Instead of calling into kernel_neon_end() and kernel_neon_begin() (and
potentially into schedule()) from the assembler code when running in
task mode and a reschedule is pending, perform only the preempt count
check in assembler, but simply return early in this case, and let the C
code deal with the consequences.
This reverts commit 7edc86cb1c.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
(cherry picked from commit 9ecc9f31d0)
Bug: 153614920
Change-Id: Iac9f43058b687182fe1bf7f47efb6b8aca128c29
Signed-off-by: Ard Biesheuvel <ardb@google.com>
Instead of calling into kernel_neon_end() and kernel_neon_begin() (and
potentially into schedule()) from the assembler code when running in
task mode and a reschedule is pending, perform only the preempt count
check in assembler, but simply return early in this case, and let the C
code deal with the consequences.
This reverts commit d82f37ab5e.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
(cherry picked from commit b2eadbf40e)
Bug: 153614920
Change-Id: Ie3f66f835bec71b29bd42fbeb9b32d36d1e0e7e4
Signed-off-by: Ard Biesheuvel <ardb@google.com>
Instead of calling into kernel_neon_end() and kernel_neon_begin() (and
potentially into schedule()) from the assembler code when running in
task mode and a reschedule is pending, perform only the preempt count
check in assembler, but simply return early in this case, and let the C
code deal with the consequences.
This reverts commit 7df8d16475.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
(cherry picked from commit 5a69e1b73d)
Bug: 153614920
Change-Id: I1393f9c5b73316ba68ca64c975a3790c31793b4a
Signed-off-by: Ard Biesheuvel <ardb@google.com>
Add a macro cond_yield that branches to a specified label when called if
the TIF_NEED_RESCHED flag is set and decreasing the preempt count would
make the task preemptible again, resulting in a schedule to occur. This
can be used by kernel mode SIMD code that keeps a lot of state in SIMD
registers, which would make chunking the input in order to perform the
cond_resched() check from C code disproportionately costly.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lore.kernel.org/r/20210203113626.220151-2-ardb@kernel.org
Signed-off-by: Will Deacon <will@kernel.org>
(cherry picked from commit d13c613f13)
Bug: 153614920
Change-Id: I066dbeef516a023c384689adaddac2047da113e9
Signed-off-by: Ard Biesheuvel <ardb@google.com>
The cipher routines in the crypto API are mostly intended for templates
implementing skcipher modes generically in software, and shouldn't be
used outside of the crypto subsystem. So move the prototypes and all
related definitions to a new header file under include/crypto/internal.
Also, let's use the new module namespace feature to move the symbol
exports into a new namespace CRYPTO_INTERNAL.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Acked-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
(cherry picked from commit 0eb76ba29d)
Bug: 153614920
Change-Id: I0ed1c7a197611f020a21d489cd15a09a650bb1e6
Signed-off-by: Ard Biesheuvel <ardb@google.com>
Allocating a cipher via the crypto API only to free it again after using
it to encrypt a single block is unnecessary in cases where the algorithm
is known at compile time. So replace this pattern with a call to the AES
library.
Cc: Ayush Sawal <ayush.sawal@chelsio.com>
Cc: Vinay Kumar Yadav <vinay.yadav@chelsio.com>
Cc: Rohit Maheshwari <rohitm@chelsio.com>
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
(cherry picked from commit a3b01ffddc)
Change-Id: I758bba01b655391044a4efbe768ca68a5c0d54e1
Signed-off-by: Ard Biesheuvel <ardb@google.com>
This was generated with
$ build/synchronize_owners common-mainline/ android-mainline common12-5.10/
Bug: 181986413
Signed-off-by: Jaegeuk Kim <jaegeuk@google.com>
Change-Id: I1416402cc4e100b950473d9a0277c34f78857638
When offlining memory the system can attempt to migrate a lot of pages, if
there are problems with migration this can flood the logs. Printing all
the data hogs the CPU and cause some RT threads to run for a long time,
which may have some bad consequences.
Rate limit the page migration warnings in order to avoid this.
Link: https://lkml.kernel.org/r/20210505140542.24935-1-georgi.djakov@linaro.org
Signed-off-by: Liam Mark <lmark@codeaurora.org>
Signed-off-by: Georgi Djakov <georgi.djakov@linaro.org>
Cc: David Hildenbrand <david@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Bug: 187917022
(cherry picked from commit 6be9af0e0f77 https: //github.com/hnaz/linux-mm.git v5.13-rc1-mmots-2021-05-10-22-15)
Change-Id: I6754d380f8c31cf9b699c242edacf593807ef2b1
Signed-off-by: Chris Goldsworthy <quic_cgoldswo@quicinc.com>
This adds QCOM_TSENSE to the db845c_gki.fragment which is
needed to avoid hard crashes seen on db845c (which crash into
USB dump mode). This was causing some troube in our testing
as we'd occasionally see hard crashes in certian CTS tests
(also reproducible with a forkbomb).
After lots of debugging we've narrowed down this config as
the change which avoids the hard crashes.
Thanks to: Vinod Koul <vinod.koul@linaro.org> and
Bjorn Andersson <bjorn.andersson@linaro.org> for helping debug
this issue!
Bug: 146449535
Reported-by: YongQin Liu <yongqin.liu@linaro.org>
Signed-off-by: John Stultz <john.stultz@linaro.org>
Change-Id: I0752c85e44028e6272c14b59fa09d2baf2e18feb
Currently user can configure UAC2 function with
parameters that violate UAC2 spec or are not supported
by UAC2 gadget implementation.
This can lead to incorrect behavior if such gadget
is connected to the host - like enumeration failure
or other issues depending on host's UAC2 driver
implementation, bringing user to a long hours
of debugging the issue.
Instead of silently accept these parameters, throw
an error if they are not valid.
Signed-off-by: Ruslan Bilovol <ruslan.bilovol@gmail.com>
Link: https://lore.kernel.org/r/1614599375-8803-4-git-send-email-ruslan.bilovol@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Bug: 187332233
(cherry picked from commit 3713d5ceb0)
Change-Id: Icc684c620e039f7dbb3591e5669506f6f69574d6
Signed-off-by: Jack Pham <jackp@codeaurora.org>
Commit f7128c7533 ("ANDROID: sched: Add vendor hooks for update_topology")
was intended for the modules to get informed when topology
initialization had completed.
However that is not sufficient, the module could be insmoded after
the hook was invoked.
For such situations create a flag for modules to detect that topology
initialization is completed.
Bug: 187234873
Change-Id: I2ed4dff9c921e3de67dd11049c80f38e3c1a64fd
Signed-off-by: Abhijeet Dharmapurikar <adharmap@codeaurora.org>
With CONFIG_LTO_CLANG_FULL, LLVM drops all CFI jump table symbols
from vmlinux, which doesn't affect kernel functionality, but can
make stack traces and other kernel output that prints out jump
table addresses harder to read.
This change works around the issue for now by adding a script that
tells kallsyms about the missing jump table symbols, even though
they don't actually exist in the symbol table, and generates a
linker script to add the missing symbols to kernel modules.
Bug: 186152035
Bug: 187415564
Change-Id: Ic3c51751c756f2f5fb2a31229e16c3397eb6e666
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Currently, most of the thermal_zones are IRQ capable and they do not need
to be updated while resuming. To improve the system performance and reduce
the resume time. Add a vendor function to check if the thermal_zone is
not IRQ capable and needs to be updated.
Bug: 170905417
Test: boot and vendor function worked properly.
Change-Id: I9389985bba29b551a7a20b55e1ed26b6c4da9b3d
Signed-off-by: David Chao <davidchao@google.com>
Currently, when reading the following properties under
ucsi-source-psy-*, src_pdos array is accessed but it only checks
for the condition num_pdos > 0.
- voltage_max
- voltage_now
- current_max
In the case if a PPM passes num_pdos greater than UCSI_MAX_PDOS,
an OOB access can be made. Fix it by checking the upper bound for
src_pdos which is UCSI_MAX_PDOS.
This issue is detected by enabling UBSAN.
Bug: 186400956
Fixes: 992a60ed0d ("usb: typec: ucsi: register with power_supply class")
Fixes: 4dbc6a4ef0 ("usb: typec: ucsi: save power data objects in PD mode")
Change-Id: Id0aca674208b23329be089671cba873baca930cf
Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com>
Not_Supported Message is acceptable in VDM AMS. Redirect the VDM state
machine to VDM_STATE_DONE when receiving Not_Supported and finish the
VDM AMS.
Also, after the loop in vdm_state_machine_work, add more conditions of
VDM states to clear the vdm_sm_running flag because those are all
stopping states when leaving the loop.
In addition, finish the VDM AMS if the port partner responds BUSY.
Fixes: 8dea75e113 ("usb: typec: tcpm: Protocol Error handling")
Fixes: 8d3a0578ad ("usb: typec: tcpm: Respond Wait if VDM state machine is running")
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Kyle Tso <kyletso@google.com>
Link: https://lore.kernel.org/r/20210507062300.1945009-3-kyletso@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit f1fbd950b5https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git usb-linus)
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ia52ad69f054b2211924ec3815b5a3d2cc09d8b88
When extcon is used in combination with dwc3, it is assumed that the dwc3
registers are untouched and as such are only configured if VBUS is valid
or ID is tied to ground.
In case VBUS is not valid or ID is floating, the registers are not
configured as such during driver initialization, causing a wrong
default state during boot.
If the registers are not in a default state, because they are for
instance touched by a boot loader, this can cause for a kernel error.
Signed-off-by: Marcel Hamer <marcel@solidxs.se>
Link: https://lore.kernel.org/r/20210427122118.1948340-1-marcel@solidxs.se
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit e17b02d497https://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git usb-linus)
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ic460a111f59e5a4fd77db0402fb297e53bbc1d9e
As part of commit e81a7018d9 ("usb: dwc3: allocate gadget structure
dynamically") the dwc3_gadget_release() was added which will free
the dwc->gadget structure upon the device's removal when
usb_del_gadget_udc() is called in dwc3_gadget_exit().
However, simply freeing the gadget results a dangling pointer
situation: the endpoints created in dwc3_gadget_init_endpoints()
have their dep->endpoint.ep_list members chained off the list_head
anchored at dwc->gadget->ep_list. Thus when dwc->gadget is freed,
the first dwc3_ep in the list now has a dangling prev pointer and
likewise for the next pointer of the dwc3_ep at the tail of the list.
The dwc3_gadget_free_endpoints() that follows will result in a
use-after-free when it calls list_del().
This was caught by enabling KASAN and performing a driver unbind.
The recent commit 568262bf54 ("usb: dwc3: core: Add shutdown
callback for dwc3") also exposes this as a panic during shutdown.
There are a few possibilities to fix this. One could be to perform
a list_del() of the gadget->ep_list itself which removes it from
the rest of the dwc3_ep chain.
Another approach is what this patch does, by splitting up the
usb_del_gadget_udc() call into its separate "del" and "put"
components. This allows dwc3_gadget_free_endpoints() to be
called before the gadget is finally freed with usb_put_gadget().
Fixes: e81a7018d9 ("usb: dwc3: allocate gadget structure dynamically")
Reviewed-by: Peter Chen <peter.chen@kernel.org>
Signed-off-by: Jack Pham <jackp@codeaurora.org>
Link: https://lore.kernel.org/r/20210501093558.7375-1-jackp@codeaurora.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit bb9c74a5bdhttps://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git usb-linus)
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I0062f1ccdda604938d8e2f68de4ae27360397d9f
The dwc2 gadget support maps and unmaps DMA buffers as necessary. When
mapping and unmapping it uses the direction of the endpoint to select
the direction of the DMA transfer, but this fails for Control OUT
transfers because the unmap occurs after the endpoint direction has
been reversed for the status phase.
A possible solution would be to unmap the buffer before the direction
is changed, but a safer, less invasive fix is to remember the buffer
direction independently of the endpoint direction.
Fixes: fe0b94abcd ("usb: dwc2: gadget: manage ep0 state in software")
Acked-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Phil Elwell <phil@raspberrypi.com>
Link: https://lore.kernel.org/r/20210506112200.2893922-1-phil@raspberrypi.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 75a41ce46bhttps://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git usb-linus)
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I1cdc8528842a33cadb0e05852b95670360db8ad3
The device event corresponding to End of Periodic Frame is only
found on older IP revisions (2.10a and prior, according to a
cursory SNPS databook search). On revisions 2.30a and newer,
including DWC3.1, the same event value and corresponding DEVTEN
bit were repurposed to indicate that the link has gone into
suspend state (U3 or L2/L1).
EOPF events had never been enabled before in this driver, and
going forward we expect current and future DWC3-based devices
won't likely to be using such old DWC3 IP revisions either.
Hence rather than keeping the deprecated EOPF macro names let's
rename them to indicate their usage for suspend events.
Acked-by: Felipe Balbi <balbi@kernel.org>
Signed-off-by: Jack Pham <jackp@codeaurora.org>
Link: https://lore.kernel.org/r/20210428090111.3370-2-jackp@codeaurora.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
(cherry picked from commit 6f26ebb79ahttps://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb.git usb-linus)
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I8351dd713ce0e60b2212a22a3b51d1139ae89779
This reverts commit fb4c1c2e9f.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I7b12040440bc2bedfbabc71d60339cb843e59570
This reverts commit 2e8b3b0b8e.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I03a41fc3ba455ded7274f2a6b5566bb369a75a86
This reverts commit 22163a8ec8.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ie256b7f1ed5b19f4600ff4c3681efaea5046adda
This reverts commit 1bbcc985d1.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I88427586de4df897c39bb3d48e8cba86ebf5c4bb
This reverts commit 1f2ef5a0f7.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I461f44ec54102dcf06ba73959a79172615cdb8fa
This reverts commit 9efd5df078.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ia22b608e4d017c6056e481d2ec819074b35810cf
This reverts commit 25ed8827cf.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I9928cdc06a26b9db3052b93067c557011e53f735
This reverts commit 85a5a6875c.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Ibe7639f73acc11b356f6b2f8f76011844c7f451b
This reverts commit f8e71c667e.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I6684b40d913a86b79899220cba823a74aea5923f
This reverts commit 2fa0387fa2.
Fixes the ABI issues in 5.10.35 that at the moment, we can't handle due
to the KABI freeze. These are not patches that mean much for android
systems, and will be reverted the next KABI "reset" point.
Bug: 161946584
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: If6aba0f06d169907827ca1f76b52adc75c84c124
Changes in 5.10.35
mips: Do not include hi and lo in clobber list for R6
netfilter: conntrack: Make global sysctls readonly in non-init netns
net: usb: ax88179_178a: initialize local variables before use
igb: Enable RSS for Intel I211 Ethernet Controller
bpf: Fix masking negation logic upon negative dst register
bpf: Fix leakage of uninitialized bpf stack under speculation
net: qrtr: Avoid potential use after free in MHI send
perf data: Fix error return code in perf_data__create_dir()
capabilities: require CAP_SETFCAP to map uid 0
perf ftrace: Fix access to pid in array when setting a pid filter
tools/cgroup/slabinfo.py: updated to work on current kernel
driver core: add a min_align_mask field to struct device_dma_parameters
swiotlb: add a IO_TLB_SIZE define
swiotlb: factor out an io_tlb_offset helper
swiotlb: factor out a nr_slots helper
swiotlb: clean up swiotlb_tbl_unmap_single
swiotlb: refactor swiotlb_tbl_map_single
swiotlb: don't modify orig_addr in swiotlb_tbl_sync_single
swiotlb: respect min_align_mask
nvme-pci: set min_align_mask
ovl: fix leaked dentry
ovl: allow upperdir inside lowerdir
ALSA: usb-audio: Add MIDI quirk for Vox ToneLab EX
USB: Add LPM quirk for Lenovo ThinkPad USB-C Dock Gen2 Ethernet
USB: Add reset-resume quirk for WD19's Realtek Hub
platform/x86: thinkpad_acpi: Correct thermal sensor allocation
perf/core: Fix unconditional security_locked_down() call
vfio: Depend on MMU
Linux 5.10.35
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: Iff7d5abe7b821f453bbe4d9dad94dfd35fe0a082
The inflight of partition 0 doesn't include inflight IOs to all
sub-partitions, since currently mq calculates inflight of specific
partition by simply camparing the value of the partition pointer.
Thus the following case is possible:
$ cat /sys/block/vda/inflight
0 0
$ cat /sys/block/vda/vda1/inflight
0 128
While single queue device (on a previous version, e.g. v3.10) has no
this issue:
$cat /sys/block/sda/sda3/inflight
0 33
$cat /sys/block/sda/inflight
0 33
Partition 0 should be specially handled since it represents the whole
disk. This issue is introduced since commit bf0ddaba65 ("blk-mq: fix
sysfs inflight counter").
Besides, this patch can also fix the inflight statistics of part 0 in
/proc/diskstats. Before this patch, the inflight statistics of part 0
doesn't include that of sub partitions. (I have marked the 'inflight'
field with asterisk.)
$cat /proc/diskstats
259 0 nvme0n1 45974469 0 367814768 6445794 1 0 1 0 *0* 111062 6445794 0 0 0 0 0 0
259 2 nvme0n1p1 45974058 0 367797952 6445727 0 0 0 0 *33* 111001 6445727 0 0 0 0 0 0
This is introduced since commit f299b7c7a9 ("blk-mq: provide internal
in-flight variant").
Fixes: bf0ddaba65 ("blk-mq: fix sysfs inflight counter")
Fixes: f299b7c7a9 ("blk-mq: provide internal in-flight variant")
Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
[axboe: adapt for 5.11 partition change]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bug: 187355247
Change-Id: I378b2cb7312a5e47d5e2ec7301dc392e6e7336d0
(cherry picked from commit b0d97557eb)
Signed-off-by: Bart Van Assche <bvanassche@google.com>
Set KMI_GENERATION=4 for 5/7 KMI update
Leaf changes summary: 2821 artifacts changed
Changed leaf types summary: 17 leaf types changed
Removed/Changed/Added functions summary: 1 Removed, 2752 Changed, 1 Added function
Removed/Changed/Added variables summary: 0 Removed, 49 Changed, 1 Added variable
1 Removed function:
[D] 'function file* filp_open(const char*, int, umode_t)'
1 Added function:
[A] 'function int __traceiter_android_vh_of_i2c_get_board_info(void*, device_node*, const char**)'
2752 functions with some sub-type change:
[C] 'function void* PDE_DATA(const inode*)' at generic.c:799:1 has some sub-type changes:
CRC (modversions) changed from 0x53dbf80f to 0x27c5b1f0
[C] 'function void __ClearPageMovable(page*)' at compaction.c:138:1 has some sub-type changes:
CRC (modversions) changed from 0xc7c1f13e to 0x22571ed2
[C] 'function void __SetPageMovable(page*, address_space*)' at compaction.c:130:1 has some sub-type changes:
CRC (modversions) changed from 0x3b98a825 to 0x5f77584f
... 2749 omitted; 2752 symbols have only CRC changes
1 Added variable:
[A] 'tracepoint __tracepoint_android_vh_of_i2c_get_board_info'
49 Changed variables:
[C] 'bus_type amba_bustype' was changed at bus.c:215:1:
CRC (modversions) changed from 0x33a71197 to 0xabc9cec3
[C] 'const address_space_operations balloon_aops' was changed at balloon_compaction.c:253:1:
CRC (modversions) changed from 0x50598639 to 0xb8eca1a2
[C] 'const clk_ops clk_fixed_factor_ops' was changed at clk-fixed-factor.c:60:1:
CRC (modversions) changed from 0xaa612693 to 0xdbcf0d4e
... 46 omitted; 49 symbols have only CRC changes
'enum android_debug_symbol at android_debug_symbols.h:9:1' changed:
type size hasn't changed
7 enumerator insertions:
'android_debug_symbol::ADS_TOTAL_CMA' value '7'
'android_debug_symbol::ADS_SLAB_CACHES' value '8'
'android_debug_symbol::ADS_SLAB_MUTEX' value '9'
'android_debug_symbol::ADS_MIN_LOW_PFN' value '10'
'android_debug_symbol::ADS_MAX_PFN' value '11'
'android_debug_symbol::ADS_PAGE_OWNER_ENABLED' value '12'
'android_debug_symbol::ADS_SLUB_DEBUG' value '13'
1 enumerator change:
'android_debug_symbol::ADS_END' from value '7' to '14' at android_debug_symbols.h:9:1
one impacted interface
'enum binder_work_type at binder_internal.h:174:1' changed:
type size hasn't changed
1 enumerator insertion:
'binder_work_type::BINDER_WORK_TRANSACTION_ONEWAY_SPAM_SUSPECT' value '3'
5 enumerator changes:
'binder_work_type::BINDER_WORK_RETURN_ERROR' from value '3' to '4' at binder_internal.h:174:1
'binder_work_type::BINDER_WORK_NODE' from value '4' to '5' at binder_internal.h:174:1
'binder_work_type::BINDER_WORK_DEAD_BINDER' from value '5' to '6' at binder_internal.h:174:1
'binder_work_type::BINDER_WORK_DEAD_BINDER_AND_CLEAR' from value '6' to '7' at binder_internal.h:174:1
'binder_work_type::BINDER_WORK_CLEAR_DEATH_NOTIFICATION' from value '7' to '8' at binder_internal.h:174:1
5 impacted interfaces
'struct binder_alloc at binder_alloc.h:96:1' changed:
type size changed from 1024 to 1088 (in bits)
1 data member insertion:
'bool oneway_spam_detected', at offset 1024 (in bits) at binder_alloc.h:115:1
5 impacted interfaces
'struct binder_buffer at binder_alloc.h:40:1' changed:
type size hasn't changed
1 data member insertion:
'unsigned int oneway_spam_suspect', at offset 4 (in bits) at binder_alloc.h:50:1
there are data member changes:
'unsigned int debug_id' offset changed (by +1 bits)
4 impacted interfaces
'struct binder_proc at binder_internal.h:436:1' changed:
type size changed from 4480 to 4672 (in bits)
1 data member insertion:
'bool oneway_spam_detection_enabled', at offset 4608 (in bits) at binder_internal.h:471:1
there are data member changes:
type 'struct binder_stats' of 'binder_proc::stats' changed:
type size changed from 1664 to 1696 (in bits)
there are data member changes:
type 'atomic_t[19]' of 'binder_stats::br' changed:
type name changed from 'atomic_t[19]' to 'atomic_t[20]'
array type size changed from 608 to 640
array type subrange 1 changed length from 19 to 20
3 ('atomic_t bc[19]' .. 'atomic_t obj_deleted[7]') offsets changed (by +32 bits)
5 impacted interfaces
7 ('list_head delivered_death' .. 'dentry* debugfs_entry') offsets changed (by +64 bits)
type 'struct binder_alloc' of 'binder_proc::alloc' changed, as reported earlier
and offset changed from 3264 to 3328 (in bits) (by +64 bits)
4 ('binder_context* context' .. 'dentry* binderfs_entry') offsets changed (by +128 bits)
5 impacted interfaces
'struct binder_stats at binder_internal.h:157:1' changed:
details were reported earlier
'struct binder_thread at binder_internal.h:507:1' changed (indirectly):
type size changed from 3264 to 3328 (in bits)
there are data member changes:
type 'struct binder_stats' of 'binder_thread::stats' changed, as reported earlier
2 ('atomic_t tmp_ref' .. 'bool is_dead') offsets changed (by +32 bits)
'task_struct* task' offset changed (by +64 bits)
4 impacted interfaces
'struct dwc3 at core.h:1080:1' changed:
type size changed from 9216 to 9984 (in bits)
5 data member insertions:
'mutex mutex', at offset 2624 (in bits) at core.h:1115:1
'u64 android_kabi_reserved1', at offset 9728 (in bits) at core.h:1302:1
'u64 android_kabi_reserved2', at offset 9792 (in bits) at core.h:1303:1
'u64 android_kabi_reserved3', at offset 9856 (in bits) at core.h:1304:1
'u64 android_kabi_reserved4', at offset 9920 (in bits) at core.h:1305:1
there are data member changes:
type 'struct dwc3_request' of 'dwc3::ep0_usb_req' changed:
type size changed from 1472 to 1600 (in bits)
2 data member insertions:
'u64 android_kabi_reserved1', at offset 1472 (in bits) at core.h:947:1
'u64 android_kabi_reserved2', at offset 1536 (in bits) at core.h:948:1
2 impacted interfaces
2 ('completion ep0_in_setup' .. 'spinlock_t lock') offsets changed (by +128 bits)
53 ('device* dev' .. 'u8 num_eps') offsets changed (by +384 bits)
type 'struct dwc3_hwparams' of 'dwc3::hwparams' changed:
type size changed from 288 to 448 (in bits)
3 data member insertions:
'u32 hwparams9', at offset 288 (in bits) at core.h:872:1
'u64 android_kabi_reserved1', at offset 320 (in bits) at core.h:874:1
'u64 android_kabi_reserved2', at offset 384 (in bits) at core.h:875:1
2 impacted interfaces
and offset changed from 8512 to 8896 (in bits) (by +384 bits)
13 ('dentry* root' .. 'u16 imod_interval') offsets changed (by +512 bits)
2 impacted interfaces
'struct dwc3_ep at core.h:696:1' changed:
type size changed from 1664 to 1792 (in bits)
2 data member insertions:
'u64 android_kabi_reserved1', at offset 1664 (in bits) at core.h:761:1
'u64 android_kabi_reserved2', at offset 1728 (in bits) at core.h:762:1
2 impacted interfaces
'struct dwc3_event_buffer at core.h:646:1' changed:
type size changed from 384 to 448 (in bits)
1 data member insertion:
'u64 android_kabi_reserved1', at offset 384 (in bits) at core.h:666:1
2 impacted interfaces
'struct dwc3_hwparams at core.h:851:1' changed:
details were reported earlier
'struct dwc3_request at core.h:902:1' changed:
details were reported earlier
'struct mod_arch_specific at module.h:17:1' changed:
type size changed from 256 to 448 (in bits)
4 data member insertions:
'const Elf64_Rela* text_relocations', at offset 256 (in bits) at module.h:25:1
'const Elf64_Rela* rodata_relocations', at offset 320 (in bits) at module.h:26:1
'int num_text_relocations', at offset 384 (in bits) at module.h:27:1
'int num_rodata_relocations', at offset 416 (in bits) at module.h:28:1
2074 impacted interfaces
'struct module at module.h:367:1' changed (indirectly):
type size hasn't changed
there are data member changes:
type 'struct mod_arch_specific' of 'module::arch' changed, as reported earlier
41 ('unsigned long int taints' .. 'u64 android_kabi_reserved4') offsets changed (by +192 bits)
2074 impacted interfaces
'struct queue_limits at blkdev.h:318:1' changed:
type size changed from 896 to 960 (in bits)
1 data member insertion:
'unsigned int bio_max_bytes', at offset 0 (in bits) at blkdev.h:319:1
there are data member changes:
27 ('unsigned long int bounce_pfn' .. 'blk_zoned_model zoned') offsets changed (by +64 bits)
2075 impacted interfaces
'struct request_queue at blkdev.h:396:1' changed (indirectly):
type size changed from 15104 to 15168 (in bits)
there are data member changes:
type 'struct queue_limits' of 'request_queue::limits' changed, as reported earlier
27 ('unsigned int required_elevator_features' .. 'u64 write_hints[5]') offsets changed (by +64 bits)
2074 impacted interfaces
'struct rproc_ops at remoteproc.h:380:1' changed:
type size hasn't changed
there are data member changes:
type 'void* (rproc*, typedef u64, typedef size_t)*' of 'rproc_ops::da_to_va' changed:
pointer type changed from: 'void* (rproc*, typedef u64, typedef size_t)*' to: 'void* (rproc*, typedef u64, typedef size_t, bool*)*'
17 impacted interfaces
Bug: 187507598
Signed-off-by: Todd Kjos <tkjos@google.com>
Change-Id: I70911d8a70cb00828dcc6ce601e97a5a4ef1eace
bio size can grow up to 4GB when muli-page bvec is enabled.
but sometimes it would lead to inefficient behaviors.
in case of large chunk direct I/O, - 32MB chunk read in user space -
all pages for 32MB would be merged to a bio structure if the pages
physical addresses are contiguous. it makes some delay to submit
until merge complete. bio max size should be limited to a proper size.
When 32MB chunk read with direct I/O option is coming from userspace,
kernel behavior is below now in do_direct_IO() loop. it's timeline.
| bio merge for 32MB. total 8,192 pages are merged.
| total elapsed time is over 2ms.
|------------------ ... ----------------------->|
| 8,192 pages merged a bio.
| at this time, first bio submit is done.
| 1 bio is split to 32 read request and issue.
|--------------->
|--------------->
|--------------->
......
|--------------->
|--------------->|
total 19ms elapsed to complete 32MB read done from device. |
If bio max size is limited with 1MB, behavior is changed below.
| bio merge for 1MB. 256 pages are merged for each bio.
| total 32 bio will be made.
| total elapsed time is over 2ms. it's same.
| but, first bio submit timing is fast. about 100us.
|--->|--->|--->|---> ... -->|--->|--->|--->|--->|
| 256 pages merged a bio.
| at this time, first bio submit is done.
| and 1 read request is issued for 1 bio.
|--------------->
|--------------->
|--------------->
......
|--------------->
|--------------->|
total 17ms elapsed to complete 32MB read done from device. |
As a result, read request issue timing is faster if bio max size is limited.
Current kernel behavior with multipage bvec, super large bio can be created.
And it lead to delay first I/O request issue.
Signed-off-by: Changheun Lee <nanich.lee@samsung.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Link: https://lore.kernel.org/r/20210503095203.29076-1-nanich.lee@samsung.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Bug: 182716953
(cherry picked from commit cd2c7545ae)
Change-Id: Ie3876daa495535dc7f856ed9a281e65d72a437c1
Signed-off-by: Bart Van Assche <bvanassche@google.com>
"dev_name" for I2C devices won't have any value set, unless they are
instantiated thru ACPI interface. As a result of this, I2C driver will
assign some dynamic name in the format "%d-%04x" (e.g. 1-001f), and
further this device names are used for kernel wakelocks.
This dynamic names are difficult to associate with actual device and
hence it will help to have an ability where vendor can assign the
device name through vendor hooks.
Bug: 147496295
Signed-off-by: Manish Varma <varmam@google.com>
Change-Id: Idb417ef5330002063a9763544f9f0364f9581276
When async binder buffer got exhausted, some normal oneway transactions
will also be discarded and may cause system or application failures. By
that time, the binder debug information we dump may not be relevant to
the root cause. And this issue is difficult to debug if without the
backtrace of the thread sending spam.
This change will send BR_ONEWAY_SPAM_SUSPECT to userspace when oneway
spamming is detected, request to dump current backtrace. Oneway spamming
will be reported only once when exceeding the threshold (target process
dips below 80% of its oneway space, and current process is responsible
for either more than 50 transactions, or more than 50% of the oneway
space). And the detection will restart when the async buffer has
returned to a healthy state.
Acked-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Hang Lu <hangl@codeaurora.org>
Link: https://lore.kernel.org/r/1617961246-4502-3-git-send-email-hangl@codeaurora.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Bug: 181190340
Change-Id: Id3d2526099bc89f04d8ad3ad6e48141b2a8f2515
(cherry picked from commit a7dc1e6f99)
Signed-off-by: Hang Lu <hangl@codeaurora.org>
The fips140.ko module relies on ELF metadata that is normally discarded
by the module loader. In order to reduce the impact of the associated
changes on non-FIPS140 kernels, the module loader will be modified to
simply copy this metadata if it encounters a module whose name is
'fips140'. For this to work, we need a couple of fields in struct
mod_arch_specific so that this module can find this data at module init
time.
Adding these fields will change the type signature of struct module, and
hence affect the stable ABI. The FIPS140 module build depends on LTO,
and so its Kconfig symbol can only be enabled if LTO is enabled as well.
This implies that adding these mod_arch_specific fields conditionally,
based on CONFIG_CRYPTO_FIPS140_INTEGRITY_CHECK, would result in an ABI
fork between LTO and !LTO builds, which is obviously undesirable.
So let's just add these fields unconditionally. This will take 24 bytes
of kernel memory per module, but given that the size of struct module
(which is the only place where struct mod_arch_specific is allocated) is
rounded up to cacheline size (128 bytes on arm64), and this change does
not push it over the edge, the net result is that no additional memory
is used (struct module is currently 1 KiB)
Bug: 153614920
Change-Id: Ia617762f37c0bf7324e899b72fd317e382e39c03
Signed-off-by: Ard Biesheuvel <ardb@google.com>
A number of systems want access to the zram flip_open functionality,
which is now called filp_open_block() to remove any possiblity of bad
drivers trying to access filesystems directly.
Fix up the two symbol lists that were referring to the wrong symbol.
Bug: 157965270
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I368da14ca739b6048dd88a38fca4af8b7f02714d
Try to mitigate potential future driver core api changes by adding
padding to a number of dwc3 structures.
Bug: 151154716
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I1826d885e3acf1f378d0e30b1ca2ddc56b16a9e4
