user/sven/linux.git/drivers/of/base.c, branch v5.4.43

of: restore old handling of cells_name=NULL in of_*_phandle_with_args()

2019-09-19T13:33:02Z

Before commit e42ee61017f5 ("of: Let of_for_each_phandle fallback to non-negative cell_count") the iterator functions calling of_for_each_phandle assumed a cell count of 0 if cells_name was NULL. This corner case was missed when implementing the fallback logic in e42ee61017f5 and resulted in an endless loop. Restore the old behaviour of of_count_phandle_with_args() and of_parse_phandle_with_args() and add a check to of_phandle_iterator_init() to prevent a similar failure as a safety precaution. of_parse_phandle_with_args_map() doesn't need a similar fix as cells_name isn't NULL there. Affected drivers are: - drivers/base/power/domain.c - drivers/base/power/domain.c - drivers/clk/ti/clk-dra7-atl.c - drivers/hwmon/ibmpowernv.c - drivers/i2c/muxes/i2c-demux-pinctrl.c - drivers/iommu/mtk_iommu.c - drivers/net/ethernet/freescale/fman/mac.c - drivers/opp/of.c - drivers/perf/arm_dsu_pmu.c - drivers/regulator/of_regulator.c - drivers/remoteproc/imx_rproc.c - drivers/soc/rockchip/pm_domains.c - sound/soc/fsl/imx-audmix.c - sound/soc/fsl/imx-audmix.c - sound/soc/meson/axg-card.c - sound/soc/samsung/tm2_wm5110.c - sound/soc/samsung/tm2_wm5110.c Thanks to Geert Uytterhoeven for reporting the issue, Peter Rosin for helping pinpoint the actual problem and the testers for confirming this fix. Fixes: e42ee61017f5 ("of: Let of_for_each_phandle fallback to non-negative cell_count") Tested-by: Marek Szyprowski Tested-by: Geert Uytterhoeven Signed-off-by: Uwe Kleine-König Signed-off-by: Rob Herring

of: Let of_for_each_phandle fallback to non-negative cell_count

2019-09-13T21:55:12Z

Referencing device tree nodes from a property allows to pass arguments. This is for example used for referencing gpios. This looks as follows: gpio_ctrl: gpio-controller { #gpio-cells = <2> ... } someothernode { gpios = <&gpio_ctrl 5 0 &gpio_ctrl 3 0>; ... } To know the number of arguments this must be either fixed, or the referenced node is checked for a $cells_name (here: "#gpio-cells") property and with this information the start of the second reference can be determined. Currently regulators are referenced with no additional arguments. To allow some optional arguments without having to change all referenced nodes this change introduces a way to specify a default cell_count. So when a phandle is parsed we check for the $cells_name property and use it as before if present. If it is not present we fall back to cells_count if non-negative and only fail if cells_count is smaller than zero. Signed-off-by: Uwe Kleine-König Signed-off-by: Rob Herring

of: Allow the iommu-map property to omit untranslated devices

2019-06-06T21:32:13Z

In PCI root complex nodes, the iommu-map property describes the IOMMU that translates each endpoint. On some platforms, the IOMMU itself is presented as a PCI endpoint (e.g. AMD IOMMU and virtio-iommu). This isn't supported by the current OF driver, which expects all endpoints to have an IOMMU. Allow the iommu-map property to have gaps. Relaxing of_map_rid() also allows the msi-map property to have gaps, which is invalid since MSIs always reach an MSI controller. In that case pci_msi_setup_msi_irqs() will return an error when attempting to find the device's MSI domain. Reviewed-by: Rob Herring Signed-off-by: Jean-Philippe Brucker Signed-off-by: Michael S. Tsirkin

of: Improve of_phandle_iterator_next() error message

2019-04-10T21:33:48Z

Understanding why of_phandle_iterator_next() returns an error can sometimes be hard to decipher based solely on the error message, a typical error example is that #foo-cells = and the phandle property used has a smaller number of cells specified, e.g.: #thermal-sensor-cells = <1>; phandle = <&scmi_sensor> instead of: phandle <&scmi_sensor 0>; Instead, make it clear what the expectations are towards debugging incorrect Device Tree faster: OF: /thermal-zones/scmi-thermal: #thermal-sensor-cells = 1, found 0 instead Signed-off-by: Florian Fainelli Signed-off-by: Rob Herring

of: __of_detach_node() - remove node from phandle cache

2018-12-21T18:42:36Z

Non-overlay dynamic devicetree node removal may leave the node in the phandle cache. Subsequent calls to of_find_node_by_phandle() will incorrectly find the stale entry. Remove the node from the cache. Add paranoia checks in of_find_node_by_phandle() as a second level of defense (do not return cached node if detached, do not add node to cache if detached). Fixes: 0b3ce78e90fc ("of: cache phandle nodes to reduce cost of of_find_node_by_phandle()") Reported-by: Michael Bringmann Cc: stable@vger.kernel.org # v4.17+ Signed-off-by: Frank Rowand Signed-off-by: Rob Herring

of: of_node_get()/of_node_put() nodes held in phandle cache

2018-12-21T18:41:58Z

The phandle cache contains struct device_node pointers. The refcount of the pointers was not incremented while in the cache, allowing use after free error after kfree() of the node. Add the proper increment and decrement of the use count. Fixes: 0b3ce78e90fc ("of: cache phandle nodes to reduce cost of of_find_node_by_phandle()") Cc: stable@vger.kernel.org # v4.17+ Signed-off-by: Frank Rowand Signed-off-by: Rob Herring

of: Use of_node_name_eq for node name comparisons

2018-12-05T20:45:13Z

Convert string compares of DT node names to use of_node_name_eq helper instead. This removes direct access to the node name pointer. Cc: Frank Rowand Cc: Pantelis Antoniou Cc: devicetree@vger.kernel.org Signed-off-by: Rob Herring

of: Use device_type helpers to access the node type

2018-11-27T17:27:04Z

Remove directly accessing device_node.type pointer and use the accessors instead. This will eventually allow removing the type pointer. Cc: Frank Rowand Cc: devicetree@vger.kernel.org Signed-off-by: Rob Herring

Merge tag 'devicetree-fixes-for-4.20-1' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux

2018-11-01T21:45:38Z

Pull Devicetree fixes from Rob Herring: - fix cpu node iterator for powerpc systems - clarify ARM CPU binding 'capacities-dmips-mhz' property calculations * tag 'devicetree-fixes-for-4.20-1' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux: of: Fix cpu node iterator to not ignore disabled cpu nodes dt-bindings: arm: Explain capacities-dmips-mhz calculations in example

of: Fix cpu node iterator to not ignore disabled cpu nodes

2018-11-01T21:16:54Z

In most cases, nodes with 'status = "disabled";' are treated as if the node is not present though it is a common bug to forget to check that. However, cpu nodes are different in that "disabled" simply means offline and the OS can bring the CPU core online. Commit f1f207e43b8a ("of: Add cpu node iterator for_each_of_cpu_node()") followed the common behavior of ignoring disabled cpu nodes. This breaks some powerpc systems (at least NXP P50XX/e5500). Fix this by dropping the status check. Fixes: 651d44f9679c ("of: use for_each_of_cpu_node iterator") Fixes: f1f207e43b8a ("of: Add cpu node iterator for_each_of_cpu_node()") Reported-by: Christian Zigotzky Tested-by: Michael Ellerman Cc: Frank Rowand Signed-off-by: Rob Herring