spi: acpi: only look up the IRQ of confirmed SPI devices
PATCH 1/1 spi: acpi: only look up the IRQ of confirmed SPI devices
When an SPI controller is registered, acpi_register_spi_device() is called for every present ACPI device in the namespace and acpi_spi_device_alloc() walks each device's _CRS. While doing so, acpi_spi_add_resource() converts the first interrupt resource it finds, even for devices that have no SpiSerialBus resource at all and are discarded right afterwards. Converting an interrupt resource registers its GSI, so this maps IRQs for arbitrary unrelated ACPI devices.
On a Dell Precision M4800 (Lynx Point, 24-pin IOAPIC) the DSDT reports Haswell-ULT LPSS child devices (\_SB.PCI0.I2C0.ACD0 and \_SB.PCI0.SDHC.WI01) as present with interrupts on GSI 37 and 38. Registering the intel-spi controller then results in:
ERROR: Unable to locate IOAPIC for GSI 37
ERROR: Unable to locate IOAPIC for GSI 38
Do what the I2C core does and only parse the interrupt resource once the device has been identified as an SPI device on this controller.
--- drivers/spi/spi.c | 36 ++++++++++++++++++++++++++++-------- 1 file changed, 28 insertions(+), 8 deletions(-)
drivers/spi/spi.c +28 -8
@@ -2995,7 +2995,6 @@ struct acpi_spi_lookup {acpi_handle device_handle;u32 max_speed_hz;u32 mode; -int irq;u8 bits_per_word;u8 chip_select;int n;
@@ -3138,17 +3137,25 @@ static int acpi_spi_add_resource(struct acpi_resource *ares, void *data)if (sb->device_polarity ==ACPI_SPI_ACTIVE_HIGH ) lookup->mode |=SPI_CS_HIGH ; } - }else if (lookup->irq <0 ) { -struct resource r; - -if (acpi_dev_resource_interrupt (ares,0 , &r)) - lookup->irq = r.start; }/* Always tell the ACPI core to skip this resource */ return 1 ; } +static int acpi_spi_add_irq_resource (struct acpi_resource *ares,void *data) +{ +int *irq = data; +struct resource r; + +if (*irq <0 && ares->type !=ACPI_RESOURCE_TYPE_SERIAL_BUS && +acpi_dev_resource_interrupt (ares,0 , &r)) + *irq = r.start; + +/* Always tell the ACPI core to skip this resource */ +return 1 ; +} +/** * acpi_spi_device_alloc - Allocate a spi device, and fill it in with ACPI information * @ctlr: controller to which the spi device belongs
@@ -3173,6 +3180,7 @@ struct spi_device *acpi_spi_device_alloc(struct spi_controller *ctlr,struct list_head resource_list;struct acpi_spi_lookup lookup = {};struct spi_device *spi; +int irq =-1 ;int ret;if (!ctlr && index ==-1 )
@@ -3180,7 +3188,6 @@ struct spi_device *acpi_spi_device_alloc(struct spi_controller *ctlr, lookup.ctlr = ctlr; lookup.device_handle =acpi_device_handle (adev); - lookup.irq =-1 ; lookup.index = index; lookup.n =0 ;
@@ -3203,6 +3210,19 @@ struct spi_device *acpi_spi_device_alloc(struct spi_controller *ctlr,if (!lookup.max_speed_hz)return ERR_PTR (-ENODEV ); +/* +* Only look up the IRQ once we know this is an SPI device for us. +* Parsing an interrupt resource maps its GSI, which must not be done +* for every unrelated ACPI device walked during enumeration. +*/ +INIT_LIST_HEAD (&resource_list); + ret =acpi_dev_get_resources (adev, &resource_list, + acpi_spi_add_irq_resource, &irq); +if (ret <0 ) +return ERR_PTR (ret); + +acpi_dev_free_resource_list (&resource_list); + spi =spi_alloc_device (lookup.ctlr);if (!spi) { dev_err(&lookup.ctlr->dev, "failed to allocate SPI device for %s\n",
@@ -3215,7 +3235,7 @@ struct spi_device *acpi_spi_device_alloc(struct spi_controller *ctlr,ACPI_COMPANION_SET (&spi->dev, adev); spi->max_speed_hz = lookup.max_speed_hz; spi->mode |= lookup.mode; - spi->irq = lookup.irq; + spi->irq = irq; spi->bits_per_word = lookup.bits_per_word;/* * By default spi->chip_select[0] will hold the physical CS number,
Signature · 1 line
2.56.0
Mark Brown
maintainerQuoted 1 line · Rosen Penev
spi: acpi: only look up the IRQ of confirmed SPI devices
Applied to
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi.git for-7.4
Thanks!
[1/1] spi: acpi: only look up the IRQ of confirmed SPI devices
https://git.kernel.org/broonie/spi/c/2d0953ab69d3
All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted.
You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed.
If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced.
Please add any relevant lists and maintainers to the CCs when replying to this mail.
Thanks,
Mark