However, when traversing the list and removing
items using gadget giveback, the DWC3 lock is briefly released,
allowing other routines to execute. There is a situation where, while
items are being removed from the cancelled_list using
dwc3_gadget_ep_cleanup_cancelled_requests(), the pullup disable
routine is running in parallel (due to UDC unbind). As the cleanup
routine removes n, and the pullup disable removes n+1, once the
cleanup retakes the DWC3 lock, it references a request who was already
removed/handled. With list debug enabled, this leads to a panic.
Ensure all instances of the macro are replaced where gadget giveback
is used.
Example call stack:
Thread#1:
__dwc3_gadget_ep_set_halt() - CLEAR HALT
-> dwc3_gadget_ep_cleanup_cancelled_requests()
->list_for_each_entry_safe()
->dwc3_gadget_giveback(n)
->dwc3_gadget_del_and_unmap_request()- n deleted[cancelled_list]
->spin_unlock
->Thread#2 executes
...
->dwc3_gadget_giveback(n+1)
->Already removed!
Thread#2:
dwc3_gadget_pullup()
->waiting for dwc3 spin_lock
...
->Thread#1 released lock
->dwc3_stop_active_transfers()
->dwc3_remove_requests()
->fetches n+1 item from cancelled_list (n removed by Thread#1)
->dwc3_gadget_giveback()
->dwc3_gadget_del_and_unmap_request()- n+1 deleted[cancelled_list]
->spin_unlock
Fixes: d4f1afe5e8 ("usb: dwc3: gadget: move requests to cancelled_list")
Signed-off-by: Wesley Cheng <quic_wcheng@quicinc.com>
Link: https://lore.kernel.org/r/20220414183521.23451-1-quic_wcheng@quicinc.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Bug: 230843866
(cherry picked from commit bf594d1d0c)
Change-Id: I357028fb51441a7effe90d115f6faa3f2034d326
Signed-off-by: Wesley Cheng <quic_wcheng@quicinc.com>
If a Setup packet is received but yet to DMA out, the controller will
not process the End Transfer command of any endpoint. Polling of its
DEPCMD.CmdAct may block setting up TRB for Setup packet, causing a
command timeout.
This may occur if the driver doesn’t service the completion interrupt of
the control status stage yet due to system latency, then it won’t
prepare TRB and start the transfer for the next Setup Stage. To the host
side, the control transfer had completed, and the host can send a new
Setup packet at this point.
In the meanwhile, if the driver receives an async call to dequeue a
request (triggering End Transfer) to any endpoint, then the driver will
service that End transfer first, blocking the control status stage
completion handler. Since no TRB is available for the Setup stage, the
Setup packet can’t be DMA’ed out and the End Transfer gets hung.
The driver must not block setting up of the Setup stage. So track and
only issue the End Transfer command only when there’s Setup TRB prepared
so that the controller can DMA out the Setup packet. Delay the End
transfer command if there's no Setup TRB available. This is applicable to
all DWC_usb3x IPs.
Co-developed-by: Wesley Cheng <quic_wcheng@quicinc.com>
Signed-off-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Signed-off-by: Wesley Cheng <quic_wcheng@quicinc.com>
Link: https://lore.kernel.org/r/20220309205402.4467-1-quic_wcheng@quicinc.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Bug: 230843866
(cherry picked from commit e4cf6580ac)
Change-Id: I7f643ae340a20f12d5f86965674aa58df1be3979
Signed-off-by: Wesley Cheng <quic_wcheng@quicinc.com>
Per Kconfig:
config USB_NET_AX88179_178A
tristate "ASIX AX88179/178A USB 3.0/2.0 to Gigabit Ethernet"
depends on USB_USBNET
select CRC32
select PHYLIB
default y
help
This option adds support for ASIX AX88179 based USB 3.0/2.0
to Gigabit Ethernet adapters.
This driver should work with at least the following devices:
* ASIX AX88179
* ASIX AX88178A
* Sitcomm LN-032
This driver creates an interface named "ethX", where X depends on
what other networking devices you have in use.
This was already enabled on 'db845c_gki.fragment',
which suggests this hardware is reasonably common
(even though I don't have a dongle that requires it).
Test: TreeHugger
Bug: 200269356
Signed-off-by: Maciej Żenczykowski <maze@google.com>
Change-Id: I9915cfb54a324f007d508a8e3d2aad1d6fc9e5de
(cherry picked from commit 3ed683cb94)
When pKVM is enabled, host memory accesses are translated by an identity
mapping at stage-2, which is populated lazily in response to synchronous
exceptions from 64-bit EL1 and EL0.
Extend this handling to cover exceptions originating from 32-bit EL0 as
well. Although these are very unlikely to occur in practice, as the
kernel typically ensures that user pages are initialised before mapping
them in, drivers could still map previously untouched device pages into
userspace and expect things to work rather than panic the system.
Cc: Quentin Perret <qperret@google.com>
Cc: Marc Zyngier <maz@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220427171332.13635-1-will@kernel.org
(cherry picked from commit 2a50fc5fd0
git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm.git fixes)
Bug: 216811181
Signed-off-by: Will Deacon <willdeacon@google.com>
Change-Id: I98ad9d9f0e2a78751ed73cc5d7c481d07a3ed1db
This fixes the kernelci error:
"ERROR: modpost: module configfs uses symbol kern_path from namespace
VFS_internal_I_am_really_a_filesystem_and_am_NOT_a_driver, but does not import it."
Fixes: 0a77fca3aa ("ANDROID: GKI: set vfs-only exports into their own namespace")
Signed-off-by: Todd Kjos <tkjos@google.com>
Change-Id: Ib4ab1b83c8c8c996b1f15c419fb8ce0549832699
Our test build is broken by KCFLAGS overrided in build.config.comm.
Since Linux Makefile supports 'export KCFLAGS=XXX' to customize the
KCFLAGS, and we should keep this functionality.
Bug: 230818006
Fixes: 6c55ca2cae ("ANDROID: Add flag to indicate compiling against ACK")
Signed-off-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
Change-Id: I9425d79697bc1fe816ce82d523f91631dee6b8f4
When ufs initializes without scmd->device->sector_size set, scsi_get_lba()
will get a wrong shift number and trigger an ubsan error. The shift
exponent 4294967286 is too large for the 64-bit type 'sector_t' (aka
'unsigned long long').
Call scsi_get_lba() only when opcode is READ_10/WRITE_10/UNMAP.
Link: https://lore.kernel.org/r/20220307111752.10465-1-peter.wang@mediatek.com
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
(cherry picked from commit 2bd3b6b759 git://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi.git for-next)
Bug: 228405696
Change-Id: I5bdf04628eff2a0c2a36c89f9934033fb3d8a392
Signed-off-by: Bart Van Assche <bvanassche@google.com>
Signed-off-by: Yee Lee <yee.lee@mediatek.com>
Add a flag: __ANDROID_COMMON_KERNEL__ which out-of-tree vendor drivers
can use to check if they are compiling against an Android Common Kernel.
These out-of-tree vendor drivers can use this flag +
LINUX_KERNEL_VERSION to determine if a feature has been backported.
Bug: 229953929
Change-Id: I832344d63f3639479784753edfb7ac405068312f
Signed-off-by: Elliot Berman <quic_eberman@quicinc.com>
There are two tracepoints in usb_gadget_connect() and
usb_gadget_disconnect(). This patch will export the tracepoints so that
vendor modules can use them.
Bug: 189130101
Change-Id: I73ace6ad7a29a835431879162cb5e5ff3d6b2239
Signed-off-by: fengmingli <mingli.feng@vivo.com>
(cherry picked from commit 6f33bce13a)
The rt sync wakeup support has a condition which relies on a field that
exists only when CONFIG_SMP is defined, causing a compilation issue.
Since sync wakeup has no real meaning on a non-SMP system, we can just
drop the CONFIG_RT_GROUP_SCHED part of the #ifdef.
Fixes: da5f3cd378 ("ANDROID: sched/rt: Add support for rt sync wakeups")
Signed-off-by: J. Avila <elavila@google.com>
Change-Id: I9b95304408d323b0c1017bd33746ecfbb2b35808
Some rt tasks undergo sync wakeup. Currently, these tasks will be placed
on other, often sleeping or otherwise idle CPUs, which can lead to
unnecessary power hits.
Bug: 157906395
Change-Id: I48864d0847bbe4f7813c842032880ad3f3b8b06b
Signed-off-by: J. Avila <elavila@google.com>
[quic_dickey@quicinc.com: Port to 5.15]
Signed-off-by: Stephen Dickey <quic_dickey@quicinc.com>
Vendor kernel modules may implement irq balancers, which could
take irq desc lock of an irq and then based on current affinity
mask or affinity hint, reconfigure the affinity of that irq.
For example : For an irq, for which affinity is broken i.e. all
the cpus in its affinity mask have gone offline. For such irqs,
we might want to reset the affinity, when the original set of
affined cpus, come back online. desc->affinity_hint can be used
for figuring out the original affinity. So, the sequence for doing
this becomes:
desc = irq_to_desc(i);
raw_spin_lock(&desc->lock);
affinity = desc->affinity_hint;
raw_spin_unlock(&desc->lock);
irq_set_affinity_hint(i, affinity);
Here, we need to release the desc lock before calling the exported
api irq_set_affinity_hint(). This creates a window where, after
unlocking desc lock and before calling irq_set_affinity_hint(),
where this setting can race with other irq_set_affinity_hint()
callers. So, export irq_do_set_affinity() symbol to provide an
api, which can be called with desc lock held.
Bug: 187157600
Change-Id: Ifad88bfaa1e7eec09c3fe5a9dd7d1d421362b41e
Signed-off-by: Neeraj Upadhyay <neeraju@codeaurora.org>
(cherry picked from commit d88c1e77fd)
GIC provides implementation specific registers, to configure
affinity of a SPI. Update the existing affinity hook to allow
vendors to configure those implementation defined settings.
Bug: 180471389
Change-Id: I273035da65eaeb346c0d8b303a722f4d8d7918d6
Signed-off-by: Neeraj Upadhyay <neeraju@codeaurora.org>
(cherry picked from commit f52f343587)
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Set the MMIO guard flag for protected vms prior to entering the guest
for the first time.
Bug: 216798684
Signed-off-by: Sebastian Ene <sebastianene@google.com>
Change-Id: I1448102ae85176d495ae7f8d6d20de4092049f0d
Export vmalloc_nr_pages which is used as part
of meminfo collection from minidump module.
Bug: 199236943
Change-Id: I4c80fe2a0712658ec46b49064fda670da84b3732
Signed-off-by: Vijayanand Jitta <quic_vjitta@quicinc.com>
Export pcpu_nr_pages symbol which is used as part
of meminfo collection from minidump module.
Bug: 199236943
Change-Id: I08262ec95a3f1be8322b9b8d2d9c4098518fc408
Signed-off-by: Vijayanand Jitta <quic_vjitta@quicinc.com>
If the file preallocated blocks and fsync'ed, we should not truncate them during
roll-forward recovery which will recover i_size correctly back.
Bug: 223740163
Fixes: d4dd19ec1e ("f2fs: do not expose unwritten blocks to user by DIO")
Cc: <stable@vger.kernel.org> # 5.17+
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
(cherry picked from commit 4d8ec91208)
Change-Id: I5e974a564667115455b53a18f31902a875d86dee
In commit f8bd6cf70d ("ANDROID: GKI: Add module load time protected
symbol lookup") the kernel/gki_module_exported.h and
kernel/gki_module_protected.h files are created, but these generated
files are not added to the .gitignore file, making them show up as added
files when building the tree.
Resolve this by adding them to the proper .gitignore file
Bug: 200082547
Fixes: f8bd6cf70d ("ANDROID: GKI: Add module load time protected symbol lookup")
Cc: Ramji Jiyani <ramjiyani@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I906ddd24bfc54d62f572fba5491e2d2025325957
To use the tracepoint in kernel module, add EXPORT_TRACE_SYMBOL_GPL to
export the dwc3_ctrl_req tracepoint
Bug: 189130101
Change-Id: Ie3245474fbd0cc18c6d41036dcf17c7bbe460814
Signed-off-by: fengmingli <mingli.feng@vivo.com>
(cherry picked from commit 0c811ab6f7)
