For HNAE3_DEVICE_VERSION_V3, a maximum of 1281 interrupt
resources are supported. To utilize these new resources,
extend the corresponding field or variable to 16bit type,
and remove the restriction of NIC client that only use a
maximum of 65 interrupt vectors. In addition, the I/O address
of the extended interrupt resources are different, so an extra
handler is needed.
Currently, the total number of interrupts is the sum of RoCE's
number and RoCE's offset (RoCE is in front of NIC), since
the number of both NIC and RoCE are same. For readability,
rewrite the corresponding field of the command, rename the
RoCE's offset field as the number of NIC interrupts, then
the total number of interrupts is sum of the number of RoCE
and NIC, and replace vport->back with hdev in
hclge_init_roce_base_info() for simplifying the code.
Signed-off-by: Yufeng Mo <moyufeng@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
For device who has device memory accessed through the PCI BAR4,
IO descriptor push of NIC and direct WQE(Work Queue Element) of
RoCE will use this device memory, so add support for mapping
this device memory, and add this info to the RoCE client whose
new feature needs.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
For DEVICE_VERSION_V1/2, there are total 1024 queues and
queue sets. For DEVICE_VERSION_V3, it increases to 1280,
and can be assigned to one pf， so remove the limitation
of 1024.
To keep compatible with DEVICE_VERSION_V1/2 and old driver
version, the queue number is split into two part:
tqp_num(range 0~1023) and ext_tqp_num(range 1024~1279).
Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Starting with iOS 14 released in September 2020, connectivity using the
personal hotspot USB tethering function of iOS devices is broken.
Communication between the host and the device (for example ICMP traffic
or DNS resolution using the DNS service running in the device itself)
works fine, but communication to endpoints further away doesn't work.
Investigation on the matter shows that no UDP and ICMP traffic from the
tethered host is reaching the Internet at all. For TCP traffic there are
exchanges between tethered host and server but packets are modified in
transit leading to impossible communication.
After some trials Matti Vuorela discovered that reducing the URB buffer
size by two bytes restored the previous behavior. While a better
solution might exist to fix the issue, since the protocol is not
publicly documented and considering the small size of the fix, let's do
that.
Tested-by: Matti Vuorela <matti.vuorela@bitfactor.fi>
Signed-off-by: Yves-Alexis Perez <corsac@corsac.net>
Link: https://lore.kernel.org/linux-usb/CAAn0qaXmysJ9vx3ZEMkViv_B19ju-_ExN8Yn_uSefxpjS6g4Lw@mail.gmail.com/
Link: https://github.com/libimobiledevice/libimobiledevice/issues/1038
Link: https://lore.kernel.org/r/20201119172439.94988-1-corsac@corsac.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
After commit 9d2e5e9eeb ("cxgb4/ch_ktls: decrypted bit is not enough")
whenever CONFIG_TLS=m and CONFIG_CHELSIO_T4=y, the following build
failure occurs:
ld: drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.o: in function
`cxgb_select_queue':
cxgb4_main.c:(.text+0x2dac): undefined reference to `tls_validate_xmit_skb'
Fix this by ensuring that if TLS is set to be a module, CHELSIO_T4 will
also be compiled as a module. As otherwise the cxgb4 driver will not be
able to access TLS' symbols.
Fixes: 9d2e5e9eeb ("cxgb4/ch_ktls: decrypted bit is not enough")
Signed-off-by: Tom Seewald <tseewald@gmail.com>
Link: https://lore.kernel.org/r/20201120192528.615-1-tseewald@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
- Fix various deficiencies in online fsck's metadata checking code.
- Fix an integer casting bug in the xattr code on 32-bit systems.
- Fix a hang in an inode walk when the inode index is corrupt.
- Fix error codes being dropped when initializing per-AG structures
- Fix nowait directio writes that partially succeed but return EAGAIN.
- Revert last week's rmap comparison patch because it was wrong.
-----BEGIN PGP SIGNATURE-----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=HoPI
-----END PGP SIGNATURE-----
Merge tag 'xfs-5.10-fixes-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux
Pull xfs fixes from Darrick Wong:
"The critical fixes are for a crash that someone reported in the xattr
code on 32-bit arm last week; and a revert of the rmap key comparison
change from last week as it was totally wrong. I need a vacation. :(
Summary:
- Fix various deficiencies in online fsck's metadata checking code
- Fix an integer casting bug in the xattr code on 32-bit systems
- Fix a hang in an inode walk when the inode index is corrupt
- Fix error codes being dropped when initializing per-AG structures
- Fix nowait directio writes that partially succeed but return EAGAIN
- Revert last week's rmap comparison patch because it was wrong"
* tag 'xfs-5.10-fixes-7' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
xfs: revert "xfs: fix rmap key and record comparison functions"
xfs: don't allow NOWAIT DIO across extent boundaries
xfs: return corresponding errcode if xfs_initialize_perag() fail
xfs: ensure inobt record walks always make forward progress
xfs: fix forkoff miscalculation related to XFS_LITINO(mp)
xfs: directory scrub should check the null bestfree entries too
xfs: strengthen rmap record flags checking
xfs: fix the minrecs logic when dealing with inode root child blocks
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEEq1nRK9aeMoq1VSgcnJ2qBz9kQNkFAl+4JHgACgkQnJ2qBz9k
QNlXOQf/YHs4q4HgBI0tsStS/U4xFtmY77Rcm1pllqH6BZPBg1vRpzfh7hZvIPMa
GceTcMAX4OmG6++fRzVgNDIuem3Jl0oDCm++pWPev+S/V06PuTu36viuFWJ3e/5g
0wDLYXRj4dRUiQtjbSkI7LAgIX1wbTANOKSZeaKFYaGHfEcFm1GkHUuHzEBVX1Jw
bRpaod3ikmjoaoI6TTZlKKnrKksSw6F5wHUiHu2ZHdZ6kQ36elwHFu8QXJCzkZ7F
F9vt4IIKq6xzEVdwDXAPjsFkPp2B4Bz+AgcSpoitg/2L5hc2d/kxgI4zvpXY8TGs
hpW6YPXEXIjhHjKX22f99ThI4BqXww==
=bBTT
-----END PGP SIGNATURE-----
Merge tag 'fsnotify_for_v5.10-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs
Pull fanotify fix from Jan Kara:
"A single fanotify fix from Amir"
* tag 'fsnotify_for_v5.10-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
fanotify: fix logic of reporting name info with watched parent
- Fix typos in seccomp selftests on powerpc and sh (Kees Cook)
- Fix PF_SUPERPRIV audit marking in seccomp and ptrace (Mickaël Salaün)
-----BEGIN PGP SIGNATURE-----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=1Mse
-----END PGP SIGNATURE-----
Merge tag 'seccomp-v5.10-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux
Pull seccomp fixes from Kees Cook:
"This gets the seccomp selftests running again on powerpc and sh, and
fixes an audit reporting oversight noticed in both seccomp and ptrace.
- Fix typos in seccomp selftests on powerpc and sh (Kees Cook)
- Fix PF_SUPERPRIV audit marking in seccomp and ptrace (Mickaël
Salaün)"
* tag 'seccomp-v5.10-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
selftests/seccomp: sh: Fix register names
selftests/seccomp: powerpc: Fix typo in macro variable name
seccomp: Set PF_SUPERPRIV when checking capability
ptrace: Set PF_SUPERPRIV when checking capability
In situations where the GPU is mostly idle, all or nearly all buffer
objects will be in the inactive list. But if the system is under memory
pressure (from something other than GPU), we could still get a lot of
shrinker calls. Which results in traversing a list of thousands of objs
and in the end finding nothing to shrink. Which isn't so efficient.
Instead split the inactive_list into two lists, one inactive objs which
are shrinkable, and a second one for those that are not. This way we
can avoid traversing objs which we know are not shrinker candidates.
v2: Fix inverted logic think-o
Signed-off-by: Rob Clark <robdclark@chromium.org>
Previously we only held obj lock in the _active_get() path, and relied
on atomic_dec_return() to not be racy in the _active_put() path where
obj lock was not held.
But this is a false sense of security. Unlike obj lifetime refcnt,
where you do not expect to *increase* the refcnt after the last put
(which would mean that something has gone horribly wrong with the
object liveness reference counting), the active_count can increase
again from zero. Racing _active_put()s and _active_get()s could leave
the obj on the wrong mm list.
But in the retire path, immediately after the _active_put(), the
_unpin_iova() would acquire obj lock. So just move the locking earlier
and rely on that to protect obj->active_count.
Fixes: c5c1643cef ("drm/msm: Drop struct_mutex from the retire path")
Signed-off-by: Rob Clark <robdclark@chromium.org>
It seems that when this was tested the happy case was more tested. A few of
the userspace apps rely on this returning negative error codes in case an
ioctl() is not available.
When running multiple ioctl() handlers or when calling an ioctl() that
doesn't exist, IIO_IOCTL_UNHANDLED is returned. In that case -EINVAL should
be returned.
Fixes: 8dedcc3eee ("iio: core: centralize ioctl() calls to the main chardev")
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20201117095154.7189-1-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The change is mostly cosmetic. This organizes the order of assignment of
the members of 'iio_buffer_fileops' to be similar to the one as defined in
the 'struct file_operations' type.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20201117103753.8450-1-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
There is no matching spi_get_drvdata() in the driver. This looks like a
left-over from before the driver was converted to device-managed functions.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20201119141806.84827-1-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
There is no matching spi_get_drvdata() in the driver. This looks like a
left-over from before the driver was converted to device-managed functions.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20201119141729.84185-1-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This seems to have been copied from a driver that calls spi_set_drvdata()
but doesn't call spi_get_drvdata().
Setting a private object on the SPI device's object isn't necessary if it
won't be accessed.
This change removes the spi_set_drvdata() call.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Tested-by: Marcelo Schmitt <marcelo.schmitt1@gmail.com>
Reviewed-by: Marcelo Schmitt <marcelo.schmitt1@gmail.com>
Link: https://lore.kernel.org/r/20201119142720.86326-1-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The iio_buffer_set_attrs() is no longer used in the drivers, so it can be
removed now.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-10-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change switches to the new iio_triggered_buffer_setup_ext()
function and removes the iio_buffer_set_attrs() call, for assigning the
HW FIFO attributes to the buffer.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-9-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change switches to the new devm_iio_triggered_buffer_setup_ext()
function and removes the iio_buffer_set_attrs() call, for assigning the
HW FIFO attributes to the buffer.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-8-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change switches to the new devm_iio_triggered_buffer_setup_ext()
function and removes the iio_buffer_set_attrs() call, for assigning the
HW FIFO attributes to the buffer.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-7-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change switches to the new iio_triggered_buffer_setup_ext()
function and removes the iio_buffer_set_attrs() call, for assigning the
HW FIFO attributes to the buffer.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-6-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change switches to the new devm_iio_triggered_buffer_setup_ext()
function and removes the iio_buffer_set_attrs() call, for assigning the
HW FIFO attributes to the buffer.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-5-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change adds a parameter to the {devm_}iio_triggered_buffer_setup()
functions to assign the extra sysfs buffer attributes that are typically
assigned via iio_buffer_set_attrs().
The functions also get renamed to iio_triggered_buffer_setup_ext() &
devm_iio_triggered_buffer_setup_ext().
For backwards compatibility the old {devm_}iio_triggered_buffer_setup()
functions are now macros wrap the new (renamed) functions with NULL for the
buffer attrs.
The aim is to remove iio_buffer_set_attrs(), so in the
iio_triggered_buffer_setup_ext() function the attributes are assigned
directly to 'buffer->attrs'.
When adding multiple IIO buffers per IIO device, it can be pretty
cumbersome to first allocate a set of buffers, then to dig them out of IIO
to assign extra attributes (with iio_buffer_set_attrs()).
Naturally, the best way would be to provide them at allocation time, which
is what this change does.
At this moment, buffers allocated with {devm_}iio_triggered_buffer_setup()
are the only ones in mainline IIO to call iio_buffer_set_attrs().
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-4-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
This change is mostly cosmetic, but it's also a pre-cursor to the
the change for 'iio_buffer_set_attrs()', where the helper gets updated to
better support multiple IIO buffers for 1 IIO device.
The only functional change is that the error message for the trigger alloc
failure is bound to the parent device vs the IIO device object.
Also, the new at91_adc_buffer_and_trigger_init() function was moved after
the definition of the 'at91_adc_fifo_attributes'.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-3-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
The iio_buffer_set_attrs() helper will be removed in this series. So, just
assign the attributes of the DMAEngine buffer logic directly.
This is IIO buffer core context, so there is direct access to the
buffer->attrs object.
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20200929125949.69934-2-alexandru.ardelean@analog.com
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Example `display-subsystem` has an incorrect compatible string.
Required properties section tells that developers should use
"fsl,imx-display-subsystem" as "compatible" string but the example
misses 'imx-' prefix.
Change example to have correct "compatible" string.
Signed-off-by: Cengiz Can <cengiz@kernel.wtf>
Link: https://lore.kernel.org/r/20201113150618.417424-1-cengiz@kernel.wtf
Signed-off-by: Rob Herring <robh@kernel.org>
The kernel uses ACPI Boot Error Record Table (BERT) to report fatal
errors that occurred in a previous boot. The MCA errors in the BERT are
reported using the x86 Processor Error Common Platform Error Record
(CPER) format. Currently, the record prints out the raw MSR values and
AMD relies on the raw record to provide MCA information.
Extract the raw MSR values of MCA registers from the BERT and feed them
into mce_log() to decode them properly.
The implementation is SMCA-specific as the raw MCA register values are
given in the register offset order of the SMCA address space.
[ bp: Massage. ]
[ Fix a build breakage in patch v1. ]
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Punit Agrawal <punit1.agrawal@toshiba.co.jp>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lkml.kernel.org/r/20201119182938.151155-1-Smita.KoralahalliChannabasappa@amd.com
In the patch to be fixed, horizontal_backporch_byte become too large
for some panel, so roll back that patch. For small hfp or hbp panel,
using vm->hfront_porch + vm->hback_porch to calculate
horizontal_backporch_byte would make it negtive, so
use horizontal_backporch_byte itself to make it positive.
Fixes: 35bf948f1e ("drm/mediatek: dsi: Fix scrolling of panel with small hfp or hbp")
Signed-off-by: CK Hu <ck.hu@mediatek.com>
Signed-off-by: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Tested-by: Bilal Wasim <bilal.wasim@imgtec.com>
'pci_set_dma_mask()' + 'pci_set_consistent_dma_mask()' can be replaced by
an equivalent 'dma_set_mask_and_coherent()' which is much less verbose.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://lore.kernel.org/r/20201121083747.1330299-1-christophe.jaillet@wanadoo.fr
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Use TEST %reg,%reg which sets the zero flag in the same way as CMP
$0,%reg, but the encoding uses one byte less.
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Andy Lutomirski <luto@kernel.org>
Link: https://lkml.kernel.org/r/20201029160258.139216-1-ubizjak@gmail.com
The only usage of the ops field in the loopback_cable struct is to call
its members, the field it self is never changed. Make it a pointer to
const. This allows us to constify two static loopback_ops structs to
allow the compiler to put them in read-only memory.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Link: https://lore.kernel.org/r/20201120231046.76758-1-rikard.falkeborn@gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Thomas Falcon says:
====================
ibmvnic: Performance improvements and other updates
The first three patches utilize a hypervisor call allowing multiple
TX and RX buffer replenishment descriptors to be sent in one operation,
which significantly reduces hypervisor call overhead. The xmit_more
and Byte Queue Limit API's are leveraged to provide this support
for TX descriptors.
The subsequent two patches remove superfluous code and members in
TX completion handling function and TX buffer structure, respectively,
and remove unused routines.
Finally, four patches which ensure that device queue memory is
cache-line aligned, resolving slowdowns observed in PCI traces,
as well as optimize the driver's NAPI polling function and
to RX buffer replenishment are provided by Dwip Banerjee.
This series provides significant performance improvements, allowing
the driver to fully utilize 100Gb NIC's.
====================
Link: https://lore.kernel.org/r/1605748345-32062-1-git-send-email-tlfalcon@linux.ibm.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reduce the amount of time spent replenishing RX buffers by only doing
so once available buffers has fallen under a certain threshold, in this
case half of the total number of buffers, or if the polling loop exits
before the packets processed is less than its budget. Non-exhaustion of
NAPI budget implies lower incoming packet pressure, allowing the leeway
to refill the buffers in preparation for any impending burst.
Signed-off-by: Dwip N. Banerjee <dnbanerg@us.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Take advantage of the additional optimizations in netdev_alloc_skb when
allocating socket buffers to be used for packet reception.
Signed-off-by: Dwip N. Banerjee <dnbanerg@us.ibm.com>
Acked-by: Lijun Pan <ljp@linux.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
If the current NAPI polling loop exits without completing it's
budget, only re-enable interrupts if there are no entries remaining
in the queue and napi_complete_done is successful. If there are entries
remaining on the queue that were missed, restart the polling loop.
Signed-off-by: Dwip N. Banerjee <dnbanerg@us.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
PCI bus slowdowns were observed on IBM VNIC devices as a result
of partial cache line writes and non-cache aligned full cache line writes.
Ensure that packet data buffers are cache-line aligned to avoid these
slowdowns.
Signed-off-by: Dwip N. Banerjee <dnbanerg@us.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
It is not longer used, so remove it.
Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Acked-by: Lijun Pan <ljp@linux.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Remove unused and superfluous code and members in
existing TX implementation and data structures.
Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Include support for the xmit_more feature utilizing the
H_SEND_SUB_CRQ_INDIRECT hypervisor call which allows the sending
of multiple subordinate Command Response Queue descriptors in one
hypervisor call via a DMA-mapped buffer. This update reduces hypervisor
calls and thus hypervisor call overhead per TX descriptor.
Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Utilize the H_SEND_SUB_CRQ_INDIRECT hypervisor call to send
multiple RX buffer descriptors to the device in one hypervisor
call operation. This change will reduce the number of hypervisor
calls and thus hypervisor call overhead needed to transmit
RX buffer descriptors to the device.
Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch introduces the infrastructure to send batched subordinate
Command Response Queue descriptors, which are used by the ibmvnic
driver to send TX frame and RX buffer descriptors.
Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Acked-by: Lijun Pan <ljp@linux.ibm.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The main purpose of the loop is to load the memory to the SCP SRAM.
If filesz is 0, can go to next program header directly.
We don't need to try to validate the FW binary for those filesz==0
segments.
Reviewed-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
Link: https://lore.kernel.org/r/20201116084413.3312631-3-tzungbi@google.com
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
It is valid if offset+length == sram_size.
For example, sram_size=100, offset=99, length=1. Accessing offset 99
with length 1 is valid.
Reviewed-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
Link: https://lore.kernel.org/r/20201116084413.3312631-2-tzungbi@google.com
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Fixes the following sparse errors on dma_alloc_coherent() and
dma_free_coherent().
On drivers/remoteproc/mtk_scp.c:559:23:
warning: incorrect type in assignment (different address spaces)
expected void [noderef] __iomem *cpu_addr
got void *
On drivers/remoteproc/mtk_scp.c:572:56:
warning: incorrect type in argument 3 (different address spaces)
expected void *cpu_addr
got void [noderef] __iomem *cpu_addr
The cpu_addr is not a __iomem address. Removes the marker.
Reviewed-by: Nicolas Boichat <drinkcat@chromium.org>
Reviewed-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Tzung-Bi Shih <tzungbi@google.com>
Link: https://lore.kernel.org/r/20201116082537.3287009-3-tzungbi@google.com
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
