| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Some PCI devices must be powered-on before they can be detected on the
bus. Introduce a simple framework reusing the existing PCI OF
infrastructure.
The way this works is: a DT node representing a PCI device connected to
the port can be matched against its power control platform driver. If
the match succeeds, the driver is responsible for powering-up the device
and calling pci_pwrctl_device_set_ready() which will trigger a PCI bus
rescan as well as subscribe to PCI bus notifications.
When the device is detected and created, we'll make it consume the same
DT node that the platform device did. When the device is bound, we'll
create a device link between it and the parent power control device.
Tested-by: Amit Pundir <[email protected]>
Tested-by: Neil Armstrong <[email protected]> # on SM8550-QRD, SM8650-QRD & SM8650-HDK
Tested-by: Caleb Connolly <[email protected]> # OnePlus 8T
Acked-by: Bjorn Helgaas <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Bartosz Golaszewski <[email protected]>
|
|
In preparation for introducing PCI device power control - a set of
library functions that will allow powering-up of PCI devices before
they're detected on the PCI bus - we need to populate the devices
defined on the device-tree.
We are reusing the platform bus as it provides us with all the
infrastructure we need to match the pwrctl drivers against the
compatibles from OF nodes.
These platform devices will be probed by the driver core and bound to
the PCI pwrctl drivers we'll introduce later.
Tested-by: Amit Pundir <[email protected]>
Tested-by: Neil Armstrong <[email protected]> # on SM8550-QRD, SM8650-QRD & SM8650-HDK
Tested-by: Caleb Connolly <[email protected]> # OnePlus 8T
Acked-by: Bjorn Helgaas <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Bartosz Golaszewski <[email protected]>
|
|
With PCI power control we deal with two struct device objects bound to
two different drivers but consuming the same OF node. We must not bind
the pinctrl twice. To that end: before setting the OF node of the newly
instantiated PCI device, check if a platform device consuming the same
OF node doesn't already exist on the platform bus and - if so - mark the
PCI device as reusing the OF node.
Tested-by: Amit Pundir <[email protected]>
Tested-by: Neil Armstrong <[email protected]> # on SM8550-QRD, SM8650-QRD & SM8650-HDK
Tested-by: Caleb Connolly <[email protected]> # OnePlus 8T
Acked-by: Bjorn Helgaas <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Bartosz Golaszewski <[email protected]>
|
|
With the introduction of PCI device power control drivers that will be
able to trigger the port rescan when probing, we need to hold the rescan
mutex during the initial pci_host_probe() too or the two could get in
each other's way.
Tested-by: Amit Pundir <[email protected]>
Tested-by: Neil Armstrong <[email protected]> # on SM8550-QRD, SM8650-QRD & SM8650-HDK
Tested-by: Caleb Connolly <[email protected]> # OnePlus 8T
Acked-by: Bjorn Helgaas <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Bartosz Golaszewski <[email protected]>
|
|
While commit 255ec3879dd4 ("phy: exynos5-usbdrd: Add 26MHz ref clk
support") correctly states that CLKRSTCTRL[7:5] doesn't need to be set
on modern Exynos platforms, SSPPLLCTL[2:0] should be programmed with
the frequency of the reference clock for the USB2.0 phy instead.
I stumbled across this while adding support for the Google Tensor
gs101, but this should apply to E850 just the same.
Do so.
Signed-off-by: André Draszik <[email protected]>
Reviewed-by: Peter Griffin <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Vinod Koul <[email protected]>
|
|
Using 0x82 seems odd, where everything else is just a sequence.
On E850, this macro isn't used (as a register value), only to assign
its value to the 'extrefclk' variable, which is otherwise unused on
that platform. Older platforms don't appear to support 26MHz in the
first place (since this macro was added for E850).
Furthermore, the downstream driver uses 0x82 to denote
USBPHY_REFCLK_DIFF_26MHZ (whatever that means exactly), but for all the
other values we match downstream's non-DIFF macros.
Update to avoid confusion. No functional change intended.
Signed-off-by: André Draszik <[email protected]>
Reviewed-by: Peter Griffin <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Vinod Koul <[email protected]>
|
|
on / not on is just a boolean flag and is a bit misleading as currently
on==1 means to turn off the power, and on==0 to turn power on.
Rename the flag and make it a bool to avoid confusion of future readers
of this code. No functional change.
While at it, fix a whitespace issue in nearby comment.
No functional change.
Signed-off-by: André Draszik <[email protected]>
Reviewed-by: Peter Griffin <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Vinod Koul <[email protected]>
|
|
The timers-howto recommends using usleep_range() and friends anytime
waiting for >= ~10us is required. Doing so can help the timer subsystem
a lot to coalesce wakeups.
Additionally, fsleep() exists as a convenient wrapper so we do not have
to think about which exact sleeping function is required in which case.
Convert all udelay() calls in this driver to use fsleep() to follow the
recommendataion.
Signed-off-by: André Draszik <[email protected]>
Reviewed-by: Peter Griffin <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Vinod Koul <[email protected]>
|
|
Most of the macros are ordered high -> low, but there are some
outliers.
Order them all uniformly from high to low. This will allow adding
additional register (field) definitions in a consistent way.
While at it, also remove some extra empty lines to group register bit
field definitions together with the relevant register. This makes the
registers easier to distinguish visually.
No functional change.
Signed-off-by: André Draszik <[email protected]>
Reviewed-by: Peter Griffin <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Vinod Koul <[email protected]>
|
|
This is the first step towards removing the P2P Device MAC.
Use ROC (which uses the AUX MAC) for P2P Device
discoveribility and action frames.
Signed-off-by: Shaul Triebitz <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140556.8c90e457abbd.I8e340759ecb299e05b1809f3d8060429c6cbbd01@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
If non-BSS and remain-on-channel (ROC) blocking were to occur
simultaneously, they'd step on each other's toes, unblocking
when not yet supported. Disentangle these bits, and ROC doesn't
need to use the non_bss_link() function then.
Fixes: a1efeb823084 ("wifi: iwlwifi: mvm: Block EMLSR when a p2p/softAP vif is active")
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140556.461fcf7b95bb.Id0d21dcb739d426ff15ec068b5df8abaab58884d@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
Stop supporting all FWs older than the max API version.
These FW versions were supported since v6.5.
Signed-off-by: Miri Korenblit <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Link: https://msgid.link/20240605140327.ad6d43fe9893.I96f769e7d5be3e6499d260451df781bd694a5142@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
When EMLSR gets unblocked, the current code checks if the last exit was
due to an EXIT reason (as opposed to a BLOCKING one), and if so, it
does nothing, as in this case a MLO scan was scheduled to run in 30
seconds.
But the code doesn't consider the time that passed from the last exit,
so if immediately after the exit a blocker occurred (e.g. non-BSS
interface), and lasts for more than 30 seconds, then the MLO scan and the
following link selection will decide not to enter EMLSR, and when the
unblocking event finally happens, the reason is still set to the EXIT one,
so it will do nothing, and we will not have the chance to re-enable EMLSR.
Fix this by checking also the time that has passed since the last exit,
only if it is less than 30 seconds, we can count on the scheduled MLO
scan.
Note that clearing the reason itself can't be done since it is needed
for the EMLSR prevention mechanism.
Fixes: 2f33561ea8f9 ("wifi: iwlwifi: mvm: trigger link selection after exiting EMLSR")
Signed-off-by: Miri Korenblit <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Link: https://msgid.link/20240605140327.58556fc4cfa9.I4c55b3cd9f20b21b37f28258d0fb6842ba413966@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
This was needed when we had multiple types of transports. Now we only
have pcie, so there is no need for this ops.
Cleanup the code such as the different trans APIs will call the pcie
function directly, instead of calling the callback,
and remove struct iwl_trans_ops.
Signed-off-by: Yedidya Benshimol <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140327.8315ff64f9f3.Ifdbc1f26d49766f7de553dcb5f613885f4ee65cc@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
The TX queue code was mostly moved out to support an internal
transport that we were never going to publish, but we're no
longer using that. Since we're also going to be dissolving
the virtual transport layer entirely, integrate the TX queue
code into the PCIe layer.
This also has a small kernel of already removing the virtual
transport function layer, since iwl_trans_send_cmd() calls
iwl_trans_pcie_send_hcmd() directly now, even if that still
calls the transport send_cmd method for now, we'll clean it
up later.
Also, not everything is renamed yet.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140327.936b13f45071.Ib219ce01a1e67bcad79d5131626db950252aaa46@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
This needs to include dbg-tlv.h since it uses the value of
IWL_FW_INI_ALLOCATION_NUM from that file.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140327.2d25691283eb.I0909621a0e293a8a21d4f1de6e5fd59c22e4b212@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
This really isn't correct to be in the opmode, do the clamping
(and power-of-2 fixup that may be necessary due to this, or even
otherwise) in the queue code. Also move down the retrying of the
allocation, it should be after all the size fixups, but also it
just makes sense, and avoids retrying same-size allocations in
the case of the BZ-family A-step workaround.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140327.000a0a1e807d.Ib822590d5aca76ff3168418ae2c139b3d43d81ed@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
When entering D3 we want to configure skip over DTIM, but
it can't use the deflink configuration, that will not even
exist. Adjust the code to handle multiple links by taking
the min skip, even if we should only have a single active
link at this point.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140327.bccf980fadb4.Idc98b9f3634f39d2fae9bd9916f5d050ccd48f95@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
Stop supporting older FWs.
Signed-off-by: Miri Korenblit <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Link: https://msgid.link/20240605140327.ff8477233010.Ic8c73bd6749cc5f8ab5297807bb0be9bd96a59fa@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
These entries are not used, remove them.
Signed-off-by: Miri Korenblit <[email protected]>
Reviewed-by: Johannes Berg <[email protected]>
Link: https://msgid.link/20240605140327.0c7c520814d5.I19cefb3d81b03a5be94c029cfffd1c8b8c437182@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
Setting a channel with 320 MHz channel width over hwsim results in an
array-index-out-of-bounds error. Fix it by adding 320 MHz to hwsim
supported channel widths.
Signed-off-by: Avraham Stern <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605135233.a766c1465566.Ib859c7233511b61b8a34022cfceeb4971c739d80@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
The control and ack force-wake registers are not accessible for
the VF drivers. To avoid changing existing code logic that tracks
woken domains, simply ignore all attempts to access control or ack
registers if we are running as a VF driver.
Signed-off-by: Michal Wajdeczko <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
|
|
The code of 'control' and 'wait' force-wake operations are very
similar for both 'wake' and 'sleep' cases. Add helpers to maximize
code reuse.
Signed-off-by: Michal Wajdeczko <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
|
|
For debug purposes it might be useful to look at the values of the
force-wake ack registers in case wake/sleep operations failures.
Move xe_gt_notice() from the caller to the helper function, where
we have the latest value of force-wake ack register available.
Signed-off-by: Michal Wajdeczko <[email protected]>
Cc: Rodrigo Vivi <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
|
|
It's too simple to deserve separate helper.
Suggested-by: Lucas De Marchi <[email protected]>
Signed-off-by: Michal Wajdeczko <[email protected]>
Reviewed-by: Francois Dugast <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
|
|
If possible, we should prefer xe_gt_notice() over drm_notice().
Signed-off-by: Michal Wajdeczko <[email protected]>
Reviewed-by: Rodrigo Vivi <[email protected]>
Cc: Lucas De Marchi <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
|
|
...to address the following warning:
drivers/hwmon/k10temp.c:273:47:
warning: declaration shadows a variable in the global scope [-Wshadow]
static umode_t k10temp_is_visible(const void *_data,
^
No functional change is intended.
Signed-off-by: Yazen Ghannam <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Acked-by: Guenter Roeck <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
...to address the following warning:
drivers/hwmon/k10temp.c:104:9:
warning: macro is not used [-Wunused-macros]
No functional change is intended.
Signed-off-by: Yazen Ghannam <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Acked-by: Guenter Roeck <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
Currently, k10temp_get_ccd_support() takes as input "pdev" and "data". However,
"pdev" is already included in "data". Furthermore, the "pdev" parameter is no
longer used in k10temp_get_ccd_support(), since its use was moved into
read_ccd_temp_reg().
Drop the "pdev" input parameter as it is no longer needed.
No functional change is intended.
Signed-off-by: Yazen Ghannam <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Mario Limonciello <[email protected]>
Acked-by: Guenter Roeck <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
The CCD temperature register is read in two places. These reads are done
using an AMD SMN access, and a number of parameters are needed for the
operation.
Move the SMN access and parameter gathering into a helper function in order to
simplify the code flow. This also has a benefit of centralizing the hardware
register access in a single place in case fixes or special decoding is required.
Signed-off-by: Yazen Ghannam <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Mario Limonciello <[email protected]>
Acked-by: Guenter Roeck <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
Check the return value of amd_smn_read() before saving a value. This
ensures invalid values aren't saved or used.
There are three cases here with slightly different behavior:
1) read_tempreg_nb_zen():
This is a function pointer which does not include a return code.
In this case, set the register value to 0 on failure. This
enforces Read-as-Zero behavior.
2) k10temp_read_temp():
This function does have return codes, so return the error code
from the failed register read. Continued operation is not
necessary, since there is no valid data from the register.
Furthermore, if the register value was set to 0, then the
following operation would underflow.
3) k10temp_get_ccd_support():
This function reads the same register from multiple CCD
instances in a loop. And a bitmask is formed if a specific bit
is set in each register instance. The loop should continue on a
failed register read, skipping the bit check.
Signed-off-by: Yazen Ghannam <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Mario Limonciello <[email protected]>
Acked-by: Guenter Roeck <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
Check the return value of amd_smn_read() before saving a value. This
ensures invalid values aren't saved. The struct umc instance is
initialized to 0 during memory allocation. Therefore, a bad read will
keep the value as 0 providing the expected Read-as-Zero behavior.
Signed-off-by: Yazen Ghannam <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Mario Limonciello <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
A number of UMC registers are read only for the purpose of debug printing. They
are not used in any calculations. Nor do they have any specific debug value.
Remove them.
Signed-off-by: Yazen Ghannam <[email protected]>
Signed-off-by: Borislav Petkov (AMD) <[email protected]>
Reviewed-by: Mario Limonciello <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
|
|
While use of fsl_ifc driver with NAND flash is fine, as the fsl_ifc_nand
driver selects FSL_IFC automatically, we need the CONFIG_FSL_IFC option to
be selectable for platforms using fsl_ifc with NOR flash.
Fixes: ea0c0ad6b6eb ("memory: Enable compile testing for most of the drivers")
Reviewed-by: Miquel Raynal <[email protected]>
Signed-off-by: Esben Haabendal <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Krzysztof Kozlowski <[email protected]>
|
|
New CPU #defines encode vendor and family as well as model.
Signed-off-by: Tony Luck <[email protected]>
Reviewed-by: Jithu Joseph <[email protected]>
Reviewed-by: Kuppuswamy Sathyanarayanan <[email protected]>
Acked-by: Hans de Goede <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Ilpo Järvinen <[email protected]>
Signed-off-by: Ilpo Järvinen <[email protected]>
|
|
The flags variable is incorrectly checked while it is still cleared and
has not been assigned any value yet.
Fix it.
Fixes: a615323f7f90 ("wifi: iwlwifi: mvm: always apply 6 GHz probe limitations")
Signed-off-by: Ayala Beker <[email protected]>
Reviewed-by: Benjamin Berg <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140556.291c33f9a283.Id651fe69828aebce177b49b2316c5780906f1b37@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
For using the ROC command, check that the ROC version
is *greater or equal* to 3, rather than *equal* to 3.
The ROC version was added to the TLV starting from
version 3.
Fixes: 67ac248e4db0 ("wifi: iwlwifi: mvm: implement ROC version 3")
Signed-off-by: Shaul Triebitz <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140327.93d86cd188ad.Iceadef5a2f3cfa4a127e94a0405eba8342ec89c1@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
Unlock the mvm mutex before returning from a
function with the mutex locked.
Fixes: a1efeb823084 ("wifi: iwlwifi: mvm: Block EMLSR when a p2p/softAP vif is active")
Signed-off-by: Shaul Triebitz <[email protected]>
Signed-off-by: Miri Korenblit <[email protected]>
Link: https://msgid.link/20240605140327.96cb956db4af.Ib468cbad38959910977b5581f6111ab0afae9880@changeid
Signed-off-by: Johannes Berg <[email protected]>
|
|
In order to let bridge chains implement HDMI connector infrastructure,
add necessary glue code to the drm_bridge_connector. In case there is a
bridge that sets DRM_BRIDGE_OP_HDMI, drm_bridge_connector will register
itself as a HDMI connector and provide proxy drm_connector_hdmi_funcs
implementation.
Note, to simplify implementation, there can be only one bridge in a
chain that sets DRM_BRIDGE_OP_HDMI. Setting more than one is considered
an error. This limitation can be lifted later, if the need arises.
Acked-by: Maxime Ripard <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Signed-off-by: Dmitry Baryshkov <[email protected]>
|
|
Turn drm_bridge_connector to using drmm_kzalloc() and
drmm_connector_init() and drop the custom destroy function. The
drm_connector_unregister() and fwnode_handle_put() are already handled
by the drm_connector_cleanup() and so are safe to be dropped.
Acked-by: Maxime Ripard <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Signed-off-by: Dmitry Baryshkov <[email protected]>
|
|
Add drm_atomic_helper_connector_hdmi_disable_audio_infoframe(), an API
to allow the driver disable sending the Audio Infoframe. This is to be
used by the drivers if setup of the infoframes is not tightly coupled
with the audio functionality and just disabling the audio playback
doesn't stop the HDMI hardware from sending the Infoframe.
Acked-by: Maxime Ripard <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Signed-off-by: Dmitry Baryshkov <[email protected]>
|
|
Based on grepping through the source code this driver appears to be
missing a call to drm_atomic_helper_shutdown() at system shutdown
time. Among other things, this means that if a panel is in use that it
won't be cleanly powered off at system shutdown time.
The fact that we should call drm_atomic_helper_shutdown() in the case
of OS shutdown/restart comes straight out of the kernel doc "driver
instance overview" in drm_drv.c.
This driver users the component model and shutdown happens in the base
driver. The "drvdata" for this driver will always be valid if
shutdown() is called and as of commit 2a073968289d
("drm/atomic-helper: drm_atomic_helper_shutdown(NULL) should be a
noop") we don't need to confirm that "drm" is non-NULL.
Suggested-by: Maxime Ripard <[email protected]>
Reviewed-by: Maxime Ripard <[email protected]>
Reviewed-by: Fei Shao <[email protected]>
Tested-by: Fei Shao <[email protected]>
Signed-off-by: Douglas Anderson <[email protected]>
Signed-off-by: Maxime Ripard <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/20240611102744.v2.1.I2b014f90afc4729b6ecc7b5ddd1f6dedcea4625b@changeid
|
|
Based on grepping through the source code, this driver appears to be
missing a call to drm_atomic_helper_shutdown() at system shutdown time.
This is important because drm_atomic_helper_shutdown() will cause
panels to get disabled cleanly which may be important for their power
sequencing. Future changes will remove any custom powering off in
individual panel drivers so the DRM drivers need to start getting this
right.
The fact that we should call drm_atomic_helper_shutdown() in the case of
OS shutdown comes straight out of the kernel doc "driver instance
overview" in drm_drv.c.
[geert: shmob_drm_remove() already calls drm_atomic_helper_shutdown]
Suggested-by: Maxime Ripard <[email protected]>
Signed-off-by: Douglas Anderson <[email protected]>
Link: https://lore.kernel.org/r/20230901164111.RFT.15.Iaf638a1d4c8b3c307a6192efabb4cbb06b195f15@changeid
[geert: s/drm_helper_force_disable_all/drm_atomic_helper_shutdown/]
Signed-off-by: Geert Uytterhoeven <[email protected]>
Reviewed-by: Laurent Pinchart <[email protected]>
Reviewed-by: Sui Jingfeng <[email protected]>
Signed-off-by: Maxime Ripard <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/17c6a5a668e5975f871b77fb1fca6711a0799d9e.1718176895.git.geert+renesas@glider.be
|
|
When multiple streams are in use, multiple TDs might be in flight when
an endpoint is stopped. We need to issue a Set TR Dequeue Pointer for
each, to ensure everything is reset properly and the caches cleared.
Change the logic so that any N>1 TDs found active for different streams
are deferred until after the first one is processed, calling
xhci_invalidate_cancelled_tds() again from xhci_handle_cmd_set_deq() to
queue another command until we are done with all of them. Also change
the error/"should never happen" paths to ensure we at least clear any
affected TDs, even if we can't issue a command to clear the hardware
cache, and complain loudly with an xhci_warn() if this ever happens.
This problem case dates back to commit e9df17eb1408 ("USB: xhci: Correct
assumptions about number of rings per endpoint.") early on in the XHCI
driver's life, when stream support was first added.
It was then identified but not fixed nor made into a warning in commit
674f8438c121 ("xhci: split handling halted endpoints into two steps"),
which added a FIXME comment for the problem case (without materially
changing the behavior as far as I can tell, though the new logic made
the problem more obvious).
Then later, in commit 94f339147fc3 ("xhci: Fix failure to give back some
cached cancelled URBs."), it was acknowledged again.
[Mathias: commit 94f339147fc3 ("xhci: Fix failure to give back some cached
cancelled URBs.") was a targeted regression fix to the previously mentioned
patch. Users reported issues with usb stuck after unmounting/disconnecting
UAS devices. This rolled back the TD clearing of multiple streams to its
original state.]
Apparently the commit author was aware of the problem (yet still chose
to submit it): It was still mentioned as a FIXME, an xhci_dbg() was
added to log the problem condition, and the remaining issue was mentioned
in the commit description. The choice of making the log type xhci_dbg()
for what is, at this point, a completely unhandled and known broken
condition is puzzling and unfortunate, as it guarantees that no actual
users would see the log in production, thereby making it nigh
undebuggable (indeed, even if you turn on DEBUG, the message doesn't
really hint at there being a problem at all).
It took me *months* of random xHC crashes to finally find a reliable
repro and be able to do a deep dive debug session, which could all have
been avoided had this unhandled, broken condition been actually reported
with a warning, as it should have been as a bug intentionally left in
unfixed (never mind that it shouldn't have been left in at all).
> Another fix to solve clearing the caches of all stream rings with
> cancelled TDs is needed, but not as urgent.
3 years after that statement and 14 years after the original bug was
introduced, I think it's finally time to fix it. And maybe next time
let's not leave bugs unfixed (that are actually worse than the original
bug), and let's actually get people to review kernel commits please.
Fixes xHC crashes and IOMMU faults with UAS devices when handling
errors/faults. Easiest repro is to use `hdparm` to mark an early sector
(e.g. 1024) on a disk as bad, then `cat /dev/sdX > /dev/null` in a loop.
At least in the case of JMicron controllers, the read errors end up
having to cancel two TDs (for two queued requests to different streams)
and the one that didn't get cleared properly ends up faulting the xHC
entirely when it tries to access DMA pages that have since been unmapped,
referred to by the stale TDs. This normally happens quickly (after two
or three loops). After this fix, I left the `cat` in a loop running
overnight and experienced no xHC failures, with all read errors
recovered properly. Repro'd and tested on an Apple M1 Mac Mini
(dwc3 host).
On systems without an IOMMU, this bug would instead silently corrupt
freed memory, making this a security bug (even on systems with IOMMUs
this could silently corrupt memory belonging to other USB devices on the
same controller, so it's still a security bug). Given that the kernel
autoprobes partition tables, I'm pretty sure a malicious USB device
pretending to be a UAS device and reporting an error with the right
timing could deliberately trigger a UAF and write to freed memory, with
no user action.
[Mathias: Commit message and code comment edit, original at:]
https://lore.kernel.org/linux-usb/[email protected]/
Fixes: e9df17eb1408 ("USB: xhci: Correct assumptions about number of rings per endpoint.")
Fixes: 94f339147fc3 ("xhci: Fix failure to give back some cached cancelled URBs.")
Fixes: 674f8438c121 ("xhci: split handling halted endpoints into two steps")
Cc: [email protected]
Cc: [email protected]
Reviewed-by: Neal Gompa <[email protected]>
Signed-off-by: Hector Martin <[email protected]>
Signed-off-by: Mathias Nyman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
As described in commit 8f873c1ff4ca ("xhci: Blacklist using streams on the
Etron EJ168 controller"), EJ188 have the same issue as EJ168, where Streams
do not work reliable on EJ188. So apply XHCI_BROKEN_STREAMS quirk to EJ188
as well.
Cc: [email protected]
Signed-off-by: Kuangyi Chiang <[email protected]>
Signed-off-by: Mathias Nyman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
As described in commit c877b3b2ad5c ("xhci: Add reset on resume quirk for
asrock p67 host"), EJ188 have the same issue as EJ168, where completely
dies on resume. So apply XHCI_RESET_ON_RESUME quirk to EJ188 as well.
Cc: [email protected]
Signed-off-by: Kuangyi Chiang <[email protected]>
Signed-off-by: Mathias Nyman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
The transferred length is set incorrectly for cancelled bulk
transfer TDs in case the bulk transfer ring stops on the last transfer
block with a 'Stop - Length Invalid' completion code.
length essentially ends up being set to the requested length:
urb->actual_length = urb->transfer_buffer_length
Length for 'Stop - Length Invalid' cases should be the sum of all
TRB transfer block lengths up to the one the ring stopped on,
_excluding_ the one stopped on.
Fix this by always summing up TRB lengths for 'Stop - Length Invalid'
bulk cases.
This issue was discovered by Alan Stern while debugging
https://bugzilla.kernel.org/show_bug.cgi?id=218890, but does not
solve that bug. Issue is older than 4.10 kernel but fix won't apply
to those due to major reworks in that area.
Tested-by: Pierre Tomon <[email protected]>
Cc: [email protected] # v4.10+
Cc: Alan Stern <[email protected]>
Signed-off-by: Mathias Nyman <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
|
|
This adds the power sequencing driver for the PMU modules present on the
Qualcomm WCN Bluetooth and Wifi chipsets. It uses the pwrseq subsystem
and knows how to match the sequencer to the consumer device by verifying
the relevant properties and DT layout. Using this driver will allow the
BT and WLAN drivers to respect the required delays between enabling the
two modules.
Tested-by: Amit Pundir <[email protected]>
Tested-by: Neil Armstrong <[email protected]> # on SM8550-QRD, SM8650-QRD & SM8650-HDK
Tested-by: Caleb Connolly <[email protected]> # OnePlus 8T
Reviewed-by: Krzysztof Kozlowski <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Bartosz Golaszewski <[email protected]>
|
|
Implement the power sequencing subsystem allowing devices to share
complex powering-up and down procedures. It's split into the consumer
and provider parts but does not implement any new DT bindings so that
the actual power sequencing is never revealed in the DT representation.
Tested-by: Amit Pundir <[email protected]>
Tested-by: Neil Armstrong <[email protected]> # on SM8550-QRD, SM8650-QRD & SM8650-HDK
Tested-by: Caleb Connolly <[email protected]> # OnePlus 8T
Acked-by: Krzysztof Kozlowski <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Bartosz Golaszewski <[email protected]>
|
|
parameters
The current cbs parameter depends on speed after uplinking,
which is not needed and will report a configuration error
if the port is not initially connected. The UAPI exposed by
tc-cbs requires userspace to recalculate the send slope anyway,
because the formula depends on port_transmit_rate (see man tc-cbs),
which is not an invariant from tc's perspective. Therefore, we
use offload->sendslope and offload->idleslope to derive the
original port_transmit_rate from the CBS formula.
Fixes: 1f705bc61aee ("net: stmmac: Add support for CBS QDISC")
Signed-off-by: Xiaolei Wang <[email protected]>
Reviewed-by: Wojciech Drewek <[email protected]>
Reviewed-by: Vladimir Oltean <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>
|