Arch maintainers, please ack/review patches.
This is a resend of a series from Frank last year[1]. I worked in Rob's
review comments to unconditionally call unflatten_device_tree() and
fixup/audit calls to of_have_populated_dt() so that behavior doesn't
change.
I need this series so I can add DT based tests in the clk framework.
Either I can merge it through the clk tree once everyone is happy, or
Rob can merge it through the DT tree and provide some branch so I can
base clk patches on it.
Changes from Frank's series[1]:
* Add a DTB loaded kunit test
* Make of_have_populated_dt() return false if the DTB isn't from the
bootloader
* Architecture calls made unconditional so that a root node is always
made
Changes from v2 (https://lore.kernel.org/r/[email protected]):
* Reorder patches to have OF changes largely first
* No longer modify initial_boot_params if ACPI=y
* Put arm64 patch back to v1
Changes from v1 (https://lore.kernel.org/r/[email protected]):
* x86 patch included
* arm64 knocks out initial dtb if acpi is in use
* keep Kconfig hidden but def_bool enabled otherwise
Frank Rowand (2):
of: Create of_root if no dtb provided by firmware
of: unittest: treat missing of_root as error instead of fixing up
Stephen Boyd (5):
of: Always unflatten in unflatten_and_copy_device_tree()
um: Unconditionally call unflatten_device_tree()
x86/of: Unconditionally call unflatten_and_copy_device_tree()
arm64: Unconditionally call unflatten_device_tree()
of: Add KUnit test to confirm DTB is loaded
arch/arm64/kernel/setup.c | 3 +-
arch/um/kernel/dtb.c | 14 ++++----
arch/x86/kernel/devicetree.c | 24 +++++++-------
drivers/of/.kunitconfig | 3 ++
drivers/of/Kconfig | 11 ++++++-
drivers/of/Makefile | 4 ++-
drivers/of/empty_root.dts | 6 ++++
drivers/of/fdt.c | 64 +++++++++++++++++++++++++++---------
drivers/of/of_test.c | 48 +++++++++++++++++++++++++++
drivers/of/platform.c | 3 --
drivers/of/unittest.c | 16 +++------
include/linux/of.h | 25 ++++++++------
12 files changed, 158 insertions(+), 63 deletions(-)
create mode 100644 drivers/of/.kunitconfig
create mode 100644 drivers/of/empty_root.dts
create mode 100644 drivers/of/of_test.c
Cc: Anton Ivanov <[email protected]>
Cc: Brendan Higgins <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: David Gow <[email protected]>
Cc: Frank Rowand <[email protected]>
Cc: Johannes Berg <[email protected]>
Cc: Richard Weinberger <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Saurabh Sengar <[email protected]>
[1] https://lore.kernel.org/r/[email protected]
base-commit: 0dd3ee31125508cd67f7e7172247f05b7fd1753a
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
We want to populate an empty DT whenever CONFIG_OF is enabled so that
overlays can be applied and the DT unit tests can be run. Make
unflatten_and_copy_device_tree() stop printing a warning if the
'initial_boot_params' pointer is NULL. Instead, simply copy the dtb if
there is one and then unflatten it. If there isn't a DT to copy, then
the call to unflatten_device_tree() is largely a no-op, so nothing
really changes here.
Cc: Rob Herring <[email protected]>
Cc: Frank Rowand <[email protected]>
Signed-off-by: Stephen Boyd <[email protected]>
---
drivers/of/fdt.c | 32 +++++++++++++++++---------------
1 file changed, 17 insertions(+), 15 deletions(-)
diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c
index bf502ba8da95..dfeba8b8ce94 100644
--- a/drivers/of/fdt.c
+++ b/drivers/of/fdt.c
@@ -1318,6 +1318,21 @@ bool __init early_init_dt_scan(void *params)
return true;
}
+static void *__init copy_device_tree(void *fdt)
+{
+ int size;
+ void *dt;
+
+ size = fdt_totalsize(fdt);
+ dt = early_init_dt_alloc_memory_arch(size,
+ roundup_pow_of_two(FDT_V17_SIZE));
+
+ if (dt)
+ memcpy(dt, fdt, size);
+
+ return dt;
+}
+
/**
* unflatten_device_tree - create tree of device_nodes from flat blob
*
@@ -1350,22 +1365,9 @@ void __init unflatten_device_tree(void)
*/
void __init unflatten_and_copy_device_tree(void)
{
- int size;
- void *dt;
+ if (initial_boot_params)
+ initial_boot_params = copy_device_tree(initial_boot_params);
- if (!initial_boot_params) {
- pr_warn("No valid device tree found, continuing without\n");
- return;
- }
-
- size = fdt_totalsize(initial_boot_params);
- dt = early_init_dt_alloc_memory_arch(size,
- roundup_pow_of_two(FDT_V17_SIZE));
-
- if (dt) {
- memcpy(dt, initial_boot_params, size);
- initial_boot_params = dt;
- }
unflatten_device_tree();
}
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
From: Frank Rowand <[email protected]>
When enabling CONFIG_OF on a platform where 'of_root' is not populated
by firmware, we end up without a root node. In order to apply overlays
and create subnodes of the root node, we need one. Create this root node
by unflattening an empty builtin dtb.
If firmware provides a flattened device tree (FDT) then the FDT is
unflattened via setup_arch(). Otherwise, the call to
unflatten(_and_copy)?_device_tree() will create an empty root node.
We make of_have_populated_dt() return true only if the DTB was loaded by
firmware so that existing callers don't change behavior after this
patch. The call in the of platform code is removed because it prevents
overlays from creating platform devices when the platform bus isn't
fully initialized.
Signed-off-by: Frank Rowand <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Cc: Rob Herring <[email protected]>
[[email protected]: Update of_have_populated_dt() to treat this empty dtb
as not populated. Drop setup_of() initcall]
Signed-off-by: Stephen Boyd <[email protected]>
---
drivers/of/Kconfig | 2 +-
drivers/of/Makefile | 2 +-
drivers/of/empty_root.dts | 6 ++++++
drivers/of/fdt.c | 32 +++++++++++++++++++++++++++++++-
drivers/of/platform.c | 3 ---
include/linux/of.h | 25 +++++++++++++++----------
6 files changed, 54 insertions(+), 16 deletions(-)
create mode 100644 drivers/of/empty_root.dts
diff --git a/drivers/of/Kconfig b/drivers/of/Kconfig
index da9826accb1b..17733285b415 100644
--- a/drivers/of/Kconfig
+++ b/drivers/of/Kconfig
@@ -54,7 +54,7 @@ config OF_FLATTREE
select CRC32
config OF_EARLY_FLATTREE
- bool
+ def_bool OF && !SPARC
select DMA_DECLARE_COHERENT if HAS_DMA && HAS_IOMEM
select OF_FLATTREE
diff --git a/drivers/of/Makefile b/drivers/of/Makefile
index eff624854575..df305348d1cb 100644
--- a/drivers/of/Makefile
+++ b/drivers/of/Makefile
@@ -2,7 +2,7 @@
obj-y = base.o cpu.o device.o module.o platform.o property.o
obj-$(CONFIG_OF_KOBJ) += kobj.o
obj-$(CONFIG_OF_DYNAMIC) += dynamic.o
-obj-$(CONFIG_OF_FLATTREE) += fdt.o
+obj-$(CONFIG_OF_FLATTREE) += fdt.o empty_root.dtb.o
obj-$(CONFIG_OF_EARLY_FLATTREE) += fdt_address.o
obj-$(CONFIG_OF_PROMTREE) += pdt.o
obj-$(CONFIG_OF_ADDRESS) += address.o
diff --git a/drivers/of/empty_root.dts b/drivers/of/empty_root.dts
new file mode 100644
index 000000000000..cf9e97a60f48
--- /dev/null
+++ b/drivers/of/empty_root.dts
@@ -0,0 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/dts-v1/;
+
+/ {
+
+};
diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c
index dfeba8b8ce94..e5a385285149 100644
--- a/drivers/of/fdt.c
+++ b/drivers/of/fdt.c
@@ -8,6 +8,7 @@
#define pr_fmt(fmt) "OF: fdt: " fmt
+#include <linux/acpi.h>
#include <linux/crash_dump.h>
#include <linux/crc32.h>
#include <linux/kernel.h>
@@ -32,6 +33,13 @@
#include "of_private.h"
+/*
+ * __dtb_empty_root_begin[] and __dtb_empty_root_end[] magically created by
+ * cmd_dt_S_dtb in scripts/Makefile.lib
+ */
+extern uint8_t __dtb_empty_root_begin[];
+extern uint8_t __dtb_empty_root_end[];
+
/*
* of_fdt_limit_memory - limit the number of regions in the /memory node
* @limit: maximum entries
@@ -1343,7 +1351,29 @@ static void *__init copy_device_tree(void *fdt)
*/
void __init unflatten_device_tree(void)
{
- __unflatten_device_tree(initial_boot_params, NULL, &of_root,
+ void *fdt = initial_boot_params;
+
+ /* Don't use the bootloader provided DTB if ACPI is enabled */
+ if (!acpi_disabled)
+ fdt = NULL;
+
+ /*
+ * Populate an empty root node when ACPI is enabled or bootloader
+ * doesn't provide one.
+ */
+ if (!fdt) {
+ fdt = (void *) __dtb_empty_root_begin;
+ /* fdt_totalsize() will be used for copy size */
+ if (fdt_totalsize(fdt) >
+ __dtb_empty_root_end - __dtb_empty_root_begin) {
+ pr_err("invalid size in dtb_empty_root\n");
+ return;
+ }
+ of_fdt_crc32 = crc32_be(~0, fdt, fdt_totalsize(fdt));
+ fdt = copy_device_tree(fdt);
+ }
+
+ __unflatten_device_tree(fdt, NULL, &of_root,
early_init_dt_alloc_memory_arch, false);
/* Get pointer to "/chosen" and "/aliases" nodes for use everywhere */
diff --git a/drivers/of/platform.c b/drivers/of/platform.c
index 126d265aa7d8..20087bb8a46b 100644
--- a/drivers/of/platform.c
+++ b/drivers/of/platform.c
@@ -549,9 +549,6 @@ static int __init of_platform_default_populate_init(void)
device_links_supplier_sync_state_pause();
- if (!of_have_populated_dt())
- return -ENODEV;
-
if (IS_ENABLED(CONFIG_PPC)) {
struct device_node *boot_display = NULL;
struct platform_device *dev;
diff --git a/include/linux/of.h b/include/linux/of.h
index 6a9ddf20e79a..52f6ad6a1c8c 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -180,11 +180,6 @@ static inline bool is_of_node(const struct fwnode_handle *fwnode)
&__of_fwnode_handle_node->fwnode : NULL; \
})
-static inline bool of_have_populated_dt(void)
-{
- return of_root != NULL;
-}
-
static inline bool of_node_is_root(const struct device_node *node)
{
return node && (node->parent == NULL);
@@ -549,11 +544,6 @@ static inline struct device_node *of_find_node_with_property(
#define of_fwnode_handle(node) NULL
-static inline bool of_have_populated_dt(void)
-{
- return false;
-}
-
static inline struct device_node *of_get_compatible_child(const struct device_node *parent,
const char *compatible)
{
@@ -1634,6 +1624,21 @@ static inline bool of_device_is_system_power_controller(const struct device_node
return of_property_read_bool(np, "system-power-controller");
}
+/**
+ * of_have_populated_dt() - Has DT been populated by bootloader
+ *
+ * Return: True if a DTB has been populated by the bootloader and it isn't the
+ * empty builtin one. False otherwise.
+ */
+static inline bool of_have_populated_dt(void)
+{
+#ifdef CONFIG_OF
+ return of_property_present(of_root, "compatible");
+#else
+ return false;
+#endif
+}
+
/*
* Overlay support
*/
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
Call this function unconditionally so that we can populate an empty DTB
on platforms that don't boot with a command line provided DTB. There's
no harm in calling unflatten_device_tree() unconditionally. If there
isn't a valid initial_boot_params dtb then unflatten_device_tree()
returns early.
Cc: Rob Herring <[email protected]>
Cc: Frank Rowand <[email protected]>
Cc: Richard Weinberger <[email protected]>
Cc: Anton Ivanov <[email protected]>
Cc: Johannes Berg <[email protected]>
Cc: <[email protected]>
Signed-off-by: Stephen Boyd <[email protected]>
---
arch/um/kernel/dtb.c | 14 +++++++-------
drivers/of/unittest.c | 4 ----
2 files changed, 7 insertions(+), 11 deletions(-)
diff --git a/arch/um/kernel/dtb.c b/arch/um/kernel/dtb.c
index 484141b06938..4954188a6a09 100644
--- a/arch/um/kernel/dtb.c
+++ b/arch/um/kernel/dtb.c
@@ -16,16 +16,16 @@ void uml_dtb_init(void)
void *area;
area = uml_load_file(dtb, &size);
- if (!area)
- return;
+ if (area) {
+ if (!early_init_dt_scan(area)) {
+ pr_err("invalid DTB %s\n", dtb);
+ memblock_free(area, size);
+ return;
+ }
- if (!early_init_dt_scan(area)) {
- pr_err("invalid DTB %s\n", dtb);
- memblock_free(area, size);
- return;
+ early_init_fdt_scan_reserved_mem();
}
- early_init_fdt_scan_reserved_mem();
unflatten_device_tree();
}
diff --git a/drivers/of/unittest.c b/drivers/of/unittest.c
index e9e90e96600e..a8b27dd16ecf 100644
--- a/drivers/of/unittest.c
+++ b/drivers/of/unittest.c
@@ -4075,10 +4075,6 @@ static int __init of_unittest(void)
add_taint(TAINT_TEST, LOCKDEP_STILL_OK);
/* adding data for unittest */
-
- if (IS_ENABLED(CONFIG_UML))
- unittest_unflatten_overlay_base();
-
res = unittest_data_add();
if (res)
return res;
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
Call this function unconditionally so that we can populate an empty DTB
on platforms that don't boot with a firmware provided or builtin DTB.
There's no harm in calling unflatten_device_tree() unconditionally here.
If there isn't a non-NULL 'initial_boot_params' pointer then
unflatten_device_tree() returns early.
Cc: Rob Herring <[email protected]>
Cc: Frank Rowand <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Dave Hansen <[email protected]>
Cc: <[email protected]>
Cc: "H. Peter Anvin" <[email protected]>
Cc: Saurabh Sengar <[email protected]>
Signed-off-by: Stephen Boyd <[email protected]>
---
arch/x86/kernel/devicetree.c | 24 +++++++++++++-----------
1 file changed, 13 insertions(+), 11 deletions(-)
diff --git a/arch/x86/kernel/devicetree.c b/arch/x86/kernel/devicetree.c
index afd09924094e..650752d112a6 100644
--- a/arch/x86/kernel/devicetree.c
+++ b/arch/x86/kernel/devicetree.c
@@ -283,22 +283,24 @@ void __init x86_flattree_get_config(void)
u32 size, map_len;
void *dt;
- if (!initial_dtb)
- return;
+ if (initial_dtb) {
+ map_len = max(PAGE_SIZE - (initial_dtb & ~PAGE_MASK), (u64)128);
- map_len = max(PAGE_SIZE - (initial_dtb & ~PAGE_MASK), (u64)128);
+ dt = early_memremap(initial_dtb, map_len);
+ size = fdt_totalsize(dt);
+ if (map_len < size) {
+ early_memunmap(dt, map_len);
+ dt = early_memremap(initial_dtb, size);
+ map_len = size;
+ }
- dt = early_memremap(initial_dtb, map_len);
- size = fdt_totalsize(dt);
- if (map_len < size) {
- early_memunmap(dt, map_len);
- dt = early_memremap(initial_dtb, size);
- map_len = size;
+ early_init_dt_verify(dt);
}
- early_init_dt_verify(dt);
unflatten_and_copy_device_tree();
- early_memunmap(dt, map_len);
+
+ if (initial_dtb)
+ early_memunmap(dt, map_len);
}
#endif
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
Call this function unconditionally so that we can populate an empty DTB
on platforms that don't boot with a firmware provided or builtin DTB.
When ACPI is in use, unflatten_device_tree() ignores the
'initial_boot_params' pointer so the live DT on those systems won't be
whatever that's pointing to. Similarly, when kexec copies the DT data
the previous kernel to the new one on ACPI systems,
of_kexec_alloc_and_setup_fdt() will ignore the live DT (the empty root
one) and copy the 'initial_boot_params' data.
Cc: Rob Herring <[email protected]>
Cc: Frank Rowand <[email protected]>
Cc: Catalin Marinas <[email protected]>
Cc: Will Deacon <[email protected]>
Cc: Mark Rutland <[email protected]>
Cc: <[email protected]>
Signed-off-by: Stephen Boyd <[email protected]>
---
arch/arm64/kernel/setup.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c
index 417a8a86b2db..ede3d59dabf0 100644
--- a/arch/arm64/kernel/setup.c
+++ b/arch/arm64/kernel/setup.c
@@ -351,8 +351,7 @@ void __init __no_sanitize_address setup_arch(char **cmdline_p)
/* Parse the ACPI tables for possible boot-time configuration */
acpi_boot_table_init();
- if (acpi_disabled)
- unflatten_device_tree();
+ unflatten_device_tree();
bootmem_init();
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
From: Frank Rowand <[email protected]>
unflatten_device_tree() now ensures that the 'of_root' node is populated
with the root of a default empty devicetree. Remove the unittest code
that created 'of_root' if it was missing. Verify that 'of_root' is valid
before attempting to attach the testcase-data subtree. Remove the
unittest code that unflattens the unittest overlay base if architecture
is UML because that is always done now.
Signed-off-by: Frank Rowand <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Cc: Rob Herring <[email protected]>
Signed-off-by: Stephen Boyd <[email protected]>
---
drivers/of/unittest.c | 12 ++++--------
1 file changed, 4 insertions(+), 8 deletions(-)
diff --git a/drivers/of/unittest.c b/drivers/of/unittest.c
index a8b27dd16ecf..742d919e8ab4 100644
--- a/drivers/of/unittest.c
+++ b/drivers/of/unittest.c
@@ -1732,20 +1732,16 @@ static int __init unittest_data_add(void)
return -EINVAL;
}
+ /* attach the sub-tree to live tree */
if (!of_root) {
- of_root = unittest_data_node;
- for_each_of_allnodes(np)
- __of_attach_node_sysfs(np);
- of_aliases = of_find_node_by_path("/aliases");
- of_chosen = of_find_node_by_path("/chosen");
- of_overlay_mutex_unlock();
- return 0;
+ pr_warn("%s: no live tree to attach sub-tree\n", __func__);
+ kfree(unittest_data);
+ return -ENODEV;
}
EXPECT_BEGIN(KERN_INFO,
"Duplicate name in testcase-data, renamed to \"duplicate-name#1\"");
- /* attach the sub-tree to live tree */
np = unittest_data_node->child;
while (np) {
struct device_node *next = np->sibling;
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
Add a KUnit test that confirms a DTB has been loaded, i.e. there is a
root node, and that the of_have_populated_dt() API works properly.
Cc: Rob Herring <[email protected]>
Cc: Frank Rowand <[email protected]>
Cc: David Gow <[email protected]>
Cc: Brendan Higgins <[email protected]>
Signed-off-by: Stephen Boyd <[email protected]>
---
drivers/of/.kunitconfig | 3 +++
drivers/of/Kconfig | 9 ++++++++
drivers/of/Makefile | 2 ++
drivers/of/of_test.c | 48 +++++++++++++++++++++++++++++++++++++++++
4 files changed, 62 insertions(+)
create mode 100644 drivers/of/.kunitconfig
create mode 100644 drivers/of/of_test.c
diff --git a/drivers/of/.kunitconfig b/drivers/of/.kunitconfig
new file mode 100644
index 000000000000..5a8fee11978c
--- /dev/null
+++ b/drivers/of/.kunitconfig
@@ -0,0 +1,3 @@
+CONFIG_KUNIT=y
+CONFIG_OF=y
+CONFIG_OF_KUNIT_TEST=y
diff --git a/drivers/of/Kconfig b/drivers/of/Kconfig
index 17733285b415..53d1b5dd89e8 100644
--- a/drivers/of/Kconfig
+++ b/drivers/of/Kconfig
@@ -37,6 +37,15 @@ config OF_UNITTEST
If unsure, say N here. This option is not safe to enable.
+config OF_KUNIT_TEST
+ tristate "Devicetree KUnit Test" if !KUNIT_ALL_TESTS
+ depends on KUNIT
+ default KUNIT_ALL_TESTS
+ help
+ This option builds KUnit unit tests for device tree infrastructure.
+
+ If unsure, say N here, but this option is safe to enable.
+
config OF_ALL_DTBS
bool "Build all Device Tree Blobs"
depends on COMPILE_TEST
diff --git a/drivers/of/Makefile b/drivers/of/Makefile
index df305348d1cb..251d33532148 100644
--- a/drivers/of/Makefile
+++ b/drivers/of/Makefile
@@ -19,4 +19,6 @@ obj-y += kexec.o
endif
endif
+obj-$(CONFIG_OF_KUNIT_TEST) += of_test.o
+
obj-$(CONFIG_OF_UNITTEST) += unittest-data/
diff --git a/drivers/of/of_test.c b/drivers/of/of_test.c
new file mode 100644
index 000000000000..71a767b42b43
--- /dev/null
+++ b/drivers/of/of_test.c
@@ -0,0 +1,48 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * KUnit tests for OF APIs
+ */
+#include <linux/module.h>
+#include <linux/of.h>
+
+#include <kunit/test.h>
+
+/*
+ * Test that the root node "/" can be found by path.
+ */
+static void dtb_root_node_found_by_path(struct kunit *test)
+{
+ struct device_node *np;
+
+ np = of_find_node_by_path("/");
+ KUNIT_EXPECT_NOT_ERR_OR_NULL(test, np);
+ of_node_put(np);
+}
+
+/*
+ * Test that the 'of_root' global variable is always populated when DT code is
+ * enabled.
+ */
+static void dtb_root_node_populates_of_root(struct kunit *test)
+{
+ KUNIT_EXPECT_NOT_ERR_OR_NULL(test, of_root);
+}
+
+static struct kunit_case dtb_test_cases[] = {
+ KUNIT_CASE(dtb_root_node_found_by_path),
+ KUNIT_CASE(dtb_root_node_populates_of_root),
+ {}
+};
+
+/*
+ * Test suite to confirm a DTB is loaded.
+ */
+static struct kunit_suite dtb_suite = {
+ .name = "dtb",
+ .test_cases = dtb_test_cases,
+};
+
+kunit_test_suites(
+ &dtb_suite,
+);
+MODULE_LICENSE("GPL");
--
https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
On Sat, 3 Feb 2024 at 03:59, Stephen Boyd <[email protected]> wrote:
>
> Add a KUnit test that confirms a DTB has been loaded, i.e. there is a
> root node, and that the of_have_populated_dt() API works properly.
>
> Cc: Rob Herring <[email protected]>
> Cc: Frank Rowand <[email protected]>
> Cc: David Gow <[email protected]>
> Cc: Brendan Higgins <[email protected]>
> Signed-off-by: Stephen Boyd <[email protected]>
> ---
This looks pretty good to me test-wise, though it still fails on m68k.
(Everything else I tried it on works, though I've definitely not tried
_every_ architecture.)
aarch64: PASSED
i386: PASSED
x86_64: PASSED
x86_64 KASAN: PASSED
powerpc64: PASSED
UML: PASSED
UML LLVM: PASSED
m68k: FAILED
> $ qemu-system-m68k -nodefaults -m 1024 -kernel .kunit-all-m68k/vmlinux -append 'kunit.enable=1 console=hvc0 kunit_shutdown=reboot' -no-reboot -nographic -serial stdio -machine virt
> [11:55:05] ===================== dtb (2 subtests) =====================
> [11:55:05] # dtb_root_node_found_by_path: EXPECTATION FAILED at drivers/of/of_test.c:18
> [11:55:05] Expected np is not null, but is
> [11:55:05] [FAILED] dtb_root_node_found_by_path
> [11:55:05] # dtb_root_node_populates_of_root: EXPECTATION FAILED at drivers/of/of_test.c:28
> [11:55:05] Expected of_root is not null, but is
> [11:55:05] [FAILED] dtb_root_node_populates_of_root
> [11:55:05] # module: of_test
> [11:55:05] # dtb: pass:0 fail:2 skip:0 total:2
> [11:55:05] # Totals: pass:0 fail:2 skip:0 total:2
> [11:55:05] ======================= [FAILED] dtb =======================
My only other question is about the test names: the mix of 'of' and
'dtb' can be a bit confusing. As is, we have:
kconfig name: OF_KUNIT_TEST
module name: of_test
suite name: dtb
test names: all start with dtb_
Given KUnit only really deals with the suite/test names directly, it's
not trivial to see that 'dtb.dtb_*' is controlled by OF_KUNIT_TEST and
in of_test if built as a module. (This is getting a bit easier now
that we have the 'module' attribute in the output, but still.)
Would 'of_dtb' work as a suite name if it's important to keep both
'of' and 'dtb'?
In general, though, this test looks good to me. Particularly if m68k
can be fixed.
Reviewed-by: David Gow <[email protected]>
Cheers,
-- David
> drivers/of/.kunitconfig | 3 +++
> drivers/of/Kconfig | 9 ++++++++
> drivers/of/Makefile | 2 ++
> drivers/of/of_test.c | 48 +++++++++++++++++++++++++++++++++++++++++
> 4 files changed, 62 insertions(+)
> create mode 100644 drivers/of/.kunitconfig
> create mode 100644 drivers/of/of_test.c
>
> diff --git a/drivers/of/.kunitconfig b/drivers/of/.kunitconfig
> new file mode 100644
> index 000000000000..5a8fee11978c
> --- /dev/null
> +++ b/drivers/of/.kunitconfig
> @@ -0,0 +1,3 @@
> +CONFIG_KUNIT=y
> +CONFIG_OF=y
> +CONFIG_OF_KUNIT_TEST=y
> diff --git a/drivers/of/Kconfig b/drivers/of/Kconfig
> index 17733285b415..53d1b5dd89e8 100644
> --- a/drivers/of/Kconfig
> +++ b/drivers/of/Kconfig
> @@ -37,6 +37,15 @@ config OF_UNITTEST
>
> If unsure, say N here. This option is not safe to enable.
>
> +config OF_KUNIT_TEST
> + tristate "Devicetree KUnit Test" if !KUNIT_ALL_TESTS
> + depends on KUNIT
> + default KUNIT_ALL_TESTS
> + help
> + This option builds KUnit unit tests for device tree infrastructure.
> +
> + If unsure, say N here, but this option is safe to enable.
> +
> config OF_ALL_DTBS
> bool "Build all Device Tree Blobs"
> depends on COMPILE_TEST
> diff --git a/drivers/of/Makefile b/drivers/of/Makefile
> index df305348d1cb..251d33532148 100644
> --- a/drivers/of/Makefile
> +++ b/drivers/of/Makefile
> @@ -19,4 +19,6 @@ obj-y += kexec.o
> endif
> endif
>
> +obj-$(CONFIG_OF_KUNIT_TEST) += of_test.o
> +
> obj-$(CONFIG_OF_UNITTEST) += unittest-data/
> diff --git a/drivers/of/of_test.c b/drivers/of/of_test.c
> new file mode 100644
> index 000000000000..71a767b42b43
> --- /dev/null
> +++ b/drivers/of/of_test.c
> @@ -0,0 +1,48 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * KUnit tests for OF APIs
> + */
> +#include <linux/module.h>
> +#include <linux/of.h>
> +
> +#include <kunit/test.h>
> +
> +/*
> + * Test that the root node "/" can be found by path.
> + */
> +static void dtb_root_node_found_by_path(struct kunit *test)
> +{
> + struct device_node *np;
> +
> + np = of_find_node_by_path("/");
> + KUNIT_EXPECT_NOT_ERR_OR_NULL(test, np);
> + of_node_put(np);
> +}
> +
> +/*
> + * Test that the 'of_root' global variable is always populated when DT code is
> + * enabled.
> + */
> +static void dtb_root_node_populates_of_root(struct kunit *test)
> +{
> + KUNIT_EXPECT_NOT_ERR_OR_NULL(test, of_root);
> +}
> +
> +static struct kunit_case dtb_test_cases[] = {
> + KUNIT_CASE(dtb_root_node_found_by_path),
> + KUNIT_CASE(dtb_root_node_populates_of_root),
> + {}
> +};
> +
> +/*
> + * Test suite to confirm a DTB is loaded.
> + */
> +static struct kunit_suite dtb_suite = {
> + .name = "dtb",
> + .test_cases = dtb_test_cases,
> +};
> +
> +kunit_test_suites(
> + &dtb_suite,
> +);
> +MODULE_LICENSE("GPL");
> --
> https://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git/
> https://git.kernel.org/pub/scm/linux/kernel/git/sboyd/spmi.git
>
> --
> You received this message because you are subscribed to the Google Groups "KUnit Development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
> To view this discussion on the web visit https://groups.google.com/d/msgid/kunit-dev/20240202195909.3458162-8-sboyd%40kernel.org.
Quoting David Gow (2024-02-02 20:10:17)
> On Sat, 3 Feb 2024 at 03:59, Stephen Boyd <[email protected]> wrote:
> >
> > Add a KUnit test that confirms a DTB has been loaded, i.e. there is a
> > root node, and that the of_have_populated_dt() API works properly.
> >
> > Cc: Rob Herring <[email protected]>
> > Cc: Frank Rowand <[email protected]>
> > Cc: David Gow <[email protected]>
> > Cc: Brendan Higgins <[email protected]>
> > Signed-off-by: Stephen Boyd <[email protected]>
> > ---
>
> This looks pretty good to me test-wise, though it still fails on m68k.
> (Everything else I tried it on works, though I've definitely not tried
> _every_ architecture.)
>
> aarch64: PASSED
> i386: PASSED
> x86_64: PASSED
> x86_64 KASAN: PASSED
> powerpc64: PASSED
> UML: PASSED
> UML LLVM: PASSED
> m68k: FAILED
> > $ qemu-system-m68k -nodefaults -m 1024 -kernel .kunit-all-m68k/vmlinux -append 'kunit.enable=1 console=hvc0 kunit_shutdown=reboot' -no-reboot -nographic -serial stdio -machine virt
> > [11:55:05] ===================== dtb (2 subtests) =====================
> > [11:55:05] # dtb_root_node_found_by_path: EXPECTATION FAILED at drivers/of/of_test.c:18
> > [11:55:05] Expected np is not null, but is
> > [11:55:05] [FAILED] dtb_root_node_found_by_path
> > [11:55:05] # dtb_root_node_populates_of_root: EXPECTATION FAILED at drivers/of/of_test.c:28
> > [11:55:05] Expected of_root is not null, but is
> > [11:55:05] [FAILED] dtb_root_node_populates_of_root
> > [11:55:05] # module: of_test
> > [11:55:05] # dtb: pass:0 fail:2 skip:0 total:2
> > [11:55:05] # Totals: pass:0 fail:2 skip:0 total:2
> > [11:55:05] ======================= [FAILED] dtb =======================
Ah yeah I forgot to mention that. m68k fails because it doesn't call the
unflatten_(and_copy)?_device_tree() function, so we don't populate a
root node on that architecture. One solution would be to make CONFIG_OF
unavailable on m68k. Or we have to make sure DT works on any
architecture. Rob, what do you prefer here?
>
>
> My only other question is about the test names: the mix of 'of' and
> 'dtb' can be a bit confusing. As is, we have:
> kconfig name: OF_KUNIT_TEST
> module name: of_test
> suite name: dtb
> test names: all start with dtb_
>
> Given KUnit only really deals with the suite/test names directly, it's
> not trivial to see that 'dtb.dtb_*' is controlled by OF_KUNIT_TEST and
> in of_test if built as a module. (This is getting a bit easier now
> that we have the 'module' attribute in the output, but still.)
>
> Would 'of_dtb' work as a suite name if it's important to keep both
> 'of' and 'dtb'?
Sure, I can add of_ prefix to the tests.
>
> In general, though, this test looks good to me. Particularly if m68k
> can be fixed.
>
> Reviewed-by: David Gow <[email protected]>
>
Thanks!
Hi Stephen,
On Mon, Feb 5, 2024 at 8:19 PM Stephen Boyd <[email protected]> wrote:
> Quoting David Gow (2024-02-02 20:10:17)
> > On Sat, 3 Feb 2024 at 03:59, Stephen Boyd <[email protected]> wrote:
> > > Add a KUnit test that confirms a DTB has been loaded, i.e. there is a
> > > root node, and that the of_have_populated_dt() API works properly.
> > >
> > > Cc: Rob Herring <[email protected]>
> > > Cc: Frank Rowand <[email protected]>
> > > Cc: David Gow <[email protected]>
> > > Cc: Brendan Higgins <[email protected]>
> > > Signed-off-by: Stephen Boyd <[email protected]>
> > > ---
> >
> > This looks pretty good to me test-wise, though it still fails on m68k.
> > (Everything else I tried it on works, though I've definitely not tried
> > _every_ architecture.)
> >
> > aarch64: PASSED
> > i386: PASSED
> > x86_64: PASSED
> > x86_64 KASAN: PASSED
> > powerpc64: PASSED
> > UML: PASSED
> > UML LLVM: PASSED
> > m68k: FAILED
> > > $ qemu-system-m68k -nodefaults -m 1024 -kernel .kunit-all-m68k/vmlinux -append 'kunit.enable=1 console=hvc0 kunit_shutdown=reboot' -no-reboot -nographic -serial stdio -machine virt
> > > [11:55:05] ===================== dtb (2 subtests) =====================
> > > [11:55:05] # dtb_root_node_found_by_path: EXPECTATION FAILED at drivers/of/of_test.c:18
> > > [11:55:05] Expected np is not null, but is
> > > [11:55:05] [FAILED] dtb_root_node_found_by_path
> > > [11:55:05] # dtb_root_node_populates_of_root: EXPECTATION FAILED at drivers/of/of_test.c:28
> > > [11:55:05] Expected of_root is not null, but is
> > > [11:55:05] [FAILED] dtb_root_node_populates_of_root
> > > [11:55:05] # module: of_test
> > > [11:55:05] # dtb: pass:0 fail:2 skip:0 total:2
> > > [11:55:05] # Totals: pass:0 fail:2 skip:0 total:2
> > > [11:55:05] ======================= [FAILED] dtb =======================
>
> Ah yeah I forgot to mention that. m68k fails because it doesn't call the
> unflatten_(and_copy)?_device_tree() function, so we don't populate a
> root node on that architecture. One solution would be to make CONFIG_OF
> unavailable on m68k. Or we have to make sure DT works on any
> architecture. Rob, what do you prefer here?
I guess the latter?
Alpha, hexagon, parisc, s390, and sparc are also lacking calls
to unflatten.*device_tree().
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68korg
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Quoting Geert Uytterhoeven (2024-02-05 11:55:29)
> On Mon, Feb 5, 2024 at 8:19 PM Stephen Boyd <[email protected]> wrote:
> > Quoting David Gow (2024-02-02 20:10:17)
> > > On Sat, 3 Feb 2024 at 03:59, Stephen Boyd <[email protected]> wrote:
> > > > Add a KUnit test that confirms a DTB has been loaded, i.e. there is a
> > > > root node, and that the of_have_populated_dt() API works properly.
> > > >
> > > > Cc: Rob Herring <[email protected]>
> > > > Cc: Frank Rowand <[email protected]>
> > > > Cc: David Gow <[email protected]>
> > > > Cc: Brendan Higgins <[email protected]>
> > > > Signed-off-by: Stephen Boyd <[email protected]>
> > > > ---
> > >
> > > This looks pretty good to me test-wise, though it still fails on m68k.
> > > (Everything else I tried it on works, though I've definitely not tried
> > > _every_ architecture.)
> > >
> > > aarch64: PASSED
> > > i386: PASSED
> > > x86_64: PASSED
> > > x86_64 KASAN: PASSED
> > > powerpc64: PASSED
> > > UML: PASSED
> > > UML LLVM: PASSED
> > > m68k: FAILED
> > > > $ qemu-system-m68k -nodefaults -m 1024 -kernel .kunit-all-m68k/vmlinux -append 'kunit.enable=1 console=hvc0 kunit_shutdown=reboot' -no-reboot -nographic -serial stdio -machine virt
> > > > [11:55:05] ===================== dtb (2 subtests) =====================
> > > > [11:55:05] # dtb_root_node_found_by_path: EXPECTATION FAILED at drivers/of/of_test.c:18
> > > > [11:55:05] Expected np is not null, but is
> > > > [11:55:05] [FAILED] dtb_root_node_found_by_path
> > > > [11:55:05] # dtb_root_node_populates_of_root: EXPECTATION FAILED at drivers/of/of_test.c:28
> > > > [11:55:05] Expected of_root is not null, but is
> > > > [11:55:05] [FAILED] dtb_root_node_populates_of_root
> > > > [11:55:05] # module: of_test
> > > > [11:55:05] # dtb: pass:0 fail:2 skip:0 total:2
> > > > [11:55:05] # Totals: pass:0 fail:2 skip:0 total:2
> > > > [11:55:05] ======================= [FAILED] dtb =======================
> >
> > Ah yeah I forgot to mention that. m68k fails because it doesn't call the
> > unflatten_(and_copy)?_device_tree() function, so we don't populate a
> > root node on that architecture. One solution would be to make CONFIG_OF
> > unavailable on m68k. Or we have to make sure DT works on any
> > architecture. Rob, what do you prefer here?
>
> I guess the latter?
> Alpha, hexagon, parisc, s390, and sparc are also lacking calls
> to unflatten.*device_tree().
>
sparc does that on purpose. Perhaps it's simplest to call
unflatten_device_tree() if of_root is still NULL after setup_arch()
returns.
---8<---
diff --git a/init/main.c b/init/main.c
index e24b0780fdff..02f5cf8be6c1 100644
--- a/init/main.c
+++ b/init/main.c
@@ -97,6 +97,8 @@
#include <linux/jump_label.h>
#include <linux/kcsan.h>
#include <linux/init_syscalls.h>
+#include <linux/of.h>
+#include <linux/of_fdt.h>
#include <linux/stackdepot.h>
#include <linux/randomize_kstack.h>
#include <net/net_namespace.h>
@@ -895,6 +897,8 @@ void start_kernel(void)
pr_notice("%s", linux_banner);
early_security_init();
setup_arch(&command_line);
+ if (!of_root)
+ unflatten_device_tree();
setup_boot_config();
setup_command_line(command_line);
setup_nr_cpu_ids();
On Fri, Feb 9, 2024 at 8:59 PM Stephen Boyd <[email protected]> wrote:
>
> Quoting Geert Uytterhoeven (2024-02-05 11:55:29)
> > On Mon, Feb 5, 2024 at 8:19 PM Stephen Boyd <[email protected]> wrote:
> > > Quoting David Gow (2024-02-02 20:10:17)
> > > > On Sat, 3 Feb 2024 at 03:59, Stephen Boyd <[email protected]> wrote:
> > > > > Add a KUnit test that confirms a DTB has been loaded, i.e. there is a
> > > > > root node, and that the of_have_populated_dt() API works properly.
> > > > >
> > > > > Cc: Rob Herring <[email protected]>
> > > > > Cc: Frank Rowand <[email protected]>
> > > > > Cc: David Gow <[email protected]>
> > > > > Cc: Brendan Higgins <[email protected]>
> > > > > Signed-off-by: Stephen Boyd <[email protected]>
> > > > > ---
> > > >
> > > > This looks pretty good to me test-wise, though it still fails on m68k.
> > > > (Everything else I tried it on works, though I've definitely not tried
> > > > _every_ architecture.)
> > > >
> > > > aarch64: PASSED
> > > > i386: PASSED
> > > > x86_64: PASSED
> > > > x86_64 KASAN: PASSED
> > > > powerpc64: PASSED
> > > > UML: PASSED
> > > > UML LLVM: PASSED
> > > > m68k: FAILED
> > > > > $ qemu-system-m68k -nodefaults -m 1024 -kernel .kunit-all-m68k/vmlinux -append 'kunit.enable=1 console=hvc0 kunit_shutdown=reboot' -no-reboot -nographic -serial stdio -machine virt
> > > > > [11:55:05] ===================== dtb (2 subtests) =====================
> > > > > [11:55:05] # dtb_root_node_found_by_path: EXPECTATION FAILED at drivers/of/of_test.c:18
> > > > > [11:55:05] Expected np is not null, but is
> > > > > [11:55:05] [FAILED] dtb_root_node_found_by_path
> > > > > [11:55:05] # dtb_root_node_populates_of_root: EXPECTATION FAILED at drivers/of/of_test.c:28
> > > > > [11:55:05] Expected of_root is not null, but is
> > > > > [11:55:05] [FAILED] dtb_root_node_populates_of_root
> > > > > [11:55:05] # module: of_test
> > > > > [11:55:05] # dtb: pass:0 fail:2 skip:0 total:2
> > > > > [11:55:05] # Totals: pass:0 fail:2 skip:0 total:2
> > > > > [11:55:05] ======================= [FAILED] dtb =======================
> > >
> > > Ah yeah I forgot to mention that. m68k fails because it doesn't call the
> > > unflatten_(and_copy)?_device_tree() function, so we don't populate a
> > > root node on that architecture. One solution would be to make CONFIG_OF
> > > unavailable on m68k. Or we have to make sure DT works on any
> > > architecture. Rob, what do you prefer here?
> >
> > I guess the latter?
> > Alpha, hexagon, parisc, s390, and sparc are also lacking calls
> > to unflatten.*device_tree().
> >
>
> sparc does that on purpose. Perhaps it's simplest to call
> unflatten_device_tree() if of_root is still NULL after setup_arch()
> returns.
>
> ---8<---
> diff --git a/init/main.c b/init/main.c
> index e24b0780fdff..02f5cf8be6c1 100644
> --- a/init/main.c
> +++ b/init/main.c
> @@ -97,6 +97,8 @@
> #include <linux/jump_label.h>
> #include <linux/kcsan.h>
> #include <linux/init_syscalls.h>
> +#include <linux/of.h>
> +#include <linux/of_fdt.h>
> #include <linux/stackdepot.h>
> #include <linux/randomize_kstack.h>
> #include <net/net_namespace.h>
> @@ -895,6 +897,8 @@ void start_kernel(void)
> pr_notice("%s", linux_banner);
> early_security_init();
> setup_arch(&command_line);
> + if (!of_root)
of_root is another thing I'd like to remove direct access to. That
check could be inside unflatten_device_tree().
> + unflatten_device_tree();
That's back to what Frank had essentially and I wanted to avoid.
I think I'd just disable the tests on the above arches and let them
opt-in. I could be convinced otherwise though.
Rob
Quoting Rob Herring (2024-02-13 09:52:00)
> On Fri, Feb 9, 2024 at 8:59 PM Stephen Boyd <[email protected]> wrote:
> >
> > ---8<---
> > diff --git a/init/main.c b/init/main.c
> > index e24b0780fdff..02f5cf8be6c1 100644
> > --- a/init/main.c
> > +++ b/init/main.c
> > @@ -97,6 +97,8 @@
> > #include <linux/jump_label.h>
> > #include <linux/kcsan.h>
> > #include <linux/init_syscalls.h>
> > +#include <linux/of.h>
> > +#include <linux/of_fdt.h>
> > #include <linux/stackdepot.h>
> > #include <linux/randomize_kstack.h>
> > #include <net/net_namespace.h>
> > @@ -895,6 +897,8 @@ void start_kernel(void)
> > pr_notice("%s", linux_banner);
> > early_security_init();
> > setup_arch(&command_line);
> > + if (!of_root)
>
> of_root is another thing I'd like to remove direct access to. That
> check could be inside unflatten_device_tree().
Ok.
>
> > + unflatten_device_tree();
>
> That's back to what Frank had essentially and I wanted to avoid.
Alright, fair enough.
>
> I think I'd just disable the tests on the above arches and let them
> opt-in. I could be convinced otherwise though.
>
Kunit folks would prefer to skip tests when dependencies aren't
satisfied. The OF_UNITTEST config already depends on !SPARC so perhaps
it's simplest to have tests skip if OF_EARLY_FLATREE=n. Then
OF_EARLY_FLATREE can be def_bool OF && !(SPARC || M68K || other arches).
The OF_UNITTEST config can depend on OF_EARLY_FLATREE instead of select
it then. This way new supporting architectures can remove themselves
from the def_bool line when they start calling unflatten_device_tree().