IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
The remove callback is only called after probe completed successfully.
In this case platform_set_drvdata() was called with a non-NULL argument
and so pdata is never NULL.
This is a preparation for making platform remove callbacks return void.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Vladimir Zapolskiy <vz@mleia.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The remove callback is only called after probe completed successfully.
In this case platform_set_drvdata() was called with a non-NULL argument
and so dd is never NULL.
This is a preparation for making platform remove callbacks return void.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The remove callback is only called after probe completed successfully.
In this case platform_set_drvdata() was called with a non-NULL argument
and so dd is never NULL.
This is a preparation for making platform remove callbacks return void.
While touching this driver remove an assignment without effect.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The remove callback is only called after probe completed successfully.
In this case platform_set_drvdata() was called with a non-NULL argument
and so dd is never NULL.
This is a preparation for making platform remove callbacks return void.
While touching this driver remove a stray empty line.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The remove callback is only called after probe completed successfully.
In this case platform_set_drvdata() was called with a non-NULL argument
and so tdes_dd is never NULL.
This is a preparation for making platform remove callbacks return void.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The remove callback is only called after probe completed successfully.
In this case platform_set_drvdata() was called with a non-NULL argument
and so sha_dd is never NULL.
This is a preparation for making platform remove callbacks return void.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The remove callback is only called after probe completed successfully.
In this case platform_set_drvdata() was called with a non-NULL argument
and so aes_dd is never NULL.
This is a preparation for making platform remove callbacks return void.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Add clk_disable_unprepare() on error path in cc_pm_resume().
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yuan Can <yuancan@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When mixing bit-field and none bit-filed in packed struct the
none bit-field starts at a distinct memory location, thus adding
an additional byte to the overall structure which is used in
memory zero-ing and other configuration calculations.
Fix this by removing the none bit-field that has a following
bit-field.
Signed-off-by: Ofer Heifetz <oferh@marvell.com>
Acked-by: Antoine Tenart <atenart@kernel.org>
Acked-by: Antoine Tenart <atenart@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When kunpeng916 encryption driver is used to deencrypt and decrypt
packets during the softirq, it is not allowed to use mutex lock.
Fixes: 915e4e8413da ("crypto: hisilicon - SEC security accelerator driver")
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The argument passed to sec_queue_aw_alloc() should be
SEC_QUEUE_AW_FROCE_NOALLOC instead of SEC_QUEUE_AR_FROCE_NOALLOC.
Signed-off-by: Jianglei Nie <niejianglei2021@163.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In nx842_pseries_init(), we should hold the reference returned by
of_find_compatible_node() and use it to call of_node_put to keep
refcount balance.
Signed-off-by: Liang He <windhl@126.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
In crypto4xx_probe(), we should hold the reference returned by
of_find_compatible_node() and use it to call of_node_put to keep
refcount balance.
Signed-off-by: Liang He <windhl@126.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The compiler complains that p8_ghash_alg isn't declared which is
because the header file aesp8-ppc.h isn't included in ghash.c.
This patch fixes the warning.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Breno Leitao <leitao@debian.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
qat_4xxx devices can be configured to allow either crypto or compression
operations. At the moment, devices are configured statically according to
the following rule:
- odd numbered devices assigned to compression services
- even numbered devices assigned to crypto services
Expose the sysfs attribute /sys/bus/pci/devices/<BDF>/qat/cfg_services
to allow to detect the configuration of a device and to change it.
The `cfg_service` attribute is only exposed for qat_4xxx devices and it
is limited to two configurations: (1) "sym;asym" for crypto services and
"dc" for compression services.
Signed-off-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Co-developed-by: Tomasz Kowallik <tomaszx.kowalik@intel.com>
Signed-off-by: Tomasz Kowallik <tomaszx.kowalik@intel.com>
Reviewed-by: Adam Guerin <adam.guerin@intel.com>
Reviewed-by: Fiona Trahe <fiona.trahe@intel.com>
Reviewed-by: Wojciech Ziemba <wojciech.ziemba@intel.com>
Reviewed-by: Vladis Dronov <vdronov@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The function adf_sriov_prepare_restart() is used in adf_sriov.c to stop
and shutdown a device preserving its configuration.
Since this function will be re-used by the logic that allows to
reconfigure the device through sysfs, move it to adf_init.c and rename
it as adf_dev_shutdown_cache_cfg();
Signed-off-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Reviewed-by: Adam Guerin <adam.guerin@intel.com>
Reviewed-by: Fiona Trahe <fiona.trahe@intel.com>
Reviewed-by: Wojciech Ziemba <wojciech.ziemba@intel.com>
Reviewed-by: Vladis Dronov <vdronov@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The function adf_cfg_add_key_value_param() allows to insert duplicates
entries in the key value store of the driver.
Change the behaviour of that function to the following policy:
- if the key doesn't exist, add it;
- if the key already exists with a different value, then delete it and
replace it with a new one containing the new value;
- if the key exists with the same value, then return without doing
anything.
The behaviour of this function has been changed in order to easily
update key-values in the driver database. In particular this is required
to update the value of the ServiceEnables key used to change the service
loaded on a device.
Signed-off-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Reviewed-by: Adam Guerin <adam.guerin@intel.com>
Reviewed-by: Fiona Trahe <fiona.trahe@intel.com>
Reviewed-by: Wojciech Ziemba <wojciech.ziemba@intel.com>
Reviewed-by: Vladis Dronov <vdronov@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Expose the device state through an attribute in sysfs and allow to
change it. This is to stop and shutdown a QAT device in order to change
its configuration.
The state attribute has been added to a newly created `qat` attribute
group which will contain all _QAT specific_ attributes.
The logic that implements the sysfs entries is part of a new file,
adf_sysfs.c. This exposes an entry point to allow the driver to create
attributes.
The function that creates the sysfs attributes is called from the probe
function of the driver and not in the state machine init function to
allow the change of states even if the device is in the down state.
In order to restore the device configuration between a transition from
down to up, the function that configures the devices has been abstracted
into the HW data structure.
The `state` attribute is only exposed for qat_4xxx devices.
Signed-off-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Co-developed-by: Tomasz Kowallik <tomaszx.kowalik@intel.com>
Signed-off-by: Tomasz Kowallik <tomaszx.kowalik@intel.com>
Reviewed-by: Adam Guerin <adam.guerin@intel.com>
Reviewed-by: Fiona Trahe <fiona.trahe@intel.com>
Reviewed-by: Wojciech Ziemba <wojciech.ziemba@intel.com>
Reviewed-by: Vladis Dronov <vdronov@redhat.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Various accelerated software implementation Kconfig values for S390 were
mistakenly placed into drivers/crypto/Kconfig, even though they're
mainly just SIMD code and live in arch/s390/crypto/ like usual. This
gives them the very unusual dependency on CRYPTO_HW, which leads to
problems elsewhere.
This patch fixes the issue by moving the Kconfig values for non-hardware
drivers into the usual place in crypto/Kconfig.
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
there is an unexpected word 'for' in the comments that need to be dropped
file - drivers/crypto/vmx/ghashp8-ppc.pl
line - 19
"# GHASH for for PowerISA v2.07."
changed to:
"# GHASH for PowerISA v2.07."
Signed-off-by: Jiang Jian <jiangjian@cdjrlc.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
there is an unexpected word 'for' in the comments that need to be dropped
file - drivers/crypto/caam/caamhash_desc.c
line - 25
* must be false for for ahash first and digest
changed to:
* must be false for ahash first and digest
Signed-off-by: Jiang Jian <jiangjian@cdjrlc.com>
Reviewed-by: Gaurav Jain <gaurav.jain@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Drop the unexpected word 'a' in the comments that need to be dropped
* This is a a cache of buffers, from which the users of CAAM QI driver
-->
* This is a cache of buffers, from which the users of CAAM QI driver
Signed-off-by: Jiang Jian <jiangjian@cdjrlc.com>
Reviewed-by: Gaurav Jain <gaurav.jain@nxp.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Replace 'is' with 'it'
file: drivers/crypto/cavium/cpt/cpt_hw_types.h
line: 268
* which if the line hits and is is dirty will cause the line to be
changed to:
* which if the line hits and it is dirty will cause the line to be
Signed-off-by: Jiang Jian <jiangjian@cdjrlc.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
there is an unexpected word "the" in the comments that need to be dropped
>- * specified in the the hw design spec. Either due to incorrect info in the
>+ * specified in the hw design spec. Either due to incorrect info in the
Signed-off-by: Jiang Jian <jiangjian@cdjrlc.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
there is an unexpected word "the" in the comments that need to be dropped
>- * The DDE is setup with the the DDE count, byte count, and address of
>+ * The DDE is setup with the DDE count, byte count, and address of
Signed-off-by: Jiang Jian <jiangjian@cdjrlc.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
This dependency looks outdated. After the previous patch, we have been able
to use this driver to encrypt some data and to create working VF on arm64.
We have not tested it yet on any big endian machine, hence the new dependency
Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Currently the QAT driver code uses a self-defined wrapper function
called get_current_node() when it wants to learn the current NUMA node.
This implementation references the topology_physical_package_id[] array,
which more or less coincidentally contains the NUMA node id, at least
on x86.
Because this is not universal, and Linux offers a direct function to
learn the NUMA node ID, replace that function with a call to
numa_node_id(), which would work everywhere.
This fixes the QAT driver operation on arm64 machines.
Reported-by: Yoan Picchi <Yoan.Picchi@arm.com>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Yoan Picchi <yoan.picchi@arm.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
On shutdown, each CCP device instance performs shutdown processing.
However, __sev_platform_shutdown_locked() uses the controlling psp
structure to obtain the pointer to the sev_device structure. However,
during driver initialization, it is possible that an error can be received
from the firmware that results in the sev_data pointer being cleared from
the controlling psp structure. The __sev_platform_shutdown_locked()
function does not check for this situation and will segfault.
While not common, this scenario should be accounted for. Add a check for a
NULL sev_device structure before attempting to use it.
Fixes: 5441a07a127f ("crypto: ccp - shutdown SEV firmware on kexec")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The ccp driver loops through the platform device resources array to get
the IRQ count for the device. With commit a1a2b7125e10 ("of/platform: Drop
static setup of IRQ resource from DT core"), the IRQ resources are no
longer stored in the platform device resource array. As a result, the IRQ
count is now always zero. This causes the driver to issue a second call to
platform_get_irq(), which fails if the IRQ count is really 1, causing the
loading of the driver to fail.
Replace looping through the resources array to count the number of IRQs
with a call to platform_irq_count().
Fixes: a1a2b7125e10 ("of/platform: Drop static setup of IRQ resource from DT core")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The base register address of V2 and V3 are different. HW V3 not needs
to change the BD err detection.
Signed-off-by: Kai Ye <yekai13@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
There is no i decrement in while (i >= 0) loop.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Fixes: 359e893e8af4 ("crypto: sun8i-ss - rework handling of IV")
Acked-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Tested-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
When the driver receives an event interrupt, the driver will enable
the event interrupt after handling all completed tasks on the function,
tasks on the function are parsed through only one thread. If the task's
user callback takes time, other tasks on the function will be blocked.
Therefore, the event irq processing is modified as follows:
1. Obtain the ID of the queue that completes the task.
2. Enable event interrupt.
3. Parse the completed tasks in the queue and call the user callback.
Enabling event interrupt in advance can quickly report pending event
interrupts and process tasks in multiple threads.
Signed-off-by: Weili Qian <qianweili@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Before stopping the function, the driver needs to flush all the remaining
work about event irq. Therefore, accelerator drivers use a private
workqueue(qm->wq) to handle event irq instead of the system workqueue.
This patch moves alloc workqueue from sec_main.c and zip_main.c to qm.c.
Signed-off-by: Weili Qian <qianweili@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The resources allocated by hisi_qm_memory_init() are released by
hisi_qm_uninit(). Add hisi_qm_memory_uninit() to release resources,
no functional change.
Signed-off-by: Weili Qian <qianweili@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The return value of 'readl_poll_timeout' is '0' or '-ETIMEDOUT'. Therefore,
change the local variable 'ret' type from 'u32' to 'int'.
Signed-off-by: Weili Qian <qianweili@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
The crypto_alloc_shash() function never returns NULL. It returns error
pointers.
Fixes: 801b7d572c0a ("crypto: sun8i-ss - add hmac(sha1)")
Signed-off-by: Peng Wu <wupeng58@huawei.com>
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Added missing checks to avoid null pointer dereference.
The patch fixes below issue reported by klocwork tool:
. Pointer 'strsep( &val, ":" )' returned from call to function 'strsep'
at line 1608 may be NULL and will be dereferenced at line 1608. Also
there are 2 similar errors on lines 1620, 1632 in otx2_cptpf_ucode.c.
Signed-off-by: Shijith Thotton <sthotton@marvell.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Added running firmware version information of AE, SE and IE components
in devlink info.
Signed-off-by: Shijith Thotton <sthotton@marvell.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Spelling mistake (triple letters) in comment.
Detected with the help of Coccinelle.
Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Spelling mistakes (triple letters) in comment.
Detected with the help of Coccinelle.
Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Returning an error value in an i2c remove callback results in an error
message being emitted by the i2c core, but otherwise it doesn't make a
difference. The device goes away anyhow and the devm cleanups are
called.
As atmel_ecc_remove() already emits an error message on failure and the
additional error message by the i2c core doesn't add any useful
information, change the return value to zero to suppress this message.
Also make the error message a bit more drastical because when the device
is still busy on remove, it's likely that it will access freed memory
soon.
This patch is a preparation for making i2c remove callbacks return void.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reviewed-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
If there is a dma_mapping_error() then return negative error codes.
Currently this code returns success.
Fixes: 801b7d572c0a ("crypto: sun8i-ss - add hmac(sha1)")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Tested-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
These failure paths should return -ENOMEM. Currently they return
success.
Fixes: 359e893e8af4 ("crypto: sun8i-ss - rework handling of IV")
Fixes: 8eec4563f152 ("crypto: sun8i-ss - do not allocate memory when handling hash requests")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Tested-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
For some sev ioctl interfaces, input may be passed that is less than or
equal to SEV_FW_BLOB_MAX_SIZE, but larger than the data that PSP
firmware returns. In this case, kmalloc will allocate memory that is the
size of the input rather than the size of the data. Since PSP firmware
doesn't fully overwrite the buffer, the sev ioctl interfaces with the
issue may return uninitialized slab memory.
Currently, all of the ioctl interfaces in the ccp driver are safe, but
to prevent future problems, change all ioctl interfaces that allocate
memory with kmalloc to use kzalloc and memset the data buffer to zero
in sev_ioctl_do_platform_status.
Fixes: 38103671aad3 ("crypto: ccp: Use the stack and common buffer for status commands")
Fixes: e799035609e15 ("crypto: ccp: Implement SEV_PEK_CSR ioctl command")
Fixes: 76a2b524a4b1d ("crypto: ccp: Implement SEV_PDH_CERT_EXPORT ioctl command")
Fixes: d6112ea0cb344 ("crypto: ccp - introduce SEV_GET_ID2 command")
Cc: stable@vger.kernel.org
Reported-by: Andy Nguyen <theflow@google.com>
Suggested-by: David Rientjes <rientjes@google.com>
Suggested-by: Peter Gonda <pgonda@google.com>
Signed-off-by: John Allen <john.allen@amd.com>
Reviewed-by: Peter Gonda <pgonda@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Enable retry for virtio-crypto-dev, so that crypto-engine
can process cipher-requests parallelly.
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: lei he <helei.sig11@bytedance.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Message-Id: <20220506131627.180784-6-pizhenwei@bytedance.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
For some akcipher operations(eg, decryption of pkcs1pad(rsa)),
the length of returned result maybe less than akcipher_req->dst_len,
we need to recalculate the actual dst_len through the virt-queue
protocol.
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: lei he <helei.sig11@bytedance.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Message-Id: <20220506131627.180784-5-pizhenwei@bytedance.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Originally, after submitting request into virtio crypto control
queue, the guest side polls the result from the virt queue. This
works like following:
CPU0 CPU1 ... CPUx CPUy
| | | |
\ \ / /
\--------spin_lock(&vcrypto->ctrl_lock)-------/
|
virtqueue add & kick
|
busy poll virtqueue
|
spin_unlock(&vcrypto->ctrl_lock)
...
There are two problems:
1, The queue depth is always 1, the performance of a virtio crypto
device gets limited. Multi user processes share a single control
queue, and hit spin lock race from control queue. Test on Intel
Platinum 8260, a single worker gets ~35K/s create/close session
operations, and 8 workers get ~40K/s operations with 800% CPU
utilization.
2, The control request is supposed to get handled immediately, but
in the current implementation of QEMU(v6.2), the vCPU thread kicks
another thread to do this work, the latency also gets unstable.
Tracking latency of virtio_crypto_alg_akcipher_close_session in 5s:
usecs : count distribution
0 -> 1 : 0 | |
2 -> 3 : 7 | |
4 -> 7 : 72 | |
8 -> 15 : 186485 |************************|
16 -> 31 : 687 | |
32 -> 63 : 5 | |
64 -> 127 : 3 | |
128 -> 255 : 1 | |
256 -> 511 : 0 | |
512 -> 1023 : 0 | |
1024 -> 2047 : 0 | |
2048 -> 4095 : 0 | |
4096 -> 8191 : 0 | |
8192 -> 16383 : 2 | |
This means that a CPU may hold vcrypto->ctrl_lock as long as 8192~16383us.
To improve the performance of control queue, a request on control queue
waits completion instead of busy polling to reduce lock racing, and gets
completed by control queue callback.
CPU0 CPU1 ... CPUx CPUy
| | | |
\ \ / /
\--------spin_lock(&vcrypto->ctrl_lock)-------/
|
virtqueue add & kick
|
---------spin_unlock(&vcrypto->ctrl_lock)------
/ / \ \
| | | |
wait wait wait wait
Test this patch, the guest side get ~200K/s operations with 300% CPU
utilization.
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Message-Id: <20220506131627.180784-4-pizhenwei@bytedance.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Originally, all of the control requests share a single buffer(
ctrl & input & ctrl_status fields in struct virtio_crypto), this
allows queue depth 1 only, the performance of control queue gets
limited by this design.
In this patch, each request allocates request buffer dynamically, and
free buffer after request, so the scope protected by ctrl_lock also
get optimized here.
It's possible to optimize control queue depth in the next step.
A necessary comment is already in code, still describe it again:
/*
* Note: there are padding fields in request, clear them to zero before
* sending to host to avoid to divulge any information.
* Ex, virtio_crypto_ctrl_request::ctrl::u::destroy_session::padding[48]
*/
So use kzalloc to allocate buffer of struct virtio_crypto_ctrl_request.
Potentially dereferencing uninitialized variables:
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Message-Id: <20220506131627.180784-3-pizhenwei@bytedance.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Use temporary variable to make code easy to read and maintain.
/* Pad cipher's parameters */
vcrypto->ctrl.u.sym_create_session.op_type =
cpu_to_le32(VIRTIO_CRYPTO_SYM_OP_CIPHER);
vcrypto->ctrl.u.sym_create_session.u.cipher.para.algo =
vcrypto->ctrl.header.algo;
vcrypto->ctrl.u.sym_create_session.u.cipher.para.keylen =
cpu_to_le32(keylen);
vcrypto->ctrl.u.sym_create_session.u.cipher.para.op =
cpu_to_le32(op);
-->
sym_create_session = &ctrl->u.sym_create_session;
sym_create_session->op_type = cpu_to_le32(VIRTIO_CRYPTO_SYM_OP_CIPHER);
sym_create_session->u.cipher.para.algo = ctrl->header.algo;
sym_create_session->u.cipher.para.keylen = cpu_to_le32(keylen);
sym_create_session->u.cipher.para.op = cpu_to_le32(op);
The new style shows more obviously:
- the variable we want to operate.
- an assignment statement in a single line.
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: zhenwei pi <pizhenwei@bytedance.com>
Message-Id: <20220506131627.180784-2-pizhenwei@bytedance.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
