usb: dwc3: gadget: when the started list is empty stop the active xfer
When we have nothing left to be queued after handling the last trb
we have to stop the current transfer. This way we can ensure that
the next request will be queued with a new and valid timestamp
and will not directly run into an missed xfer.
Reviewed-by: Thinh Nguyen <thinhn@synopsys.com> Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Michael Olbrich [Wed, 1 Jul 2020 18:24:52 +0000 (20:24 +0200)]
usb: dwc3: gadget: make starting isoc transfers more robust
Currently __dwc3_gadget_start_isoc must be called very shortly after
XferNotReady. Otherwise the frame number is outdated and start transfer
will fail, even with several retries.
DSTS provides the lower 14 bit of the frame number. Use it in combination
with the frame number provided by XferNotReady to guess the current frame
number. This will succeed unless more than one 14 rollover has happened
since XferNotReady.
Start transfer might still fail if the frame number is increased
immediately after DSTS is read. So retries are still needed.
Don't drop the current request if this happens. This way it is not lost and
can be used immediately to try again with the next frame number.
With this change, __dwc3_gadget_start_isoc is still not successfully in all
cases bit it increases the acceptable delay after XferNotReady
significantly.
Reviewed-by: Thinh Nguyen <thinhn@synopsys.com> Signed-off-by: Michael Olbrich <m.olbrich@pengutronix.de> Signed-off-by: Michael Tretter <m.tretter@pengutronix.de> Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Zqiang [Tue, 30 Jun 2020 05:23:31 +0000 (13:23 +0800)]
usb: gadget: function: printer: Interface is disabled and returns error
After the device is disconnected from the host side, the interface of
the device is reset. If the userspace operates the device again,
an error code should be returned.
Signed-off-by: Zqiang <qiang.zhang@windriver.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Ruslan Bilovol [Fri, 3 Jul 2020 13:49:03 +0000 (16:49 +0300)]
usb: gadget: f_uac2: fix AC Interface Header Descriptor wTotalLength
As per UAC2 spec (ch. 4.7.2), wTotalLength of AC Interface
Header Descriptor "includes the combined length of this
descriptor header and all Clock Source, Unit and Terminal
descriptors."
Thus add its size to its wTotalLength.
Also after recent changes wTotalLength is calculated
dynamically, update static definition of uac2_ac_header_descriptor
accordingly
Fixes: 132fcb460839 ("usb: gadget: Add Audio Class 2.0 Driver") Signed-off-by: Ruslan Bilovol <ruslan.bilovol@gmail.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
There were some review comments after the patch was integrated.
Address those.
Fixes: 1883a934e156 ("dt-bindings: usb: convert keystone-usb.txt to YAML") Reviewed-by: Rob Herring <robh@kernel.org> Signed-off-by: Roger Quadros <rogerq@ti.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
The BDC clock is optional and we may get an -EPROBE_DEFER error code
which would not be propagated correctly, fix this by using
devm_clk_get_optional().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Al Cooper <alcooperx@gmail.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
GISB bus error kernel panics have been observed during S2 transition
tests on the 7271t platform. The errors are a result of the BDC
interrupt handler trying to access BDC register space after the
system's suspend callbacks have completed.
Adding a suspend hook to the BDC driver that halts the controller before
S2 entry thus preventing unwanted access to the BDC register space during
this transition.
Signed-off-by: Danesh Petigara <danesh.petigara@broadcom.com> Signed-off-by: Al Cooper <alcooperx@gmail.com> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Al Cooper [Wed, 22 Jul 2020 17:07:44 +0000 (13:07 -0400)]
usb: bdc: driver runs out of buffer descriptors on large ADB transfers
Version v1.0.40 of the Android host ADB software increased maximum
transfer sizes from 256K to 1M. Since the STB ADB gadget driver
requests only 16K at a time, the BDC driver ran out of buffer
descriptors (BDs) if the queuing happens faster than the incoming
16K transfers. This issue is fixed by doubling the number of BDs
that can be queued so that the entire 1M request can be queued
without running out of buffers.
Signed-off-by: Al Cooper <alcooperx@gmail.com> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Al Cooper [Wed, 22 Jul 2020 17:07:43 +0000 (13:07 -0400)]
usb: bdc: Adb shows offline after resuming from S2
On Android systems, After temporarily putting device to S2 by
short pressing the power button on the remote, the display turns
off. Then press the power button to turn the display back up. Adb
devices would show the devices is offline. It needs a physical
disconnect of the usb cable or power cycle to bring the device
back online. The device is operational otherwise.
The problem is that during S2 resume, the ADB gadget driver could
not link back with the BDC driver because the endpoint flags were
cleared. The fix is to clear the endpoint flags for the disconnect
case only and not for S2 exit.
Signed-off-by: Al Cooper <alcooperx@gmail.com> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Sasi Kumar [Wed, 22 Jul 2020 17:07:42 +0000 (13:07 -0400)]
bdc: Fix bug causing crash after multiple disconnects
Multiple connects/disconnects can cause a crash on the second
disconnect. The driver had a problem where it would try to send
endpoint commands after it was disconnected which is not allowed
by the hardware. The fix is to only allow the endpoint commands
when the endpoint is connected. This will also fix issues that
showed up when using configfs to create gadgets.
Signed-off-by: Sasi Kumar <sasi.kumar@broadcom.com> Signed-off-by: Al Cooper <alcooperx@gmail.com> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Al Cooper [Wed, 22 Jul 2020 17:07:40 +0000 (13:07 -0400)]
dt-bindings: usb: bdc: Update compatible strings
Remove "brcm,bdc-v0.16" because it was never used on any system.
Add "brcm,bdc-udc-v2" which exists for any STB system with BDC.
Acked-by: Rob Herring <robh@kernel.org> Signed-off-by: Al Cooper <alcooperx@gmail.com> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
USB: PHY: JZ4770: Unify code style and simplify code.
1.Modify the macro definition to unify "#define USBPCR_XXXX n"
into the "#define USBPCR_XXXX (n << USBPCR_XXXX_LSB)" style,
so as to unify the code style in the "jz4770_phy_init()" and
simplify the code.
2.Remove unused macro definitions to simplify the code.
Tested-by: 周正 (Zhou Zheng) <sernia.zhou@foxmail.com> Suggested-by: Paul Cercueil <paul@crapouillou.net> Signed-off-by: 周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Peter Chen [Wed, 22 Jul 2020 03:06:19 +0000 (11:06 +0800)]
usb: cdns3: gadget: always zeroed TRB buffer when enable endpoint
During the endpoint dequeue operation, it changes dequeued TRB as link
TRB, when the endpoint is disabled and re-enabled, the DMA fetches the
TRB before the link TRB, after it handles current TRB, the DMA pointer
will advance to the TRB after link TRB, but enqueue and dequene
variables don't know it due to no hardware interrupt at the time, when
the next TRB is added to link TRB position, the DMA will not handle
this TRB due to its pointer is already at the next TRB. See the trace
log like below:
Marek Szyprowski [Thu, 16 Jul 2020 12:09:48 +0000 (14:09 +0200)]
usb: dwc2: Fix error path in gadget registration
When gadget registration fails, one should not call usb_del_gadget_udc().
Ensure this by setting gadget->udc to NULL. Also in case of a failure
there is no need to disable low-level hardware, so return immiedetly
instead of jumping to error_init label.
This fixes the following kernel NULL ptr dereference on gadget failure
(can be easily triggered with g_mass_storage without any module
parameters):
dwc2 12480000.hsotg: dwc2_check_params: Invalid parameter besl=1
dwc2 12480000.hsotg: dwc2_check_params: Invalid parameter g_np_tx_fifo_size=1024
dwc2 12480000.hsotg: EPs: 16, dedicated fifos, 7808 entries in SPRAM
Mass Storage Function, version: 2009/09/11
LUN: removable file: (no medium)
no file given for LUN0
g_mass_storage 12480000.hsotg: failed to start g_mass_storage: -22
8<--- cut here ---
Unable to handle kernel NULL pointer dereference at virtual address 00000104
pgd = (ptrval)
[00000104] *pgd=00000000
Internal error: Oops: 805 [#1] PREEMPT SMP ARM
Modules linked in:
CPU: 0 PID: 12 Comm: kworker/0:1 Not tainted 5.8.0-rc5 #3133
Hardware name: Samsung Exynos (Flattened Device Tree)
Workqueue: events deferred_probe_work_func
PC is at usb_del_gadget_udc+0x38/0xc4
LR is at __mutex_lock+0x31c/0xb18
...
Process kworker/0:1 (pid: 12, stack limit = 0x(ptrval))
Stack: (0xef121db0 to 0xef122000)
...
[<c076bf3c>] (usb_del_gadget_udc) from [<c0726bec>] (dwc2_hsotg_remove+0x10/0x20)
[<c0726bec>] (dwc2_hsotg_remove) from [<c0711208>] (dwc2_driver_probe+0x57c/0x69c)
[<c0711208>] (dwc2_driver_probe) from [<c06247c0>] (platform_drv_probe+0x6c/0xa4)
[<c06247c0>] (platform_drv_probe) from [<c0621df4>] (really_probe+0x200/0x48c)
[<c0621df4>] (really_probe) from [<c06221e8>] (driver_probe_device+0x78/0x1fc)
[<c06221e8>] (driver_probe_device) from [<c061fcd4>] (bus_for_each_drv+0x74/0xb8)
[<c061fcd4>] (bus_for_each_drv) from [<c0621b54>] (__device_attach+0xd4/0x16c)
[<c0621b54>] (__device_attach) from [<c0620c98>] (bus_probe_device+0x88/0x90)
[<c0620c98>] (bus_probe_device) from [<c06211b0>] (deferred_probe_work_func+0x3c/0xd0)
[<c06211b0>] (deferred_probe_work_func) from [<c0149280>] (process_one_work+0x234/0x7dc)
[<c0149280>] (process_one_work) from [<c014986c>] (worker_thread+0x44/0x51c)
[<c014986c>] (worker_thread) from [<c0150b1c>] (kthread+0x158/0x1a0)
[<c0150b1c>] (kthread) from [<c0100114>] (ret_from_fork+0x14/0x20)
Exception stack(0xef121fb0 to 0xef121ff8)
...
---[ end trace 9724c2fc7cc9c982 ]---
While fixing this also fix the double call to dwc2_lowlevel_hw_disable()
if dr_mode is set to USB_DR_MODE_PERIPHERAL. In such case low-level
hardware is already disabled before calling usb_add_gadget_udc(). That
function correctly preserves low-level hardware state, there is no need
for the second unconditional dwc2_lowlevel_hw_disable() call.
Fixes: 207324a321a8 ("usb: dwc2: Postponed gadget registration to the udc class driver") Acked-by: Minas Harutyunyan <hminas@synopsys.com> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Implement udc->pullup callback, so that udc_connect/disconnect work.
This is needed for composite gadget, as it assumes udc_disconnect()
actually works and calls its ->disconnect callback.
Acked-by: Cristian Birsan <cristian.birsan@microchip.com> Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com> Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Marek Szyprowski [Thu, 16 Jul 2020 11:55:25 +0000 (13:55 +0200)]
usb: gadget: udc: Flush pending work also in error path
When binding an UDC driver to the pending gadget fails in
check_pending_gadget_drivers(), the usb_add_gadget_udc_release() function
ends without waiting for the usb_gadget_state_work to finish, what in
turn might cause the whole struct usb_gadget being freed by the caller
before the usb_gadget_state_work being executed.
This can be observed on some boards with USB Mass Storage gadget
compiled-in and kernel booted without the needed module parameters:
dwc2 12480000.hsotg: dwc2_check_params: Invalid parameter besl=1
dwc2 12480000.hsotg: dwc2_check_params: Invalid parameter g_np_tx_fifo_size=1024
dwc2 12480000.hsotg: EPs: 16, dedicated fifos, 7808 entries in SPRAM
Mass Storage Function, version: 2009/09/11
LUN: removable file: (no medium)
no file given for LUN0
g_mass_storage 12480000.hsotg: failed to start g_mass_storage: -22
dwc2: probe of 12480000.hsotg failed with error -22
8<--- cut here ---
Unable to handle kernel NULL pointer dereference at virtual address 00000004
pgd = (ptrval)
[00000004] *pgd=00000000
Internal error: Oops: 5 [#1] PREEMPT SMP ARM
Modules linked in:
CPU: 1 PID: 88 Comm: kworker/1:2 Not tainted 5.8.0-rc5-next-20200715-00062-gc5bb489ae825-dirty #8792
Hardware name: Samsung Exynos (Flattened Device Tree)
Workqueue: 0x0 (rcu_gp)
PC is at process_one_work+0x44/0x7dc
...
Process kworker/1:2 (pid: 88, stack limit = 0x(ptrval))
Stack: (0xed9f1f00 to 0xed9f2000)
...
[<c0148590>] (process_one_work) from [<c0148d6c>] (worker_thread+0x44/0x51c)
[<c0148d6c>] (worker_thread) from [<c01500c0>] (kthread+0x158/0x1a0)
[<c01500c0>] (kthread) from [<c0100114>] (ret_from_fork+0x14/0x20)
Exception stack(0xed9f1fb0 to 0xed9f1ff8)
...
---[ end trace 5033c1326a62e5f3 ]---
note: kworker/1:2[88] exited with preempt_count 1
Fix this by flushing pending work in error path.
Reviewed-by: Peter Chen <peter.chen@nxp.com> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Peter Chen [Fri, 17 Jul 2020 10:13:17 +0000 (18:13 +0800)]
usb: cdns3: gadget: improve the set_configuration handling
- Delete the duplicated EP_CMD_ERDY and EP_CMD_REQ_CMPL setting
- Prepare the next setup before setting EP_CMD_ERDY and EP_CMD_REQ_CMPL,
it could avoid a bug that DMA hang at EP0 OUT for DEV_VER_NXP_V1
- Delete the duplicated cdns3_set_hw_configuration calling at
cdns3_req_ep0_set_configuration, the composite.c will handle
this request, and call .ep0_queue back, and at .ep_queue it will
call cdns3_set_hw_configuration.
- Move cdns3_allow_enable_l1 into cdns3_set_hw_configuration since
it is part of this function.
Signed-off-by: Peter Chen <peter.chen@nxp.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Dan Robertson [Mon, 13 Jul 2020 16:05:22 +0000 (12:05 -0400)]
usb: dwc3: meson-g12a: fix shared reset control use
The reset is a shared reset line, but reset_control_reset is still used
and reset_control_deassert is not guaranteed to have been called before
the first reset_control_assert call. When suspending the following
warning may be seen:
Fixes: 6d9fa35a347a87 ("usb: dwc3: meson-g12a: get the reset as shared") Signed-off-by: Dan Robertson <dan@dlrobertson.com> Signed-off-by: Felipe Balbi <balbi@kernel.org>
Amelie Delaunay [Tue, 16 Jun 2020 14:07:16 +0000 (16:07 +0200)]
usb: dwc2: don't use ID/Vbus detection if usb-role-switch on STM32MP15 SoCs
If usb-role-switch is present in the device tree, it means that ID and Vbus
signals are not connected to the OTG controller but to an external
component (GPIOs, Type-C controller). In this configuration, usb role
switch is used to force valid sessions on STM32MP15 SoCs.
Amelie Delaunay [Tue, 16 Jun 2020 14:07:15 +0000 (16:07 +0200)]
usb: dwc2: override PHY input signals with usb role switch support
This patch adds support for usb role switch to dwc2, by using overriding
control of the PHY voltage valid and ID input signals.
iddig signal (ID) can be overridden:
- when setting GUSBCFG_FORCEHOSTMODE, iddig input pin is overridden with 1;
- when setting GUSBCFG_FORCEDEVMODE, iddig input pin is overridden with 0.
avalid/bvalid/vbusvalid signals can be overridden respectively with:
- GOTGCTL_AVALOEN + GOTGCTL_AVALOVAL
- GOTGCTL_BVALOEN + GOTGCTL_BVALOVAL
- GOTGCTL_VBVALEN + GOTGCTL_VBVALOVAL
It is possible to determine valid sessions thanks to usb role switch:
- if USB_ROLE_NONE then !avalid && !bvalid && !vbusvalid
- if USB_ROLE_DEVICE then !avalid && bvalid && vbusvalid
- if USB_ROLE_HOST then avalid && !bvalid && vbusvalid
Merge tag 'thunderbolt-for-v5.9' of git://git.kernel.org/pub/scm/linux/kernel/git/westeri/thunderbolt into usb-next
Mika writes:
thunderbolt: Changes for v5.9 merge window
This includes following Thunderbolt/USB4 changes for v5.9 merge window:
* Improvements around NHI (Native Host Interface) HopID allocation
* Improvements to tunneling and USB3 bandwidth management support
* Add KUnit tests for path walking and tunneling
* Initial support for USB4 retimer firmware upgrade
* Implement Thunderbolt device firmware upgrade mechanism that runs
the NVM image authentication when the device is disconnected.
* A couple of small non-critical fixes
* tag 'thunderbolt-for-v5.9' of git://git.kernel.org/pub/scm/linux/kernel/git/westeri/thunderbolt: (32 commits)
thunderbolt: Fix old style declaration warning
thunderbolt: Add support for authenticate on disconnect
thunderbolt: Add support for separating the flush to SPI and authenticate
thunderbolt: Ensure left shift of 512 does not overflow a 32 bit int
thunderbolt: Add support for on-board retimers
thunderbolt: Implement USB4 port sideband operations for retimer access
thunderbolt: Retry USB4 block read operation
thunderbolt: Generalize usb4_switch_do_[read|write]_data()
thunderbolt: Split common NVM functionality into a separate file
thunderbolt: Add Intel USB-IF ID to the NVM upgrade supported list
thunderbolt: Add KUnit tests for tunneling
thunderbolt: Add USB3 bandwidth management
thunderbolt: Make tb_port_get_link_speed() available to other files
thunderbolt: Implement USB3 bandwidth negotiation routines
thunderbolt: Increase DP DPRX wait timeout
thunderbolt: Report consumed bandwidth in both directions
thunderbolt: Make usb4_switch_map_pcie_down() also return enabled ports
thunderbolt: Make usb4_switch_map_usb3_down() also return enabled ports
thunderbolt: Do not tunnel USB3 if link is not USB4
thunderbolt: Add DP IN resources for all routers
...
The OMAP1 OHCI driver is using the legacy GPIO API to grab some
random GPIO lines. One is from the TPS65010 chip and used for
power, another one is for overcurrent and while the driver picks
this line it doesn't watch it at all.
Convert the driver and the OMAP1 OSK board file to pass these
two GPIOs as machine described GPIO descriptors.
I noticed the overcurrent GPIO line is not really used in the
code so dropped in a little comment for other developers.
Cc: Janusz Krzysztofik <jmkrzyszt@gmail.com> Cc: Tony Lindgren <tony@atomide.com> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Link: https://lore.kernel.org/r/20200720135524.100374-2-linus.walleij@linaro.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The OMAP1 was using static locals to hold the clock handles
which is uncommon and does not scale. Create a private data
struct and use that to hold the clocks.
Cc: Janusz Krzysztofik <jmkrzyszt@gmail.com> Cc: Tony Lindgren <tony@atomide.com> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Link: https://lore.kernel.org/r/20200720135524.100374-1-linus.walleij@linaro.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Rationale:
Reduces attack surface on kernel devs opening the links for MITM
as HTTPS traffic is much harder to manipulate.
Deterministic algorithm:
For each file:
If not .svg:
For each line:
If doesn't contain `\bxmlns\b`:
For each link, `\bhttp://[^# \t\r\n]*(?:\w|/)`:
If neither `\bgnu\.org/license`, nor `\bmozilla\.org/MPL\b`:
If both the HTTP and HTTPS versions
return 200 OK and serve the same content:
Replace HTTP with HTTPS.
Bixuan Cui [Fri, 17 Jul 2020 00:22:53 +0000 (08:22 +0800)]
usb: usbtest: reduce stack usage in test_queue
Fix the warning: [-Werror=-Wframe-larger-than=]
drivers/usb/misc/usbtest.c: In function 'test_queue':
drivers/usb/misc/usbtest.c:2148:1:
warning: the frame size of 1232 bytes is larger than 1024 bytes
usb: typec: tcpm: Stay in BIST mode till hardreset or unattached
Port starts to toggle when transitioning to unattached state.
This is incorrect while in BIST mode.
6.4.3.1 BIST Carrier Mode
Upon receipt of a BIST Message, with a BIST Carrier Mode BIST Data Object,
the UUT Shall send out a continuous string of BMC encoded alternating "1"s
and “0”s. The UUT Shall exit the Continuous BIST Mode within
tBISTContMode of this Continuous BIST Mode being enabled(see
Section 6.6.7.2).
6.4.3.2 BIST Test Data
Upon receipt of a BIST Message, with a BIST Test Data BIST Data Object,
the UUT Shall return a GoodCRC Message and Shall enter a test mode in which
it sends no further Messages except for GoodCRC Messages in response to
received Messages. See Section 5.9.2 for the definition of the Test Data
Frame. The test Shall be ended by sending Hard Reset Signaling to reset the
UUT.
usb: typec: tcpci: Support BIST test data mode for compliance.
Quoting from TCPCI spec:
"Setting this bit to 1 is intended to be used only when a USB compliance
tester is using USB BIST Test Data to test the PHY layer of the TCPC. The
TCPM should clear this bit when a disconnect is detected.
0: Normal Operation. Incoming messages enabled by RECEIVE_DETECT
passed to TCPM via Alert.
1: BIST Test Mode. Incoming messages enabled by RECEIVE_DETECT
result in GoodCRC response but may not be passed to the TCPM via
Alert."
usb: typec: intel_pmc_mux: Avoid connect request on disconnect
When pmc_usb_mux_set() is invoked when a device is disconnected, a valid
scenario is for state->alt == NULL and state->mode == TYPEC_STATE_USB.
In such cases, if a pmc_usb_disconnect() has already been issued (from
either pmc_usb_set_orientation() when orientation ==
TYPEC_ORIENTATION_NONE, or pmc_usb_set_role() when role ==
USB_ROLE_NONE), a pmc_usb_connect() will be issued despite no peripheral
being present.
This confuses the PMC and leads to all subsequent PMC IPC requests
returning errors due to timeout.
To prevent this, return early if the port orientation or role is already
set to none.
usb: core: fix quirks_param_set() writing to a const pointer
The function quirks_param_set() takes as argument a const char* pointer
to the new value of the usbcore.quirks parameter. It then casts this
pointer to a non-const char* pointer and passes it to the strsep()
function, which overwrites the value.
Fix this by creating a copy of the value using kstrdup() and letting
that copy be written to by strsep().
Andy Shevchenko [Thu, 2 Jul 2020 14:30:45 +0000 (17:30 +0300)]
usb: hcd: Try MSI interrupts on PCI devices
It appears that some platforms share same IRQ line between several devices,
some of which are EHCI and OHCI controllers. This is neither practical nor
performance-wise, especially in the case when they are supporting MSI.
In order to improve the situation try to allocate MSI and fallback to legacy
IRQ if no MSI available.
Merge tag 'perf-tools-fixes-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into master
Pull perf tooling fixes from Arnaldo Carvalho de Melo:
- Update hashmap.h from libbpf and kvm.h from x86's kernel UAPI.
- Set opt->set in libsubcmd's OPT_CALLBACK_SET(). This fixes
'perf record --switch-output-event event-name' usage"
* tag 'perf-tools-fixes-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
tools arch kvm: Sync kvm headers with the kernel sources
perf tools: Sync hashmap.h with libbpf's
libsubcmd: Fix OPT_CALLBACK_SET()
Merge tag 'x86-urgent-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into master
Pull x86 fixes from Thomas Gleixner:
"A pile of fixes for x86:
- Fix the I/O bitmap invalidation on XEN PV, which was overlooked in
the recent ioperm/iopl rework. This caused the TSS and XEN's I/O
bitmap to get out of sync.
- Use the proper vectors for HYPERV.
- Make disabling of stack protector for the entry code work with GCC
builds which enable stack protector by default. Removing the option
is not sufficient, it needs an explicit -fno-stack-protector to
shut it off.
- Mark check_user_regs() noinstr as it is called from noinstr code.
The missing annotation causes it to be placed in the text section
which makes it instrumentable.
- Add the missing interrupt disable in exc_alignment_check()
- Fixup a XEN_PV build dependency in the 32bit entry code
- A few fixes to make the Clang integrated assembler happy
- Move EFI stub build to the right place for out of tree builds
- Make prepare_exit_to_usermode() static. It's not longer called from
ASM code"
* tag 'x86-urgent-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/boot: Don't add the EFI stub to targets
x86/entry: Actually disable stack protector
x86/ioperm: Fix io bitmap invalidation on Xen PV
x86: math-emu: Fix up 'cmp' insn for clang ias
x86/entry: Fix vectors to IDTENTRY_SYSVEC for CONFIG_HYPERV
x86/entry: Add compatibility with IAS
x86/entry/common: Make prepare_exit_to_usermode() static
x86/entry: Mark check_user_regs() noinstr
x86/traps: Disable interrupts in exc_aligment_check()
x86/entry/32: Fix XEN_PV build dependency
Merge tag 'timers-urgent-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into master
Pull timer fixes from Thomas Gleixner:
"Two fixes for the timer wheel:
- A timer which is already expired at enqueue time can set the
base->next_expiry value backwards. As a consequence base->clk can
be set back as well. This can lead to timers expiring early. Add a
sanity check to prevent this.
- When a timer is queued with an expiry time beyond the wheel
capacity then it should be queued in the bucket of the last wheel
level which is expiring last.
The code adjusted the expiry time to the maximum wheel capacity,
which is only correct when the wheel clock is 0. Aside of that the
check whether the delta is larger than wheel capacity does not
check the delta, it checks the expiry value itself. As a result
timers can expire at random.
Fix this by checking the right variable and adjust expiry time so
it becomes base->clock plus capacity which places it into the
outmost bucket in the last wheel level"
* tag 'timers-urgent-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
timer: Fix wheel index calculation on last level
timer: Prevent base->clk from moving backward
Merge tag 'sched-urgent-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into master
Pull scheduler fixes from Thomas Gleixner:
"A set of scheduler fixes:
- Plug a load average accounting race which was introduced with a
recent optimization casing load average to show bogus numbers.
- Fix the rseq CPU id initialization for new tasks. sched_fork() does
not update the rseq CPU id so the id is the stale id of the parent
task, which can cause user space data corruption.
- Handle a 0 return value of task_h_load() correctly in the load
balancer, which does not decrease imbalance and therefore pulls
until the maximum number of loops is reached, which might be all
tasks just created by a fork bomb"
* tag 'sched-urgent-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: handle case of task_h_load() returning 0
sched: Fix unreliable rseq cpu_id for new tasks
sched: Fix loadavg accounting race
Merge tag 'irq-urgent-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip into master
Pull irq fixes from Thomas Gleixner:
"Two fixes for the interrupt subsystem:
- Make the handling of the firmware node consistent and do not free
the node after the domain has been created successfully. The core
code stores a pointer to it which can lead to a use after free or
double free.
This used to "work" because the pointer was not stored when the
initial code was written, but at some point later it was required
to store it. Of course nobody noticed that the existing users break
that way.
- Handle affinity setting on inactive interrupts correctly when
hierarchical irq domains are enabled.
When interrupts are inactive with the modern hierarchical irqdomain
design, the interrupt chips are not necessarily in a state where
affinity changes can be handled. The legacy irq chip design allowed
this because interrupts are immediately fully initialized at
allocation time. X86 has a hacky workaround for this, but other
implementations do not.
This cased malfunction on GIC-V3. Instead of playing whack a mole
to find all affected drivers, change the core code to store the
requested affinity setting and then establish it when the interrupt
is allocated, which makes the X86 hack go away"
* tag 'irq-urgent-2020-07-19' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
genirq/affinity: Handle affinity setting on inactive interrupts correctly
irqdomain/treewide: Keep firmware node unconditionally allocated
Merge tag 'usb-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb into master
Pull USB fixes from Greg KH:
"Here are a few small USB fixes, and one thunderbolt fix, for 5.8-rc6.
Nothing huge in here, just the normal collection of gadget, dwc2/3,
serial, and other minor USB driver fixes and id additions. Full
details are in the shortlog.
All of these have been in linux-next for a while with no reported
issues"
* tag 'usb-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
USB: serial: iuu_phoenix: fix memory corruption
USB: c67x00: fix use after free in c67x00_giveback_urb
usb: gadget: function: fix missing spinlock in f_uac1_legacy
usb: gadget: udc: atmel: fix uninitialized read in debug printk
usb: gadget: udc: atmel: remove outdated comment in usba_ep_disable()
usb: dwc2: Fix shutdown callback in platform
usb: cdns3: trace: fix some endian issues
usb: cdns3: ep0: fix some endian issues
usb: gadget: udc: gr_udc: fix memleak on error handling path in gr_ep_init()
usb: gadget: fix langid kernel-doc warning in usbstring.c
usb: dwc3: pci: add support for the Intel Jasper Lake
usb: dwc3: pci: add support for the Intel Tiger Lake PCH -H variant
usb: chipidea: core: add wakeup support for extcon
USB: serial: option: add Quectel EG95 LTE modem
thunderbolt: Fix path indices used in USB3 tunnel discovery
USB: serial: ch341: add new Product ID for CH340
USB: serial: option: add GosunCn GM500 series
USB: serial: cypress_m8: enable Simply Automated UPB PIM
Merge tag 'dma-mapping-5.8-6' of git://git.infradead.org/users/hch/dma-mapping into master
Pull dma-mapping fixes from Christoph Hellwig:
"Ensure we always have fully addressable memory in the dma coherent
pool (Nicolas Saenz Julienne)"
* tag 'dma-mapping-5.8-6' of git://git.infradead.org/users/hch/dma-mapping:
dma-pool: do not allocate pool memory from CMA
dma-pool: make sure atomic pool suits device
dma-pool: introduce dma_guess_pool()
dma-pool: get rid of dma_in_atomic_pool()
dma-direct: provide function to check physical memory area validity
vmlinux-objs-y is added to targets, which currently means that the EFI
stub gets added to the targets as well. It shouldn't be added since it
is built elsewhere.
This confuses Makefile.build which interprets the EFI stub as a target
$(obj)/$(objtree)/drivers/firmware/efi/libstub/lib.a
and will create drivers/firmware/efi/libstub/ underneath
arch/x86/boot/compressed, to hold this supposed target, if building
out-of-tree. [0]
Fix this by pulling the stub out of vmlinux-objs-y into efi-obj-y.
[0] See scripts/Makefile.build near the end:
# Create directories for object files if they do not exist
Kees Cook [Fri, 26 Jun 2020 20:34:25 +0000 (13:34 -0700)]
x86/entry: Actually disable stack protector
Some builds of GCC enable stack protector by default. Simply removing
the arguments is not sufficient to disable stack protector, as the stack
protector for those GCC builds must be explicitly disabled. Remove the
argument removals and add -fno-stack-protector. Additionally include
missed x32 argument updates, and adjust whitespace for readability.
Fixes: 20355e5f73a7 ("x86/entry: Exclude low level entry code from sanitizing") Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lkml.kernel.org/r/202006261333.585319CA6B@keescook
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi into master
Pull SCSI fix from James Bottomley:
"One small driver fix. Although the one liner makes it sound like a
cosmetic change, it's a regression fix for the megaraid_sas driver"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: megaraid_sas: Remove undefined ENABLE_IRQ_POLL macro
Merge tag 'hwmon-for-v5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging into master
Pull hwmon fixes from Guenter Roeck:
- Using SCT on some Tohsiba drives causes firmware hangs. Disable its
use in the drivetemp driver.
- Handle potential buffer overflows in scmi and aspeed-pwm-tacho
driver.
- Energy reporting does not work well on all AMD CPUs. Restrict
amd_energy to known working models.
- Enable reading the CPU temperature on NCT6798D using undocumented
registers.
- Fix read errors seen if PEC is enabled in adm1275 driver.
- Fix setting the pwm1_enable in emc2103 driver.
* tag 'hwmon-for-v5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging:
hwmon: (drivetemp) Avoid SCT usage on Toshiba DT01ACA family drives
hwmon: (scmi) Fix potential buffer overflow in scmi_hwmon_probe()
hwmon: (nct6775) Accept PECI Calibration as temperature source for NCT6798D
hwmon: (adm1275) Make sure we are reading enough data for different chips
hwmon: (emc2103) fix unable to change fan pwm1_enable attribute
hwmon: (amd_energy) match for supported models
hwmon: (aspeed-pwm-tacho) Avoid possible buffer overflow
Merge tag 'riscv-for-linus-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux into master
Pull RISC-V fixes from Palmer Dabbelt:
"Two fixes:
- 16KiB kernel stacks on rv64, which fixes a lot of crashes.
- Rolling an mmiowb() into the scheduler, which when combined with
Will's fix to the mmiowb()-on-spinlock should fix the PREEMPT
issues we've been seeing"
* tag 'riscv-for-linus-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
RISC-V: Upgrade smp_mb__after_spinlock() to iorw,iorw
riscv: use 16KB kernel stack on 64-bit
Merge tag 'powerpc-5.8-7' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux into master
Pull powerpc fixes from Michael Ellerman:
"Some more powerpc fixes for 5.8:
- A fix to the VAS code we merged this cycle, to report the proper
error code to userspace for address translation failures. And a
selftest update to match.
- Another fix for our pkey handling of PROT_EXEC mappings.
- A fix for a crash when booting a "secure VM" under an ultravisor
with certain numbers of CPUs.
Thanks to: Aneesh Kumar K.V, Haren Myneni, Laurent Dufour, Sandipan
Das, Satheesh Rajendran, Thiago Jung Bauermann"
* tag 'powerpc-5.8-7' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
selftests/powerpc: Use proper error code to check fault address
powerpc/vas: Report proper error code for address translation failure
powerpc/pseries/svm: Fix incorrect check for shared_lppaca_size
powerpc/book3s64/pkeys: Fix pkey_access_permitted() for execute disable pkey
hwmon: (drivetemp) Avoid SCT usage on Toshiba DT01ACA family drives
It has been observed that Toshiba DT01ACA family drives have
WRITE FPDMA QUEUED command timeouts and sometimes just freeze until
power-cycled under heavy write loads when their temperature is getting
polled in SCT mode. The SMART mode seems to be fine, though.
Let's make sure we don't use SCT mode for these drives then.
While only the 3 TB model was actually caught exhibiting the problem let's
play safe here to avoid data corruption and extend the ban to the whole
family.
Andy Lutomirski [Fri, 17 Jul 2020 23:53:55 +0000 (16:53 -0700)]
x86/ioperm: Fix io bitmap invalidation on Xen PV
tss_invalidate_io_bitmap() wasn't wired up properly through the pvop
machinery, so the TSS and Xen's io bitmap would get out of sync
whenever disabling a valid io bitmap.
Add a new pvop for tss_invalidate_io_bitmap() to fix it.
This is XSA-329.
Fixes: 22fe5b0439dd ("x86/ioperm: Move TSS bitmap update to exit to user work") Signed-off-by: Andy Lutomirski <luto@kernel.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Juergen Gross <jgross@suse.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/d53075590e1f91c19f8af705059d3ff99424c020.1595030016.git.luto@kernel.org
Merge tag 'nfs-for-5.8-3' of git://git.linux-nfs.org/projects/anna/linux-nfs into master
Pull NFS client fixes from Anna Schumaker:
"A few more NFS client bugfixes for Linux 5.8:
NFS:
- Fix interrupted slots by using the SEQUENCE operation
SUNRPC:
- revert d03727b248d0 to fix unkillable IOs
xprtrdma:
- Fix double-free in rpcrdma_ep_create()
- Fix recursion into rpcrdma_xprt_disconnect()
- Fix return code from rpcrdma_xprt_connect()
- Fix handling of connect errors
- Fix incorrect header size calculations"
* tag 'nfs-for-5.8-3' of git://git.linux-nfs.org/projects/anna/linux-nfs:
SUNRPC reverting d03727b248d0 ("NFSv4 fix CLOSE not waiting for direct IO compeletion")
xprtrdma: fix incorrect header size calculations
NFS: Fix interrupted slots by sending a solo SEQUENCE operation
xprtrdma: Fix handling of connect errors
xprtrdma: Fix return code from rpcrdma_xprt_connect()
xprtrdma: Fix recursion into rpcrdma_xprt_disconnect()
xprtrdma: Fix double-free in rpcrdma_ep_create()
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux into master
Pull arm64 fixes from Will Deacon:
"A batch of arm64 fixes.
Although the diffstat is a bit larger than we'd usually have at this
stage, a decent amount of it is the addition of comments describing
our syscall tracing behaviour, and also a sweep across all the modular
arm64 PMU drivers to make them rebust against unloading and unbinding.
There are a couple of minor things kicking around at the moment (CPU
errata and module PLTs for very large modules), but I'm not expecting
any significant changes now for us in 5.8.
- Fix kernel text addresses for relocatable images booting using EFI
and with KASLR disabled so that they match the vmlinux ELF binary.
- Fix unloading and unbinding of PMU driver modules.
- Fix generic mmiowb() when writeX() is called from preemptible
context (reported by the riscv folks).
- Fix ptrace hardware single-step interactions with signal handlers,
system calls and reverse debugging.
- Fix reporting of 64-bit x0 register for 32-bit tasks via
'perf_regs'.
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
drivers/perf: Prevent forced unbinding of PMU drivers
asm-generic/mmiowb: Allow mmiowb_set_pending() when preemptible()
arm64: Use test_tsk_thread_flag() for checking TIF_SINGLESTEP
arm64: ptrace: Use NO_SYSCALL instead of -1 in syscall_trace_enter()
arm64: syscall: Expand the comment about ptrace and syscall(-1)
arm64: ptrace: Add a comment describing our syscall entry/exit trap ABI
arm64: compat: Ensure upper 32 bits of x0 are zero on syscall return
arm64: ptrace: Override SPSR.SS when single-stepping is enabled
arm64: ptrace: Consistently use pseudo-singlestep exceptions
drivers/perf: Fix kernel panic when rmmod PMU modules during perf sampling
efi/libstub/arm64: Retain 2MB kernel Image alignment if !KASLR
Thomas Gleixner [Fri, 17 Jul 2020 16:00:02 +0000 (18:00 +0200)]
genirq/affinity: Handle affinity setting on inactive interrupts correctly
Setting interrupt affinity on inactive interrupts is inconsistent when
hierarchical irq domains are enabled. The core code should just store the
affinity and not call into the irq chip driver for inactive interrupts
because the chip drivers may not be in a state to handle such requests.
X86 has a hacky workaround for that but all other irq chips have not which
causes problems e.g. on GIC V3 ITS.
Instead of adding more ugly hacks all over the place, solve the problem in
the core code. If the affinity is set on an inactive interrupt then:
- Store it in the irq descriptors affinity mask
- Update the effective affinity to reflect that so user space has
a consistent view
- Don't call into the irq chip driver
This is the core equivalent of the X86 workaround and works correctly
because the affinity setting is established in the irq chip when the
interrupt is activated later on.
Note, that this is only effective when hierarchical irq domains are enabled
by the architecture. Doing it unconditionally would break legacy irq chip
implementations.
For hierarchial irq domains this works correctly as none of the drivers can
have a dependency on affinity setting in inactive state by design.
Remove the X86 workaround as it is not longer required.
When an expiration delta falls into the last level of the wheel, that delta
has be compared against the maximum possible delay and reduced to fit in if
necessary.
However instead of comparing the delta against the maximum, the code
compares the actual expiry against the maximum. Then instead of fixing the
delta to fit in, it sets the maximum delta as the expiry value.
This can result in various undesired outcomes, the worst possible one
being a timer expiring 15 days ahead to fire immediately.
Fixes: 500462a9de65 ("timers: Switch to a non-cascading wheel") Signed-off-by: Frederic Weisbecker <frederic@kernel.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/20200717140551.29076-2-frederic@kernel.org
SUNRPC reverting d03727b248d0 ("NFSv4 fix CLOSE not waiting for direct IO compeletion")
Reverting commit d03727b248d0 "NFSv4 fix CLOSE not waiting for
direct IO compeletion". This patch made it so that fput() by calling
inode_dio_done() in nfs_file_release() would wait uninterruptably
for any outstanding directIO to the file (but that wait on IO should
be killable).
The problem the patch was also trying to address was REMOVE returning
ERR_ACCESS because the file is still opened, is supposed to be resolved
by server returning ERR_FILE_OPEN and not ERR_ACCESS.
Signed-off-by: Olga Kornievskaia <kolga@netapp.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Merge tag 'fuse-fixes-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse into master
Pull fuse fixes from Miklos Szeredi:
- two regressions in this cycle caused by the conversion of writepage
list to an rb_tree
- two regressions in v5.4 cause by the conversion to the new mount API
- saner behavior of fsconfig(2) for the reconfigure case
- an ancient issue with FS_IOC_{GET,SET}FLAGS ioctls
* tag 'fuse-fixes-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
fuse: Fix parameter for FS_IOC_{GET,SET}FLAGS
fuse: don't ignore errors from fuse_writepages_fill()
fuse: clean up condition for writepage sending
fuse: reject options on reconfigure via fsconfig(2)
fuse: ignore 'data' argument of mount(..., MS_REMOUNT)
fuse: use ->reconfigure() instead of ->remount_fs()
fuse: fix warning in tree_insert() and clean up writepage insertion
fuse: move rb_erase() before tree_insert()
Merge tag 'ovl-fixes-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs into master
Pull overlayfs fixes from Miklos Szeredi:
- fix a regression introduced in v4.20 in handling a regenerated
squashfs lower layer
- two regression fixes for this cycle, one of which is Oops inducing
- miscellaneous issues
* tag 'ovl-fixes-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
ovl: fix lookup of indexed hardlinks with metacopy
ovl: fix unneeded call to ovl_change_flags()
ovl: fix mount option checks for nfs_export with no upperdir
ovl: force read-only sb on failure to create index dir
ovl: fix regression with re-formatted lower squashfs
ovl: fix oops in ovl_indexdir_cleanup() with nfs_export=on
ovl: relax WARN_ON() when decoding lower directory file handle
ovl: remove not used argument in ovl_check_origin
ovl: change ovl_copy_up_flags static
ovl: inode reference leak in ovl_is_inuse true case.
Merge tag 'spi-fix-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi into master
Pull spi fixes from Mark Brown:
"A couple of small driver specific fixes for fairly minor issues"
* tag 'spi-fix-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: spi-sun6i: sun6i_spi_transfer_one(): fix setting of clock rate
spi: mediatek: use correct SPI_CFG2_REG MACRO
Merge tag 'regulator-fix-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator into master
Pull regulator fixes from Mark Brown:
"The more substantial fix here is the rename of the da903x driver which
fixes a collision with the parent MFD driver name which caused issues
when things were built as modules.
There's also a fix for a mislableled regulator on the pmi8994 which is
quite important for systems with that device"
* tag 'regulator-fix-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
MAINTAINERS: remove obsolete entry after file renaming
regulator: rename da903x to da903x-regulator
regulator: qcom_smd: Fix pmi8994 label
Merge tag 'regmap-fix-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap into master
Pull regmap fixes from Mark Brown:
"A couple of substantial fixes here, one from Doug which fixes the
debugfs code for MMIO regmaps (fortunately not the common case) and
one from Marc fixing lookups of multiple regmaps for the same device
(a very unusual case).
There's also a fix for Kconfig to ensure we enable SoundWire properly"
* tag 'regmap-fix-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap:
regmap: debugfs: Don't sleep while atomic for fast_io regmaps
regmap: add missing dependency on SoundWire
regmap: dev_get_regmap_match(): fix string comparison
RISC-V: Upgrade smp_mb__after_spinlock() to iorw,iorw
While digging through the recent mmiowb preemption issue it came up that
we aren't actually preventing IO from crossing a scheduling boundary.
While it's a bit ugly to overload smp_mb__after_spinlock() with this
behavior, it's what PowerPC is doing so there's some precedent.
Merge tag 'renesas-fixes-for-v5.8-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-devel into arm/fixes
Renesas fixes for v5.8
- Replace CONFIG_PCIE_RCAR by CONFIG_PCIE_RCAR_HOST in the defconfig,
to unblock a planned Kconfig change.
* tag 'renesas-fixes-for-v5.8-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-devel:
arm64: defconfig: Enable CONFIG_PCIE_RCAR_HOST
Merge tag 'sound-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound into master
Pull sound fixes from Takashi Iwai:
"No surprise here, just a few device-specific small fixes: two fixes
for USB LINE6 and one for USB-audio drivers wrt syzkaller fuzzer
issues, while the rest are all HD-audio Realtek quirks"
* tag 'sound-5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda/realtek - fixup for yet another Intel reference board
ALSA: hda/realtek - Enable Speaker for ASUS UX563
ALSA: hda/realtek - Enable Speaker for ASUS UX533 and UX534
ALSA: hda/realtek: Enable headset mic of Acer TravelMate B311R-31 with ALC256
ALSA: hda/realtek: enable headset mic of ASUS ROG Zephyrus G14(G401) series with ALC289
ALSA: hda/realtek - change to suitable link model for ASUS platform
ALSA: usb-audio: Fix race against the error recovery URB submission
ALSA: line6: Sync the pending work cancel at disconnection
ALSA: line6: Perform sanity check for each URB creation
tools arch kvm: Sync kvm headers with the kernel sources
To pick up the changes from:
83d31e5271ac ("KVM: nVMX: fixes for preemption timer migration")
That don't entail changes in tooling.
This silences these tools/perf build warnings:
Warning: Kernel ABI header at 'tools/arch/x86/include/uapi/asm/kvm.h' differs from latest version at 'arch/x86/include/uapi/asm/kvm.h'
diff -u tools/arch/x86/include/uapi/asm/kvm.h arch/x86/include/uapi/asm/kvm.h
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
b2f9f1535bb9 ("libbpf: Fix libbpf hashmap on (I)LP32 architectures")
Silencing this warning:
Warning: Kernel ABI header at 'tools/perf/util/hashmap.h' differs from latest version at 'tools/lib/bpf/hashmap.h'
diff -u tools/perf/util/hashmap.h tools/lib/bpf/hashmap.h
I'll eventually update the warning to remove the "Kernel ABI" part
and instead state libbpf when noticing that the original is at
"tools/lib/something".
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Andrii Nakryiko <andriin@fb.com> Cc: Jakub Bogusz <qboosh@pld-linux.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org>
Ian Rogers <irogers@google.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Ravi Bangoria [Fri, 19 Jun 2020 13:34:11 +0000 (19:04 +0530)]
libsubcmd: Fix OPT_CALLBACK_SET()
Any option macro with _SET suffix should set opt->set variable which is
not happening for OPT_CALLBACK_SET(). This is causing issues with perf
record --switch-output-event. Fix that.
Before:
# ./perf record --overwrite -e sched:*switch,syscalls:sys_enter_mmap \
--switch-output-event syscalls:sys_enter_mmap
^C[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.297 MB perf.data (657 samples) ]
After:
$ ./perf record --overwrite -e sched:*switch,syscalls:sys_enter_mmap \
--switch-output-event syscalls:sys_enter_mmap
[ perf record: dump data: Woken up 1 times ]
[ perf record: Dump perf.data.2020061918144542 ]
[ perf record: dump data: Woken up 1 times ]
[ perf record: Dump perf.data.2020061918144608 ]
[ perf record: dump data: Woken up 1 times ]
[ perf record: Dump perf.data.2020061918144660 ]
^C[ perf record: dump data: Woken up 1 times ]
[ perf record: Dump perf.data.2020061918144784 ]
[ perf record: Woken up 0 times to write data ]
[ perf record: Dump perf.data.2020061918144803 ]
[ perf record: Captured and wrote 0.419 MB perf.data.<timestamp> ]
Fixes: 636eb4d001b1 ("libsubcmd: Introduce OPT_CALLBACK_SET()") Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lore.kernel.org/lkml/20200619133412.50705-1-ravi.bangoria@linux.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Qi Liu [Fri, 17 Jul 2020 08:49:23 +0000 (16:49 +0800)]
drivers/perf: Prevent forced unbinding of PMU drivers
Forcefully unbinding PMU drivers during perf sampling will lead to
a kernel panic, because the perf upper-layer framework call a NULL
pointer in this situation.
To solve this issue, "suppress_bind_attrs" should be set to true, so
that bind/unbind can be disabled via sysfs and prevent unbinding PMU
drivers during perf sampling.
Will Deacon [Thu, 16 Jul 2020 11:28:16 +0000 (12:28 +0100)]
asm-generic/mmiowb: Allow mmiowb_set_pending() when preemptible()
Although mmiowb() is concerned only with serialising MMIO writes occuring
in contexts where a spinlock is held, the call to mmiowb_set_pending()
from the MMIO write accessors can occur in preemptible contexts, such
as during driver probe() functions where ordering between CPUs is not
usually a concern, assuming that the task migration path provides the
necessary ordering guarantees.
Unfortunately, the default implementation of mmiowb_set_pending() is not
preempt-safe, as it makes use of a a per-cpu variable to track its
internal state. This has been reported to generate the following splat
on riscv:
Although it's possible to fix the driver in this case, other splats have
been seen from other drivers, including the infamous 8250 UART, and so
it's better to address this problem in the mmiowb core itself.
Fix mmiowb_set_pending() by using the raw_cpu_ptr() to get at the mmiowb
state and then only updating the 'mmiowb_pending' field if we are not
preemptible (i.e. we have a non-zero nesting count).
Cc: Arnd Bergmann <arnd@arndb.de> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Guo Ren <guoren@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Reported-by: Palmer Dabbelt <palmer@dabbelt.com> Reported-by: Emil Renner Berthing <kernel@esmil.dk> Tested-by: Emil Renner Berthing <kernel@esmil.dk> Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com> Acked-by: Palmer Dabbelt <palmerdabbelt@google.com> Link: https://lore.kernel.org/r/20200716112816.7356-1-will@kernel.org Signed-off-by: Will Deacon <will@kernel.org>
Merge tag 'drm-fixes-2020-07-17-1' of git://anongit.freedesktop.org/drm/drm into master
Pull drm fixes from Dave Airlie:
"Weekly fixes pull, big bigger than I'd normally like, but they are
fairly scattered and small individually.
The vmwgfx one is a black screen regression, otherwise the largest is
an MST encoder fix for amdgpu which results in a WARN in some cases,
and a scattering of i915 fixes.
I'm tracking two regressions at the moment that hopefully we get
nailed down this week for rc7.
dma-buf:
- sleeping atomic fix
amdgpu:
- Fix a race condition with KIQ
- Preemption fix
- Fix handling of fake MST encoders
- OLED panel fix
- Handle allocation failure in stream construction
- Renoir SMC fix
- SDMA 5.x fix
i915:
- FBC w/a stride fix
- Fix use-after-free fix on module reload
- Ignore irq enabling on the virtual engines to fix device sleep
- Use GTT when saving/restoring engine GPR
- Fix selftest sort function
vmwgfx:
- black screen fix
aspeed:
- fbcon init warn fix"
* tag 'drm-fixes-2020-07-17-1' of git://anongit.freedesktop.org/drm/drm:
drm/amdgpu/sdma5: fix wptr overwritten in ->get_wptr()
drm/amdgpu/powerplay: Modify SMC message name for setting power profile mode
drm/amd/display: handle failed allocation during stream construction
drm/amd/display: OLED panel backlight adjust not work with external display connected
drm/amdgpu/display: create fake mst encoders ahead of time (v4)
drm/amdgpu: fix preemption unit test
drm/amdgpu/gfx10: fix race condition for kiq
drm/i915: Recalculate FBC w/a stride when needed
drm/i915: Move cec_notifier to intel_hdmi_connector_unregister, v2.
drm/i915/gt: Only swap to a random sibling once upon creation
drm/i915/gt: Ignore irq enabling on the virtual engines
drm/i915/perf: Use GTT when saving/restoring engine GPR
drm/i915/selftests: Fix compare functions provided for sorting
drm/vmwgfx: fix update of display surface when resolution changes
dmabuf: use spinlock to access dmabuf->name
drm/aspeed: Call drm_fbdev_generic_setup after drm_dev_register
Vincent Guittot [Fri, 10 Jul 2020 15:24:26 +0000 (17:24 +0200)]
sched/fair: handle case of task_h_load() returning 0
task_h_load() can return 0 in some situations like running stress-ng
mmapfork, which forks thousands of threads, in a sched group on a 224 cores
system. The load balance doesn't handle this correctly because
env->imbalance never decreases and it will stop pulling tasks only after
reaching loop_max, which can be equal to the number of running tasks of
the cfs. Make sure that imbalance will be decreased by at least 1.
misfit task is the other feature that doesn't handle correctly such
situation although it's probably more difficult to face the problem
because of the smaller number of CPUs and running tasks on heterogenous
system.
We can't simply ensure that task_h_load() returns at least one because it
would imply to handle underflow in other places.
Merge tag 'omap-for-v5.8/fixes-rc5-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap into arm/fixes
Fixes for omaps for v5.8-rc cycle
Few fixes for issues noticed during testing:
- Two DEBUG_ATOMIC_SLEEP fixes for ti-sysc interconnect target module
driver
- A regression fix for ti-sysc no-idle handling that caused issues
compared to earlier platform data based booting
- A fix for memory leak for omap_hwmod_allocate_module
- Fix d_can driver probe for am437x
* tag 'omap-for-v5.8/fixes-rc5-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
ARM: dts: Fix dcan driver probe failed on am437x platform
ARM: OMAP2+: Fix possible memory leak in omap_hwmod_allocate_module
bus: ti-sysc: Do not disable on suspend for no-idle
bus: ti-sysc: Fix sleeping function called from invalid context for RTC quirk
bus: ti-sysc: Fix wakeirq sleeping function called from invalid context
Merge tag 'imx-fixes-5.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux into arm/fixes
i.MX fixes for 5.8, round 2:
- A couple of fixes on i.MX platform device registration code to stop
the use of invalid IRQ 0.
- Fix a regression seen on ls1021a platform, caused by commit 52102a3ba6a61 ("soc: imx: move cpu code to drivers/soc/imx").
- Fix a misconfiguration of audio SSI on imx6qdl-gw551x board.
* tag 'imx-fixes-5.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux:
soc: imx: check ls1021a
ARM: imx: Remove imx_add_imx_dma() unused irq_err argument
ARM: imx: Provide correct number of resources when registering gpio devices
ARM: dts: imx6qdl-gw551x: fix audio SSI
Douglas Anderson [Wed, 15 Jul 2020 23:46:15 +0000 (16:46 -0700)]
regmap: debugfs: Don't sleep while atomic for fast_io regmaps
If a regmap has "fast_io" set then its lock function uses a spinlock.
That doesn't work so well with the functions:
* regmap_cache_only_write_file()
* regmap_cache_bypass_write_file()
Both of the above functions have the pattern:
1. Lock the regmap.
2. Call:
debugfs_write_file_bool()
copy_from_user()
__might_fault()
__might_sleep()
Let's reorder things a bit so that we do all of our sleepable
functions before we grab the lock.