On 5/7/2014 7:51 PM, Frank Rowand wrote: > On 5/7/2014 2:48 PM, Rob Herring wrote: >> From: Rob Herring <robh@xxxxxxxxxx> >> >> Add a test case for nodes which have the same name and same >> non-translatable unit address. > > If I apply patch 1 and 2 without applying 3 and 4 then console > warnings are printed, but from a different area of code than > the original problem reported. This probably is not a big deal, > but I'm trying to figure out if I can modify the test to also > show the original problem. If you want to add a test that triggers the same stack trace as the orginally reported problem, a patch is below. It would apply between your original patch 1 and patch 2. > > The test case also properly reports the failure. > > Once all 4 patches are applied, then the test case passes. > > Thus: > > Tested-by: Frank Rowand <frank.rowand@xxxxxxxxxxxxxx> > >> >> Signed-off-by: Rob Herring <robh@xxxxxxxxxx> >> --- >> drivers/of/selftest.c | 23 ++++++++++++++++++ >> drivers/of/testcase-data/testcases.dtsi | 1 + >> drivers/of/testcase-data/tests-platform.dtsi | 35 ++++++++++++++++++++++++++++ >> 3 files changed, 59 insertions(+) >> create mode 100644 drivers/of/testcase-data/tests-platform.dtsi >> > > < snip > > > From: Frank Rowand <frank.rowand@xxxxxxxxxxxxxx> Add another test case to of_selftest_platform_populate(). This case triggers the same stack trace from of_platform_populate() to sysfs_warn_dup() as seen in the case reported by https://lkml.org/lkml/2014/4/23/312. Signed-off-by: Frank Rowand <frank.rowand@xxxxxxxxxxxxxx> --- drivers/of/selftest.c | 27 ++++++++++++++++++++++ drivers/of/testcase-data/tests-platform.dtsi | 33 +++++++++++++++++++++++++++ 2 files changed, 60 insertions(+) Index: b/drivers/of/testcase-data/tests-platform.dtsi =================================================================== --- a/drivers/of/testcase-data/tests-platform.dtsi +++ b/drivers/of/testcase-data/tests-platform.dtsi @@ -31,5 +31,38 @@ }; }; }; + + test-master { + compatible = "test-master"; + #address-cells = <1>; + #size-cells = <0>; + + test-device@0 { + compatible = "test-device"; + reg = <0x0>; + + #address-cells = <1>; + #size-cells = <0>; + + sub-dev@100 { + compatible = "test-sub-device"; + reg = <0x100>; + }; + }; + + test-device@1 { + compatible = "test-device"; + reg = <0x1>; + + #address-cells = <1>; + #size-cells = <0>; + + sub-dev@100 { + compatible = "test-sub-device"; + reg = <0x100>; + }; + }; + }; }; }; + Index: b/drivers/of/selftest.c =================================================================== --- a/drivers/of/selftest.c +++ b/drivers/of/selftest.c @@ -436,6 +436,7 @@ static void __init of_selftest_platform_ { .compatible = "test-device", }, {} }; + struct platform_device *pdev; np = of_find_node_by_path("/testcase-data/platform-tests"); if (!np) { @@ -447,6 +448,32 @@ static void __init of_selftest_platform_ rc = of_platform_populate(child, match, NULL, NULL); selftest(!rc, "Could not create device for node '%s'\n", child->name); } + + np = of_find_node_by_path("/testcase-data/test-master"); + if (!np) { + pr_err("No /testcase-data/test-master node in device tree\n"); + return; + } + + for_each_child_of_node(np, child) { + pdev = of_device_alloc(child, NULL, NULL); + if (pdev) { + pdev->dev.bus = &platform_bus_type; + if (of_device_add(pdev) != 0) { + platform_device_put(pdev); + pdev = NULL; + } + } + selftest(pdev, "Could not allocate device for node '%s'\n", + child->full_name); + + if (pdev) { + rc = of_platform_populate(child, NULL, NULL, &pdev->dev); + selftest(!rc, "Could not populate node '%s'\n", + child->full_name); + } + } + } static int __init of_selftest(void) -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
