[PATCH] avoid compiler warnings

[PATCH] avoid compiler warnings

[Jan Beulich]

Pointers should not be casted to u32 as this results in compiler warnings
on 64-bit platforms.

Signed-off-by: Jan Beulich <jbeulich@novell.com>

--- linux-2.6.19-rc5/drivers/acpi/executer/exmutex.c	2006-09-20 05:42:06.000000000 +0200
+++ 2.6.19-rc5-acpi-warnings/drivers/acpi/executer/exmutex.c	2006-11-06 09:10:16.000000000 +0100
@@ -266,10 +266,10 @@ acpi_ex_release_mutex(union acpi_operand
 	     walk_state->thread->thread_id)
 	    && (obj_desc->mutex.os_mutex != ACPI_GLOBAL_LOCK)) {
 		ACPI_ERROR((AE_INFO,
-			    "Thread %X cannot release Mutex [%4.4s] acquired by thread %X",
-			    (u32) walk_state->thread->thread_id,
+			    "Thread %lX cannot release Mutex [%4.4s] acquired by thread %lX",
+			    (unsigned long) walk_state->thread->thread_id,
 			    acpi_ut_get_node_name(obj_desc->mutex.node),
-			    (u32) obj_desc->mutex.owner_thread->thread_id));
+			    (unsigned long) obj_desc->mutex.owner_thread->thread_id));
 		return_ACPI_STATUS(AE_AML_NOT_OWNER);
 	}
 
--- linux-2.6.19-rc5/drivers/acpi/utilities/utmutex.c	2006-09-20 05:42:06.000000000 +0200
+++ 2.6.19-rc5-acpi-warnings/drivers/acpi/utilities/utmutex.c	2006-11-06 09:10:16.000000000 +0100
@@ -258,8 +258,8 @@ acpi_status acpi_ut_acquire_mutex(acpi_m
 		acpi_gbl_mutex_info[mutex_id].thread_id = this_thread_id;
 	} else {
 		ACPI_EXCEPTION((AE_INFO, status,
-				"Thread %X could not acquire Mutex [%X]",
-				(u32) this_thread_id, mutex_id));
+				"Thread %lX could not acquire Mutex [%X]",
+				(unsigned long) this_thread_id, mutex_id));
 	}
 
 	return (status);

Re: [PATCH] avoid compiler warnings

[Mikael Pettersson]

Jan Beulich writes:
 > Pointers should not be casted to u32 as this results in compiler warnings
 > on 64-bit platforms.
 > 
 > Signed-off-by: Jan Beulich <jbeulich@novell.com>
 > 
 > --- linux-2.6.19-rc5/drivers/acpi/executer/exmutex.c	2006-09-20 05:42:06.000000000 +0200
 > +++ 2.6.19-rc5-acpi-warnings/drivers/acpi/executer/exmutex.c	2006-11-06 09:10:16.000000000 +0100
 > @@ -266,10 +266,10 @@ acpi_ex_release_mutex(union acpi_operand
 >  	     walk_state->thread->thread_id)
 >  	    && (obj_desc->mutex.os_mutex != ACPI_GLOBAL_LOCK)) {
 >  		ACPI_ERROR((AE_INFO,
 > -			    "Thread %X cannot release Mutex [%4.4s] acquired by thread %X",
 > -			    (u32) walk_state->thread->thread_id,
 > +			    "Thread %lX cannot release Mutex [%4.4s] acquired by thread %lX",
 > +			    (unsigned long) walk_state->thread->thread_id,
 >  			    acpi_ut_get_node_name(obj_desc->mutex.node),
 > -			    (u32) obj_desc->mutex.owner_thread->thread_id));
 > +			    (unsigned long) obj_desc->mutex.owner_thread->thread_id));
 >  		return_ACPI_STATUS(AE_AML_NOT_OWNER);
 >  	}
 >  
 > --- linux-2.6.19-rc5/drivers/acpi/utilities/utmutex.c	2006-09-20 05:42:06.000000000 +0200
 > +++ 2.6.19-rc5-acpi-warnings/drivers/acpi/utilities/utmutex.c	2006-11-06 09:10:16.000000000 +0100
 > @@ -258,8 +258,8 @@ acpi_status acpi_ut_acquire_mutex(acpi_m
 >  		acpi_gbl_mutex_info[mutex_id].thread_id = this_thread_id;
 >  	} else {
 >  		ACPI_EXCEPTION((AE_INFO, status,
 > -				"Thread %X could not acquire Mutex [%X]",
 > -				(u32) this_thread_id, mutex_id));
 > +				"Thread %lX could not acquire Mutex [%X]",
 > +				(unsigned long) this_thread_id, mutex_id));
 >  	}
 >  
 >  	return (status);

NAK. Use "%p" for formatting pointers. No casts needed.

Re: [PATCH] avoid compiler warnings

[Jan Beulich]

>>Pointers should not be casted to u32 as this results in compiler warnings
>>on 64-bit platforms.
>
>NAK. Use "%p" for formatting pointers. No casts needed.

Indeed, how did I not see this... While at this, I saw that there were a few
more instances that needed fixing (they weren't actively generating warnings
because of the build settings).

Signed-off-by: Jan Beulich <jbeulich@novell.com>

--- linux-2.6.19-rc5/drivers/acpi/executer/exmutex.c	2006-09-20 05:42:06.000000000 +0200
+++ 2.6.19-rc5-acpi-warnings/drivers/acpi/executer/exmutex.c	2006-11-15 17:22:39.000000000 +0100
@@ -266,10 +266,10 @@ acpi_ex_release_mutex(union acpi_operand
 	     walk_state->thread->thread_id)
 	    && (obj_desc->mutex.os_mutex != ACPI_GLOBAL_LOCK)) {
 		ACPI_ERROR((AE_INFO,
-			    "Thread %X cannot release Mutex [%4.4s] acquired by thread %X",
-			    (u32) walk_state->thread->thread_id,
+			    "Thread %p cannot release Mutex [%4.4s] acquired by thread %p",
+			    walk_state->thread->thread_id,
 			    acpi_ut_get_node_name(obj_desc->mutex.node),
-			    (u32) obj_desc->mutex.owner_thread->thread_id));
+			    obj_desc->mutex.owner_thread->thread_id));
 		return_ACPI_STATUS(AE_AML_NOT_OWNER);
 	}
 
--- linux-2.6.19-rc5/drivers/acpi/utilities/utmutex.c	2006-09-20 05:42:06.000000000 +0200
+++ 2.6.19-rc5-acpi-warnings/drivers/acpi/utilities/utmutex.c	2006-11-15 17:24:31.000000000 +0100
@@ -222,7 +222,7 @@ acpi_status acpi_ut_acquire_mutex(acpi_m
 			if (acpi_gbl_mutex_info[i].thread_id == this_thread_id) {
 				if (i == mutex_id) {
 					ACPI_ERROR((AE_INFO,
-						    "Mutex [%s] already acquired by this thread [%X]",
+						    "Mutex [%s] already acquired by this thread [%p]",
 						    acpi_ut_get_mutex_name
 						    (mutex_id),
 						    this_thread_id));
@@ -231,7 +231,7 @@ acpi_status acpi_ut_acquire_mutex(acpi_m
 				}
 
 				ACPI_ERROR((AE_INFO,
-					    "Invalid acquire order: Thread %X owns [%s], wants [%s]",
+					    "Invalid acquire order: Thread %p owns [%s], wants [%s]",
 					    this_thread_id,
 					    acpi_ut_get_mutex_name(i),
 					    acpi_ut_get_mutex_name(mutex_id)));
@@ -243,23 +243,23 @@ acpi_status acpi_ut_acquire_mutex(acpi_m
 #endif
 
 	ACPI_DEBUG_PRINT((ACPI_DB_MUTEX,
-			  "Thread %X attempting to acquire Mutex [%s]\n",
-			  (u32) this_thread_id, acpi_ut_get_mutex_name(mutex_id)));
+			  "Thread %p attempting to acquire Mutex [%s]\n",
+			  this_thread_id, acpi_ut_get_mutex_name(mutex_id)));
 
 	status = acpi_os_acquire_mutex(acpi_gbl_mutex_info[mutex_id].mutex,
 				       ACPI_WAIT_FOREVER);
 	if (ACPI_SUCCESS(status)) {
 		ACPI_DEBUG_PRINT((ACPI_DB_MUTEX,
-				  "Thread %X acquired Mutex [%s]\n",
-				  (u32) this_thread_id,
+				  "Thread %p acquired Mutex [%s]\n",
+				  this_thread_id,
 				  acpi_ut_get_mutex_name(mutex_id)));
 
 		acpi_gbl_mutex_info[mutex_id].use_count++;
 		acpi_gbl_mutex_info[mutex_id].thread_id = this_thread_id;
 	} else {
 		ACPI_EXCEPTION((AE_INFO, status,
-				"Thread %X could not acquire Mutex [%X]",
-				(u32) this_thread_id, mutex_id));
+				"Thread %p could not acquire Mutex [%X]",
+				this_thread_id, mutex_id));
 	}
 
 	return (status);
@@ -285,7 +285,7 @@ acpi_status acpi_ut_release_mutex(acpi_m
 
 	this_thread_id = acpi_os_get_thread_id();
 	ACPI_DEBUG_PRINT((ACPI_DB_MUTEX,
-			  "Thread %X releasing Mutex [%s]\n", (u32) this_thread_id,
+			  "Thread %p releasing Mutex [%s]\n", this_thread_id,
 			  acpi_ut_get_mutex_name(mutex_id)));
 
 	if (mutex_id > ACPI_MAX_MUTEX) {

Re: [PATCH] avoid compiler warnings

[Mikael Pettersson]

On Wed, 15 Nov 2006 16:47:59 +0000, Jan Beulich wrote:
> >>Pointers should not be casted to u32 as this results in compiler warnings
> >>on 64-bit platforms.
> >
> >NAK. Use "%p" for formatting pointers. No casts needed.
> 
> Indeed, how did I not see this... While at this, I saw that there were a few
> more instances that needed fixing (they weren't actively generating warnings
> because of the build settings).

Tested on x86-64 and i386, and it did kill the warnings w/o noticeable regressions.

However, for some reason (config settings?) you didn't fix the two cast warnings
in utdebug.c; the appended patch fixes them too.

Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>

--- linux-2.6.19-rc5/drivers/acpi/utilities/utdebug.c.~1~	2006-09-20 19:28:46.000000000 +0200
+++ linux-2.6.19-rc5/drivers/acpi/utilities/utdebug.c	2006-11-15 21:22:27.000000000 +0100
@@ -180,8 +180,8 @@ acpi_ut_debug_print(u32 requested_debug_
 	if (thread_id != acpi_gbl_prev_thread_id) {
 		if (ACPI_LV_THREADS & acpi_dbg_level) {
 			acpi_os_printf
-			    ("\n**** Context Switch from TID %X to TID %X ****\n\n",
-			     (u32) acpi_gbl_prev_thread_id, (u32) thread_id);
+			    ("\n**** Context Switch from TID %p to TID %p ****\n\n",
+			     acpi_gbl_prev_thread_id, thread_id);
 		}
 
 		acpi_gbl_prev_thread_id = thread_id;