kernel/linux.git/drivers/mfd, branch linux-5.11.y

mfd: stm32-timers: Avoid clearing auto reload register

2021-05-14T08:49:58+00:00

[ Upstream commit 4917e498c6894ba077867aff78f82cffd5ffbb5c ] The ARR register is cleared unconditionally upon probing, after the maximum value has been read. This initial condition is rather not intuitive, when considering the counter child driver. It rather expects the maximum value by default: - The counter interface shows a zero value by default for 'ceiling' attribute. - Enabling the counter without any prior configuration makes it doesn't count. The reset value of ARR register is the maximum. So Choice here is to backup it, and restore it then, instead of clearing its value. It also fixes the initial condition seen by the counter driver. Fixes: d0f949e220fd ("mfd: Add STM32 Timers driver") Signed-off-by: Fabrice Gasnier Acked-by: William Breathitt Gray Signed-off-by: Lee Jones Signed-off-by: Sasha Levin

mfd: intel_pmt: Fix nuisance messages and handling of disabled capabilities

2021-05-14T08:49:37+00:00

[ Upstream commit a1a5c1c3df282dc122508a17500317266ef19e46 ] Some products will be available that have PMT capabilities that are not supported. Remove the warnings in this instance to avoid nuisance messages and confusion. Also return an error code for capabilities that are disabled by quirk to prevent them from keeping the driver loaded if only disabled capabilities are found. Fixes: 4f8217d5b0ca ("mfd: Intel Platform Monitoring Technology support") Signed-off-by: David E. Box Reviewed-by: Hans de Goede Signed-off-by: Lee Jones Signed-off-by: Sasha Levin

mfd: stmpe: Revert "Constify static struct resource"

2021-05-14T08:49:32+00:00

commit cb9e880a797a77c21c0f0e7ccd553da8eb4870af upstream. In stmpe_devices_init(), the start and end field of these structs are modified, so they can not be const. Add a comment to those structs that lacked it to reduce the risk that this happens again. This reverts commit 8d7b3a6dac4eae22c58b0853696cbd256966741b. Fixes: 8d7b3a6dac4e ("mfd: stmpe: Constify static struct resource") Reported-by: Andy Shevchenko Signed-off-by: Rikard Falkeborn Reviewed-by: Andy Shevchenko Signed-off-by: Lee Jones Signed-off-by: Greg Kroah-Hartman

mfd: arizona: Fix rumtime PM imbalance on error

2021-05-12T06:37:29+00:00

[ Upstream commit fe6df2b48043bbe1e852b2320501d3b169363c35 ] pm_runtime_get_sync() will increase the rumtime PM counter even it returns an error. Thus a pairing decrement is needed to prevent refcount leak. Fix this by replacing this API with pm_runtime_resume_and_get(), which will not change the runtime PM counter on error. Signed-off-by: Dinghao Liu Acked-by: Charles Keepax Signed-off-by: Lee Jones Signed-off-by: Sasha Levin

mfd: da9063: Support SMBus and I2C mode

2021-05-12T06:37:29+00:00

[ Upstream commit 586478bfc9f7e16504d6f64cf18bcbdf6fd0cbc9 ] By default the PMIC DA9063 2-wire interface is SMBus compliant. This means the PMIC will automatically reset the interface when the clock signal ceases for more than the SMBus timeout of 35 ms. If the I2C driver / device is not capable of creating atomic I2C transactions, a context change can cause a ceasing of the clock signal. This can happen if for example a real-time thread is scheduled. Then the DA9063 in SMBus mode will reset the 2-wire interface. Subsequently a write message could end up in the wrong register. This could cause unpredictable system behavior. The DA9063 PMIC also supports an I2C compliant mode for the 2-wire interface. This mode does not reset the interface when the clock signal ceases. Thus the problem depicted above does not occur. This patch tests for the bus functionality "I2C_FUNC_I2C". It can reasonably be assumed that the bus cannot obey SMBus timings if this functionality is set. SMBus commands most probably are emulated in this case which is prone to the latency issue described above. This patch enables the I2C bus mode if I2C_FUNC_I2C is set or otherwise keeps the default SMBus mode. Signed-off-by: Hubert Streidl Signed-off-by: Mark Jonas Reviewed-by: Wolfram Sang Signed-off-by: Lee Jones Signed-off-by: Sasha Levin

mfd: intel_quark_i2c_gpio: Revert "Constify static struct resources"

2021-03-30T12:30:26+00:00

[ Upstream commit a61f4661fba404418a7c77e86586dc52a58a93c6 ] The structures are used as place holders, so they are modified at run-time. Obviously they may not be constants. BUG: unable to handle page fault for address: d0643220 ... CPU: 0 PID: 110 Comm: modprobe Not tainted 5.11.0+ #1 Hardware name: Intel Corp. QUARK/GalileoGen2, BIOS 0x01000200 01/01/2014 EIP: intel_quark_mfd_probe+0x93/0x1c0 [intel_quark_i2c_gpio] This partially reverts the commit c4a164f41554d2899bed94bdcc499263f41787b4. While at it, add a comment to avoid similar changes in the future. Fixes: c4a164f41554 ("mfd: Constify static struct resources") Signed-off-by: Andy Shevchenko Reviewed-by: Rikard Falkeborn Tested-by: Tong Zhang Signed-off-by: Lee Jones Signed-off-by: Sasha Levin

mfd: gateworks-gsc: Fix interrupt type

2021-03-04T11:15:37+00:00

commit 8d9bf3c3e1451fc8de7b590040a868ade26d6b22 upstream. The Gateworks System Controller has an active-low interrupt. Fix the interrupt request type. Cc: Fixes: d85234994b2f ("mfd: Add Gateworks System Controller core driver") Signed-off-by: Tim Harvey Signed-off-by: Lee Jones Signed-off-by: Greg Kroah-Hartman

mfd: wm831x-auxadc: Prevent use after free in wm831x_auxadc_read_irq()

2021-03-04T11:14:53+00:00

[ Upstream commit 26783d74cc6a440ee3ef9836a008a697981013d0 ] The "req" struct is always added to the "wm831x->auxadc_pending" list, but it's only removed from the list on the success path. If a failure occurs then the "req" struct is freed but it's still on the list, leading to a use after free. Fixes: 78bb3688ea18 ("mfd: Support multiple active WM831x AUXADC conversions") Signed-off-by: Dan Carpenter Acked-by: Charles Keepax Signed-off-by: Lee Jones Signed-off-by: Sasha Levin

mfd: altera-sysmgr: Fix physical address storing more

2021-03-04T11:14:52+00:00

[ Upstream commit b0b5b16b78cea1b2b990a69ab8e07a42ccf7a2ed ] A recent fix improved the way the resource gets passed to the low-level accessors, but left one warning that appears in configurations with a resource_size_t that is wider than a pointer: In file included from drivers/mfd/altera-sysmgr.c:19: drivers/mfd/altera-sysmgr.c: In function 'sysmgr_probe': drivers/mfd/altera-sysmgr.c:148:40: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast] 148 | regmap = devm_regmap_init(dev, NULL, (void *)res->start, | ^ include/linux/regmap.h:646:6: note: in definition of macro '__regmap_lockdep_wrapper' 646 | fn(__VA_ARGS__, &_key, \ | ^~~~~~~~~~~ drivers/mfd/altera-sysmgr.c:148:12: note: in expansion of macro 'devm_regmap_init' 148 | regmap = devm_regmap_init(dev, NULL, (void *)res->start, | ^~~~~~~~~~~~~~~~ I had tried a different approach that would store the address in the private data as a phys_addr_t, but the easiest solution now seems to be to add a double cast to shut up the warning. As the address is passed to an inline assembly, it is guaranteed to not be wider than a register anyway. Fixes: d9ca7801b6e5 ("mfd: altera-sysmgr: Fix physical address storing hacks") Signed-off-by: Arnd Bergmann Signed-off-by: Lee Jones Signed-off-by: Sasha Levin

mfd: bd9571mwv: Use devm_mfd_add_devices()

2021-03-04T11:14:35+00:00

[ Upstream commit c58ad0f2b052b5675d6394e03713ee41e721b44c ] To remove mfd devices when unload this driver, should use devm_mfd_add_devices() instead. Fixes: d3ea21272094 ("mfd: Add ROHM BD9571MWV-M MFD PMIC driver") Signed-off-by: Yoshihiro Shimoda Acked-for-MFD-by: Lee Jones Reviewed-by: Geert Uytterhoeven Reviewed-by: Matti Vaittinen Signed-off-by: Lee Jones Signed-off-by: Sasha Levin