Use a more standard form for the VT version number comments.
One slight oddball case is the dp_max_link_rate that had two
version numbers (216/230) and a platform name (GLK). The
story goes that the field was introduced in the spec in
version 216, along with a note that it's used on CNL+. Later
in version 230 the definition of the bit was changed in
bacakwards incompatible ways and the CNL note disappeard.
For us the original CNL+ note in the header got changed to
to GLK+ when all CNL support was dropped from the codebase.
We do still need (and have) handling for both the 216+ and
the 230+ defintions (parse_bdb_216_dp_max_link_rate() vs.
parse_bdb_230_dp_max_link_rate()).
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220715202044.11153-2-ville.syrjala@linux.intel.com
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Query eswitch functions returns information of the external host
PF(if it exists). It can be used to check if DEVX is running on ECPF.
Reviewed-by: Erez Shitrit <erezsh@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Bodong Wang <bodong@mellanox.com>
Link: https://lore.kernel.org/r/4265925178ab3224dc1d3e3784bb312d808edca5.1661763785.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The cited commit removed from the cleanup flow of umr the checks
if the resources were created. This could lead to null-ptr-deref
in case that we had failure in mlx5_ib_stage_ib_reg_init stage.
Fix it by adding new state to the umr that can say if the resources
were created or not and check it in the umr cleanup flow before
destroying the resources.
Fixes: 04876c12c1 ("RDMA/mlx5: Move init and cleanup of UMR to umr.c")
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Link: https://lore.kernel.org/r/4cfa61386cf202e9ce330e8d228ce3b25a36326e.1661763459.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When accessing Ports Performance Counters Register (PPCNT),
local port must be one if it is Function-Per-Port HCA that
HCA_CAP.num_ports is 1.
The offending patch can change the local port to other values
when accessing PPCNT after enabling switchdev mode. The following
syndrome will be printed:
# cat /sys/class/infiniband/rdmap4s0f0/ports/2/counters/*
# dmesg
mlx5_core 0000:04:00.0: mlx5_cmd_check:756:(pid 12450): ACCESS_REG(0x805) op_mod(0x1) failed, status bad parameter(0x3), syndrome (0x1e5585)
Fix it by setting local port to one for Function-Per-Port HCA.
Fixes: 210b1f7807 ("IB/mlx5: When not in dual port RoCE mode, use provided port as native")
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Chris Mi <cmi@nvidia.com>
Link: https://lore.kernel.org/r/6c5086c295c76211169e58dbd610fb0402360bab.1661763459.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When the RDMA auxiliary driver probes, it sets its profile based on
devlink driverinit value. The latter might not be in sync with FW yet
(In case devlink reload is not performed), thus causing a mismatch
between RDMA driver and FW. This results in the following FW syndrome
when the RDMA driver tries to adjust RoCE state, which fails the probe:
"0xC1F678 | modify_nic_vport_context: roce_en set on a vport that
doesn't support roce"
To prevent this, select the PF profile based on FW RoCE capability
instead of relying on devlink driverinit value.
To provide backward compatibility of the RoCE disable feature, on older
FW's where roce_rw is not set (FW RoCE capability is read-only), keep
the current behavior e.g., rely on devlink driverinit value.
Fixes: fbfa97b4d7 ("net/mlx5: Disable roce at HCA level")
Reviewed-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Link: https://lore.kernel.org/r/cb34ce9a1df4a24c135cb804db87f7d2418bd6cc.1661763459.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Fix a nested dead lock as part of ODP flow by using mmput_async().
From the below call trace [1] can see that calling mmput() once we have
the umem_odp->umem_mutex locked as required by
ib_umem_odp_map_dma_and_lock() might trigger in the same task the
exit_mmap()->__mmu_notifier_release()->mlx5_ib_invalidate_range() which
may dead lock when trying to lock the same mutex.
Moving to use mmput_async() will solve the problem as the above
exit_mmap() flow will be called in other task and will be executed once
the lock will be available.
[1]
[64843.077665] task:kworker/u133:2 state:D stack: 0 pid:80906 ppid:
2 flags:0x00004000
[64843.077672] Workqueue: mlx5_ib_page_fault mlx5_ib_eqe_pf_action [mlx5_ib]
[64843.077719] Call Trace:
[64843.077722] <TASK>
[64843.077724] __schedule+0x23d/0x590
[64843.077729] schedule+0x4e/0xb0
[64843.077735] schedule_preempt_disabled+0xe/0x10
[64843.077740] __mutex_lock.constprop.0+0x263/0x490
[64843.077747] __mutex_lock_slowpath+0x13/0x20
[64843.077752] mutex_lock+0x34/0x40
[64843.077758] mlx5_ib_invalidate_range+0x48/0x270 [mlx5_ib]
[64843.077808] __mmu_notifier_release+0x1a4/0x200
[64843.077816] exit_mmap+0x1bc/0x200
[64843.077822] ? walk_page_range+0x9c/0x120
[64843.077828] ? __cond_resched+0x1a/0x50
[64843.077833] ? mutex_lock+0x13/0x40
[64843.077839] ? uprobe_clear_state+0xac/0x120
[64843.077860] mmput+0x5f/0x140
[64843.077867] ib_umem_odp_map_dma_and_lock+0x21b/0x580 [ib_core]
[64843.077931] pagefault_real_mr+0x9a/0x140 [mlx5_ib]
[64843.077962] pagefault_mr+0xb4/0x550 [mlx5_ib]
[64843.077992] pagefault_single_data_segment.constprop.0+0x2ac/0x560
[mlx5_ib]
[64843.078022] mlx5_ib_eqe_pf_action+0x528/0x780 [mlx5_ib]
[64843.078051] process_one_work+0x22b/0x3d0
[64843.078059] worker_thread+0x53/0x410
[64843.078065] ? process_one_work+0x3d0/0x3d0
[64843.078073] kthread+0x12a/0x150
[64843.078079] ? set_kthread_struct+0x50/0x50
[64843.078085] ret_from_fork+0x22/0x30
[64843.078093] </TASK>
Fixes: 36f30e486d ("IB/core: Improve ODP to use hmm_range_fault()")
Reviewed-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/74d93541ea533ef7daec6f126deb1072500aeb16.1661251841.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Alex Elder says:
====================
net: ipa: start using transaction IDs
A previous group of patches added ID fields to track the state of
transactions:
https://lore.kernel.org/netdev/20220831224017.377745-1-elder@linaro.org
This series starts using those IDs instead of the lists used
previously. Most of this series involves reworking the function
that determines which transaction is the "last", which determines
when a channel has been quiesed. The last patch is mainly used to
prove that the new index method of tracking transaction state is
equivalent to the previous use of lists.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The completed transaction list is used in gsi_channel_trans_complete()
to return the next transaction in completed state.
Add some temporary checks to verify the transaction indicated by the
completed ID matches the one first in this list.
Similarly, we use the pending and completed transaction lists when
cancelling pending transactions in gsi_channel_trans_cancel_pending().
Add temporary checks there to verify the transactions indicated by
IDs match those tracked by these lists.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Do a little more refactoring in gsi_channel_trans_last() to simplify
it further. The resulting code should behave exactly as before.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using a little logic we can simplify gsi_channel_trans_last().
The first condition in that function looks like this:
if (trans_info->allocated_id != trans_info->free_id)
And if that's false, we proceed to the next one:
if (trans_info->committed_id != trans_info->allocated_id)
Failure of the first test implies:
trans_info->allocated_id == trans_info->free_id
And therefore, the second one can be rewritten this way:
if (trans_info->committed_id != trans_info->free_id)
Substituting free_id for allocated_id and committed_id can also be
done in the code blocks executed when these conditions yield true.
The net result is that all three blocks for TX endpoints can be
consolidated into just one.
The two blocks of code at the end of that function (used for both TX
and RX channels) can be similarly consolidated into a single block.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Always use transaction IDs when finding the "last" transaction to
await when quiescing a channel. This basically extends what was
done in the previous patch to all other transaction state IDs.
As a result we are no longer updating any transaction lists inside
gsi_channel_trans_last(), so there's no need to take the spinlock.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the allocated and free transaction IDs to determine whether the
"last" transaction used for quiescing a channel is in allocated
state. The last allocated transaction that has not been committed
(if any) immediately precedes the first free transaction in the
transaction array.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
When quiescing a channel, we find the "last" transaction, which is
the latest one to have been allocated. (New transaction allocation
will have been prevented by the time this is called.)
Currently we do this by looking for the first non-empty transaction
list in each state, then return the last entry from that last.
Instead, determine the last entry in each list (if any) and return
that entry if found.
Temporarily (locally) introduce list_last_entry_or_null() as a
helper for this, mirroring list_first_entry_or_null(). This macro
definition will be removed by an upcoming patch.
Remove the temporary warnings added by the previous commit.
Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that we have devm_clk_get_optional_enabled(), we don't have to
open-code it.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It's not clear where these entries came from, and as I wrote in the
comment: Not even Realtek's r8101 driver knows these chip id's.
So remove the comment.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It's not clear why XID's 380 and 381..387 ever got different chip
version id's. VER_12 and VER_17 are handled exactly the same.
Therefore merge handling under the VER_17 umbrella.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Drop another couple of pointless pr_info() calls, and drop a number of
instances of hcd_name variables that are no longer referenced now that
they are no longer printed to the log at driver registration time.
Fixes: 10174220f5 ("usb: reduce kernel log spam on driver registration")
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lore.kernel.org/r/20220905111740.352348-1-ardb@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I would like to stop exporting OF-specific gpiod_get_from_of_node()
so that gpiolib can be cleaned a bit, so let's switch to the generic
fwnode property API.
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Link: https://lore.kernel.org/r/20220903-gpiod_get_from_of_node-remove-v1-5-b29adfb27a6c@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I would like to stop exporting OF-specific devm_gpiod_get_from_of_node()
so that gpiolib can be cleaned a bit, so let's switch to the generic
device property API.
I believe that the only reason the driver, instead of the standard
devm_gpiod_get(), used devm_gpiod_get_from_of_node() is because it
wanted to set up a pretty consumer name for the GPIO, and we now have
a special API for that.
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Link: https://lore.kernel.org/r/20220903-gpiod_get_from_of_node-remove-v1-4-b29adfb27a6c@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The ohci retaining bogus hardware states cause usb disconnect devices
connected before hibernation(s4), this issue occur when ohci-platform
driver build as a module and the built-in ohci-platform driver will
re probe and re enumerate the devices, so there will be no such problem.
Avoid retaining bogus hardware states during resume-from-hibernation.
Previously we had reset the hardware as part of preparing to reinstate
the snapshot image. But we can do better now with the new PM framework,
since we know exactly which resume operations are from hibernation.
According to the commit 'cd1965db05 ("USB: ohci: move ohci_pci_{
suspend,resume} to ohci-hcd.c")' and commit '6ec4beb5c7 ("USB: new
flag for resume-from-hibernation")', the flag "hibernated" is for
resume-from-hibernation and it should be true when usb resume from disk.
When this flag "hibernated" is set, the drivers will reset the hardware
to get rid of any existing state and make sure resume from hibernation
re-enumerates everything for ohci.
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Yinbo Zhu <zhuyinbo@loongson.cn>
Link: https://lore.kernel.org/r/20220902063639.17875-1-zhuyinbo@loongson.cn
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Sometimes drivers are missing logs when they return EINVAL.
Printing the failure here in common code can help understand where
EINVAL is coming from.
All other atomic_check() calls in this file already have similar
logging.
v2: add missing newlines
Signed-off-by: Simon Ser <contact@emersion.fr>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220829151451.152114-1-contact@emersion.fr
Kuyo reports that the pattern of using debugfs_remove(debugfs_lookup())
leaks a dentry and with a hotplug stress test, the machine eventually
runs out of memory.
Fix this up by using the newly created debugfs_lookup_and_remove() call
instead which properly handles the dentry reference counting logic.
Cc: Major Chen <major.chen@samsung.com>
Cc: stable <stable@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ben Segall <bsegall@google.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Daniel Bristot de Oliveira <bristot@redhat.com>
Cc: Valentin Schneider <vschneid@redhat.com>
Cc: Matthias Brugger <matthias.bgg@gmail.com>
Reported-by: Kuyo Chang <kuyo.chang@mediatek.com>
Tested-by: Kuyo Chang <kuyo.chang@mediatek.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220902123107.109274-2-gregkh@linuxfoundation.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is a very common pattern of using
debugfs_remove(debufs_lookup(..)) which results in a dentry leak of the
dentry that was looked up. Instead of having to open-code the correct
pattern of calling dput() on the dentry, create
debugfs_lookup_and_remove() to handle this pattern automatically and
properly without any memory leaks.
Cc: stable <stable@kernel.org>
Reported-by: Kuyo Chang <kuyo.chang@mediatek.com>
Tested-by: Kuyo Chang <kuyo.chang@mediatek.com>
Link: https://lore.kernel.org/r/YxIaQ8cSinDR881k@kroah.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Python likes to send an empty string for some sysfs files, including the
driver_override field. When commit 23d99baf9d ("PCI: Use
driver_set_override() instead of open-coding") moved the PCI core to use
the driver core function instead of hand-rolling their own handler, this
showed up as a regression from some userspace tools, like DPDK.
Fix this up by actually looking at the length of the string first
instead of trusting that userspace got it correct.
Fixes: 23d99baf9d ("PCI: Use driver_set_override() instead of open-coding")
Cc: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: stable <stable@kernel.org>
Reported-by: Stephen Hemminger <stephen@networkplumber.org>
Tested-by: Huisong Li <lihuisong@huawei.com>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20220901163734.3583106-1-gregkh@linuxfoundation.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When registering a connector, the kernel sends a hotplug uevent in
drm_connector_register(). When unregistering a connector, drivers
are expected to send a uevent as well. However, user-space has no way
to figure out that the connector isn't registered anymore: it'll still
be reported in GETCONNECTOR IOCTLs.
The documentation for DRM_CONNECTOR_UNREGISTERED states:
> The connector […] has since been unregistered and removed from
> userspace, or the connector was unregistered before it had a chance
> to be exposed to userspace
Signed-off-by: Simon Ser <contact@emersion.fr>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220801133754.461037-1-contact@emersion.fr
The 'enable-active-low' property is not a valid one.
Only 'enable-active-high' is valid, and when this property is absent
the gpio regulator will act as active low by default.
Remove the invalid 'enable-active-low' property.
Fixes: b33a22a1e7 ("arm64: dts: rockchip: add basic dts for Pine64 Quartz64-A")
Signed-off-by: Fabio Estevam <festevam@denx.de>
Link: https://lore.kernel.org/r/20220827175140.1696699-2-festevam@denx.de
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
The 'enable-active-low' property is not a valid one.
Only 'enable-active-high' is valid, and when this property is absent
the gpio regulator will act as active low by default.
Remove the invalid 'enable-active-low' property.
Fixes: 2c66fc34e9 ("arm64: dts: rockchip: add RK3399-Q7 (Puma) SoM")
Signed-off-by: Fabio Estevam <festevam@denx.de>
Link: https://lore.kernel.org/r/20220827175140.1696699-1-festevam@denx.de
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Asus UM325UAZ has GPIO 18 programmed as both an interrupt and a wake
source, but confirmed with internal team on this design this pin is
floating and shouldn't have been programmed. This causes lots of
spurious IRQs on the system and horrendous battery life.
Add a quirk to ignore attempts to program this pin on this system.
Reported-by: Pavel Krc <reg.krn@pkrc.net>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216208
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
gpiolib-acpi already had support for ignoring a pin for wakeup, but
if an OEM configures a floating pin as an interrupt source then
stopping it from being a wakeup won't do much good to stop the
interrupt storm.
Add support for a module parameter and quirk infrastructure to
ignore interrupts as well.
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Add support for Radxa ROCK 4C+ SBC.
Key differences of 4C+ compared to previous ROCK Pi 4.
- Rockchip RK3399-T SoC
- DP from 4C replaced with micro HDMI 2K@60fps
- 4-lane MIPI DSI with 1920*1080
- RK817 Audio codec
Also, an official naming convention from Radxa mention to remove
Pi from board name, so this 4C+ is named as Radxa ROCK 4C+ not
Radxa ROCK Pi 4C+.
Signed-off-by: Stephen Chen <stephen@radxa.com>
Signed-off-by: Manoj Sai <abbaraju.manojsai@amarulasolutions.com>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Link: https://lore.kernel.org/r/20220902065057.97425-3-jagan@amarulasolutions.com
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
RK3399-T is down-clocked version of RK3399 SoC operated at 1.5GHz.
Add CPU operating points table for it.
Signed-off-by: FUKAUMI Naoki <naoki@radxa.com>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Link: https://lore.kernel.org/r/20220902065057.97425-2-jagan@amarulasolutions.com
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Document the dt-bindings for Radxa ROCK 4C+ SBC.
Key differences of 4C+ compared to previous ROCK Pi 4.
- Rockchip RK3399-T SoC
- DP from 4C replaced with micro HDMI 2K@60fps
- 4-lane MIPI DSI with 1920*1080
- RK817 Audio codec
Also, an official naming convention from Radxa mention to remove
Pi from board name, so this 4C+ is named as Radxa ROCK 4C+ not
Radxa ROCK Pi 4C+.
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
Link: https://lore.kernel.org/r/20220902065057.97425-1-jagan@amarulasolutions.com
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
The property "vbus-supply" was copied from the vendor kernel but is not
available in mainstream. Use correct property "phy-supply".
Fixes: d6cfb110b0 ("arm64: dts: rockchip: add usb3 support to rk3568-evb1-v10")
Signed-off-by: Michael Riesch <michael.riesch@wolfvision.net>
Link: https://lore.kernel.org/r/20220905064335.104650-2-michael.riesch@wolfvision.net
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
The property "vbus-supply" was copied from the vendor kernel but is not
available in mainstream. Use correct property "phy-supply".
Fixes: 254a1f6a29 ("arm64: dts: rockchip: add usb3 support to the radxa rock3 model a")
Signed-off-by: Michael Riesch <michael.riesch@wolfvision.net>
Link: https://lore.kernel.org/r/20220905064335.104650-1-michael.riesch@wolfvision.net
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Add power domains support for RV1126 SoC.
Cypto, VO and NPU domains will add it future patches.
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com>
Signed-off-by: Jagan Teki <jagan@edgeble.ai>
Link: https://lore.kernel.org/r/20220818124132.125304-4-jagan@edgeble.ai
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
The SRv6 layer allows defining HMAC data that can later be used to sign IPv6
Segment Routing Headers. This configuration is realised via netlink through
four attributes: SEG6_ATTR_HMACKEYID, SEG6_ATTR_SECRET, SEG6_ATTR_SECRETLEN and
SEG6_ATTR_ALGID. Because the SECRETLEN attribute is decoupled from the actual
length of the SECRET attribute, it is possible to provide invalid combinations
(e.g., secret = "", secretlen = 64). This case is not checked in the code and
with an appropriately crafted netlink message, an out-of-bounds read of up
to 64 bytes (max secret length) can occur past the skb end pointer and into
skb_shared_info:
Breakpoint 1, seg6_genl_sethmac (skb=<optimized out>, info=<optimized out>) at net/ipv6/seg6.c:208
208 memcpy(hinfo->secret, secret, slen);
(gdb) bt
#0 seg6_genl_sethmac (skb=<optimized out>, info=<optimized out>) at net/ipv6/seg6.c:208
#1 0xffffffff81e012e9 in genl_family_rcv_msg_doit (skb=skb@entry=0xffff88800b1f9f00, nlh=nlh@entry=0xffff88800b1b7600,
extack=extack@entry=0xffffc90000ba7af0, ops=ops@entry=0xffffc90000ba7a80, hdrlen=4, net=0xffffffff84237580 <init_net>, family=<optimized out>,
family=<optimized out>) at net/netlink/genetlink.c:731
#2 0xffffffff81e01435 in genl_family_rcv_msg (extack=0xffffc90000ba7af0, nlh=0xffff88800b1b7600, skb=0xffff88800b1f9f00,
family=0xffffffff82fef6c0 <seg6_genl_family>) at net/netlink/genetlink.c:775
#3 genl_rcv_msg (skb=0xffff88800b1f9f00, nlh=0xffff88800b1b7600, extack=0xffffc90000ba7af0) at net/netlink/genetlink.c:792
#4 0xffffffff81dfffc3 in netlink_rcv_skb (skb=skb@entry=0xffff88800b1f9f00, cb=cb@entry=0xffffffff81e01350 <genl_rcv_msg>)
at net/netlink/af_netlink.c:2501
#5 0xffffffff81e00919 in genl_rcv (skb=0xffff88800b1f9f00) at net/netlink/genetlink.c:803
#6 0xffffffff81dff6ae in netlink_unicast_kernel (ssk=0xffff888010eec800, skb=0xffff88800b1f9f00, sk=0xffff888004aed000)
at net/netlink/af_netlink.c:1319
#7 netlink_unicast (ssk=ssk@entry=0xffff888010eec800, skb=skb@entry=0xffff88800b1f9f00, portid=portid@entry=0, nonblock=<optimized out>)
at net/netlink/af_netlink.c:1345
#8 0xffffffff81dff9a4 in netlink_sendmsg (sock=<optimized out>, msg=0xffffc90000ba7e48, len=<optimized out>) at net/netlink/af_netlink.c:1921
...
(gdb) p/x ((struct sk_buff *)0xffff88800b1f9f00)->head + ((struct sk_buff *)0xffff88800b1f9f00)->end
$1 = 0xffff88800b1b76c0
(gdb) p/x secret
$2 = 0xffff88800b1b76c0
(gdb) p slen
$3 = 64 '@'
The OOB data can then be read back from userspace by dumping HMAC state. This
commit fixes this by ensuring SECRETLEN cannot exceed the actual length of
SECRET.
Reported-by: Lucas Leong <wmliang.tw@gmail.com>
Tested: verified that EINVAL is correctly returned when secretlen > len(secret)
Fixes: 4f4853dc1c ("ipv6: sr: implement API to control SR HMAC structure")
Signed-off-by: David Lebrun <dlebrun@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On system suspend when system memory is low then i915_gem_obj_copy_ttm()
could fail trying to backup a lmem obj. GEM_WARN_ON() is not enough,
suspend shouldn't continue if i915_ttm_backup() throws an error.
v2: Keep the fdo issue till we have a igt test(Matt).
v3: Use %pe(Andrzej)
References: https://gitlab.freedesktop.org/drm/intel/-/issues/6529
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Suggested-by: Chris P Wilson <chris.p.wilson@intel.com>
Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220901172217.18392-1-nirmoy.das@intel.com
Maxime Chevallier says:
====================
net: altera: tse: phylink conversion
This is V4 of a series converting the Altera TSE driver to phylink,
introducing a new PCS driver along the way.
The Altera TSE can be built with a SGMII/1000BaseX PCS, allowing to use
SFP ports with this MAC, which is the end goal of adding phylink support
and a proper PCS driver.
The PCS itself can either be mapped in the MAC's register space, in that
case, it's accessed through 32 bits registers, with the higher 16 bits
always 0. Alternatively, it can sit on its own register space, exposing
16 bits registers, some of which ressemble the standard PHY registers.
To tackle that rework, several things needs updating, starting by the DT
binding, since we add support for a new register range for the PCS.
Hence, the first patch of the series is a conversion to YAML of the
existing binding.
Then, patch 2 does a bit of simple cleanup to the TSE driver, using nice
reverse xmas tree definitions.
Patch 3 adds the actual PCS driver, as a standalone driver. Some future
series will then reuse that PCS driver from the dwmac-socfpga driver,
which implements support for this exact PCS too, allowing to share the
code nicely.
Patch 4 is then a phylink conversion of the altera_tse driver, to use
this new PCS driver.
Finally, patch 5 updates the newly converted DT binding to support the
pcs register range.
This series contains bits and pieces for this conversion, please tell me if
you want me to send it as individual patches.
V4 Changes:
- Add missing MODULE_* macros to the TSE PCS driver
V3 Changes:
- YAML binding conversion changes and PCS addition changes thanks to
Krzysztof's reviews
V2 Changes :
- Fixed the binding after the YAML conversion
- Added a pcs_validate() callback
- Introduced a comment to justify a soft reset for the PCS
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Some implementations of the TSE have their PCS as an external bloc,
exposed at its own register range. Document this, and add a new example
showing a case using the pcs and the new phylink conversion to connect
an sfp port to a TSE mac.
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert the Altera Triple Speed Ethernet Controller to phylink.
This controller supports MII, GMII and RGMII with its MAC, and
SGMII + 1000BaseX through a small embedded PCS.
The PCS itself has a register set very similar to what is found in a
typical 802.3 ethernet PHY, but this register set memory-mapped instead
of lying on an mdio bus.
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The Altera Triple Speed Ethernet has a SGMII/1000BaseC PCS that can be
integrated in several ways. It can either be part of the TSE MAC's
address space, accessed through 32 bits accesses on the mapped mdio
device 0, or through a dedicated 16 bits register set.
This driver allows using the TSE PCS outside of altera TSE's driver,
since it can be used standalone by other MACs.
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make the driver code cleaner through a strictly cosmetic change, using
he reverse xmas tree variable declaration ordering.
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>