Masahiro Yamada [Thu, 31 Mar 2022 08:47:09 +0000 (09:47 +0100)]
ARM: 9189/1: decompressor: fix unneeded rebuilds of library objects
Since commit 251cc826be7d ("ARM: 9154/1: decompressor: do not copy source
files while building"), the following three are rebuilt every time.
AS arch/arm/boot/compressed/lib1funcs.o
AS arch/arm/boot/compressed/ashldi3.o
AS arch/arm/boot/compressed/bswapsdi2.o
Move the "OBJS += ..." line up so these objects are added to 'targets'.
Fixes: 251cc826be7d ("ARM: 9154/1: decompressor: do not copy source files while building") Reported-by: Russell King (Oracle) <[email protected]> Signed-off-by: Masahiro Yamada <[email protected]> Signed-off-by: Russell King (Oracle) <[email protected]>
Xiaolong Huang [Wed, 30 Mar 2022 14:22:14 +0000 (15:22 +0100)]
rxrpc: fix some null-ptr-deref bugs in server_key.c
Some function calls are not implemented in rxrpc_no_security, there are
preparse_server_key, free_preparse_server_key and destroy_server_key.
When rxrpc security type is rxrpc_no_security, user can easily trigger a
null-ptr-deref bug via ioctl. So judgment should be added to prevent it
When splitting a value entry, we may need to add the new nodes to the LRU
list and remove the parent node from the LRU list. The WARN_ON checks
in shadow_lru_isolate() catch this oversight. This bug was latent
until we stopped splitting folios in shrink_page_list() with commit 820c4e2e6f51 ("mm/vmscan: Free non-shmem folios without splitting them").
That allows the creation of large shadow entries, and subsequently when
trying to page in a small page, we will split the large shadow entry
in __filemap_add_folio().
Jakob Koschel [Thu, 31 Mar 2022 09:12:18 +0000 (11:12 +0200)]
block: use dedicated list iterator variable
To move the list iterator variable into the list_for_each_entry_*()
macro in the future it should be avoided to use the list iterator
variable after the loop body.
To *never* use the list iterator variable after the loop it was
concluded to use a separate iterator variable instead of a
found boolean [1].
David Howells [Wed, 30 Mar 2022 14:39:16 +0000 (15:39 +0100)]
rxrpc: Fix call timer start racing with call destruction
The rxrpc_call struct has a timer used to handle various timed events
relating to a call. This timer can get started from the packet input
routines that are run in softirq mode with just the RCU read lock held.
Unfortunately, because only the RCU read lock is held - and neither ref or
other lock is taken - the call can start getting destroyed at the same time
a packet comes in addressed to that call. This causes the timer - which
was already stopped - to get restarted. Later, the timer dispatch code may
then oops if the timer got deallocated first.
Fix this by trying to take a ref on the rxrpc_call struct and, if
successful, passing that ref along to the timer. If the timer was already
running, the ref is discarded.
The timer completion routine can then pass the ref along to the call's work
item when it queues it. If the timer or work item where already
queued/running, the extra ref is discarded.
Guangbin Huang [Wed, 30 Mar 2022 13:45:06 +0000 (21:45 +0800)]
net: hns3: fix software vlan talbe of vlan 0 inconsistent with hardware
When user delete vlan 0, as driver will not delete vlan 0 for hardware in
function hclge_set_vlan_filter_hw(), so vlan 0 in software vlan talbe should
not be deleted.
Fixes: fe4144d47eef ("net: hns3: sync VLAN filter entries when kill VLAN ID failed") Signed-off-by: Guangbin Huang <[email protected]> Signed-off-by: Paolo Abeni <[email protected]>
Yufeng Mo [Wed, 30 Mar 2022 13:45:05 +0000 (21:45 +0800)]
net: hns3: fix the concurrency between functions reading debugfs
Currently, the debugfs mechanism is that all functions share a
global variable to save the pointer for obtaining data. When
different functions concurrently access the same file node,
repeated release exceptions occur. Therefore, the granularity
of the pointer for storing the obtained data is adjusted to be
private for each function.
Fixes: 5e69ea7ee2a6 ("net: hns3: refactor the debugfs process") Signed-off-by: Yufeng Mo <[email protected]> Signed-off-by: Guangbin Huang <[email protected]> Signed-off-by: Paolo Abeni <[email protected]>
====================
docs: update and move the netdev-FAQ
A section of documentation for tree-specific process quirks had
been created a while back. There's only one tree in it, so far,
the tip tree, but the contents seem to answer similar questions
as we answer in the netdev-FAQ. Move the netdev-FAQ.
Take this opportunity to touch up and update a few sections.
v3: remove some confrontational? language from patch 7
v2: remove non-git in patch 3
add patch 5
====================
Jakub Kicinski [Wed, 30 Mar 2022 04:24:54 +0000 (21:24 -0700)]
docs: netdev: move the patch marking section up
We want people to mark their patches with net and net-next in the subject.
Many miss doing that. Move the FAQ section which points that out up, and
place it after the section which enumerates the trees, that seems like
a pretty logical place for it. Since the two sections are together we
can remove a little bit (not too much) of the repetition.
v2: also remove the text for non-git setups, we want people to use git.
can: gs_usb: gs_make_candev(): fix memory leak for devices with extended bit timing configuration
Some CAN-FD capable devices offer extended bit timing information for
the data bit timing. The information must be read with an USB control
message. The memory for this message is allocated but not free()ed (in
the non error case). This patch adds the missing free.
Pavel Skripkin [Sun, 13 Mar 2022 10:09:03 +0000 (13:09 +0300)]
can: mcba_usb: properly check endpoint type
Syzbot reported warning in usb_submit_urb() which is caused by wrong
endpoint type. We should check that in endpoint is actually present to
prevent this warning.
Found pipes are now saved to struct mcba_priv and code uses them
directly instead of making pipes in place.
Hangyu Hua [Fri, 11 Mar 2022 08:02:08 +0000 (16:02 +0800)]
can: mcba_usb: mcba_usb_start_xmit(): fix double dev_kfree_skb in error path
There is no need to call dev_kfree_skb() when usb_submit_urb() fails
because can_put_echo_skb() deletes original skb and
can_free_echo_skb() deletes the cloned skb.
Hangyu Hua [Fri, 11 Mar 2022 08:06:14 +0000 (16:06 +0800)]
can: usb_8dev: usb_8dev_start_xmit(): fix double dev_kfree_skb() in error path
There is no need to call dev_kfree_skb() when usb_submit_urb() fails
because can_put_echo_skb() deletes original skb and
can_free_echo_skb() deletes the cloned skb.
Hangyu Hua [Mon, 28 Feb 2022 08:36:39 +0000 (16:36 +0800)]
can: ems_usb: ems_usb_start_xmit(): fix double dev_kfree_skb() in error path
There is no need to call dev_kfree_skb() when usb_submit_urb() fails
beacause can_put_echo_skb() deletes the original skb and
can_free_echo_skb() deletes the cloned skb.
can: m_can: m_can_tx_handler(): fix use after free of skb
can_put_echo_skb() will clone skb then free the skb. Move the
can_put_echo_skb() for the m_can version 3.0.x directly before the
start of the xmit in hardware, similar to the 3.1.x branch.
Tom Rix [Sat, 19 Mar 2022 15:31:28 +0000 (08:31 -0700)]
can: mcp251xfd: mcp251xfd_register_get_dev_id(): fix return of error value
Clang static analysis reports this issue:
| mcp251xfd-core.c:1813:7: warning: The left operand
| of '&' is a garbage value
| FIELD_GET(MCP251XFD_REG_DEVID_ID_MASK, dev_id),
| ^ ~~~~~~
dev_id is set in a successful call to mcp251xfd_register_get_dev_id().
Though the status of calls made by mcp251xfd_register_get_dev_id() are
checked and handled, their status' are not returned. So return err.
In commit 42bf50a1795a ("can: isotp: support MSG_TRUNC flag when
reading from socket") a new check for recvmsg flags has been
introduced that only checked for the flags that are handled in
isotp_recvmsg() itself.
This accidentally removed the MSG_PEEK feature flag which is processed
later in the call chain in __skb_try_recv_from_queue().
Add MSG_PEEK to the set of valid flags to restore the feature.
Niklas Cassel [Tue, 8 Mar 2022 13:28:05 +0000 (14:28 +0100)]
riscv: dts: canaan: Fix SPI3 bus width
According to the K210 Standalone SDK Programming guide:
https://canaan-creative.com/wp-content/uploads/2020/03/kendryte_standalone_programming_guide_20190311144158_en.pdf
Section 15.4.3.3:
SPI0 and SPI1 supports: standard, dual, quad and octal transfers.
SPI3 supports: standard, dual and quad transfers (octal is not supported).
In order to support quad transfers (Quad SPI), SPI3 must have four IO wires
connected to the SPI flash.
Update the device tree to specify the correct bus width.
Tested on maix bit, maix dock and maixduino, which all have the same
SPI flash (gd25lq128d) connected to SPI3. maix go is untested, but it
would not make sense for this k210 board to be designed differently.
The LLVM make variable allows a developer to quickly switch between the
GNU and LLVM tools. However, it does not handle versioned binaries, such
as the ones shipped by Debian, as LLVM=1 just defines the tool variables
with the unversioned binaries.
There was some discussion during the review of the patch that introduces
LLVM=1 around versioned binaries, ultimately coming to the conclusion
that developers can just add the folder that contains the unversioned
binaries to their PATH, as Debian's versioned suffixed binaries are
really just symlinks to the unversioned binaries in /usr/lib/llvm-#/bin:
However, that can be cumbersome to developers who are constantly testing
series with different toolchains and versions. It is simple enough to
support these versioned binaries directly in the Kbuild system by
allowing the developer to specify the version suffix with LLVM=, which
is shorter than the above suggestion:
$ make ... LLVM=-14
It does not change the meaning of LLVM=1 (which will continue to use
unversioned binaries) and it does not add too much additional complexity
to the existing $(LLVM) code, while allowing developers to quickly test
their series with different versions of the whole LLVM suite of tools.
Some developers may build LLVM from source but not add the binaries to
their PATH, as they may not want to use that toolchain systemwide.
Support those developers by allowing them to supply the directory that
the LLVM tools are available in, as it is no more complex to support
than the version suffix change above.
$ make ... LLVM=/path/to/llvm/
Update and reorder the documentation to reflect these new additions.
At the same time, notate that LLVM=0 is not the same as just omitting it
altogether, which has confused people in the past.
Masahiro Yamada [Sat, 5 Mar 2022 12:56:05 +0000 (21:56 +0900)]
kbuild: add --target to correctly cross-compile UAPI headers with Clang
When you compile-test UAPI headers (CONFIG_UAPI_HEADER_TEST=y) with
Clang, they are currently compiled for the host target (likely x86_64)
regardless of the given ARCH=.
In fact, some exported headers include libc headers. For example,
include/uapi/linux/agpgart.h includes <stdlib.h> after being exported.
The header search paths should match to the target we are compiling
them for.
Pick up the --target triple from KBUILD_CFLAGS in the same ways as
commit 7f58b487e9ff ("kbuild: make Clang build userprogs for target
architecture").
Masahiro Yamada [Sun, 6 Mar 2022 07:25:35 +0000 (16:25 +0900)]
fixdep: use fflush() and ferror() to ensure successful write to files
Currently, fixdep checks the return value from (v)printf(), but it does
not ensure the complete write to the .cmd file.
printf() just writes data to the internal buffer, which usually succeeds.
(Of course, it may fail for another reason, for example when the file
descriptor is closed, but that is another story.)
When the buffer (4k?) is full, an actual write occurs, and printf() may
really fail. One of typical cases is "No space left on device" when the
disk is full.
The data remaining in the buffer will be pushed out to the file when
the program exits, but we never know if it is successful.
One straight-forward fix would be to add the following code at the end
of the program.
ret = fflush(stdout);
if (ret < 0) {
/* error handling */
}
However, it is tedious to check the return code in all the call sites
of printf(), fflush(), fclose(), and whatever can cause actual writes
to the end device. Doing that lets the program bail out at the first
failure but is usually not worth the effort.
Instead, let's check the error status from ferror(). This is 'sticky',
so you need to check it just once. You still need to call fflush().
Randy Dunlap [Wed, 30 Mar 2022 01:20:25 +0000 (18:20 -0700)]
net: sparx5: uses, depends on BRIDGE or !BRIDGE
Fix build errors when BRIDGE=m and SPARX5_SWITCH=y:
riscv64-linux-ld: drivers/net/ethernet/microchip/sparx5/sparx5_switchdev.o: in function `.L305':
sparx5_switchdev.c:(.text+0xdb0): undefined reference to `br_vlan_enabled'
riscv64-linux-ld: drivers/net/ethernet/microchip/sparx5/sparx5_switchdev.o: in function `.L283':
sparx5_switchdev.c:(.text+0xee0): undefined reference to `br_vlan_enabled'
Jakub Kicinski [Thu, 31 Mar 2022 02:14:11 +0000 (19:14 -0700)]
Merge branch 'wireguard-patches-for-5-18-rc1'
Jason A. Donenfeld says:
====================
wireguard patches for 5.18-rc1
Here's a small set of fixes for the next net push:
1) Pipacs reported a CFI violation in a cleanup routine, which he
triggered using grsec's RAP. I haven't seen reports of this yet from
the Android/CFI world yet, but it's only a matter of time there.
2) A small rng cleanup to the self test harness to make it initialize
faster on 5.18.
3) Wang reported and fixed a skb leak for CONFIG_IPV6=n.
4) After Wang's fix for the direct leak, I investigated how that code
path even could be hit, and found that the netlink layer still
handles IPv6 endpoints, when it probably shouldn't.
====================
wireguard: socket: ignore v6 endpoints when ipv6 is disabled
The previous commit fixed a memory leak on the send path in the event
that IPv6 is disabled at compile time, but how did a packet even arrive
there to begin with? It turns out we have previously allowed IPv6
endpoints even when IPv6 support is disabled at compile time. This is
awkward and inconsistent. Instead, let's just ignore all things IPv6,
the same way we do other malformed endpoints, in the case where IPv6 is
disabled.
Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld <[email protected]> Signed-off-by: Jakub Kicinski <[email protected]>
In function wg_socket_send_buffer_as_reply_to_skb() or wg_socket_send_
buffer_to_peer(), the semantics of send6() is required to free skb. But
when CONFIG_IPV6 is disable, kfree_skb() is missing. This patch adds it
to fix this bug.
Signed-off-by: Wang Hai <[email protected]> Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Signed-off-by: Jason A. Donenfeld <[email protected]> Signed-off-by: Jakub Kicinski <[email protected]>
The seed_rng() function was written to work across lots of old kernels,
back when WireGuard used a big compatibility layer. Now that things have
evolved, we can vastly simplify this, by just marking the RNG as seeded.
wireguard: queueing: use CFI-safe ptr_ring cleanup function
We make too nuanced use of ptr_ring to entirely move to the skb_array
wrappers, but we at least should avoid the naughty function pointer cast
when cleaning up skbs. Otherwise RAP/CFI will honk at us. This patch
uses the __skb_array_destroy_skb wrapper for the cleanup, rather than
directly providing kfree_skb, which is what other drivers in the same
situation do too.
Reported-by: PaX Team <[email protected]> Fixes: 886fcee939ad ("wireguard: receive: use ring buffer for incoming handshakes") Signed-off-by: Jason A. Donenfeld <[email protected]> Signed-off-by: Jakub Kicinski <[email protected]>
Palmer Dabbelt [Wed, 30 Mar 2022 23:17:54 +0000 (16:17 -0700)]
RISC-V CPU Idle Support
This series adds RISC-V CPU Idle support using SBI HSM suspend function.
The RISC-V SBI CPU idle driver added by this series is highly inspired
from the ARM PSCI CPU idle driver.
Special thanks Sandeep Tripathy for providing early feeback on SBI HSM
support in all above projects (RISC-V SBI specification, OpenSBI, and
Linux RISC-V).
* palmer/riscv-idle:
RISC-V: Enable RISC-V SBI CPU Idle driver for QEMU virt machine
dt-bindings: Add common bindings for ARM and RISC-V idle states
cpuidle: Add RISC-V SBI CPU idle driver
cpuidle: Factor-out power domain related code from PSCI domain driver
RISC-V: Add SBI HSM suspend related defines
RISC-V: Add arch functions for non-retentive suspend entry/exit
RISC-V: Rename relocate() and make it global
RISC-V: Enable CPU_IDLE drivers
Kees Cook [Thu, 24 Feb 2022 06:04:11 +0000 (22:04 -0800)]
riscv: Rename "sp_in_global" to "current_stack_pointer"
To follow the existing per-arch conventions, rename "sp_in_global" to
"current_stack_pointer". This will let it be used in non-arch places
(like HARDENED_USERCOPY).
Linus Torvalds [Wed, 30 Mar 2022 22:11:26 +0000 (15:11 -0700)]
Merge tag 'for-5.18/parisc-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux
Pull more parisc architecture updates from Helge Deller:
- Revert a patch to the invalidate/flush vmap routines which broke
kernel patching functions on older PA-RISC machines.
- Fix the kernel patching code wrt locking and flushing. Works now on
B160L machine as well.
- Fix CPU IRQ affinity for LASI, WAX and Dino chips
- Add CPU hotplug support
- Detect the hppa-suse-linux-gcc compiler when cross-compiling
* tag 'for-5.18/parisc-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
parisc: Fix patch code locking and flushing
parisc: Find a new timesync master if current CPU is removed
parisc: Move common_stext into .text section when CONFIG_HOTPLUG_CPU=y
parisc: Rewrite arch_cpu_idle_dead() for CPU hotplugging
parisc: Implement __cpu_die() and __cpu_disable() for CPU hotplugging
parisc: Add PDC locking functions for rendezvous code
parisc: Move disable_sr_hashing_asm() into .text section
parisc: Move CPU startup-related functions into .text section
parisc: Move store_cpu_topology() into text section
parisc: Switch from GENERIC_CPU_DEVICES to GENERIC_ARCH_TOPOLOGY
parisc: Ensure set_firmware_width() is called only once
parisc: Add constants for control registers and clean up mfctl()
parisc: Detect hppa-suse-linux-gcc compiler for cross-building
parisc: Clean up cpu_check_affinity() and drop cpu_set_affinity_irq()
parisc: Fix CPU affinity for Lasi, WAX and Dino chips
Revert "parisc: Fix invalidate/flush vmap routines"
Linus Torvalds [Wed, 30 Mar 2022 22:06:31 +0000 (15:06 -0700)]
Merge tag 'modules-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof/linux
Pull module update from Luis Chamberlain:
"There is only one patch which qualifies for modules for v5.18-rc1 and
its a small fix from Dan Carpenter for lib/test_kmod module.
The rest of the changes are too major and landed in modules-testing
too late for inclusion. The good news is that most of the major
changes for v5.19 is going to be tested very early through linux-next.
This simple fix is all we have for modules for v5.18-rc1"
* tag 'modules-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof/linux:
lib/test: use after free in register_test_dev_kmod()
Jonathan Corbet [Fri, 25 Mar 2022 21:51:35 +0000 (15:51 -0600)]
docs: Add a document on how to fix a messy diffstat
A branch with merges in will sometimes create a diffstat containing a lot
of unrelated work at "git request-pull" time. Create a document based on
Linus's advice (found in the links below) and add it to the maintainer
manual in the hope of saving some wear on Linus's keyboard going forward.
Akira Yokosawa [Tue, 29 Mar 2022 06:07:02 +0000 (15:07 +0900)]
docs: sphinx/requirements: Limit jinja2<3.1
jinja2 release 3.1.0 (March 24, 2022) broke Sphinx<4.0.
This looks like the result of deprecating Python 3.6.
It has been tested against Sphinx 4.3.0 and later.
Setting an upper limit of <3.1 to junja2 can unbreak Sphinx<4.0
including Sphinx 2.4.4.
Zheng Yongjun [Tue, 29 Mar 2022 09:08:00 +0000 (09:08 +0000)]
net: dsa: felix: fix possible NULL pointer dereference
As the possible failure of the allocation, kzalloc() may return NULL
pointer.
Therefore, it should be better to check the 'sgi' in order to prevent
the dereference of NULL pointer.
Lars Ellenberg [Wed, 30 Mar 2022 18:55:51 +0000 (20:55 +0200)]
drbd: fix potential silent data corruption
Scenario:
---------
bio chain generated by blk_queue_split().
Some split bio fails and propagates its error status to the "parent" bio.
But then the (last part of the) parent bio itself completes without error.
We would clobber the already recorded error status with BLK_STS_OK,
causing silent data corruption.
Reproducer:
-----------
How to trigger this in the real world within seconds:
DRBD on top of degraded parity raid,
small stripe_cache_size, large read_ahead setting.
Drop page cache (sysctl vm.drop_caches=1, fadvise "DONTNEED",
umount and mount again, "reboot").
Cause significant read ahead.
Large read ahead request is split by blk_queue_split().
Parts of the read ahead that are already in the stripe cache,
or find an available stripe cache to use, can be serviced.
Parts of the read ahead that would need "too much work",
would need to wait for a "stripe_head" to become available,
are rejected immediately.
For larger read ahead requests that are split in many pieces, it is very
likely that some "splits" will be serviced, but then the stripe cache is
exhausted/busy, and the remaining ones will be rejected.
Jackie Liu [Wed, 30 Mar 2022 12:09:21 +0000 (20:09 +0800)]
MIPS: rb532: move GPIOD definition into C-files
My kernel robot reports build error from drivers/iio/adc/da9150-gpadc.c,
drivers/iio/adc/da9150-gpadc.c:254:13: error: ‘DA9150_GPADC_CHAN_0x08’
undeclared here (not in a function); did you mean ‘DA9150_GPADC_CHAN_TBAT’?
254 | .channel = DA9150_GPADC_CHAN_##_id,
We define GPIOD in rb.h, in fact it should only be used in gpio.c, but
it affects the driver da9150-gpadc.c which goes against the original
intention of the design, just move it to its scope.
Xiaoke Wang [Fri, 25 Mar 2022 11:49:41 +0000 (19:49 +0800)]
MIPS: lantiq: check the return value of kzalloc()
kzalloc() is a memory allocation function which can return NULL when
some internal memory errors happen. So it is better to check the
return value of it to prevent potential wrong memory access or
memory leak.
Xiaoke Wang [Fri, 25 Mar 2022 10:57:18 +0000 (18:57 +0800)]
mips: sgi-ip22: add a check for the return of kzalloc()
kzalloc() is a memory allocation function which can return NULL when
some internal memory errors happen. So it is better to check it to
prevent potential wrong memory access.
Linus Torvalds [Wed, 30 Mar 2022 18:00:33 +0000 (11:00 -0700)]
Merge tag 'pwm/for-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm
Pull pwm updates from Thierry Reding:
"This contains conversions of some more drivers to the atomic API as
well as the addition of new chip support for some existing drivers.
There are also various minor fixes and cleanups across the board, from
drivers to device tree bindings"
* tag 'pwm/for-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/thierry.reding/linux-pwm: (45 commits)
pwm: rcar: Simplify multiplication/shift logic
dt-bindings: pwm: renesas,tpu: Do not require pwm-cells twice
dt-bindings: pwm: tiehrpwm: Do not require pwm-cells twice
dt-bindings: pwm: tiecap: Do not require pwm-cells twice
dt-bindings: pwm: samsung: Do not require pwm-cells twice
dt-bindings: pwm: intel,keembay: Do not require pwm-cells twice
dt-bindings: pwm: brcm,bcm7038: Do not require pwm-cells twice
dt-bindings: pwm: toshiba,visconti: Include generic PWM schema
dt-bindings: pwm: renesas,pwm: Include generic PWM schema
dt-bindings: pwm: sifive: Include generic PWM schema
dt-bindings: pwm: rockchip: Include generic PWM schema
dt-bindings: pwm: mxs: Include generic PWM schema
dt-bindings: pwm: iqs620a: Include generic PWM schema
dt-bindings: pwm: intel,lgm: Include generic PWM schema
dt-bindings: pwm: imx: Include generic PWM schema
dt-bindings: pwm: allwinner,sun4i-a10: Include generic PWM schema
pwm: pwm-mediatek: Beautify error messages text
pwm: pwm-mediatek: Allocate clk_pwms with devm_kmalloc_array
pwm: pwm-mediatek: Simplify error handling with dev_err_probe()
pwm: brcmstb: Remove useless locking
...
Linus Torvalds [Wed, 30 Mar 2022 17:58:28 +0000 (10:58 -0700)]
Merge tag 'regulator-fix-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator
Pull regulator fixes from Mark Brown:
"A couple of fixes for the rt4831 driver which fix features that didn't
work due to incomplete description of the register configuration"
* tag 'regulator-fix-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
regulator: rt4831: Add active_discharge_on to fix discharge API
regulator: rt4831: Add bypass mask to fix set_bypass API work
Linus Torvalds [Wed, 30 Mar 2022 17:54:49 +0000 (10:54 -0700)]
Merge tag 'dmaengine-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine
Pull dmaengine updates from Vinod Koul:
"This time we have bunch of driver updates and some new device support.
New support:
- Document RZ/V2L and RZ/G2UL dma binding
- TI AM62x k3-udma and k3-psil support
Updates:
- Yaml conversion for Mediatek uart apdma schema
- Removal of DMA-32 fallback configuration for various drivers
- imx-sdma updates for channel restart"
* tag 'dmaengine-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (23 commits)
dmaengine: hisi_dma: fix MSI allocate fail when reload hisi_dma
dmaengine: dw-axi-dmac: cleanup comments
dmaengine: fsl-dpaa2-qdma: Drop comma after SoC match table sentinel
dt-bindings: dma: Convert mtk-uart-apdma to DT schema
dmaengine: ppc4xx: Make use of the helper macro LIST_HEAD()
dmaengine: idxd: Remove useless DMA-32 fallback configuration
dmaengine: qcom_hidma: Remove useless DMA-32 fallback configuration
dmaengine: sh: Kconfig: Add ARCH_R9A07G054 dependency for RZ_DMAC config option
dmaengine: ti: k3-psil: Add AM62x PSIL and PDMA data
dmaengine: ti: k3-udma: Add AM62x DMSS support
dmaengine: ti: cleanup comments
dmaengine: imx-sdma: clean up some inconsistent indenting
dmaengine: Revert "dmaengine: shdma: Fix runtime PM imbalance on error"
dmaengine: idxd: restore traffic class defaults after wq reset
dmaengine: altera-msgdma: Remove useless DMA-32 fallback configuration
dmaengine: stm32-dma: set dma_device max_sg_burst
dmaengine: imx-sdma: fix cyclic buffer race condition
dmaengine: imx-sdma: restart cyclic channel if needed
dmaengine: iot: Remove useless DMA-32 fallback configuration
dmaengine: ptdma: handle the cases based on DMA is complete
...
Linus Torvalds [Wed, 30 Mar 2022 17:50:48 +0000 (10:50 -0700)]
Merge tag 'rproc-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/remoteproc/linux
Pull remoteproc updates from Bjorn Andersson:
"In the remoteproc core, it's now possible to mark the sysfs attributes
read only on a per-instance basis, which is then used by the TI wkup
M3 driver.
Also, the rproc_shutdown() interface propagates errors to the caller
and an array underflow is fixed in the debugfs interface. The
rproc_da_to_va() API is moved to the public API to allow e.g. child
rpmsg devices to acquire pointers to memory shared with the remote
processor.
The TI K3 R5F and DSP drivers gains support for attaching to instances
already started by the bootloader, aka IPC-only mode.
The Mediatek remoteproc driver gains support for the MT8186 SCP. The
driver's probe function is reordered and moved to use the devres
version of rproc_alloc() to save a few gotos. The driver's probe
function is also transitioned to use dev_err_probe() to provide better
debug support.
Support for the Qualcomm SC7280 Wireless Subsystem (WPSS) is
introduced. The Hexagon based remoteproc drivers gains support for
voting for interconnect bandwidth during launch of the remote
processor. The modem subsystem (MSS) driver gains support for probing
the BAM-DMUX driver, which provides the network interface towards the
modem on a set of older Qualcomm platforms. In addition a number a bug
fixes are introduces in the Qualcomm drivers.
Lastly Qualcomm ADSP DeviceTree binding is converted to YAML format,
to allow validation of DeviceTree source files"
* tag 'rproc-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/remoteproc/linux: (22 commits)
remoteproc: qcom_q6v5_mss: Create platform device for BAM-DMUX
remoteproc: qcom: q6v5_wpss: Add support for sc7280 WPSS
dt-bindings: remoteproc: qcom: Add SC7280 WPSS support
dt-bindings: remoteproc: qcom: adsp: Convert binding to YAML
remoteproc: k3-dsp: Add support for IPC-only mode for all K3 DSPs
remoteproc: k3-dsp: Refactor mbox request code in start
remoteproc: k3-r5: Add support for IPC-only mode for all R5Fs
remoteproc: k3-r5: Refactor mbox request code in start
remoteproc: Change rproc_shutdown() to return a status
remoteproc: qcom: q6v5: Add interconnect path proxy vote
remoteproc: mediatek: Support mt8186 scp
dt-bindings: remoteproc: mediatek: Add binding for mt8186 scp
remoteproc: qcom_q6v5_mss: Fix some leaks in q6v5_alloc_memory_region
remoteproc: qcom_wcnss: Add missing of_node_put() in wcnss_alloc_memory_region
remoteproc: qcom: Fix missing of_node_put in adsp_alloc_memory_region
remoteproc: move rproc_da_to_va declaration to remoteproc.h
remoteproc: wkup_m3: Set sysfs_read_only flag
remoteproc: Introduce sysfs_read_only flag
remoteproc: Fix count check in rproc_coredump_write()
remoteproc: mtk_scp: Use dev_err_probe() where possible
...
Linus Torvalds [Wed, 30 Mar 2022 17:47:48 +0000 (10:47 -0700)]
Merge tag 'hwlock-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/remoteproc/linux
Pull hwspinlock updates from Bjorn Andersson:
"This updates sprd and srm32 drivers to use struct_size() instead of
their open-coded equivalents. It also cleans up the omap dt-bindings
example"
* tag 'hwlock-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/remoteproc/linux:
hwspinlock: sprd: Use struct_size() helper in devm_kzalloc()
hwspinlock: stm32: Use struct_size() helper in devm_kzalloc()
dt-bindings: hwlock: omap: Remove redundant binding example
Linus Torvalds [Wed, 30 Mar 2022 17:43:19 +0000 (10:43 -0700)]
Merge tag 'rpmsg-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/remoteproc/linux
Pull rpmsg updates from Bjorn Andersson:
"The major part of the rpmsg changes for v5.18 relates to improvements
in the rpmsg char driver, which now allow automatically attaching to
rpmsg channels as well as initiating new communication channels from
the Linux side.
The SMD driver is moved to arch_initcall with the purpose of
registering root clocks earlier during boot.
Also in the SMD driver, a workaround for the resource power management
(RPM) channel is introduced to resolve an issue where both the RPM and
Linux side waits for the other to close the communication established
by the bootloader - this unblocks support for clocks and regulators on
some older Qualcomm platforms"
* tag 'rpmsg-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/remoteproc/linux:
rpmsg: ctrl: Introduce new RPMSG_CREATE/RELEASE_DEV_IOCTL controls
rpmsg: char: Introduce the "rpmsg-raw" channel
rpmsg: char: Add possibility to use default endpoint of the rpmsg device
rpmsg: char: Refactor rpmsg_chrdev_eptdev_create function
rpmsg: Update rpmsg_chrdev_register_device function
rpmsg: Move the rpmsg control device from rpmsg_char to rpmsg_ctrl
rpmsg: Create the rpmsg class in core instead of in rpmsg char
rpmsg: char: Export eptdev create and destroy functions
rpmsg: char: treat rpmsg_trysend() ENOMEM as EAGAIN
rpmsg: qcom_smd: Fix redundant channel->registered assignment
rpmsg: use struct_size over open coded arithmetic
rpmsg: smd: allow opening rpm_requests even if already opened
rpmsg: qcom_smd: Promote to arch_initcall
Linus Torvalds [Wed, 30 Mar 2022 17:36:41 +0000 (10:36 -0700)]
Merge tag 'i3c/for-5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux
Pull i3c updates from Alexandre Belloni:
- support dynamic addition of i2c devices
* tag 'i3c/for-5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/i3c/linux:
i3c: fix uninitialized variable use in i2c setup
i3c: support dynamically added i2c devices
i3c: remove i2c board info from i2c_dev_desc
Linus Torvalds [Wed, 30 Mar 2022 17:11:04 +0000 (10:11 -0700)]
Merge tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk updates from Stephen Boyd:
"There's one large change in the core clk framework here. We change how
clk_set_rate_range() works so that the frequency is re-evaulated each
time the rate is changed. Previously we wouldn't let clk providers see
a rate that was different if it was still within the range, which
could be bad for power if the clk could run slower when a range
expands. Now the clk provider can decide to do something differently
when the constraints change. This broke Nvidia's clk driver so we had
to wait for the fix for that to bake a little more in -next.
The rate range patch series also introduced a kunit suite for the clk
framework that we're going to extend in the next release. It already
made it easy to find corner cases in the rate range patches so I'm
excited to see it cover more clk code and increase our confidence in
core framework patches in the future. I also added a kunit test for
the basic clk gate code and that work will continue to cover more
basic clk types: muxes, dividers, etc.
Beyond the core code we have the usual set of clk driver updates and
additions. Qualcomm again dominates the diffstat here with lots more
SoCs being supported and i.MX follows afer that with a similar number
of SoCs gaining clk drivers. Beyond those large additions there's
drivers being modernized to use clk_parent_data so we can move away
from global string names for all the clks in an SoC. Finally there's
lots of little fixes all over the clk drivers for typos, warnings, and
missing clks that aren't critical and get batched up waiting for the
next merge window to open. Nothing super big stands out in the driver
pile. Full details are below.
Core:
- Make clk_set_rate_range() re-evaluate the limits each time
- Introduce various clk_set_rate_range() tests
- Add clk_drop_range() to drop a previously set range
New Drivers:
- i.MXRT1050 clock driver and bindings
- i.MX8DXL clock driver and bindings
- i.MX93 clock driver and bindings
- NCO blocks on Apple SoCs
- Audio clks on StarFive JH7100 RISC-V SoC
- Add support for the new Renesas RZ/V2L SoC
- Qualcomm SDX65 A7 PLL
- Qualcomm SM6350 GPU clks
- Qualcomm SM6125, SM6350, QCS2290 display clks
- Qualcomm MSM8226 multimedia clks
Updates:
- Kunit tests for clk-gate implementation
- Terminate arrays with sentinels and make that clearer
- Cleanup SPDX tags
- Fix typos in comments
- Mark mux table as const in clk-mux
- Make the all_lists array const
- Convert Cirrus Logic CS2000P driver to regmap, yamlify DT binding
and add support for dynamic mode
- Clock configuration on Microchip PolarFire SoCs
- Free allocations on probe error in Mediatek clk driver
- Modernize Mediatek clk driver by consolidating code
- Add watchdog (WDT), I2C, and pin function controller (PFC) clocks
on Renesas R-Car S4-8
- Improve the clocks for the Rockchip rk3568 display outputs
(parenting, pll-rates)
- Use of_device_get_match_data() instead of open-coding on Rockchip
rk3568
- Reintroduce the expected fractional-divider behaviour that
disappeared with the addition of CLK_FRAC_DIVIDER_POWER_OF_TWO_PS
- Remove SYS PLL 1/2 clock gates for i.MX8M*
- Remove AUDIO MCLK ROOT from i.MX7D
- Add fracn gppll clock type used by i.MX93
- Add new composite clock for i.MX93
- Add missing media mipi phy ref clock for i.MX8MP
- Fix off by one in imx_lpcg_parse_clks_from_dt()
- Rework for the imx pll14xx
- sama7g5: One low priority fix for GCLK of PDMC
- Add DMA engine (SYS-DMAC) clocks on Renesas R-Car S4-8
- Add MOST (MediaLB I/F) clocks on Renesas R-Car E3 and D3
- Add CAN-FD clocks on Renesas R-Car V3U
- Qualcomm SC8280XP RPMCC
- Add some missing clks on Qualcomm MSM8992/MSM8994/MSM8998 SoCs
- Rework Qualcomm GCC bindings and convert SDM845 camera bindig to
YAML
- Convert various Qualcomm drivers to use clk_parent_data
- Remove test clocks from various Qualcomm drivers
- Crypto engine clks on Qualcomm IPQ806x + more freqs for SDCC/NSS
- Qualcomm SM8150 EMAC, PCIe, UFS GDSCs
- Better pixel clk frequency support on Qualcomm RCG2 clks"
* tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux: (227 commits)
clk: zynq: Update the parameters to zynq_clk_register_periph_clk
clk: zynq: trivial warning fix
clk: Drop the rate range on clk_put()
clk: test: Test clk_set_rate_range on orphan mux
clk: Initialize orphan req_rate
dt-bindings: clock: drop useless consumer example
dt-bindings: clock: renesas: Make example 'clocks' parsable
clk: qcom: gcc-msm8994: Fix gpll4 width
dt-bindings: clock: fix dt_binding_check error for qcom,gcc-other.yaml
clk: rs9: Add Renesas 9-series PCIe clock generator driver
clk: fixed-factor: Introduce devm_clk_hw_register_fixed_factor_index()
clk: visconti: prevent array overflow in visconti_clk_register_gates()
dt-bindings: clk: rs9: Add Renesas 9-series I2C PCIe clock generator
clk: sifive: Move all stuff into SoCs header files from C files
clk: sifive: Add SoCs prefix in each SoCs-dependent data
riscv: dts: Change the macro name of prci in each device node
dt-bindings: change the macro name of prci in header files and example
clk: sifive: duplicate the macro definitions for the time being
clk: qcom: sm6125-gcc: fix typos in comments
clk: ti: clkctrl: fix typos in comments
...
Linus Torvalds [Wed, 30 Mar 2022 17:04:11 +0000 (10:04 -0700)]
Merge tag 'libnvdimm-for-5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull libnvdimm updates from Dan Williams:
"The update for this cycle includes the deprecation of block-aperture
mode and a new perf events interface for the papr_scm nvdimm driver.
The perf events approach was acked by PeterZ.
- Add perf support for nvdimm events, initially only for 'papr_scm'
devices.
- Deprecate the 'block aperture' support in libnvdimm, it only ever
existed in the specification, not in shipping product"
* tag 'libnvdimm-for-5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
nvdimm/blk: Fix title level
MAINTAINERS: remove section LIBNVDIMM BLK: MMIO-APERTURE DRIVER
powerpc/papr_scm: Fix build failure when
drivers/nvdimm: Fix build failure when CONFIG_PERF_EVENTS is not set
nvdimm/region: Delete nd_blk_region infrastructure
ACPI: NFIT: Remove block aperture support
nvdimm/namespace: Delete nd_namespace_blk
nvdimm/namespace: Delete blk namespace consideration in shared paths
nvdimm/blk: Delete the block-aperture window driver
nvdimm/region: Fix default alignment for small regions
docs: ABI: sysfs-bus-nvdimm: Document sysfs event format entries for nvdimm pmu
powerpc/papr_scm: Add perf interface support
drivers/nvdimm: Add perf interface to expose nvdimm performance stats
drivers/nvdimm: Add nvdimm pmu structure
Guo Xuenan [Wed, 30 Mar 2022 16:49:28 +0000 (09:49 -0700)]
fs: fix an infinite loop in iomap_fiemap
when get fiemap starting from MAX_LFS_FILESIZE, (maxbytes - *len) < start
will always true , then *len set zero. because of start offset is beyond
file size, for erofs filesystem it will always return iomap.length with
zero,iomap iterate will enter infinite loop. it is necessary cover this
corner case to avoid this situation.
Carlos Llamas [Tue, 29 Mar 2022 20:18:15 +0000 (20:18 +0000)]
loop: fix ioctl calls using compat_loop_info
Support for cryptoloop was deleted in commit 47e9624616c8 ("block:
remove support for cryptoloop and the xor transfer"), making the usage
of loop_info->lo_encrypt_type obsolete. However, this member was also
removed from the compat_loop_info definition and this breaks userspace
ioctl calls for 32-bit binaries and CONFIG_COMPAT=y.
This patch restores the compat_loop_info->lo_encrypt_type member and
marks it obsolete as well as in the uapi header definitions.
Jakob Koschel [Thu, 24 Mar 2022 07:13:01 +0000 (08:13 +0100)]
ksmbd: replace usage of found with dedicated list iterator variable
To move the list iterator variable into the list_for_each_entry_*()
macro in the future it should be avoided to use the list iterator
variable after the loop body.
To *never* use the list iterator variable after the loop it was
concluded to use a separate iterator variable instead of a
found boolean [1].
This removes the need to use a found variable and simply checking if
the variable was set, can determine if the break/goto was hit.
Takashi Iwai [Wed, 30 Mar 2022 12:09:03 +0000 (14:09 +0200)]
ALSA: pcm: Fix potential AB/BA lock with buffer_mutex and mmap_lock
syzbot caught a potential deadlock between the PCM
runtime->buffer_mutex and the mm->mmap_lock. It was brought by the
recent fix to cover the racy read/write and other ioctls, and in that
commit, I overlooked a (hopefully only) corner case that may take the
revert lock, namely, the OSS mmap. The OSS mmap operation
exceptionally allows to re-configure the parameters inside the OSS
mmap syscall, where mm->mmap_mutex is already held. Meanwhile, the
copy_from/to_user calls at read/write operations also take the
mm->mmap_lock internally, hence it may lead to a AB/BA deadlock.
A similar problem was already seen in the past and we fixed it with a
refcount (in commit b248371628aa). The former fix covered only the
call paths with OSS read/write and OSS ioctls, while we need to cover
the concurrent access via both ALSA and OSS APIs now.
This patch addresses the problem above by replacing the buffer_mutex
lock in the read/write operations with a refcount similar as we've
used for OSS. The new field, runtime->buffer_accessing, keeps the
number of concurrent read/write operations. Unlike the former
buffer_mutex protection, this protects only around the
copy_from/to_user() calls; the other codes are basically protected by
the PCM stream lock. The refcount can be a negative, meaning blocked
by the ioctls. If a negative value is seen, the read/write aborts
with -EBUSY. In the ioctl side, OTOH, they check this refcount, too,
and set to a negative value for blocking unless it's already being
accessed.
Thomas Gleixner [Thu, 24 Mar 2022 13:47:14 +0000 (14:47 +0100)]
x86/fpu/xstate: Handle supervisor states in XSTATE permissions
The size calculation in __xstate_request_perm() fails to take supervisor
states into account because the permission bitmap is only relevant for user
states.
Up to 5.17 this does not matter because there are no supervisor states
supported, but the (re-)enabling of ENQCMD makes them available.
Thomas Gleixner [Thu, 24 Mar 2022 13:47:13 +0000 (14:47 +0100)]
x86/fpu/xsave: Handle compacted offsets correctly with supervisor states
So far the cached fixed compacted offsets worked, but with (re-)enabling
of ENQCMD this does no longer work with KVM fpstate.
KVM does not have supervisor features enabled for the guest FPU, which
means that KVM has then a different XSAVE area layout than the host FPU
state. This in turn breaks the copy from/to UABI functions when invoked for
a guest state.
Remove the pre-calculated compacted offsets and calculate the offset
of each component at runtime based on the XCOMP_BV field in the XSAVE
header.
The runtime overhead is not interesting because these copy from/to UABI
functions are not used in critical fast paths. KVM uses them to save and
restore FPU state during migration. The host uses them for ptrace and for
the slow path of 32bit signal handling.
Mohan Kumar [Tue, 29 Mar 2022 15:59:40 +0000 (21:29 +0530)]
ALSA: hda: Avoid unsol event during RPM suspending
There is a corner case with unsol event handling during codec runtime
suspending state. When the codec runtime suspend call initiated, the
codec->in_pm atomic variable would be 0, currently the codec runtime
suspend function calls snd_hdac_enter_pm() which will just increments
the codec->in_pm atomic variable. Consider unsol event happened just
after this step and before snd_hdac_leave_pm() in the codec runtime
suspend function. The snd_hdac_power_up_pm() in the unsol event
flow in hdmi_present_sense_via_verbs() function would just increment
the codec->in_pm atomic variable without calling pm_runtime_get_sync
function.
As codec runtime suspend flow is already in progress and in parallel
unsol event is also accessing the codec verbs, as soon as codec
suspend flow completes and clocks are switched off before completing
the unsol event handling as both functions doesn't wait for each other.
This will result in below errors
[ 589.428020] tegra-hda 3510000.hda: azx_get_response timeout, switching
to polling mode: last cmd=0x505f2f57
[ 589.428344] tegra-hda 3510000.hda: spurious response 0x80000074:0x5,
last cmd=0x505f2f57
[ 589.428547] tegra-hda 3510000.hda: spurious response 0x80000065:0x5,
last cmd=0x505f2f57
To avoid this, the unsol event flow should not perform any codec verb
related operations during RPM_SUSPENDING state.
Kai-Heng Feng [Wed, 30 Mar 2022 06:13:33 +0000 (14:13 +0800)]
ALSA: hda/realtek: Fix audio regression on Mi Notebook Pro 2020
Commit 5aec98913095 ("ALSA: hda/realtek - ALC236 headset MIC recording
issue") is to solve recording issue met on AL236, by matching codec
variant ALC269_TYPE_ALC257 and ALC269_TYPE_ALC256.
This match can be too broad and Mi Notebook Pro 2020 is broken by the
patch.
Instead, use codec ID to be narrow down the scope, in order to make
ALC256 unaffected.
Linus Torvalds [Wed, 30 Mar 2022 06:29:18 +0000 (23:29 -0700)]
fs: fix fd table size alignment properly
Jason Donenfeld reports that my commit 1c24a186398f ("fs: fd tables have
to be multiples of BITS_PER_LONG") doesn't work, and the reason is an
embarrassing brown-paper-bag bug.
Yes, we want to align the number of fds to BITS_PER_LONG, and yes, the
reason they might not be aligned is because the incoming 'max_fd'
argument might not be aligned.
But aligining the argument - while simple - will cause a "infinitely
big" maxfd (eg NR_OPEN_MAX) to just overflow to zero. Which most
definitely isn't what we want either.
The obvious fix was always just to do the alignment last, but I had
moved it earlier just to make the patch smaller and the code look
simpler. Duh. It certainly made _me_ look simple.
We've added 16 non-merge commits during the last 1 day(s) which contain
a total of 24 files changed, 354 insertions(+), 187 deletions(-).
The main changes are:
1) x86 specific bits of fprobe/rethook, from Masami and Peter.
2) ice/xsk fixes, from Maciej and Magnus.
3) Various small fixes, from Andrii, Yonghong, Geliang and others.
* https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
selftests/bpf: Fix clang compilation errors
ice: xsk: Fix indexing in ice_tx_xsk_pool()
ice: xsk: Stop Rx processing when ntc catches ntu
ice: xsk: Eliminate unnecessary loop iteration
xsk: Do not write NULL in SW ring at allocation failure
x86,kprobes: Fix optprobe trampoline to generate complete pt_regs
x86,rethook: Fix arch_rethook_trampoline() to generate a complete pt_regs
x86,rethook,kprobes: Replace kretprobe with rethook on x86
kprobes: Use rethook for kretprobe if possible
bpftool: Fix generated code in codegen_asserts
selftests/bpf: fix selftest after random: Urandom_read tracepoint removal
bpf: Fix maximum permitted number of arguments check
bpf: Sync comments for bpf_get_stack
fprobe: Fix sparse warning for acccessing __rcu ftrace_hash
fprobe: Fix smatch type mismatch warning
bpf/bpftool: Add unprivileged_bpf_disabled check against value of 2
====================
Linus Torvalds [Wed, 30 Mar 2022 01:55:37 +0000 (18:55 -0700)]
Merge tag 'nfs-for-5.18-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client updates from Trond Myklebust:
"Highlights include:
Features:
- Switch NFS to use readahead instead of the obsolete readpages.
- Readdir fixes to improve cacheability of large directories when
there are multiple readers and writers.
- Readdir performance improvements when doing a seekdir() immediately
after opening the directory (common when re-exporting NFS).
- NFS swap improvements from Neil Brown.
- Loosen up memory allocation to permit direct reclaim and write back
in cases where there is no danger of deadlocking the writeback code
or NFS swap.
- Avoid sillyrename when the NFSv4 server claims to support the
necessary features to recover the unlinked but open file after
reboot.
Bugfixes:
- Patch from Olga to add a mount option to control NFSv4.1 session
trunking discovery, and default it to being off.
- Fix a lockup in nfs_do_recoalesce().
- Two fixes for list iterator variables being used when pointing to
the list head.
- Fix a kernel memory scribble when reading from a non-socket
transport in /sys/kernel/sunrpc.
- Fix a race where reconnecting to a server could leave the TCP
socket stuck forever in the connecting state.
- Patch from Neil to fix a shutdown race which can leave the SUNRPC
transport timer primed after we free the struct xprt itself.
- Patch from Xin Xiong to fix reference count leaks in the NFSv4.2
copy offload.
- Sunrpc patch from Olga to avoid resending a task on an offlined
transport.
Cleanups:
- Patches from Dave Wysochanski to clean up the fscache code"
* tag 'nfs-for-5.18-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (91 commits)
NFSv4/pNFS: Fix another issue with a list iterator pointing to the head
NFS: Don't loop forever in nfs_do_recoalesce()
SUNRPC: Don't return error values in sysfs read of closed files
SUNRPC: Do not dereference non-socket transports in sysfs
NFSv4.1: don't retry BIND_CONN_TO_SESSION on session error
SUNRPC don't resend a task on an offlined transport
NFS: replace usage of found with dedicated list iterator variable
SUNRPC: avoid race between mod_timer() and del_timer_sync()
pNFS/files: Ensure pNFS allocation modes are consistent with nfsiod
pNFS/flexfiles: Ensure pNFS allocation modes are consistent with nfsiod
NFSv4/pnfs: Ensure pNFS allocation modes are consistent with nfsiod
NFS: Avoid writeback threads getting stuck in mempool_alloc()
NFS: nfsiod should not block forever in mempool_alloc()
SUNRPC: Make the rpciod and xprtiod slab allocation modes consistent
SUNRPC: Fix unx_lookup_cred() allocation
NFS: Fix memory allocation in rpc_alloc_task()
NFS: Fix memory allocation in rpc_malloc()
SUNRPC: Improve accuracy of socket ENOBUFS determination
SUNRPC: Replace internal use of SOCKWQ_ASYNC_NOSPACE
SUNRPC: Fix socket waits for write buffer space
...
Dave Chinner [Wed, 30 Mar 2022 01:22:02 +0000 (18:22 -0700)]
xfs: drop async cache flushes from CIL commits.
Jan Kara reported a performance regression in dbench that he
bisected down to commit bad77c375e8d ("xfs: CIL checkpoint
flushes caches unconditionally").
Whilst developing the journal flush/fua optimisations this cache was
part of, it appeared to made a significant difference to
performance. However, now that this patchset has settled and all the
correctness issues fixed, there does not appear to be any
significant performance benefit to asynchronous cache flushes.
In fact, the opposite is true on some storage types and workloads,
where additional cache flushes that can occur from fsync heavy
workloads have measurable and significant impact on overall
throughput.
Local dbench testing shows little difference on dbench runs with
sync vs async cache flushes on either fast or slow SSD storage, and
no difference in streaming concurrent async transaction workloads
like fs-mark.
Performance and cache flush behaviour is restored to pre-regression
levels.
As such, we can now consider the async cache flush mechanism an
unnecessary exercise in premature optimisation and hence we can
now remove it and the infrastructure it requires completely.
Fixes: bad77c375e8d ("xfs: CIL checkpoint flushes caches unconditionally") Reported-and-tested-by: Jan Kara <[email protected]> Signed-off-by: Dave Chinner <[email protected]> Reviewed-by: Darrick J. Wong <[email protected]> Signed-off-by: Darrick J. Wong <[email protected]>
Dave Chinner [Wed, 30 Mar 2022 01:22:02 +0000 (18:22 -0700)]
xfs: shutdown during log recovery needs to mark the log shutdown
When a checkpoint writeback is run by log recovery, corruption
propagated from the log can result in writeback verifiers failing
and calling xfs_force_shutdown() from
xfs_buf_delwri_submit_buffers().
This results in the mount being marked as shutdown, but the log does
not get marked as shut down because:
/*
* If this happens during log recovery then we aren't using the runtime
* log mechanisms yet so there's nothing to shut down.
*/
if (!log || xlog_in_recovery(log))
return false;
If there are other buffers that then fail (say due to detecting the
mount shutdown), they will now hang in xfs_do_force_shutdown()
waiting for the log to shut down like this:
xlog_force_shutdown() always needs to mark the log as shut down,
regardless of whether recovery is in progress or not, so that
multiple calls to xfs_force_shutdown() during recovery don't end
up waiting for the log to be shut down like this.
Dave Chinner [Wed, 30 Mar 2022 01:22:01 +0000 (18:22 -0700)]
xfs: xfs_trans_commit() path must check for log shutdown
If a shut races with xfs_trans_commit() and we have shut down the
filesystem but not the log, we will still cancel the transaction.
This can result in aborting dirty log items instead of committing and
pinning them whilst the log is still running. Hence we can end up
with dirty, unlogged metadata that isn't in the AIL in memory that
can be flushed to disk via writeback clustering.
This was discovered from a g/388 trace where an inode log item was
having IO completed on it and it wasn't in the AIL, hence tripping
asserts xfs_ail_check(). Inode cluster writeback started long after
the filesystem shutdown started, and long after the transaction
containing the dirty inode was aborted and the log item marked
XFS_LI_ABORTED. The inode was seen as dirty and unpinned, so it
was flushed. IO completion tried to remove the inode from the AIL,
at which point stuff went bad:
XFS (pmem1): Log I/O Error (0x6) detected at xfs_fs_goingdown+0xa3/0xf0 (fs/xfs/xfs_fsops.c:500). Shutting down filesystem.
XFS: Assertion failed: in_ail, file: fs/xfs/xfs_trans_ail.c, line: 67
XFS (pmem1): Please unmount the filesystem and rectify the problem(s)
Workqueue: xfs-buf/pmem1 xfs_buf_ioend_work
RIP: 0010:assfail+0x27/0x2d
Call Trace:
<TASK>
xfs_ail_check+0xa8/0x180
xfs_ail_delete_one+0x3b/0xf0
xfs_buf_inode_iodone+0x329/0x3f0
xfs_buf_ioend+0x1f8/0x530
xfs_buf_ioend_work+0x15/0x20
process_one_work+0x1ac/0x390
worker_thread+0x56/0x3c0
kthread+0xf6/0x120
ret_from_fork+0x1f/0x30
</TASK>
xfs_trans_commit() needs to check log state for shutdown, not mount
state. It cannot abort dirty log items while the log is still
running as dirty items must remained pinned in memory until they are
either committed to the journal or the log has shut down and they
can be safely tossed away. Hence if the log has not shut down, the
xfs_trans_commit() path must allow completed transactions to commit
to the CIL and pin the dirty items even if a mount shutdown has
started.
Dave Chinner [Wed, 30 Mar 2022 01:22:01 +0000 (18:22 -0700)]
xfs: xfs_do_force_shutdown needs to block racing shutdowns
When we call xfs_forced_shutdown(), the caller often expects the
filesystem to be completely shut down when it returns. However,
if we have racing xfs_forced_shutdown() calls, the first caller sets
the mount shutdown flag then goes to shutdown the log. The second
caller sees the mount shutdown flag and returns immediately - it
does not wait for the log to be shut down.
Unfortunately, xfs_forced_shutdown() is used in some places that
expect it to completely shut down the filesystem before it returns
(e.g. xfs_trans_log_inode()). As such, returning before the log has
been shut down leaves us in a place where the transaction failed to
complete correctly but we still call xfs_trans_commit(). This
situation arises because xfs_trans_log_inode() does not return an
error and instead calls xfs_force_shutdown() to ensure that the
transaction being committed is aborted.
Unfortunately, we have a race condition where xfs_trans_commit()
needs to check xlog_is_shutdown() because it can't abort log items
before the log is shut down, but it needs to use xfs_is_shutdown()
because xfs_forced_shutdown() does not block waiting for the log to
shut down.
To fix this conundrum, first we make all calls to
xfs_forced_shutdown() block until the log is also shut down. This
means we can then safely use xfs_forced_shutdown() as a mechanism
that ensures the currently running transaction will be aborted by
xfs_trans_commit() regardless of the shutdown check it uses.
Dave Chinner [Wed, 30 Mar 2022 01:22:01 +0000 (18:22 -0700)]
xfs: log shutdown triggers should only shut down the log
We've got a mess on our hands.
1. xfs_trans_commit() cannot cancel transactions because the mount is
shut down - that causes dirty, aborted, unlogged log items to sit
unpinned in memory and potentially get written to disk before the
log is shut down. Hence xfs_trans_commit() can only abort
transactions when xlog_is_shutdown() is true.
2. xfs_force_shutdown() is used in places to cause the current
modification to be aborted via xfs_trans_commit() because it may be
impractical or impossible to cancel the transaction directly, and
hence xfs_trans_commit() must cancel transactions when
xfs_is_shutdown() is true in this situation. But we can't do that
because of #1.
3. Log IO errors cause log shutdowns by calling xfs_force_shutdown()
to shut down the mount and then the log from log IO completion.
4. xfs_force_shutdown() can result in a log force being issued,
which has to wait for log IO completion before it will mark the log
as shut down. If #3 races with some other shutdown trigger that runs
a log force, we rely on xfs_force_shutdown() silently ignoring #3
and avoiding shutting down the log until the failed log force
completes.
5. To ensure #2 always works, we have to ensure that
xfs_force_shutdown() does not return until the the log is shut down.
But in the case of #4, this will result in a deadlock because the
log Io completion will block waiting for a log force to complete
which is blocked waiting for log IO to complete....
So the very first thing we have to do here to untangle this mess is
dissociate log shutdown triggers from mount shutdowns. We already
have xlog_forced_shutdown, which will atomically transistion to the
log a shutdown state. Due to internal asserts it cannot be called
multiple times, but was done simply because the only place that
could call it was xfs_do_force_shutdown() (i.e. the mount shutdown!)
and that could only call it once and once only. So the first thing
we do is remove the asserts.
We then convert all the internal log shutdown triggers to call
xlog_force_shutdown() directly instead of xfs_force_shutdown(). This
allows the log shutdown triggers to shut down the log without
needing to care about mount based shutdown constraints. This means
we shut down the log independently of the mount and the mount may
not notice this until it's next attempt to read or modify metadata.
At that point (e.g. xfs_trans_commit()) it will see that the log is
shutdown, error out and shutdown the mount.
To ensure that all the unmount behaviours and asserts track
correctly as a result of a log shutdown, propagate the shutdown up
to the mount if it is not already set. This keeps the mount and log
state in sync, and saves a huge amount of hassle where code fails
because of a log shutdown but only checks for mount shutdowns and
hence ends up doing the wrong thing. Cleaning up that mess is
an exercise for another day.
This enables us to address the other problems noted above in
followup patches.
Dave Chinner [Wed, 30 Mar 2022 01:22:00 +0000 (18:22 -0700)]
xfs: run callbacks before waking waiters in xlog_state_shutdown_callbacks
Brian reported a null pointer dereference failure during unmount in
xfs/006. He tracked the problem down to the AIL being torn down
before a log shutdown had completed and removed all the items from
the AIL. The failure occurred in this path while unmount was
proceeding in another task:
This is processing an EIO error to a log write, and it's
triggering a force shutdown. This causes the log to be shut down,
and then it is running attached iclog callbacks from the shutdown
context. That means the fs and log has already been marked as
xfs_is_shutdown/xlog_is_shutdown and so high level code will abort
(e.g. xfs_trans_commit(), xfs_log_force(), etc) with an error
because of shutdown.
The umount would have been blocked waiting for a log force
completion inside xfs_log_cover() -> xfs_sync_sb(). The first thing
for this situation to occur is for xfs_sync_sb() to exit without
waiting for the iclog buffer to be comitted to disk. The
above trace is the completion routine for the iclog buffer, and
it is shutting down the filesystem.
xlog_state_shutdown_callbacks() does this:
{
struct xlog_in_core *iclog;
LIST_HEAD(cb_list);
spin_lock(&log->l_icloglock);
iclog = log->l_iclog;
do {
if (atomic_read(&iclog->ic_refcnt)) {
/* Reference holder will re-run iclog callbacks. */
continue;
}
list_splice_init(&iclog->ic_callbacks, &cb_list);
>>>>>> wake_up_all(&iclog->ic_write_wait);
>>>>>> wake_up_all(&iclog->ic_force_wait);
} while ((iclog = iclog->ic_next) != log->l_iclog);
This wakes any thread waiting on IO completion of the iclog (in this
case the umount log force) before shutdown processes all the pending
callbacks. That means the xfs_sync_sb() waiting on a sync
transaction in xfs_log_force() on iclog->ic_force_wait will get
woken before the callbacks attached to that iclog are run. This
results in xfs_sync_sb() returning an error, and so unmount unblocks
and continues to run whilst the log shutdown is still in progress.
Normally this is just fine because the force waiter has nothing to
do with AIL operations. But in the case of this unmount path, the
log force waiter goes on to tear down the AIL because the log is now
shut down and so nothing ever blocks it again from the wait point in
xfs_log_cover().
Hence it's a race to see who gets to the AIL first - the unmount
code or xlog_cil_process_committed() killing the superblock buffer.
To fix this, we just have to change the order of processing in
xlog_state_shutdown_callbacks() to run the callbacks before it wakes
any task waiting on completion of the iclog.
Dave Chinner [Wed, 30 Mar 2022 01:22:00 +0000 (18:22 -0700)]
xfs: shutdown in intent recovery has non-intent items in the AIL
generic/388 triggered a failure in RUI recovery due to a corrupted
btree record and the system then locked up hard due to a subsequent
assert failure while holding a spinlock cancelling intents:
XFS (pmem1): Corruption of in-memory data (0x8) detected at xfs_do_force_shutdown+0x1a/0x20 (fs/xfs/xfs_trans.c:964). Shutting down filesystem.
XFS (pmem1): Please unmount the filesystem and rectify the problem(s)
XFS: Assertion failed: !xlog_item_is_intent(lip), file: fs/xfs/xfs_log_recover.c, line: 2632
Call Trace:
<TASK>
xlog_recover_cancel_intents.isra.0+0xd1/0x120
xlog_recover_finish+0xb9/0x110
xfs_log_mount_finish+0x15a/0x1e0
xfs_mountfs+0x540/0x910
xfs_fs_fill_super+0x476/0x830
get_tree_bdev+0x171/0x270
? xfs_init_fs_context+0x1e0/0x1e0
xfs_fs_get_tree+0x15/0x20
vfs_get_tree+0x24/0xc0
path_mount+0x304/0xba0
? putname+0x55/0x60
__x64_sys_mount+0x108/0x140
do_syscall_64+0x35/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
Essentially, there's dirty metadata in the AIL from intent recovery
transactions, so when we go to cancel the remaining intents we assume
that all objects after the first non-intent log item in the AIL are
not intents.
This is not true. Intent recovery can log new intents to continue
the operations the original intent could not complete in a single
transaction. The new intents are committed before they are deferred,
which means if the CIL commits in the background they will get
inserted into the AIL at the head.
Hence if we shut down the filesystem while processing intent
recovery, the AIL may have new intents active at the current head.
Hence this check:
/*
* We're done when we see something other than an intent.
* There should be no intents left in the AIL now.
*/
if (!xlog_item_is_intent(lip)) {
#ifdef DEBUG
for (; lip; lip = xfs_trans_ail_cursor_next(ailp, &cur))
ASSERT(!xlog_item_is_intent(lip));
#endif
break;
}
in both xlog_recover_process_intents() and
log_recover_cancel_intents() is simply not valid. It was valid back
when we only had EFI/EFD intents and didn't chain intents, but it
hasn't been valid ever since intent recovery could create and commit
new intents.
Given that crashing the mount task like this pretty much prevents
diagnosing what went wrong that lead to the initial failure that
triggered intent cancellation, just remove the checks altogether.
projects / linux.git / log