Performance regression reported with NFS/RDMA using Omnipath,
bisected to commit e084ee673c ("svcrdma: Add Write chunk WRs to
the RPC's Send WR chain").
Tracing on the server reports:
nfsd-7771 [060] 1758.891809: svcrdma_sq_post_err:
cq.id=205 cid=226 sc_sq_avail=13643/851 status=-12
sq_post_err reports ENOMEM, and the rdma->sc_sq_avail (13643) is
larger than rdma->sc_sq_depth (851). The number of available Send
Queue entries is always supposed to be smaller than the Send Queue
depth. That seems like a Send Queue accounting bug in svcrdma.
As it's getting to be late in the 6.9-rc cycle, revert this commit.
It can be revisited in a subsequent kernel release.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=218743
Fixes: e084ee673c ("svcrdma: Add Write chunk WRs to the RPC's Send WR chain")
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The IP core only has breaking changes when there major version changes.
Hence, only match the major number. This is also in line with the other
core ADI has upstream. The current check for erroring out
'expected_version > current_version"' is then wrong as we could just
increase the core major with breaking changes and that would go
unnoticed.
Fixes: ef04070692 ("iio: adc: adi-axi-adc: add support for AXI ADC IP core")
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-ad9467-new-features-v1-2-3e7628ff6d5e@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
struct Scsi_Host private data contains pointer to struct ctlr_info.
Restore allocation of only 8 bytes to store pointer in struct Scsi_Host
private data area.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: bbbd254991 ("scsi: hpsa: Fix allocation size for scsi_host_alloc()")
Signed-off-by: Yuri Karpov <YKarpov@ispras.ru>
Link: https://lore.kernel.org/r/20240312170447.743709-1-YKarpov@ispras.ru
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sparse reports:
fs/xfs/xfs_log_cil.c:1127:1: warning: context imbalance in 'xlog_cil_push_work' - different lock contexts for basic block
fs/xfs/xfs_log_cil.c:1380:1: warning: context imbalance in 'xlog_cil_push_background' - wrong count at exit
fs/xfs/xfs_log_cil.c:1623:9: warning: context imbalance in 'xlog_cil_commit' - unexpected unlock
xlog_cil_push_background() has a locking annotations for an rw_sem.
Sparse does not track lock contexts for rw_sems, so the
annotation generates false warnings. Remove the annotation.
xlog_wait_on_iclog() drops the log->l_ic_loglock. The function has a
sparse annotation, but the prototype in xfs_log_priv.h does not.
Hence the warning from xlog_cil_push_work() which calls
xlog_wait_on_iclog(). Add the missing annotation.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Chandan Babu R <chandanbabu@kernel.org>
The AD9739A is a 14-bit, 2.5 GSPS high performance RF DACs that are capable
of synthesizing wideband signals from DC up to 3 GHz.
A dual-port, source synchronous, LVDS interface simplifies the digital
interface with existing FGPA/ASIC technology. On-chip controllers are used
to manage external and internal clock domain variations over temperature to
ensure reliable data transfer from the host to the DAC core.
Co-developed-by: Dragos Bogdan <dragos.bogdan@analog.com>
Signed-off-by: Dragos Bogdan <dragos.bogdan@analog.com>
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-iio-backend-axi-dac-v4-10-5ca45b4de294@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Support the Analog Devices Generic AXI DAC IP core. The IP core is used
for interfacing with digital-to-analog (DAC) converters that require either
a high-speed serial interface (JESD204B/C) or a source synchronous parallel
interface (LVDS/CMOS). Typically (for such devices) SPI will be used for
configuration only, while this IP core handles the streaming of data into
memory via DMA.
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-iio-backend-axi-dac-v4-9-5ca45b4de294@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This adds the needed backend ops for supporting a backend inerfacing
with an high speed dac. The new ops are:
* data_source_set();
* set_sampling_freq();
* extend_chan_spec();
* ext_info_set();
* ext_info_get().
Also to note the new helpers that are meant to be used by the backends
when extending an IIO channel (adding extended info):
* iio_backend_ext_info_set();
* iio_backend_ext_info_get().
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-iio-backend-axi-dac-v4-8-5ca45b4de294@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Use the iio_dma_buffer_write() and iio_dma_buffer_space_available()
functions provided by the buffer-dma core, to enable write support in
the buffer-dmaengine code.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-iio-backend-axi-dac-v4-5-5ca45b4de294@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Update the devm_iio_dmaengine_buffer_setup() function to support
specifying the buffer direction.
Update the iio_dmaengine_buffer_submit() function to handle input
buffers as well as output buffers.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-iio-backend-axi-dac-v4-4-5ca45b4de294@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
As of commit 7d1d865181 ("[SCSI] libsas: fix false positive 'device
attached' conditions"), reset the phy->entacted_sas_addr address to a
zero-address when the link rate is less than 1.5G.
Currently we find that when a new device is attached, and the link rate is
less than 1.5G, but the device type is not NO_DEVICE, for example: the link
rate is SAS_PHY_RESET_IN_PROGRESS and the device type is stp. After setting
the phy->entacted_sas_addr address to the zero address, the port will
continue to be created for the phy with the zero-address, and other phys
with the zero-address will be tried to be added to the new port:
[562240.051197] sas: ex 500e004aaaaaaa1f phy19:U:0 attached: 0000000000000000 (no device)
// phy19 is deleted but still on the parent port's phy_list
[562240.062536] sas: ex 500e004aaaaaaa1f phy0 new device attached
[562240.062616] sas: ex 500e004aaaaaaa1f phy00:U:5 attached: 0000000000000000 (stp)
[562240.062680] port-7:7:0: trying to add phy phy-7:7:19 fails: it's already part of another port
Therefore, it should be the same as sas_get_phy_attached_dev(). Only when
device_type is SAS_PHY_UNUSED, sas_address is set to the 0 address.
Fixes: 7d1d865181 ("[SCSI] libsas: fix false positive 'device attached' conditions")
Signed-off-by: Xingui Yang <yangxingui@huawei.com>
Link: https://lore.kernel.org/r/20240312141103.31358-5-yangxingui@huawei.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
We found that when ex_phy was attached and added to the parent wide port,
ex_phy->port was not set, resulting in sas_unregister_devs_sas_addr() not
calling sas_port_delete_phy() when deleting the phy, and the deleted phy
was still on the parent wide port's phy_list.
When we use sas_port_add_ex_phy() to set ex_phy->port to solve the above
problem, we find that after all the phys of the parent_port are removed and
the number of phy becomes 0, the parent_port will not be set to NULL. This
causes the freed parent port to be used when attaching a new ex_phy in
sas_ex_add_parent_port().
Use sas_port_add_ex_phy() instead of sas_port_add_phy() to set ex_phy->port
when ex_phy is added to the parent port, and set ex_dev->parent_port to
NULL when the number of phy on the port becomes 0.
Signed-off-by: Xingui Yang <yangxingui@huawei.com>
Link: https://lore.kernel.org/r/20240312141103.31358-4-yangxingui@huawei.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Move sas_add_parent_port() to sas_expander.c and rename it to
sas_ex_add_parent_port() as it is only used in this file.
Signed-off-by: Xingui Yang <yangxingui@huawei.com>
Link: https://lore.kernel.org/r/20240312141103.31358-3-yangxingui@huawei.com
Reviewed-by: John Garry <john.g.garry@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
This moves the process of adding ex_phy to a port into a new helper.
Signed-off-by: Xingui Yang <yangxingui@huawei.com>
Link: https://lore.kernel.org/r/20240312141103.31358-2-yangxingui@huawei.com
Reviewed-by: John Garry <john.g.garry@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Adding write support to the buffer-dma code is easy - the write()
function basically needs to do the exact same thing as the read()
function: dequeue a block, read or write the data, enqueue the block
when entirely processed.
Therefore, the iio_buffer_dma_read() and the new iio_buffer_dma_write()
now both call a function iio_buffer_dma_io(), which will perform this
task.
Note that we preemptively reset block->bytes_used to the buffer's size
in iio_dma_buffer_request_update(), as in the future the
iio_dma_buffer_enqueue() function won't reset it.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-iio-backend-axi-dac-v4-3-5ca45b4de294@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Change its name to iio_dma_buffer_usage(), as this function can be used
both for the .data_available and the .space_available callbacks.
Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-iio-backend-axi-dac-v4-2-5ca45b4de294@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This brings the DMA buffer API more in line with what we have in the
triggered buffer. There's no need of having both
devm_iio_dmaengine_buffer_setup() and devm_iio_dmaengine_buffer_alloc().
Hence we introduce the new iio_dmaengine_buffer_setup() that together
with devm_iio_dmaengine_buffer_setup() should be all we need.
Note that as part of this change iio_dmaengine_buffer_alloc() is again
static and the axi-adc was updated accordingly.
Signed-off-by: Nuno Sa <nuno.sa@analog.com>
Link: https://lore.kernel.org/r/20240419-iio-backend-axi-dac-v4-1-5ca45b4de294@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Both functions `dps310_get_pres_precision` and
`dps310_get_temp_precision` provide the oversampling rate by calling the
`BIT()` macro. However, to look up the corresponding scale factor, we
need the register value itself. Currently, this is achieved by undoing
the calculation of the oversampling rate with `ilog2()`.
Simplify the two functions for getting the scale factor and directly
use the register content for the lookup.
Signed-off-by: Thomas Haemmerle <thomas.haemmerle@leica-geosystems.com>
Link: https://lore.kernel.org/r/20240415105030.1161770-5-thomas.haemmerle@leica-geosystems.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Align error handling with `dps310_calculate_temp`, where it's not
possible to differentiate between errors and valid calculations by
checking if the returned value is negative.
Signed-off-by: Thomas Haemmerle <thomas.haemmerle@leica-geosystems.com>
Link: https://lore.kernel.org/r/20240415105030.1161770-3-thomas.haemmerle@leica-geosystems.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The current implementation interprets negative values returned from
`dps310_calculate_temp` as error codes.
This has a side effect that when negative temperature values are
calculated, they are interpreted as error.
Fix this by using the return value only for error handling and passing a
pointer for the value.
Fixes: ba6ec48e76 ("iio: Add driver for Infineon DPS310")
Signed-off-by: Thomas Haemmerle <thomas.haemmerle@leica-geosystems.com>
Link: https://lore.kernel.org/r/20240415105030.1161770-2-thomas.haemmerle@leica-geosystems.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
We currently insist on disabling PAuth on vcpu_load(), and get to
enable it on first guest use of an instruction or a key (ignoring
the NV case for now).
It isn't clear at all what this is trying to achieve: guests tend
to use PAuth when available, and nothing forces you to expose it
to the guest if you don't want to. This also isn't totally free:
we take a full GPR save/restore between host and guest, only to
write ten 64bit registers. The "value proposition" escapes me.
So let's forget this stuff and enable PAuth eagerly if exposed to
the guest. This results in much simpler code. Performance wise,
that's not bad either (tested on M2 Pro running a fully automated
Debian installer as the workload):
- On a non-NV guest, I can see reduction of 0.24% in the number
of cycles (measured with perf over 10 consecutive runs)
- On a NV guest (L2), I see a 2% reduction in wall-clock time
(measured with 'time', as M2 doesn't have a PMUv3 and NV
doesn't support it either)
So overall, a much reduced complexity and a (small) performance
improvement.
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-16-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Now that we (hopefully) correctly handle ERETAx, drop the masking
of the PAuth feature (something that was not even complete, as
APA3 and AGA3 were still exposed).
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-15-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Now that we have some emulation in place for ERETA[AB], we can
plug it into the exception handling machinery.
As for a bare ERET, an "easy" ERETAx instruction is processed as
a fixup, while something that requires a translation regime
transition or an exception delivery is left to the slow path.
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-14-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
FEAT_NV has the interesting property of relying on ERET being
trapped. An added complexity is that it also traps ERETAA and
ERETAB, meaning that the Pointer Authentication aspect of these
instruction must be emulated.
Add an emulation of Pointer Authentication, limited to ERETAx
(always using SP_EL2 as the modifier and ELR_EL2 as the pointer),
using the Generic Authentication instructions.
The emulation, however small, is placed in its own compilation
unit so that it can be avoided if the configuration doesn't
include it (or the toolchan in not up to the task).
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-13-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Pointer Authentication comes in many flavors, and a faithful emulation
relies on correctly handling the flavour implemented by the HW.
For this, provide a new kvm_has_pauth() that checks whether we
expose to the guest a particular level of support. This checks
across all 3 possible authentication algorithms (Q5, Q3 and IMPDEF).
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-12-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
In order for a L1 hypervisor to correctly handle PAuth instructions,
it must observe traps caused by a L1 PAuth instruction when
HCR_EL2.API==0. Since we already handle the case for API==1 as
a fixup, only the exception injection case needs to be handled.
Rework the kvm_handle_ptrauth() callback to reinject the trap
in this case. Note that APK==0 is already handled by the exising
triage_sysreg_trap() helper.
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-11-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Although KVM couples API and APK for simplicity, the architecture
makes no such requirement, and the two can be independently set or
cleared.
Check for which of the two possible reasons we have trapped here,
and if the corresponding L1 control bit isn't set, delegate the
handling for forwarding.
Otherwise, set this exact bit in HCR_EL2 and resume the guest.
Of course, in the non-NV case, we keep setting both bits and
be done with it. Note that the entry core already saves/restores
the keys should any of the two control bits be set.
This results in a bit of rework, and the removal of the (trivial)
vcpu_ptrauth_enable() helper.
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-10-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
If the L1 hypervisor decides to trap ERETs while running L2,
make sure we don't try to emulate it, just like we wouldn't
if it had its NV bit set.
The exception will be reinjected from the core handler.
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-9-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
A significant part of the FEAT_NV extension is to trap ERET
instructions so that the hypervisor gets a chance to switch
from a vEL2 L1 guest to an EL1 L2 guest.
But this also has the unfortunate consequence of trapping ERET
in unsuspecting circumstances, such as staying at vEL2 (interrupt
handling while being in the guest hypervisor), or returning to host
userspace in the case of a VHE guest.
Although we already make some effort to handle these ERET quicker
by not doing the put/load dance, it is still way too far down the
line for it to be efficient enough.
For these cases, it would ideal to ERET directly, no question asked.
Of course, we can't do that. But the next best thing is to do it as
early as possible, in fixup_guest_exit(), much as we would handle
FPSIMD exceptions.
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-8-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Honor the trap forwarding bits for both ERET and SMC, using a new
helper that checks for common conditions.
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Co-developed-by: Jintack Lim <jintack.lim@linaro.org>
Signed-off-by: Jintack Lim <jintack.lim@linaro.org>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-7-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Add the HCR_EL2 configuration for FEAT_NV2, adding the required
bits for running a guest hypervisor, and overall merging the
allowed bits provided by the guest.
This heavily replies on unavaliable features being sanitised
when the HCR_EL2 shadow register is accessed, and only a couple
of bits must be explicitly disabled.
Non-NV guests are completely unaffected by any of this.
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-6-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
It has become obvious that HCR_EL2.NV serves the exact same use
as VCPU_HYP_CONTEXT, only in an architectural way. So just drop
the flag for good.
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-5-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
PAuth comes it two parts: address authentication, and generic
authentication. So far, KVM mandates that both are implemented.
PAuth also comes in three flavours: Q5, Q3, and IMPDEF. Only one
can be implemented for any of address and generic authentication.
Crucially, the architecture doesn't mandate that address and generic
authentication implement the *same* flavour. This would make
implementing ERETAx very difficult for NV, something we are not
terribly keen on.
So only allow PAuth support for KVM on systems that are not totally
insane. Which is so far 100% of the known HW.
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-4-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
The ESR_ELx_ERET_ISS_ERET* macros are a bit confusing:
- ESR_ELx_ERET_ISS_ERET really indicates that we have trapped an
ERETA* instruction, as opposed to an ERET
- ESR_ELx_ERET_ISS_ERETA really indicates that we have trapped
an ERETAB instruction, as opposed to an ERETAA.
We could repaint those to make more sense, but these are the
names that are present in the ARM ARM, and we are sentimentally
attached to those.
Instead, add two new helpers:
- esr_iss_is_eretax() being true tells you that you need to
authenticate the ERET
- esr_iss_is_eretab() tells you that you need to use the B key
instead of the A key
Following patches will make use of these primitives.
Suggested-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-3-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
The unsuspecting kernel tinkerer can be easily confused into
writing something that looks like this:
ikey.lo = __vcpu_sys_reg(vcpu, SYS_APIAKEYLO_EL1);
which seems vaguely sensible, until you realise that the second
parameter is the encoding of a sysreg, and not the index into
the vcpu sysreg file... Debugging what happens in this case is
an interesting exercise in head<->wall interactions.
As they often say: "Any resemblance to actual persons, living
or dead, or actual events is purely coincidental".
In order to save people's time, add some compile-time hardening
that will at least weed out the "stupidly out of range" values.
This will *not* catch anything that isn't a compile-time constant.
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Reviewed-by: Oliver Upton <oliver.upton@linux.dev>
Link: https://lore.kernel.org/r/20240419102935.1935571-2-maz@kernel.org
Signed-off-by: Marc Zyngier <maz@kernel.org>
Add support for the GPADC for the Allwinner H616. It is identical to
the existing ADC for the D1/T113s/R329/T507 SoCs.
Signed-off-by: Chris Morgan <macromorgan@hotmail.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://lore.kernel.org/r/20240417170423.20640-3-macroalpha82@gmail.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The -next commit f8fef2fa41 (tty: msm_serial: use
dmaengine_prep_slave_sg()), switched to using dma_map_sg(). But the
return value of dma_map_sg() is special: it returns number of elements
mapped. And not a standard error value.
The commit also forgot to reset dma->tx_sg in case of this failure.
Fix both these mistakes.
Thanks to Marek who helped debugging this.
Signed-off-by: Jiri Slaby (SUSE) <jirislaby@kernel.org>
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Link: https://lore.kernel.org/r/20240419080931.30949-1-jirislaby@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fix a missing bounds check in superblock validation.
Note that we don't yet have repair code for this case - repair code for
individual items is generally low priority, since the whole superblock
is checksummed, validated prior to write, and we have backups.
Reported-by: lei lu <llfamsec@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Petr Machata says:
====================
mlxsw: Fixes
This patchset fixes the following issues:
- During driver de-initialization the driver unregisters the EMAD
response trap by setting its action to DISCARD. However the manual
only permits TRAP and FORWARD, and future firmware versions will
enforce this.
In patch #1, suppress the error message by aligning the driver to the
manual and use a FORWARD (NOP) action when unregistering the trap.
- The driver queries the Management Capabilities Mask (MCAM) register
during initialization to understand if certain features are supported.
However, not all firmware versions support this register, leading to
the driver failing to load.
Patches #2 and #3 fix this issue by treating an error in the register
query as an indication that the feature is not supported.
v2:
- Patch #2:
- Make mlxsw_env_max_module_eeprom_len_query() void
====================
Link: https://lore.kernel.org/r/cover.1713446092.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The driver queries the Management Capabilities Mask (MCAM) register
during initialization to understand if a new and deeper reset flow is
supported.
However, not all firmware versions support this register, leading to the
driver failing to load.
Fix by treating an error in the register query as an indication that the
feature is not supported.
Fixes: f257c73e53 ("mlxsw: pci: Add support for new reset flow")
Reported-by: Tim 'mithro' Ansell <me@mith.ro>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/ee968c49d53bac96a4c66d1b09ebbd097d81aca5.1713446092.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The driver queries the Management Capabilities Mask (MCAM) register
during initialization to understand if it can read up to 128 bytes from
transceiver modules.
However, not all firmware versions support this register, leading to the
driver failing to load.
Fix by treating an error in the register query as an indication that the
feature is not supported.
Fixes: 1f4aea1f72 ("mlxsw: core_env: Read transceiver module EEPROM in 128 bytes chunks")
Reported-by: Tim 'mithro' Ansell <me@mith.ro>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/0afa8b2e8bac178f5f88211344429176dcc72281.1713446092.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The device's manual (PRM - Programmer's Reference Manual) classifies the
trap that is used to deliver EMAD responses as an "event trap". Among
other things, it means that the only actions that can be associated with
the trap are TRAP and FORWARD (NOP).
Currently, during driver de-initialization the driver unregisters the
trap by setting its action to DISCARD, which violates the above
guideline. Future firmware versions will prevent such misuses by
returning an error. This does not prevent the driver from working, but
an error will be printed to the kernel log during module removal /
devlink reload:
mlxsw_spectrum 0000:03:00.0: Reg cmd access status failed (status=7(bad parameter))
mlxsw_spectrum 0000:03:00.0: Reg cmd access failed (reg_id=7003(hpkt),type=write)
Suppress the error message by aligning the driver to the manual and use
a FORWARD (NOP) action when unregistering the trap.
Fixes: 4ec14b7634 ("mlxsw: Add interface to access registers and process events")
Cc: Jiri Pirko <jiri@resnulli.us>
Cc: Amit Cohen <amcohen@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://lore.kernel.org/r/753a89e14008fde08cb4a2c1e5f537b81d8eb2d6.1713446092.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When bringing down the TX rings we flush the rings but forget to
reclaimed the flushed packets. This leads to a memory leak since we
do not free the dma mapped buffers. This also leads to tx control
block corruption when bringing down the interface for power
management.
Fixes: 490cb41200 ("net: bcmasp: Add support for ASP2.0 Ethernet controller")
Signed-off-by: Justin Chen <justin.chen@broadcom.com>
Acked-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20240418180541.2271719-1-justin.chen@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Florian Westphal says:
====================
testing: make netfilter selftests functional in vng environment
This is the second batch of the netfilter selftest move.
Changes since v1:
- makefile and kernel config are updated to have all required features
- fix makefile with missing bits to make kselftest-install work
- test it via vng as per
https://github.com/linux-netdev/nipa/wiki/How-to-run-netdev-selftests-CI-style
(Thanks Jakub!)
- squash a few fixes, e.g. nft_queue.sh v1 had a race w. NFNETLINK_QUEUE=m
- add a settings file with 8m timeout, for nft_concat_range.sh sake.
That script can be sped up a bit, I think, but its not contained in
this batch yet.
- toss the first two bogus rebase artifacts (Matthieu Baerts)
scripts are moved to lib.sh infra. This allows to use busywait helper
and ditch various 'sleep 2' all over the place.
Tested on Fedora 39:
vng --build --config tools/testing/selftests/net/netfilter/config
make -C tools/testing/selftests/ TARGETS=net/netfilter
vng -v --run . --user root --cpus 2 -- \
make -C tools/testing/selftests TARGETS=net/netfilter run_tests
... all tests pass except nft_audit.sh which SKIPs due to nft version mismatch
(Fedora is on nft 1.0.7 which lacks reset keyword support).
Missing/WIP bits:
- speed up nf_concat_range.sh test
- extend flowtable selftest
- shellcheck fixups for remaining scripts
====================
Link: https://lore.kernel.org/r/20240418152744.15105-1-fw@strlen.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Jakub reports the Makefile missed a few updates to make kselftest-install
work for the netfilter tests and points out that config file lacks many
dependencies such as VETH support.
The settings file (timeout 8m) is added for nft_concat_range.sh script
which can take several minutes to complete.
Fixes: 3f189349e5 ("selftests: netfilter: move to net subdir")
Reported-by: Jakub Kicinski <kuba@kernel.org>
Closes: https://lore.kernel.org/all/20240412175413.04e5e616@kernel.org/
Signed-off-by: Florian Westphal <fw@strlen.de>
Link: https://lore.kernel.org/r/20240418152744.15105-13-fw@strlen.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>