This patch marks the struct regmap_bus tcan4x5x_bus as constant.
Reviewed-by: Dan Murphy <dmurphy@ti.com>
Tested-by: Sean Nyekjaer <sean@geanix.com>
Link: https://lore.kernel.org/r/20201215231746.1132907-6-mkl@pengutronix.de
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This is a preparation patch to move the regmap related code into a seperate
file. This patch removes the tcan4x5x.c to tcan4x5x-core.c.
Reviewed-by: Dan Murphy <dmurphy@ti.com>
Tested-by: Sean Nyekjaer <sean@geanix.com>
Link: https://lore.kernel.org/r/20201215231746.1132907-4-mkl@pengutronix.de
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This patch beautifies the indention of the tcan4x5x_of_match and
tcan4x5x_id_table.
Reviewed-by: Dan Murphy <dmurphy@ti.com>
Tested-by: Sean Nyekjaer <sean@geanix.com>
Link: https://lore.kernel.org/r/20201215231746.1132907-3-mkl@pengutronix.de
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
This patch replaces the DEVICE_NAME macro by KBUILD_MODNAME and removed the
superfluous DEVICE_NAME.
Reviewed-by: Dan Murphy <dmurphy@ti.com>
Tested-by: Sean Nyekjaer <sean@geanix.com>
Link: https://lore.kernel.org/r/20201215231746.1132907-2-mkl@pengutronix.de
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
It's not used anywhere downstream (and never had been, AFAICS).
Theoretically, fs/binfmt_elf.c does use it, but only in the
non-regset coredump handling and all biarch architectures
end up with that ifdefed out.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
with mips converted to use of fs/config_binfmt_elf.c, there's no
need to keep selects of that thing all over arch/* - we can simply
turn into def_bool y if COMPAT && BINFMT_ELF (in fs/Kconfig.binfmt)
and get rid of all selects.
Several architectures got those selects wrong (e.g. you could
end up with sparc64 sans BINFMT_ELF, with select violating
dependencies, etc.)
Randy Dunlap has spotted some of those; IMO this is simpler than
his fix, but it depends upon the stuff that would need to be
backported, so we might end up using his variant for -stable.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Like amd64, mips has two 32bit ABIs - o32 and n32. Unlike amd64,
it does not use compat_binfmt_elf.c for either of those; each
of those ABIs has a binfmt handler of its own, both very similar
to fs/compat_binfmt_elf.c. And the same technics as we use on
amd64 can be used to make fs/compat_binfmt_elf.c handle both.
* merge elfo32_check_arch() with elfn32_check_arch(),
make that serve as compat_elf_check_arch(). Note that
SET_PERSONALITY2() is already the same for all ABI variants -
it looks at the elf header to choose the flags to set.
* add asm/elfcore-compat.h, using the bigger (n32) variant
of elf32_prstatus as compat_elf_prstatus there.
* make PRSTATUS_SIZE() and SET_PR_FPVALID() choose the
right layout, same as done for amd64. test_thread_flag(TIF_32BIT_REGS)
is used as the predicate.
Voila - we are rid of binfmt_elf{n,o}32.c; fs/compat_binfmt_elf.c is
used, same as for all other ELF-supporting 64bit architectures that
need 32bit compat.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
normal mips one is just fine - it's only used after we'd done
SET_PERSONALITY2() and by that point TASK_SIZE will yield the
right value
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
elf_caddr_t: unused since 2002
jiffies_to_timeval: unused since 2015
TASK_SIZE: used only downstream of SET_PERSONALITY2(), and after that
point the normal definition results in TASK_SIZE32 just fine.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
It's really trivial - the only wrinkle is making sure that
compiler knows that ia32-related side of COMPAT_ARCH_DLINFO
is dead code on such configs (we don't get there without
having passed compat_elf_check_arch(), and on such configs
that'll fail for ia32 binary).
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
To get rid of hardcoded size/offset in those macros we need to have
a definition of i386 variant of struct elf_prstatus. However, we can't
do that in asm/compat.h - the types needed for that are not there and
adding an include of asm/user32.h into asm/compat.h would cause a lot
of mess.
That could be conveniently done in elfcore-compat.h, but currently there
is nowhere to put arch-dependent parts of it - no asm/elfcore-compat.h.
So we introduce a new file (asm/elfcore-compat.h, present on architectures
that have CONFIG_ARCH_HAS_ELFCORE_COMPAT set, currently only on x86),
have it pulled by linux/elfcore-compat.h and move the definitions there.
As a side benefit, we don't need to worry about accidental inclusion of
that file into binfmt_elf.c itself, so we don't need the dance with
COMPAT_PRSTATUS_SIZE, etc. - only fs/compat_binfmt_elf.c will see
that header.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Preparations to doing i386 compat elf_prstatus sanely - rather than duplicating
the beginning of compat_elf_prstatus, take these fields into a separate
structure (compat_elf_prstatus_common), so that it could be reused. Due to
the incestous relationship between binfmt_elf.c and compat_binfmt_elf.c we
need the same shape change done to native struct elf_prstatus, gathering the
fields prior to pr_reg into a new structure (struct elf_prstatus_common).
Fortunately, offset of pr_reg is always a multiple of 16 with no padding
right before it, so it's possible to turn all the stuff prior to it into
a single member without disturbing the layout.
[build fix from Geert Uytterhoeven folded in]
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Tegra Quad SPI controller hardware supports sending dummy bytes based
on programmed dummy clock cycles after the actual transfer bytes.
This patch adds this support of hardware dummy bytes transfer and
skips transfer of dummy bytes from the software.
For dummy cycles more than Tegra Quad SPI hardware maximum dummy
cycles limit, driver transfers dummy bytes from the software.
Signed-off-by: Sowjanya Komatineni <skomatineni@nvidia.com>
Link: https://lore.kernel.org/r/1608585459-17250-7-git-send-email-skomatineni@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch marks dummy transfer by setting dummy_data bit to 1.
Controllers supporting dummy transfer by hardware use this bit field
to skip software transfer of dummy bytes and use hardware dummy bytes
transfer.
Signed-off-by: Sowjanya Komatineni <skomatineni@nvidia.com>
Link: https://lore.kernel.org/r/1608585459-17250-6-git-send-email-skomatineni@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Tegra210 QSPI clock output has divider DIV2_SEL which will be enabled
when using DDR interface mode.
This patch adds clock ID for this to dt-binding.
Acked-by: Rob Herring <robh@kernel.org>
Signed-off-by: Sowjanya Komatineni <skomatineni@nvidia.com>
Link: https://lore.kernel.org/r/1608585459-17250-2-git-send-email-skomatineni@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Double Transfer Rate (DTR) mode transfers data twice per clock cycle.
Add support for parsing DTR ops and set up the registers to allow it.
Most SPI NOR flashes expect 2 byte commands. Parse the 2-byte opcode
from SPI MEM and set it up in the CQSPI_REG_OP_EXT_LOWER register.
Increment the delay needed before issuing indirect writes because larger
delay is needed for DTR mode. With the current delay some writes end up
missing.
Signed-off-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20201222184425.7028-8-p.yadav@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
When performing a direct read via DMA the timeout for completion is set
equal to the read length. This is fine for larger reads. For a small
read like the Read Status Register command, the timeout would be 1 or 2
milliseconds. This is not enough to cover the overhead needed in setting
up DMA.
Make sure the timeout is at least 500 ms to allow DMA ample time to
finish. For reads larger than 500 bytes, the timeout will continue to be
equal to the read length.
Signed-off-by: Pratyush Yadav <p.yadav@ti.com>
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/r/20201222184425.7028-7-p.yadav@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The default SPI MEM supports_op hook rejects DTR ops by default. Add a
simple supports_op hook that very closely imitates the SPI MEM one. It
will be extended in later commits to allow DTR ops.
Signed-off-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20201222184425.7028-6-p.yadav@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
SPI MEM deals with dummy bytes but the controller deals with dummy
cycles. Multiplying bytes by 8 is correct if the dummy phase uses 1S
mode since 1 byte will be sent in 8 cycles. But if the dummy phase uses
4S mode then 1 byte will be sent in 2 cycles.
To correctly translate dummy bytes to dummy cycles, the dummy buswidth
also needs to be taken into account. Divide 8 by the buswidth to get the
correct multiplier for getting the number of cycles.
Signed-off-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20201222184425.7028-5-p.yadav@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
If a command does not have an address phase it goes via the STIG path.
The dummy cycles are not initialized for the STIG commands. As a result,
STIG commands with dummy cycles will not work.
Initialize the dummy cycle field before issuing the STIG command to make
sure it is sent correctly. Move the code to calculate dummy cycle value
to a separate function so it is not repeated twice. DTR support will add
some more logic here to it is worth it to extract it out in a function.
Signed-off-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20201222184425.7028-4-p.yadav@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The controller can only support up to 31 dummy cycles. If the command
requires more it falls back to using 31. This command is likely to fail
because the correct number of cycles are not waited upon. Rather than
silently issuing an incorrect command, fail loudly so the caller can get
a chance to find out the command can't be supported by the controller.
Fixes: 1406234105 ("mtd: spi-nor: Add driver for Cadence Quad SPI Flash Controller")
Signed-off-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20201222184425.7028-3-p.yadav@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
As of commit 9326e4f1e5 ("spi: Limit the spi device max speed to
controller's max speed"), the SPI device max speed is set to the
controller's max speed if it is larger. The Cadence QSPI controller does
not set the controller's max speed so it is left at its initial value of
0. This means the SPI device max speed is always set to 0.
The SPI device max speed is used to calculate the baud rate divider when
performing an operation. If this speed is 0, the default divider of 32
is used. No matter what speed is specified by the device tree property
'spi-max-frequency', the device will always operate at ref_clk / 32.
Fix this by setting master->max_speed_hz to the ref clock speed so the
correct divider can be calculated.
Signed-off-by: Pratyush Yadav <p.yadav@ti.com>
Link: https://lore.kernel.org/r/20201222184425.7028-2-p.yadav@ti.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In mtrr_type_lookup(), if the input memory address region is not in the
MTRR, over 4GB, and not over the top of memory, a write-back attribute
is returned. These condition checks are for ensuring the input memory
address region is actually mapped to the physical memory.
However, if the end address is just aligned with the top of memory,
the condition check treats the address is over the top of memory, and
write-back attribute is not returned.
And this hits in a real use case with NVDIMM: the nd_pmem module tries
to map NVDIMMs as cacheable memories when NVDIMMs are connected. If a
NVDIMM is the last of the DIMMs, the performance of this NVDIMM becomes
very low since it is aligned with the top of memory and its memory type
is uncached-minus.
Move the input end address change to inclusive up into
mtrr_type_lookup(), before checking for the top of memory in either
mtrr_type_lookup_{variable,fixed}() helpers.
[ bp: Massage commit message. ]
Fixes: 0cc705f56e ("x86/mm/mtrr: Clean up mtrr_type_lookup()")
Signed-off-by: Ying-Tsun Huang <ying-tsun.huang@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20201215070721.4349-1-ying-tsun.huang@amd.com
While no information about the H6 RSB controller is included in the
datasheet or manual, the vendor BSP and power management blob both
reference the RSB clock parent and register address. These values were
verified by experimentation.
Since this clock/reset are added late, the specifier is added at the end
to maintain the existing DT binding. The code is kept in register order.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Acked-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Gate the clock to save power while the controller is idle.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Acked-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Since system firmware is likely to use the RSB bus to communicate with a
PMIC while the system is suspended, we cannot make any assumptions about
the controller state after resuming. Thus it is important to completely
reinitialize the controller.
The RSB bus needs to be ready as soon as IRQs are enabled, to handle
wakeup event IRQs coming from the PMIC. Thus it uses NOIRQ callbacks.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Acked-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
This separates the resource acquisition from the hardware initialization
phase, so the hardware initialization can be repeated after system
suspend/resume. The same is done for the exit/remove function, except
that there is no resource deallocation phase due to the use of devres.
The requested RSB clock frequency is stored in `struct sunxi_rsb` so it
will be available when reinitializing the hardware.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Acked-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
For some reason, this driver's OF match table was placed above the
probe/remove functions, far away from the platform_driver definition.
Adding device PM ops would move the table even farther away. Let's move
it to the usual place, right before the platform_driver.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Acked-by: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
We've observed crashes due to an empty cpu mask in
hyperv_flush_tlb_others. Obviously the cpu mask in question is changed
between the cpumask_empty call at the beginning of the function and when
it is actually used later.
One theory is that an interrupt comes in between and a code path ends up
changing the mask. Move the check after interrupt has been disabled to
see if it fixes the issue.
Signed-off-by: Wei Liu <wei.liu@kernel.org>
Cc: stable@kernel.org
Link: https://lore.kernel.org/r/20210105175043.28325-1-wei.liu@kernel.org
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
Commit eff8728fe6 ("vmlinux.lds.h: Add PGO and AutoFDO input
sections") added ".text.unlikely.*" and ".text.hot.*" due to an LLVM
change [1].
After another LLVM change [2], these sections are seen in some PowerPC
builds, where there is a orphan section warning then build failure:
$ make -skj"$(nproc)" \
ARCH=powerpc CROSS_COMPILE=powerpc64le-linux-gnu- LLVM=1 O=out \
distclean powernv_defconfig zImage.epapr
ld.lld: warning: kernel/built-in.a(panic.o):(.text.unlikely.) is being placed in '.text.unlikely.'
...
ld.lld: warning: address (0xc000000000009314) of section .text is not a multiple of alignment (256)
...
ERROR: start_text address is c000000000009400, should be c000000000008000
ERROR: try to enable LD_HEAD_STUB_CATCH config option
ERROR: see comments in arch/powerpc/tools/head_check.sh
...
Explicitly handle these sections like in the main linker script so
there is no more build failure.
[1]: https://reviews.llvm.org/D79600
[2]: https://reviews.llvm.org/D92493
Fixes: 83a092cf95 ("powerpc: Link warning for orphan sections")
Cc: stable@vger.kernel.org
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://github.com/ClangBuiltLinux/linux/issues/1218
Link: https://lore.kernel.org/r/20210104205952.1399409-1-natechancellor@gmail.com
The only used argument in this function is the "req".
Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Reviewed-by: Minwoo Im <minwoo.im.dev@gmail.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
When a queue is in NVMET_RDMA_Q_CONNECTING state, it may has some
requests at rsp_wait_list. In case a disconnect occurs at this
state, no one will empty this list and will return the requests to
free_rsps list. Normally nvmet_rdma_queue_established() free those
requests after moving the queue to NVMET_RDMA_Q_LIVE state, but in
this case __nvmet_rdma_queue_disconnect() is called before. The
crash happens at nvmet_rdma_free_rsps() when calling
list_del(&rsp->free_list), because the request exists only at
the wait list. To fix the issue, simply clear rsp_wait_list when
destroying the queue.
Signed-off-by: Israel Rukshin <israelr@nvidia.com>
Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
There are no callers for nvme_reset_ctrl_sync() and
nvme_alloc_request_qid() so that we keep the symbols exported.
Unexport those functions, mark them static and update the header file
respectively.
Signed-off-by: Minwoo Im <minwoo.im.dev@gmail.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
While handling the completion queue, keep a local copy of the command id
from the DMA-accessible completion entry. This silences a time-of-check
to time-of-use (TOCTOU) warning from KF/x[1], with respect to a
Thunderclap[2] vulnerability analysis. The double-read impact appears
benign.
There may be a theoretical window for @command_id to be used as an
adversary-controlled array-index-value for mounting a speculative
execution attack, but that mitigation is saved for a potential follow-on.
A man-in-the-middle attack on the data payload is out of scope for this
analysis and is hopefully mitigated by filesystem integrity mechanisms.
[1] https://github.com/intel/kernel-fuzzer-for-xen-project
[2] http://thunderclap.io/thunderclap-paper-ndss2019.pdf
Signed-off-by: Lalithambika Krishna Kumar <lalithambika.krishnakumar@intel.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
We may send a request (with or without its data) from two paths:
1. From our I/O context nvme_tcp_io_work which is triggered from:
- queue_rq
- r2t reception
- socket data_ready and write_space callbacks
2. Directly from queue_rq if the send_list is empty (because we want to
save the context switch associated with scheduling our io_work).
However, given that now we have the send_mutex, we may run into a race
condition where none of these contexts will send the pending payload to
the controller. Both io_work send path and queue_rq send path
opportunistically attempt to acquire the send_mutex however queue_rq only
attempts to send a single request, and if io_work context fails to
acquire the send_mutex it will complete without rescheduling itself.
The race can trigger with the following sequence:
1. queue_rq sends request (no incapsule data) and blocks
2. RX path receives r2t - prepares data PDU to send, adds h2cdata PDU
to the send_list and schedules io_work
3. io_work triggers and cannot acquire the send_mutex - because of (1),
ends without self rescheduling
4. queue_rq completes the send, and completes
==> no context will send the h2cdata - timeout.
Fix this by having queue_rq sending as much as it can from the send_list
such that if it still has any left, its because the socket buffer is
full and the socket write_space callback will trigger, thus guaranteeing
that a context will be scheduled to send the h2cdata PDU.
Fixes: db5ad6b7f8 ("nvme-tcp: try to send request in queue_rq context")
Reported-by: Potnuri Bharat Teja <bharat@chelsio.com>
Reported-by: Samuel Jones <sjones@kalrayinc.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Tested-by: Potnuri Bharat Teja <bharat@chelsio.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
A system with more than one of these SSDs will only have one usable.
Hence the kernel fails to detect nvme devices due to duplicate cntlids.
[ 6.274554] nvme nvme1: Duplicate cntlid 33 with nvme0, rejecting
[ 6.274566] nvme nvme1: Removing after probe failure status: -22
Adding the NVME_QUIRK_IGNORE_DEV_SUBNQN quirk to resolves the issue.
Signed-off-by: Gopal Tiwari <gtiwari@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Kernel robot had the following warnings:
>> fcloop.c:1506:6: warning: %x in format string (no. 1) requires
>> 'unsigned int *' but the argument type is 'signed int *'.
>> [invalidScanfArgType_int]
>> if (sscanf(buf, "%x:%d:%d", &opcode, &starting, &amount) != 3)
>> ^
Resolve by changing opcode from and int to an unsigned int
and
>> fcloop.c:1632:32: warning: Uninitialized variable: lport [uninitvar]
>> ret = __wait_localport_unreg(lport);
>> ^
>> fcloop.c:1615:28: warning: Uninitialized variable: nport [uninitvar]
>> ret = __remoteport_unreg(nport, rport);
>> ^
These aren't actual issues as the values are assigned prior to use.
It appears the tool doesn't understand list_first_entry_or_null().
Regardless, quiet the tool by initializing the pointers to NULL at
declaration.
Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Recent patches changed calling sequences. nvme_fc_abort_outstanding_ios
used to be called from a timeout or work context. Now it is being called
in an io completion context, which can be an interrupt handler.
Unfortunately, the abort outstanding ios routine attempts to stop nvme
queues and nested routines that may try to sleep, which is in conflict
with the interrupt handler.
Correct replacing the direct call with a work element scheduling, and the
abort outstanding ios routine will be called in the work element.
Fixes: 95ced8a2c7 ("nvme-fc: eliminate terminate_io use by nvme_fc_error_recovery")
Signed-off-by: James Smart <james.smart@broadcom.com>
Reported-by: Daniel Wagner <dwagner@suse.de>
Tested-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cristian is actively developing new features and more involved than me.
So add Cristian as a designated reviewer. Also add the newly added scmi
regulator driver to the list.
Link: https://lore.kernel.org/r/20210105151945.406093-1-sudeep.holla@arm.com
Cc: Cristian Marussi <cristian.marussi@arm.com>
Acked-by: Cristian Marussi <cristian.marussi@arm.com>
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
The SMC/HVC SCMI transport is modified to allow the completion of an
SCMI message to be indicated by an interrupt rather than the return of
the smc/hvc call. This accommodates the existing behavior of the
BrcmSTB SCMI "platform".
Link: https://lore.kernel.org/r/20201222145603.40192-3-jim2101024@gmail.com
Signed-off-by: Jim Quinlan <jim2101024@gmail.com>
[sudeep.holla: added call to reinit_completion, whitespace cleanup, dropped
irrelavant info in the commit log]
Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>