Now that KVM does NOT gift async #PF workers a "struct kvm" reference,
don't bother skipping "done" workers when flushing/canceling queued
workers, as the deadlock that was being fudged around can no longer occur.
When workers, i.e. async_pf_execute(), were gifted a referenced, it was
possible for a worker to put the last reference and trigger VM destruction,
i.e. trigger flushing of a workqueue from a worker in said workqueue.
Note, there is no actual lock, the deadlock was that a worker will be
stuck waiting for itself (the workqueue code simulates a lock/unlock via
lock_map_{acquire,release}()).
Skipping "done" workers isn't problematic per se, but using work->vcpu as
a "done" flag is confusing, e.g. it's not clear that async_pf.lock is
acquired to protect the work->vcpu, NOT the processing of async_pf.queue
(which is protected by vcpu->mutex).
This reverts commit 22583f0d9c.
Suggested-by: Xu Yilun <yilun.xu@linux.intel.com>
Link: https://lore.kernel.org/r/20240423191649.2885257-1-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Default halt_poll_ns_shrink value of 0 always resets polling interval
to 0 on an un-successful poll where vcpu wakeup is not received. This is
mostly to avoid pointless polling for more number of shorter intervals. But
disabled shrink assumes vcpu wakeup is less likely to be received in
subsequent shorter polling intervals. Another side effect of 0 shrink value
is that, even on a successful poll if total block time was greater than
current polling interval, the polling interval starts over from 0 instead
of shrinking by a factor.
Enabling shrink with value of 2 allows the polling interval to gradually
decrement in case of un-successful poll events as well. This gives a fair
chance for successful polling events in subsequent polling intervals rather
than resetting it to 0 and starting over from grow_start.
Below kvm stat log snippet shows interleaved growth and shrinking of
polling interval:
87162647182125: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 10000 (grow 0)
87162647637763: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (grow 10000)
87162649627943: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 40000 (grow 20000)
87162650892407: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (shrink 40000)
87162651540378: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 40000 (grow 20000)
87162652276768: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (shrink 40000)
87162652515037: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 40000 (grow 20000)
87162653383787: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (shrink 40000)
87162653627670: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 10000 (shrink 20000)
87162653796321: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 20000 (grow 10000)
87162656171645: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 10000 (shrink 20000)
87162661607487: kvm_halt_poll_ns: vcpu 0: halt_poll_ns 0 (shrink 10000)
Having both grow and shrink enabled creates a balance in polling interval
growth and shrink behavior. Tests show improved successful polling attempt
ratio which contribute to VM performance. Power penalty is quite negligible
as shrunk polling intervals create bursts of very short durations.
Performance assessment results show 3-6% improvements in CPU+GPU, Memory
and Storage Android VM workloads whereas 5-9% improvement in average FPS of
gaming VM workloads.
Power penalty is below 1% where host OS is either idle or running a
native workload having 2 VMs enabled. CPU/GPU intensive gaming workloads
as well do not show any increased power overhead with shrink enabled.
Co-developed-by: Rajendran Jaishankar <jaishankar.rajendran@intel.com>
Signed-off-by: Rajendran Jaishankar <jaishankar.rajendran@intel.com>
Signed-off-by: Parshuram Sangle <parshuram.sangle@intel.com>
Link: https://lore.kernel.org/r/20231102154628.2120-2-parshuram.sangle@intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
After
faf01aef05 ("KVM: PPC: Merge powerpc's debugfs entry content into generic entry")
kvm_debugfs_dir is not used anywhere else outside of kvm_main.c
Unexport it and make it static.
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20240515150804.9354-1-bp@kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
The call to cc_platform_has() triggers a fault and system crash if call depth
tracking is active because the GS segment has been reset by load_segments() and
GS_BASE is now 0 but call depth tracking uses per-CPU variables to operate.
Call cc_platform_has() earlier in the function when GS is still valid.
[ bp: Massage. ]
Fixes: 5d8213864a ("x86/retbleed: Add SKL return thunk")
Signed-off-by: David Kaplan <david.kaplan@amd.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Cc: <stable@kernel.org>
Link: https://lore.kernel.org/r/20240603083036.637-1-bp@kernel.org
Rename console_replay_all() to console_try_replay_all() to make
clear that the implementation is best effort. Also, the function
should not be called in NMI context as it takes locks, so update
the comment in code.
Fixes: 693f75b91a ("printk: Add function to replay kernel log on consoles")
Fixes: 1b743485e2 ("tty/sysrq: Replay kernel log messages on consoles via sysrq")
Suggested-by: Petr Mladek <pmladek@suse.com>
Signed-off-by: Shimoyashiki Taichi <taichi.shimoyashiki@sony.com>
Signed-off-by: Sreenath Vijayan <sreenath.vijayan@sony.com>
Link: https://lore.kernel.org/r/Zlguq/wU21Z8MqI4@sreenath.vijayan@sony.com
Reviewed-by: Petr Mladek <pmladek@suse.com>
Signed-off-by: Petr Mladek <pmladek@suse.com>
The iterator variable dst cannot be NULL and the if check can be removed.
Remove it and fix the following Coccinelle/coccicheck warning reported
by itnull.cocci:
ERROR: iterator variable bound on line 762 cannot be NULL
Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Toke Høiland-Jørgensen <toke@redhat.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: https://lore.kernel.org/bpf/20240529101900.103913-2-thorsten.blum@toblux.com
make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in lib/test_bpf.o
Add the missing invocation of the MODULE_DESCRIPTION() macro.
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240531-md-lib-test_bpf-v1-1-868e4bd2f9ed@quicinc.com
The spi-pxa2xx.c is bloated with a platform driver code while
pretending to provide a core functionality. Make it real core
library by splitting out the platform driver to a separate file.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://msgid.link/r/20240530151117.1130792-11-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In preparation of the extracting platform driver from spi-pxa2xx.c
split the probe and remove functions so we have bus independent
and platform device ones.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://msgid.link/r/20240530151117.1130792-10-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
pxa2xx_spi_fw_translate_cs() checks for the ACPI companion device
presence along with the SSP type. But the SSP type is uniquely
determines the case. Hence remove the superflous check.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://msgid.link/r/20240530151117.1130792-9-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The mmio_base can't be NULL at this point. It's either checked
in both pxa_ssp_probe() and pxa2xx_spi_init_ssp() or correctly
provided by PCI core. Hence, remove duplicate check which is
a dead code.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://msgid.link/r/20240530151117.1130792-8-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Print DMA burst size only when DMA is enabled to avoid making
a false impression that DMA is enabled when it may be not.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://msgid.link/r/20240530151117.1130792-7-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Since the ACPI enumerated devices provide a property with SSP type,
there is no more necessity to bear the copy of them in the ID table.
Drop the driver data in ACPI ID table.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://msgid.link/r/20240530151117.1130792-4-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The old Intel platforms, such as Intel Braswell, also provide
the property of SSP type. Reorganize the pxa2xx_spi_init_pdata()
to take that into account.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://msgid.link/r/20240530151117.1130792-3-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In the error path or remove path the reference counter in PXA SSP driver
may be dropped before the other resources, that were allocated after
bumbing the reference counter. This breaks reversed order of freeing and
might have an undesired side effects. Prevent this from happening by
wrapping pxa_ssp_request() to be device managed resource.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://msgid.link/r/20240530151117.1130792-2-andriy.shevchenko@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Follow the convention already in use by other platforms on naming the
sof_dev_desc struct by dropping the underscore between adl and n
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://msgid.link/r/20240603072544.5215-1-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The "ADC2_BCS Disable Switch" control does basically nothing: sets field
in driver's private data structure which is never read again to
configure hardware. Drop the control as it has no effect.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://msgid.link/r/20240523124808.42923-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
All four microphone bias registers have similar layout: VOUT control
mask is exactly the same. Use one define, just like older driver
wcd938x is doing, to make the code a bit simpler.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://msgid.link/r/20240523125049.43713-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Now graph_link_init() is parsing dai_link related property.
Let's handle mclk_fs on it, too.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/87v82yfrv0.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Now graph_link_init() is parsing dai_link related property.
Let's handle mclk_fs on it, too.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/87y17ufrv8.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Current audio-graph-card is checking if the node name was "ports" or
not when parsing the property.
if (of_node_name_eq(ports, "ports"))
of_xxx(ports, ...);
Now, it is using new port_to_ports() which will be NULL if the node
doesn't have "ports", and each of_xxx functions will do nothing if
node was NULL.
Now we don't need to check ports node name. Let's remove and cleanup it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/871q5mh6fv.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Current audio-graph-card is using of_get_parent() to get "port" from
"ep", or get "ports" from "port". But it is difficlut to understand,
and "ports" might not exist.
This patch adds ep_to_port() to get "port" from "ep", and port_to_ports()
to get "ports" from "port". "ports" will be NULL if not exist.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/8734q2h6fz.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Now graph_link_init() is parsing dai_link related property.
Let's handle mclk_fs on it, too.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/874jaih6g3.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Current dai_link related property are parsed and enabled only on CPU
port node (A)(b)(c). OTOH, Audio Graph Card2 supports many connections
like Multi-CPU, DPCM, Codec2Codec today. For example in Multi-CPU case,
it will be checked via (X) -> (B) -> (b) process, but (X) / (B) part
property is not parsed.
>From dai_link related settings point of view, (B) (C) part and Codec
port also enabled is more viscerally understandable, and useful.
card2 {
(X) links = <&snd-cpu (A)
&snd-multi (B)
&snd-dpcm (C)
...>
multi {
ports {
(B) snd-multi: port { ... };
...
};
};
dpcm {
ports {
(C) snd-dpcm: port { ... };
...
};
};
codec2codec {
...
};
};
cpu_device {
ports {
(A) snd-cpu: port { ... };
(b) mcpu: port { ... };
(c) dcpu: port { ... };
}
};
One note here is that if it was Multi-CPU/Codec case, 1st port only
enabled to have property it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/875xuyh6g7.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Current audio-graph-card2 is checking if the node name was "ports" or
not when parsing the property.
if (of_node_name_eq(ports, "ports"))
of_xxx(ports, ...);
Now, it is using new port_to_ports() which will be NULL if the node
doesn't have "ports", and each of_xxx functions will do nothing if
node was NULL.
Now we don't need to check ports node name. Let's remove and cleanup it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/877cfeh6gb.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Current audio-graph-card2 is using of_get_parent() to get "port" from
"ep", or get "ports" from "port". But it is difficlut to understand,
and "ports" might not exist.
This patch adds ep_to_port() to get "port" from "ep", and port_to_ports()
to get "ports" from "port". "ports" will be NULL if not exist.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/878qzuh6gf.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
soc-pcm.c :: soc_get_playback_capture() will indicate error
if both playback_only / capture_only were true.
Thus, graph_util_parse_link_direction() which setup playback_only /
capture_only don't need to check it.
And, its return value is not used on existing driver. Let's remove it.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://msgid.link/r/87a5kah6gm.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
'cx2072x_eq_ctrl' has been unused since the original
commit a497a43637 ("ASoC: Add support for Conexant CX2072X CODEC").
Remove it.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://msgid.link/r/20240601225446.183505-4-linux@treblig.org
Signed-off-by: Mark Brown <broonie@kernel.org>
'wm0010_spi_msg' has been unused since the original
commit e3523e0186 ("ASoC: wm0010: Add initial wm0010 DSP driver").
Remove it.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://msgid.link/r/20240601225446.183505-3-linux@treblig.org
Signed-off-by: Mark Brown <broonie@kernel.org>
'rx_macro_reg_mask_val' is unused since the original
commit af3d54b997 ("ASoC: codecs: lpass-rx-macro: add support for
lpass rx macro").
Remove it.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Link: https://msgid.link/r/20240601225446.183505-2-linux@treblig.org
Signed-off-by: Mark Brown <broonie@kernel.org>
make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in sound/soc/fsl/imx-pcm-dma.o
Add the missing invocation of the MODULE_DESCRIPTION() macro.
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://msgid.link/r/20240602-md-snd-fsl-imx-pcm-dma-v1-1-e7efc33c6bf3@quicinc.com
Signed-off-by: Mark Brown <broonie@kernel.org>
make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in sound/soc/mxs/snd-soc-mxs-pcm.o
Add the missing invocation of the MODULE_DESCRIPTION() macro.
Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Link: https://msgid.link/r/20240602-md-snd-soc-mxs-pcm-v1-1-1e663d11328d@quicinc.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Not having linux-arm-msm@ in cc for audio-related changes for Qualcomm
platforms means that interested parties can easily miss the patches. Add
corresponding L: entry so that linux-arm-msm ML gets CC'ed for audio
patches too.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Acked-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://msgid.link/r/20240531-asoc-qcom-cc-lamsm-v1-1-f026ad618496@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>
We just return 0 after the skip_tlv label. No need to use a label.
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://msgid.link/r/20240603073224.14726-3-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
sof_ipc4_dma_config_tlv{} is for Audio DSP firmware only.
Don't set it in dspless mode.
Fixes: 17386cb1b4 ("ASoC: SOF: Intel: hda-dai: set dma_stream_channel_map device")
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://msgid.link/r/20240603073224.14726-2-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Correct typo in bpftool profiler and change all instances of 'MATRICS' to
'METRICS' in the profiler.bpf.c file.
Signed-off-by: Swan Beaujard <beaujardswan@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Quentin Monnet <qmo@kernel.org>
Link: https://lore.kernel.org/bpf/20240602225812.81171-1-beaujardswan@gmail.com
'key_t' is unused in a couple of files since the original commit 60dd49ea65
("selftests/bpf: Add test for bpf array map iterators"). Remove it.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240602234112.225107-3-linux@treblig.org
'scale_test_def' is unused since commit 3762a39ce8 ("selftests/bpf: Split out
bpf_verif_scale selftests into multiple tests"). Remove it.
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20240602234112.225107-2-linux@treblig.org
Zba extension is very useful for generating addresses that index into array
of basic data types. This patch introduces sh2add and sh3add helpers for
RV32 and RV64 respectively, to accelerate addressing for array of unsigned
long data.
Signed-off-by: Xiao Wang <xiao.w.wang@intel.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Björn Töpel <bjorn@kernel.org>
Link: https://lore.kernel.org/bpf/20240524075543.4050464-3-xiao.w.wang@intel.com
I accidentally picked up an earlier version of this patch, which had
already landed via mm. The patch I picked up contains a bug, which I
kept as I thought it was a fix. So let's just revert it.
This reverts commit 4c6c002042.
Fixes: 4c6c002042 ("riscv: mm: accelerate pagefault when badaccess")
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Reviewed-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Link: https://lore.kernel.org/r/20240530164451.21336-1-palmer@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
On riscv32, it is possible for the last page in virtual address space
(0xfffff000) to be allocated. This page overlaps with PTR_ERR, so that
shouldn't happen.
There is already some code to ensure memblock won't allocate the last page.
However, buddy allocator is left unchecked.
Fix this by reserving physical memory that would be mapped at virtual
addresses greater than 0xfffff000.
Reported-by: Björn Töpel <bjorn@kernel.org>
Closes: https://lore.kernel.org/linux-riscv/878r1ibpdn.fsf@all.your.base.are.belong.to.us
Fixes: 76d2a0493a ("RISC-V: Init and Halt Code")
Signed-off-by: Nam Cao <namcao@linutronix.de>
Cc: <stable@vger.kernel.org>
Tested-by: Björn Töpel <bjorn@rivosinc.com>
Reviewed-by: Björn Töpel <bjorn@rivosinc.com>
Reviewed-by: Mike Rapoport (IBM) <rppt@kernel.org>
Link: https://lore.kernel.org/r/20240425115201.3044202-1-namcao@linutronix.de
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
