Skip to contentSkip to conversation

of: fix sparse warnings in of_find_next_cache_node

PATCH In review Rob Herring 2 participants 6 messages May 4, 2017 – May 5, 2017

Subsystems (2)
lore
Share & tools
lore

Rob Herring

author maintainer
#0 lore raw

PATCH 1/3 of: fix sparse warnings in of_find_next_cache_node

sparse gives a warning that 'handle' is not a __be32:

../drivers/of/base.c:2261:61: warning: incorrect type in argument 1 (different base types)
../drivers/of/base.c:2261:61: expected restricted __be32 const [usertype] *p
../drivers/of/base.c:2261:61: got unsigned int const [usertype] *[assigned] handle

We could just change the type, but the code can be improved by using of_parse_phandle instead of open coding it with of_get_property and of_find_node_by_phandle.

---
 drivers/of/base.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)
drivers/of/base.c +6 -7
index d7c4629a3a2d..016f9d77d64d 100644
@@ -2250,15 +2250,14 @@ EXPORT_SYMBOL_GPL(of_console_check);
  */
 struct device_node *of_find_next_cache_node(const struct device_node *np)
 {
-	struct device_node *child;
-	const phandle *handle;
+	struct device_node *child, *cache_node;
 
-	handle = of_get_property(np, "l2-cache", NULL);
-	if (!handle)
-		handle = of_get_property(np, "next-level-cache", NULL);
+	cache_node = of_parse_phandle(np, "l2-cache", 0);
+	if (!cache_node)
+		cache_node = of_parse_phandle(np, "next-level-cache", 0);
 
-	if (handle)
-		return of_find_node_by_phandle(be32_to_cpup(handle));
+	if (cache_node)
+		return cache_node;
 
 	/* OF on pmac has nodes instead of properties named "l2-cache"
 	 * beneath CPU nodes.
Signature · 6 lines
2.11.0

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author maintainer
#1 lore raw

PATCH 2/3 of: fix sparse warning in of_pci_range_parser_one

sparse gives the following warning for 'pci_space':

../drivers/of/address.c:266:26: warning: incorrect type in assignment (different base types)
../drivers/of/address.c:266:26: expected unsigned int [unsigned] [usertype] pci_space
../drivers/of/address.c:266:26: got restricted __be32 const [usertype] <noident>

It appears that pci_space is only ever accessed on powerpc, so the endian swap is often not needed.

---
 drivers/of/address.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/of/address.c +1 -1
index 02b2903fe9d2..72914cdfce2a 100644
@@ -263,7 +263,7 @@ struct of_pci_range *of_pci_range_parser_one(struct of_pci_range_parser *parser,
 	if (!parser->range || parser->range + parser->np > parser->end)
 		return NULL;
 
-	range->pci_space = parser->range[0];
+	range->pci_space = be32_to_cpup(parser->range);
 	range->flags = of_bus_pci_get_flags(parser->range);
 	range->pci_addr = of_read_number(parser->range + 1, ns);
 	range->cpu_addr = of_translate_address(parser->node,
Signature · 1 line
2.11.0
Reviewed-by on v1 2/3
#2 lore raw

Re: [PATCH 2/3] of: fix sparse warning in of_pci_range_parser_one

Quoted 21 lines · Rob Herring

sparse gives the following warning for 'pci_space':

../drivers/of/address.c:266:26: warning: incorrect type in assignment (different base types)
../drivers/of/address.c:266:26: expected unsigned int [unsigned] [usertype] pci_space
../drivers/of/address.c:266:26: got restricted __be32 const [usertype] <noident>

It appears that pci_space is only ever accessed on powerpc, so the endian swap is often not needed.

---
 drivers/of/address.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/of/address.c +0 -0
index 02b2903fe9d2..72914cdfce2a 100644

drivers/of/address.c struct of_pci_range *of_pci_range_parser_one(struct of_pci_range_parser *parser, · line 269

Quoted 8 lines
drivers/of/address.c +1 -1
@@ -263,7 +263,7 @@ struct of_pci_range *of_pci_range_parser_one(struct of_pci_range_parser *parser,
 	if (!parser->range || parser->range + parser->np > parser->end)
 		return NULL;
 
-	range->pci_space = parser->range[0];
+	range->pci_space = be32_to_cpup(parser->range);
 	range->flags = of_bus_pci_get_flags(parser->range);
 	range->pci_addr = of_read_number(parser->range + 1, ns);
 	range->cpu_addr = of_translate_address(parser->node,

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

author maintainer
#3 lore raw

PATCH 3/3 of: fix sparse warnings in fdt, irq, and resolver code

In reply to Rob Herring

sparse generates the following warnings in drivers/of/:

../drivers/of/fdt.c:63:36: warning: cast to restricted __be32 ../drivers/of/fdt.c:68:33: warning: cast to restricted __be32 ../drivers/of/irq.c:105:88: warning: incorrect type in initializer (different base types)
../drivers/of/irq.c:105:88: expected restricted __be32
../drivers/of/irq.c:105:88: got int
../drivers/of/irq.c:526:35: warning: incorrect type in assignment (different modifiers)
../drivers/of/irq.c:526:35: expected int ( *const [usertype] irq_init_cb )( ... )
../drivers/of/irq.c:526:35: got void const *const data
../drivers/of/resolver.c:95:42: warning: incorrect type in assignment (different base types)
../drivers/of/resolver.c:95:42: expected unsigned int [unsigned] [usertype] <noident>
../drivers/of/resolver.c:95:42: got restricted __be32 [usertype] <noident>

All these are harmless type mismatches fixed by adjusting the types.

---
 drivers/of/fdt.c       | 4 ++--
 drivers/of/irq.c       | 2 +-
 drivers/of/resolver.c  | 2 +-
 include/linux/of_irq.h | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)
drivers/of/fdt.c +2 -2
index e33f7818bc6c..a0972219ccfc 100644
@@ -48,8 +48,8 @@ void of_fdt_limit_memory(int limit)
 	const void *val;
 	int nr_address_cells = OF_ROOT_NODE_ADDR_CELLS_DEFAULT;
 	int nr_size_cells = OF_ROOT_NODE_SIZE_CELLS_DEFAULT;
-	const uint32_t *addr_prop;
-	const uint32_t *size_prop;
+	const __be32 *addr_prop;
+	const __be32 *size_prop;
 	int root_offset;
 	int cell_size;
 
drivers/of/irq.c +1 -1
index 7c56b72d1dc6..d11437cb1187 100644
@@ -102,7 +102,7 @@ int of_irq_parse_raw(const __be32 *addr, struct of_phandle_args *out_irq)
 	struct device_node *ipar, *tnode, *old = NULL, *newpar = NULL;
 	__be32 initial_match_array[MAX_PHANDLE_ARGS];
 	const __be32 *match_array = initial_match_array;
-	const __be32 *tmp, *imap, *imask, dummy_imask[] = { [0 ... MAX_PHANDLE_ARGS] = ~0 };
+	const __be32 *tmp, *imap, *imask, dummy_imask[] = { [0 ... MAX_PHANDLE_ARGS] = cpu_to_be32(~0) };
 	u32 intsize = 1, addrsize, newintsize = 0, newaddrsize = 0;
 	int imaplen, match, i, rc = -EINVAL;
 
drivers/of/resolver.c +1 -1
index 7ae9863cb0a4..771f4844c781 100644
@@ -92,7 +92,7 @@ static void adjust_overlay_phandles(struct device_node *overlay,
 		if (phandle == OF_PHANDLE_ILLEGAL)
 			continue;
 
-		*(uint32_t *)prop->value = cpu_to_be32(overlay->phandle);
+		*(__be32 *)prop->value = cpu_to_be32(overlay->phandle);
 	}
 
 	for_each_child_of_node(overlay, child)
include/linux/of_irq.h +1 -1
index 1e0deb8e8494..ec6b11deb773 100644
@@ -8,7 +8,7 @@
 #include <linux/ioport.h>
 #include <linux/of.h>
 
-typedef int (*of_irq_init_cb_t)(struct device_node *, struct device_node *);
+typedef int const (*of_irq_init_cb_t)(struct device_node *, struct device_node *);
 
 /*
  * Workarounds only applied to 32bit powermac machines
Signature · 6 lines
2.11.0

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Reviewed-by on v1 3/3
#4 lore raw

Re: [PATCH 3/3] of: fix sparse warnings in fdt, irq, and resolver code

Quoted 71 lines · Rob Herring

sparse generates the following warnings in drivers/of/:

../drivers/of/fdt.c:63:36: warning: cast to restricted __be32 ../drivers/of/fdt.c:68:33: warning: cast to restricted __be32 ../drivers/of/irq.c:105:88: warning: incorrect type in initializer (different base types)
../drivers/of/irq.c:105:88: expected restricted __be32
../drivers/of/irq.c:105:88: got int
../drivers/of/irq.c:526:35: warning: incorrect type in assignment (different modifiers)
../drivers/of/irq.c:526:35: expected int ( *const [usertype] irq_init_cb )( ... )
../drivers/of/irq.c:526:35: got void const *const data
../drivers/of/resolver.c:95:42: warning: incorrect type in assignment (different base types)
../drivers/of/resolver.c:95:42: expected unsigned int [unsigned] [usertype] <noident>
../drivers/of/resolver.c:95:42: got restricted __be32 [usertype] <noident>

All these are harmless type mismatches fixed by adjusting the types.

---
 drivers/of/fdt.c       | 4 ++--
 drivers/of/irq.c       | 2 +-
 drivers/of/resolver.c  | 2 +-
 include/linux/of_irq.h | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)
drivers/of/fdt.c +2 -2
index e33f7818bc6c..a0972219ccfc 100644
@@ -48,8 +48,8 @@ void of_fdt_limit_memory(int limit)
 	const void *val;
 	int nr_address_cells = OF_ROOT_NODE_ADDR_CELLS_DEFAULT;
 	int nr_size_cells = OF_ROOT_NODE_SIZE_CELLS_DEFAULT;
-	const uint32_t *addr_prop;
-	const uint32_t *size_prop;
+	const __be32 *addr_prop;
+	const __be32 *size_prop;
 	int root_offset;
 	int cell_size;
 
drivers/of/irq.c +1 -1
index 7c56b72d1dc6..d11437cb1187 100644
@@ -102,7 +102,7 @@ int of_irq_parse_raw(const __be32 *addr, struct of_phandle_args *out_irq)
 	struct device_node *ipar, *tnode, *old = NULL, *newpar = NULL;
 	__be32 initial_match_array[MAX_PHANDLE_ARGS];
 	const __be32 *match_array = initial_match_array;
-	const __be32 *tmp, *imap, *imask, dummy_imask[] = { [0 ... MAX_PHANDLE_ARGS] = ~0 };
+	const __be32 *tmp, *imap, *imask, dummy_imask[] = { [0 ... MAX_PHANDLE_ARGS] = cpu_to_be32(~0) };
 	u32 intsize = 1, addrsize, newintsize = 0, newaddrsize = 0;
 	int imaplen, match, i, rc = -EINVAL;
 
drivers/of/resolver.c +1 -1
index 7ae9863cb0a4..771f4844c781 100644
@@ -92,7 +92,7 @@ static void adjust_overlay_phandles(struct device_node *overlay,
 		if (phandle == OF_PHANDLE_ILLEGAL)
 			continue;
 
-		*(uint32_t *)prop->value = cpu_to_be32(overlay->phandle);
+		*(__be32 *)prop->value = cpu_to_be32(overlay->phandle);
 	}
 
 	for_each_child_of_node(overlay, child)
include/linux/of_irq.h +0 -0
index 1e0deb8e8494..ec6b11deb773 100644

include/linux/of_irq.h line 14

Quoted 8 lines
include/linux/of_irq.h +1 -1
@@ -8,7 +8,7 @@
 #include <linux/ioport.h>
 #include <linux/of.h>
 
-typedef int (*of_irq_init_cb_t)(struct device_node *, struct device_node *);
+typedef int const (*of_irq_init_cb_t)(struct device_node *, struct device_node *);
 
 /*
  * Workarounds only applied to 32bit powermac machines

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Reviewed-by on v1 1/3
#5 lore raw
In reply to Rob Herring
Quoted 32 lines · Rob Herring

sparse gives a warning that 'handle' is not a __be32:

../drivers/of/base.c:2261:61: warning: incorrect type in argument 1 (different base types)
../drivers/of/base.c:2261:61: expected restricted __be32 const [usertype] *p
../drivers/of/base.c:2261:61: got unsigned int const [usertype] *[assigned] handle

We could just change the type, but the code can be improved by using of_parse_phandle instead of open coding it with of_get_property and of_find_node_by_phandle.

---
 drivers/of/base.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)
drivers/of/base.c +2 -5
index d7c4629a3a2d..016f9d77d64d 100644
@@ -2250,15 +2250,14 @@ EXPORT_SYMBOL_GPL(of_console_check);
  */
 struct device_node *of_find_next_cache_node(const struct device_node *np)
 {
-	struct device_node *child;
-	const phandle *handle;
+	struct device_node *child, *cache_node;
 
-	handle = of_get_property(np, "l2-cache", NULL);
-	if (!handle)
-		handle = of_get_property(np, "next-level-cache", NULL);
+	cache_node = of_parse_phandle(np, "l2-cache", 0);

drivers/of/base.c EXPORT_SYMBOL_GPL(of_console_check); · line 2263

Quoted 10 lines
drivers/of/base.c +4 -2
+	if (!cache_node)
+		cache_node = of_parse_phandle(np, "next-level-cache", 0);
 
-	if (handle)
-		return of_find_node_by_phandle(be32_to_cpup(handle));
+	if (cache_node)
+		return cache_node;
 
 	/* OF on pmac has nodes instead of properties named "l2-cache"
 	 * beneath CPU nodes.

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html