11 May, 2017
1 commit
-
Pull virtio updates from Michael Tsirkin:
"Fixes, cleanups, performanceA bunch of changes to virtio, most affecting virtio net. Also ptr_ring
batched zeroing - first of batching enhancements that seems ready."* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
s390/virtio: change maintainership
tools/virtio: fix spelling mistake: "wakeus" -> "wakeups"
virtio_net: tidy a couple debug statements
ptr_ring: support testing different batching sizes
ringtest: support test specific parameters
ptr_ring: batch ring zeroing
virtio: virtio_driver doc
virtio_net: don't reset twice on XDP on/off
virtio_net: fix support for small rings
virtio_net: reduce alignment for buffers
virtio_net: rework mergeable buffer handling
virtio_net: allow specifying context for rx
virtio: allow extra context per descriptor
tools/virtio: fix build breakage
virtio: add context flag to find vqs
virtio: wrap find_vqs
ringtest: fix an assert statement
03 May, 2017
1 commit
-
Allows maintaining extra context per vq. For ease of use, passing in
NULL is legal and disables the feature for all vqs.Includes fixes by Christian for s390, acked by Cornelia.
Signed-off-by: Christian Borntraeger
Acked-by: Cornelia Huck
Signed-off-by: Michael S. Tsirkin
29 Mar, 2017
1 commit
-
Remove the standalone SMD implementation as we have transitioned the
client drivers to use the RPMSG based one.Also remove all dependencies on QCOM_SMD from Kconfig files, in order to
keep them selectable in the absence of the removed symbol.Acked-by: Andy Gross
Signed-off-by: Bjorn Andersson
Signed-off-by: David S. Miller
21 Mar, 2017
1 commit
-
qcom_smd_register_edge() is provided by either QCOM_SMD or RPMSG_QCOM_SMD,
and if both of them are disabled, it does nothing.The check for the PIL drivers however only checks for QCOM_SMD, so it breaks
with QCOM_SMD=n && RPMSG_QCOM_SMD=m:drivers/remoteproc/built-in.o: In function `smd_subdev_remove':
qcom_wcnss_iris.c:(.text+0x231c): undefined reference to `qcom_smd_unregister_edge'
drivers/remoteproc/built-in.o: In function `smd_subdev_probe':
qcom_wcnss_iris.c:(.text+0x2344): undefined reference to `qcom_smd_register_edge'
drivers/remoteproc/built-in.o: In function `smd_subdev_probe':
qcom_q6v5_pil.c:(.text+0x3538): undefined reference to `qcom_smd_register_edge'
qcom_q6v5_pil.c:(.text+0x3538): relocation truncated to fit: R_AARCH64_CALL26 against undefined symbol `qcom_smd_register_edge'This clarifies the Kconfig dependency.
Fixes: 4b48921a8f74 ("remoteproc: qcom: Use common SMD edge handler")
Signed-off-by: Arnd Bergmann
Signed-off-by: Bjorn Andersson
03 Mar, 2017
1 commit
-
Pull vhost updates from Michael Tsirkin:
"virtio, vhost: optimizations, fixesLooks like a quiet cycle for vhost/virtio, just a couple of minor
tweaks. Most notable is automatic interrupt affinity for blk and scsi.
Hopefully other devices are not far behind"* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
virtio-console: avoid DMA from stack
vhost: introduce O(1) vq metadata cache
virtio_scsi: use virtio IRQ affinity
virtio_blk: use virtio IRQ affinity
blk-mq: provide a default queue mapping for virtio device
virtio: provide a method to get the IRQ affinity mask for a virtqueue
virtio: allow drivers to request IRQ affinity when creating VQs
virtio_pci: simplify MSI-X setup
virtio_pci: don't duplicate the msix_enable flag in struct pci_dev
virtio_pci: use shared interrupts for virtqueues
virtio_pci: remove struct virtio_pci_vq_info
vhost: try avoiding avail index access when getting descriptor
virtio_mmio: expose header to userspace
28 Feb, 2017
1 commit
-
Add a struct irq_affinity pointer to the find_vqs methods, which if set
is used to tell the PCI layer to create the MSI-X vectors for our I/O
virtqueues with the proper affinity from the start. Compared to after
the fact affinity hints this gives us an instantly working setup and
allows to allocate the irq descritors node-local and avoid interconnect
traffic. Last but not least this will allow blk-mq queues are created
based on the interrupt affinity for storage drivers.Signed-off-by: Christoph Hellwig
Reviewed-by: Jason Wang
Signed-off-by: Michael S. Tsirkin
22 Feb, 2017
1 commit
-
In the transition from using rproc_da_to_va(), the type of the load
offset became unsigned. This causes the subsequent check to let negative
values less than p_memsz + mem_size through and we write outside of the
buffer.Change the type back to a signed value to catch this.
Fixes: 7f0dd07a9b29 ("remoteproc: qcom: mdt_loader: Refactor MDT loader")
Fixes: e7fd25226295 ("remoteproc: qcom: q6v5: Decouple driver from MDT loader")
Reported-by: Dan Carpenter
Reported-by: Stanimir Varbanov
Acked-by: Andy Gross
Signed-off-by: Bjorn Andersson
07 Feb, 2017
13 commits
-
This patch provides virtio communication support based on mailbox
for ST co-processors.Signed-off-by: Loic Pallardy
Signed-off-by: Ludovic Barre
Signed-off-by: Lee Jones
Signed-off-by: Bjorn Andersson -
Associated clock is prepared in st_rproc_parse_dt function.
it should be unprepared in case of error during probing.Signed-off-by: Loic Pallardy
Signed-off-by: Bjorn Andersson -
The functions rproc_add_virtio_devices() and rproc_fw_config_virtio()
are reduced to trigger auto-boot only. Modify these function names and
related comments to reflect their current state.This patch does not add any functional change.
Signed-off-by: Sarangdhar Joshi
Signed-off-by: Bjorn Andersson -
The rproc_add_virtio_devices() requests firmware asynchronously and
triggers boot if the auto_boot flag is set. However, this
asynchronous call seems to be redundant for non auto-boot scenario
since the rproc_boot() would call request_firmware() anyways. Move
the auto_boot check to rproc_add() so that a redundant call to
_request_firmware can be avoided for non auto-boot case.Signed-off-by: Sarangdhar Joshi
Signed-off-by: Bjorn Andersson -
SCM call to check whether Peripheral Authentication Service (PAS) is
supported returns false for ADSP on MSM8996. Drop this call from the
driver so that the probe() function succeeds for ADSP PIL device.Signed-off-by: Sarangdhar Joshi
Signed-off-by: Bjorn Andersson -
Call the common SMD edge handler to instantiate subdevices to bring
associated SMD edges up and down as the remoteproc is started and
stopped.Signed-off-by: Bjorn Andersson
-
Move the SMD edge handling to the Qualcomm common file to make it
reusable for other Qualcomm remoteproc drivers.Signed-off-by: Bjorn Andersson
-
With the remoteproc parts cleaned out of the MDT loader we can move it
to drivers/soc/qcom.Acked-by: Andy Gross
Signed-off-by: Bjorn Andersson -
Pushing the SCM calls into the MDT loader reduces duplication in the
callers and allows for non-remoteproc clients to use the helper for
parsing and loading MDT files.Cc: Andy Gross
Signed-off-by: Bjorn Andersson -
The "fw" firmware object is passed from the remoteproc core and should
not be overwritten, as that results in leaked buffers and a double free
of the the last firmware object.Fixes: 051fb70fd4ea ("remoteproc: qcom: Driver for the self-authenticating Hexagon v5")
Cc: stable@vger.kernel.org
Signed-off-by: Bjorn Andersson -
In preparation for moving the mdt loader out of remoteproc let's move
the somewhat unrelated resource table dummy helper to a Qualcomm
"common" file.Signed-off-by: Bjorn Andersson
-
Rather than duplicating half of the MDT loader in the validation step
move the entire MDT parser into the q6v5 driver. This allows us to make
the shared MDT-loader call the SCM PAS operations directly which
simplifies the client code and allows for better reuse of the code.Cc: Avaneesh Kumar Dwivedi
Signed-off-by: Bjorn Andersson -
The Q6V5 in MSM8916 doesn't have a mss supply, so remove this and update
the code to support cases without proxy or active supplies.Signed-off-by: Bjorn Andersson
01 Feb, 2017
1 commit
-
The recently added initialization is rather unusual because it uses a constructor for
a variable-length array to assign a constant structure to a member that uses a fixed-length
array. This confuses clang and breaks the build.drivers/remoteproc/qcom_q6v5_pil.c:1024:18: error: incompatible pointer types initializing 'const char *' with an expression of type
:%s 'struct qcom_mss_reg_res [4]' [-Werror,-Wincompatible-pointer-types]
.proxy_supply = (struct qcom_mss_reg_res[]) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/remoteproc/qcom_q6v5_pil.c:1024:18: warning: suggest braces around initialization of subobject [-Wmissing-braces]
.proxy_supply = (struct qcom_mss_reg_res[]) {
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~We can either turn this constructor into a regular initializer by removing
the 'struct qcom_mss_reg_res[])', or we can make the array variable length.The latter approach is used for the arrays of strings in the same structure,
so let's use that here too.Fixes: 19f902b53b47 ("remoteproc: qcom: Initialize and enable proxy and active regulators.")
Signed-off-by: Arnd Bergmann
Signed-off-by: Bjorn Andersson
31 Jan, 2017
6 commits
-
firmware_loading_complete is used to synchronize operations
on rproc while asynchronous firmware loading is in progress.
However, rproc_boot() no longer waits on
firmware_loading_complete. Hence drop this completion
variable altogether and handle the race between rproc_del()
and rproc_boot() using new state RPROC_DELETED.The request_firmware_nowait() will hold the reference to
rproc device by using a get_device()/put_device(), so the
rproc struct will remain valid even when we return from
rproc_del() before the asynchronous call to
rproc_fw_config_virtio() completes.CC: Loic Pallardy
CC: Bjorn Andersson
Signed-off-by: Sarangdhar Joshi
Signed-off-by: Bjorn Andersson -
Add new state RPROC_DELETED to handle synchronization
between rproc_del() and other operations on rproc. This
state represents the rproc device that has been "deleted".CC: Loic Pallardy
CC: Bjorn Andersson
Signed-off-by: Sarangdhar Joshi
Signed-off-by: Bjorn Andersson -
The "remoteproc{0,1...}" sysfs entries are added in
rproc_add() and deleted in rproc_type_release() instead of
in rproc_del(). That leaves these lingering entries sticking
around after we return from rproc_del(). Move the
rproc_delete_debug_dir() to rproc_del() to fix this.Signed-off-by: Sarangdhar Joshi
Signed-off-by: Bjorn Andersson -
This patch add slpi remoteproc support in existing adsp rproc driver.
Signed-off-by: Avaneesh Kumar Dwivedi
[bjorn: documented aggre2 and px-supply]
Signed-off-by: Bjorn Andersson -
This patch add additional clock and regulator resource which are
initialized based on compatible and has no impact on existing driver
working. This resourse addition enable the existing driver to handle.
low pass sensor processor device also.Signed-off-by: Avaneesh Kumar Dwivedi
Signed-off-by: Bjorn Andersson -
This patch initialize certain driver related data based on compatible
string. This enable driver to handle more than one similar device in
by differentiating in probe their private data.Signed-off-by: Avaneesh Kumar Dwivedi
Signed-off-by: Bjorn Andersson
19 Jan, 2017
5 commits
-
Declare rproc_ops structures as const as they are only passed as an
argument to the function rproc_alloc. This argument is of type const, so
rproc_ops structures having this property can be declared const too.
Done using Coccinelle:@r1 disable optional_qualifier @
identifier i;
position p;
@@
static struct rproc_ops i@p = {...};@ok1@
identifier r1.i;
position p;
@@
rproc_alloc(...,&i@p,...)@bad@
position p!={r1.p,ok1.p};
identifier r1.i;
@@
i@p@depends on !bad disable optional_qualifier@
identifier r1.i;
@@
+const
struct rproc_ops i;File size details:
Size of the file remoteproc/da8xx_remoteproc.o remains the same before and
after applying the changes.text data bss dec hex filename
1312 100 4 1416 588 remoteproc/da8xx_remoteproc.o
1312 100 4 1416 588 remoteproc/da8xx_remoteproc.o970 240 0 1210 4ba remoteproc/omap_remoteproc.o
1002 192 0 1194 4aa remoteproc/omap_remoteproc.o1901 240 0 2141 85d remoteproc/st_remoteproc.o
1933 192 0 2125 84d remoteproc/st_remoteproc.o1288 96 0 1384 568 remoteproc/st_slim_rproc.o
1320 64 0 1384 568 remoteproc/st_slim_rproc.o2121 240 0 2361 939 remoteproc/wkup_m3_rproc.o
2161 192 0 2353 931 remoteproc/wkup_m3_rproc.oSigned-off-by: Bhumika Goyal
Signed-off-by: Bjorn Andersson -
Certain regulators need voting by rproc on behalf of hexagon only during
restart operation but certain regulator need to be voted till hexagon
is up, these regulators are identified as proxy and active regulators
respectively. This patch provide interface to initialize, enable and
disable proxy and active regulators separately.Signed-off-by: Avaneesh Kumar Dwivedi
[bjorn: dropped disable of proxy regulators from stop]
Signed-off-by: Bjorn Andersson -
Certain clocks need voting by rproc on behalf of hexagon only during
restart operation but certain clocks need to be voted till hexagon is
up, these clocks are identified as proxy and active clocks respectively.
This patch provide interface to initialize, enable and disable proxy and
active clocks separately.Signed-off-by: Avaneesh Kumar Dwivedi
[bjorn: dropped disable of proxy clocks on stop]
Signed-off-by: Bjorn Andersson -
MSS rproc loader need chip specific resources initialization during probe
to load and boot modem firmware, this need compatible string based
differentiation in resources to be initialized. This patch add and provide
a template struct whose fields represent all those resources which are
needed to load and boot modem fw and which may differ from chip to chip.
This patch also add new compatible string for msm8916, msm8974 platform.Signed-off-by: Avaneesh Kumar Dwivedi
Signed-off-by: Bjorn Andersson -
Top level config option without any kind of help... is kind of
strange. Remote processors could also mean some kind of distributed
computing...Signed-off-by: Pavel Machek
Signed-off-by: Bjorn Andersson
30 Dec, 2016
2 commits
-
Following any fw_rsc_vdev entries in the resource table are two variable
length arrays, the first one reference vring resources and the second
one is the virtio config space. The virtio config space is used by
virtio to communicate status and configuration changes and must as such
be shared with the remote.The reverted commit incorrectly made any changes to the virtio config
space only affect the local copy, in an attempt to allowing memory
protection of the shared resource table.This reverts commit cda8529346935fc86f476999ac4fbfe4e17abf11.
Signed-off-by: Bjorn Andersson -
Commit 2b45cef5868a ("remoteproc: Further extend the vdev life cycle")
extends kref support for vdev management.
It introduces a regression when following sequence is executed:
rproc_boot --> rproc_shutdown --> rproc_boot
Second rproc_boot call crashes on register_virtio_device as device
is already existing.
Issue is previous vdev is never released when rproc is stop because
associated refcount is too high.kref_get introduces is not needed as kref_init already initializes
krefcount to 1 because it considers associated variable as used.
This introduces a misalignment between kref_get and kref_put calls.Fixes: 2b45cef5868a ("remoteproc: Further extend the vdev life cycle")
Signed-off-by: Loic Pallardy
Signed-off-by: Bjorn Andersson
10 Dec, 2016
5 commits
-
The adsp-pil driver relies on SCM and causes a build error without it:
ERROR: "qcom_scm_pas_supported" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_is_available" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_pas_auth_and_reset" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_pas_shutdown" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_pas_mem_setup" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_pas_init_image" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!This adds a 'select', as SCM is a silent Kconfig symbol that gets
enabled implicitly by all its users.Fixes: b9e718e950c3 ("remoteproc: Introduce Qualcomm ADSP PIL")
Signed-off-by: Arnd Bergmann
Signed-off-by: Bjorn Andersson -
In the event that rproc_boot() is called before the firmware loaded
completion has been flagged it will wait with the mutex held,
obstructing the request_firmware_nowait() callback from completing the
wait.As rproc_fw_config_virtio() has been reduced to only triggering
auto-boot there is no longer a reason for waiting in rproc_boot(), so
drop this.Cc: Sarangdhar Joshi
Signed-off-by: Bjorn Andersson -
Back in July 2014 I asked around what was the intended target
platform for the STE Modem remoteproc driver, so that I could add the
proper hardware dependency to its config option. The answer I got was
that there was no known publicly available hardware needing it and it
was unlikely that there ever would.So I think it's time to delete this driver to lower the maintenance
burden.Signed-off-by: Jean Delvare
Cc: Linus Walleij
Cc: Ohad Ben-Cohen
Cc: Bjorn Andersson
Cc: Suman Anna
Acked-by: Linus Walleij
Acked-by: Loic Pallardy
Signed-off-by: Bjorn Andersson -
Warning users that remoteproc and it's binary format are under
development doesn't serve much of a purpose. Different drivers support
different image formats and the resource table has a version field that
would need to be bumped when incompatible changes are introduced.So lets drop this warning to clean up the kernel log.
Signed-off-by: Bjorn Andersson
-
The adsp-pil driver relies on SCM and causes a build error without it:
ERROR: "qcom_scm_pas_supported" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_is_available" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_pas_auth_and_reset" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_pas_shutdown" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_pas_mem_setup" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!
ERROR: "qcom_scm_pas_init_image" [drivers/remoteproc/qcom_adsp_pil.ko] undefined!This adds a 'select', as SCM is a silent Kconfig symbol that gets
enabled implicitly by all its users.Fixes: b9e718e950c3 ("remoteproc: Introduce Qualcomm ADSP PIL")
Signed-off-by: Arnd Bergmann
Signed-off-by: Bjorn Andersson
03 Dec, 2016
1 commit
-
Add linux/sizes.h to prevent build failure on non ARM architectures
as:CC [M] drivers/remoteproc/qcom_mdt_loader.o
In file included from include/linux/cache.h:4:0,
from include/linux/printk.h:8,
from include/linux/kernel.h:13,
from include/asm-generic/bug.h:13,
from arch/x86/include/asm/bug.h:35,
from include/linux/bug.h:4,
from include/linux/thread_info.h:11,
from arch/x86/include/asm/elf.h:7,
from include/linux/elf.h:4,
from drivers/remoteproc/qcom_mdt_loader.c:18:
drivers/remoteproc/qcom_mdt_loader.c: In function ‘qcom_mdt_parse’:
drivers/remoteproc/qcom_mdt_loader.c:90:52: error: ‘SZ_4K’ undeclared
(first use in this function)Reviewed-by: Stephen Boyd
Signed-off-by: Stanimir Varbanov
Signed-off-by: Bjorn Andersson
