powerpc: Make BUG_ON & WARN_ON play nice with compile-time optimisations
Subsystems (2)
PATCH 1/4 powerpc: Make BUG_ON & WARN_ON play nice with compile-time optimisations
Change BUG_ON and WARN_ON to give the compiler a chance to perform compile-time optimsations. Depending on the complexity of the condition, the compiler may not do this very well, so if it's important check the object code.
Current GCC's (4.x) produce good code as long as the condition does not include a function call, including a static inline.
---
include/asm-powerpc/bug.h | 30 ++++++++++++++++++++++++++++-- 1 files changed, 28 insertions(+), 2 deletions(-)
include/asm-powerpc/bug.h +28 -2
@@ -30,6 +30,12 @@ struct bug_entry *find_bug(unsigned long#ifdef CONFIG_BUG +/* +* BUG_ON() and WARN_ON() do their best to cooperate with compile-time +* optimisations. However depending on the complexity of the condition +* some compiler versions may not produce optimal results. +*/ +#define BUG () do { \ __asm__ __volatile__( \ "1: twi 31,0,0\n" \
@@ -40,17 +46,36 @@ struct bug_entry *find_bug(unsigned long } while (0)#define BUG_ON (x) do { \ - __asm__ __volatile__( \ + if (__builtin_constant_p(x)) { \ + if (x) \ + BUG(); \ + } else { \ + __asm__ __volatile__( \ "1: "PPC_TLNEI" %0,0\n" \ ".section __bug_table,\"a\"\n" \ "\t"PPC_LONG" 1b,%1,%2,%3\n" \ ".previous" \ : : "r" ((long)(x)), "i" (__LINE__), \ "i" (__FILE__), "i" (__FUNCTION__)); \ + } \ } while (0) -#define WARN_ON (x) do { \ +#define WARN () do { \ __asm__ __volatile__( \ + "1: twi 31,0,0\n" \ + ".section __bug_table,\"a\"\n" \ + "\t"PPC_LONG" 1b,%0,%1,%2\n" \ + ".previous" \ + : : "i" (__LINE__ + BUG_WARNING_TRAP), \ + "i" (__FILE__), "i" (__FUNCTION__)); \ +} while (0) + +#define WARN_ON (x) do { \ + if (__builtin_constant_p(x)) { \ + if (x) \ + WARN(); \ + } else { \ + __asm__ __volatile__( \ "1: "PPC_TLNEI" %0,0\n" \ ".section __bug_table,\"a\"\n" \ "\t"PPC_LONG" 1b,%1,%2,%3\n" \
@@ -58,6 +83,7 @@ struct bug_entry *find_bug(unsigned long : : "r" ((long)(x)), \"i" (__LINE__ +BUG_WARNING_TRAP ), \ "i" (__FILE__), "i" (__FUNCTION__)); \ + } \ }while (0 ) #define HAVE_ARCH_BUG
PATCH 2/4 powerpc: Change firmware_has_feature() to a macro
So that we can use firmware_has_feature() in a BUG_ON() and have the compiler elide the code entirely if the feature can never be set, change firmware_has_feature to a macro. Unfortunate, but necessary at least until GCC bug #26724 is fixed.
---
include/asm-powerpc/firmware.h | 8 +++----- 1 files changed, 3 insertions(+), 5 deletions(-)
include/asm-powerpc/firmware.h +3 -5
@@ -84,11 +84,9 @@ enum {*/ extern unsigned long ppc64_firmware_features; -static inline unsigned long firmware_has_feature (unsigned long feature) -{ -return (FW_FEATURE_ALWAYS & feature) || - (FW_FEATURE_POSSIBLE & ppc64_firmware_features & feature); -} +#define firmware_has_feature (feature) \ + ((FW_FEATURE_ALWAYS & (feature)) || \ + (FW_FEATURE_POSSIBLE & ppc64_firmware_features & (feature)))extern void system_reset_fwnmi (void );extern void machine_check_fwnmi (void );
PATCH 3/4 powerpc: Rename and export ppc64_firmware_features
In reply to Michael EllermanWe need to export ppc64_firmware_features for modules. Before we do that I think we should probably rename it to powerpc_firmware_features.
---
arch/powerpc/kernel/firmware.c | 2 +- arch/powerpc/platforms/iseries/setup.c | 4 ++-- arch/powerpc/platforms/pseries/firmware.c | 2 +- arch/powerpc/platforms/pseries/setup.c | 2 +- include/asm-powerpc/firmware.h | 5 +++-- 5 files changed, 8 insertions(+), 7 deletions(-)
arch/powerpc/kernel/firmware.c +1 -1
@@ -17,4 +17,4 @@#include <asm/firmware.h> -unsigned long ppc64_firmware_features; +unsigned long powerpc_firmware_features;
arch/powerpc/platforms/iseries/setup.c +2 -2
@@ -680,8 +680,8 @@ static int __init iseries_probe(int platif (PLATFORM_ISERIES_LPAR != platform)return 0 ; - ppc64_firmware_features |=FW_FEATURE_ISERIES ; - ppc64_firmware_features |=FW_FEATURE_LPAR ; + powerpc_firmware_features |=FW_FEATURE_ISERIES ; + powerpc_firmware_features |=FW_FEATURE_LPAR ;return 1 ; }
arch/powerpc/platforms/pseries/firmware.c +1 -1
@@ -91,7 +91,7 @@ void __init fw_feature_init(void)continue ;/* we have a match */ - ppc64_firmware_features |= + powerpc_firmware_features |= firmware_features_table[i].val;break ; }
arch/powerpc/platforms/pseries/setup.c +1 -1
@@ -386,7 +386,7 @@ static int __init pSeries_probe(int plat*/ if (platform ==PLATFORM_PSERIES_LPAR ) - ppc64_firmware_features |=FW_FEATURE_LPAR ; + powerpc_firmware_features |=FW_FEATURE_LPAR ;return 1 ; }
include/asm-powerpc/firmware.h +3 -2
@@ -82,11 +82,12 @@ enum {/* This is used to identify firmware features which are available * to the kernel. */ -extern unsigned long ppc64_firmware_features; +extern unsigned long powerpc_firmware_features; +EXPORT_SYMBOL_GPL (powerpc_firmware_features);#define firmware_has_feature (feature) \ ((FW_FEATURE_ALWAYS & (feature)) || \ - (FW_FEATURE_POSSIBLE & ppc64_firmware_features & (feature))) + (FW_FEATURE_POSSIBLE & powerpc_firmware_features & (feature)))extern void system_reset_fwnmi (void );extern void machine_check_fwnmi (void );
Re: [PATCH 3/4] powerpc: Rename and export ppc64_firmware_features
Quoted 2 lines · On Thu, 23 Mar 2006 23:33:11 +1100 Michael Ellerman <[email protected]>…
+EXPORT_SYMBOL_GPL(powerpc_firmware_features);
This should be in firmware.c not a header file.
Signature · 3 lines
Cheers, Stephen Rothwell [email protected] http://www.canb.auug.org.au/~sfr/
PATCH 1/1 powerpc: Rename and export ppc64_firmware_features
We need to export ppc64_firmware_features for modules. Before we do that I think we should probably rename it to powerpc_firmware_features.
---
arch/powerpc/kernel/firmware.c | 3 ++- arch/powerpc/platforms/iseries/setup.c | 4 ++-- arch/powerpc/platforms/pseries/firmware.c | 2 +- arch/powerpc/platforms/pseries/setup.c | 2 +- include/asm-powerpc/firmware.h | 4 ++-- 5 files changed, 8 insertions(+), 7 deletions(-)
arch/powerpc/kernel/firmware.c +2 -1
@@ -17,4 +17,5 @@#include <asm/firmware.h> -unsigned long ppc64_firmware_features; +unsigned long powerpc_firmware_features; +EXPORT_SYMBOL_GPL (powerpc_firmware_features);
arch/powerpc/platforms/iseries/setup.c +2 -2
@@ -680,8 +680,8 @@ static int __init iseries_probe(int platif (PLATFORM_ISERIES_LPAR != platform)return 0 ; - ppc64_firmware_features |=FW_FEATURE_ISERIES ; - ppc64_firmware_features |=FW_FEATURE_LPAR ; + powerpc_firmware_features |=FW_FEATURE_ISERIES ; + powerpc_firmware_features |=FW_FEATURE_LPAR ;return 1 ; }
arch/powerpc/platforms/pseries/firmware.c +1 -1
@@ -91,7 +91,7 @@ void __init fw_feature_init(void)continue ;/* we have a match */ - ppc64_firmware_features |= + powerpc_firmware_features |= firmware_features_table[i].val;break ; }
arch/powerpc/platforms/pseries/setup.c +1 -1
@@ -386,7 +386,7 @@ static int __init pSeries_probe(int plat*/ if (platform ==PLATFORM_PSERIES_LPAR ) - ppc64_firmware_features |=FW_FEATURE_LPAR ; + powerpc_firmware_features |=FW_FEATURE_LPAR ;return 1 ; }
include/asm-powerpc/firmware.h +2 -2
@@ -82,11 +82,11 @@ enum {/* This is used to identify firmware features which are available * to the kernel. */ -extern unsigned long ppc64_firmware_features; +extern unsigned long powerpc_firmware_features;#define firmware_has_feature (feature) \ ((FW_FEATURE_ALWAYS & (feature)) || \ - (FW_FEATURE_POSSIBLE & ppc64_firmware_features & (feature))) + (FW_FEATURE_POSSIBLE & powerpc_firmware_features & (feature)))extern void system_reset_fwnmi (void );extern void machine_check_fwnmi (void );
PATCH 4/4 powerpc: Cope with duplicate node & property names in /proc/device-tree
In reply to Michael EllermanVarious dodgy firmware might give us nodes and/or properties in the device tree with conflicting names. That's generally ok, except for when we export the device tree via /proc, so check when we're creating the proc device tree and munge names accordingly.
Tested on a faked device tree with kexec, would be good if someone with actual bogus firmware could try it, but just for completeness.
---
fs/proc/proc_devtree.c | 93 +++++++++++++++++++++++++++++++++++++------------ 1 files changed, 72 insertions(+), 21 deletions(-)
fs/proc/proc_devtree.c +72 -21
@@ -52,7 +52,8 @@ static int property_read_proc(char *page* Add a property to a node */ static struct proc_dir_entry * -__proc_device_tree_add_prop (struct proc_dir_entry *de,struct property *pp) +__proc_device_tree_add_prop (struct proc_dir_entry *de,struct property *pp, +char *name) {struct proc_dir_entry *ent;
@@ -60,14 +61,14 @@ __proc_device_tree_add_prop(struct proc_* Unfortunately proc_register puts each new entry * at the beginning of the list. So we rearrange them. */ - ent =create_proc_read_entry (pp->name, -strncmp (pp->name,"security-" ,9 ) + ent =create_proc_read_entry (name, +strncmp (name,"security-" ,9 ) ?S_IRUGO :S_IRUSR , de, property_read_proc, pp);if (ent ==NULL )return NULL ; -if (!strncmp (pp->name,"security-" ,9 )) +if (!strncmp (name,"security-" ,9 )) ent->size =0 ;/* don't leak number of password chars */ else ent->size = pp->length;
@@ -78,7 +79,7 @@ __proc_device_tree_add_prop(struct proc_void proc_device_tree_add_prop (struct proc_dir_entry *pde,struct property *prop) { -__proc_device_tree_add_prop (pde, prop); +__proc_device_tree_add_prop (pde, prop, prop->name); } void proc_device_tree_remove_prop(struct proc_dir_entry *pde,
@@ -106,6 +107,61 @@ void proc_device_tree_update_prop(struct }/* +* Various dodgy firmware might give us nodes and/or properties with +* conflicting names. That's generally ok, except for exporting via /proc, +* so munge names here to ensure they're unique. +*/ + +static int duplicate_name (struct proc_dir_entry *de,char *name) +{ +struct proc_dir_entry *ent; + +for (ent = de->subdir; ent !=NULL ; ent = ent->next) +if (strcmp (ent->name, name) ==0 ) +return 1 ; + +return 0 ; +} + +static char *fixup_name (struct device_node *np,struct proc_dir_entry *de, +char *name) +{ +char *fixed_name; +int fixup_len =strlen (name) +2 +1 ;/* name + #x + \0 */ +int i =1 , size; + +realloc: + fixed_name =kmalloc (fixup_len,GFP_KERNEL ); +if (fixed_name ==NULL ) { +printk (KERN_ERR "device-tree: Out of memory trying to fixup " +"name \"%s\"\n" , name); +return name; + } + +retry: + size =snprintf (fixed_name, fixup_len,"%s#%d" , name, i); + size++;/* account for NULL */ + +if (size > fixup_len) { +/* We ran out of space, free and reallocate. */ +kfree (fixed_name); + fixup_len = size; + goto realloc; + } + +if (duplicate_name (de, fixed_name)) { +/* Multiple duplicates. Retry with a different offset. */ + i++; + goto retry; + } + +printk (KERN_WARNING "device-tree: Duplicate name in %s, " +"renamed to \"%s\"\n" , np->full_name, fixed_name); + +return fixed_name; +} + +/* * Process a node, adding entries for its children and its properties. */ void proc_device_tree_add_node(struct device_node *np,
@@ -118,35 +174,30 @@ void proc_device_tree_add_node(struct deset_node_proc_entry (np, de);for (child =NULL ; (child =of_get_next_child (np, child));) { +/* Use everything after the last slash, or the full name */ p =strrchr (child->full_name,'/' );if (!p) p = child->full_name;else ++p; + +if (duplicate_name (de, p)) + p =fixup_name (np, de, p); + ent =proc_mkdir (p, de);if (ent ==0 )break ;proc_device_tree_add_node (child, ent); }of_node_put (child); +for (pp = np->properties; pp !=0 ; pp = pp->next) { -/* -* Yet another Apple device-tree bogosity: on some machines, -* they have properties & nodes with the same name. Those -* properties are quite unimportant for us though, thus we -* simply "skip" them here, but we do have to check. -*/ -for (ent = de->subdir; ent !=NULL ; ent = ent->next) -if (!strcmp (ent->name, pp->name)) -break ; -if (ent !=NULL ) { -printk (KERN_WARNING "device-tree: property \"%s\" name" -" conflicts with node in %s\n" , pp->name, - np->full_name); -continue ; - } + p = pp->name; + +if (duplicate_name (de, p)) + p =fixup_name (np, de, p); - ent =__proc_device_tree_add_prop (de, pp); + ent =__proc_device_tree_add_prop (de, pp, p);if (ent ==0 )break ; }