[PATCH] lib: use correct format string for find-bit tests

[PATCH] lib: use correct format string for find-bit tests

[Arnd Bergmann]

The cycles_t definition is architecture specific, which causes
a link error on all architectures that use a 'long long' or 'int'
type for it:

lib/test_find_bit.c: In function 'test_find_last_bit':
include/linux/kern_levels.h:5:18: error: format '%ld' expects argument of type 'long int', but argument 2 has type 'cycles_t {aka long long unsigned int}' [-Werror=format=]

This adds an explicit cast to 'u64' for it, which lets us use
'%llu' everywhere.

Fixes: 09588b1f1d58 ("lib: test module for find_*_bit() functions")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 lib/test_find_bit.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/lib/test_find_bit.c b/lib/test_find_bit.c
index f3b4d34e11dd..f4394a36f9aa 100644
--- a/lib/test_find_bit.c
+++ b/lib/test_find_bit.c
@@ -51,7 +51,8 @@ static int __init test_find_first_bit(void *bitmap, unsigned long len)
 		__clear_bit(i, bitmap);
 	}
 	cycles = get_cycles() - cycles;
-	pr_err("find_first_bit:\t\t%ld cycles,\t%ld iterations\n", cycles, cnt);
+	pr_err("find_first_bit:\t\t%llu cycles,\t%ld iterations\n",
+	       (u64)cycles, cnt);
 
 	return 0;
 }
@@ -65,7 +66,8 @@ static int __init test_find_next_bit(const void *bitmap, unsigned long len)
 	for (cnt = i = 0; i < BITMAP_LEN; cnt++)
 		i = find_next_bit(bitmap, BITMAP_LEN, i) + 1;
 	cycles = get_cycles() - cycles;
-	pr_err("find_next_bit:\t\t%ld cycles,\t%ld iterations\n", cycles, cnt);
+	pr_err("find_next_bit:\t\t%llu cycles,\t%ld iterations\n",
+	       (u64)cycles, cnt);
 
 	return 0;
 }
@@ -79,8 +81,8 @@ static int __init test_find_next_zero_bit(const void *bitmap, unsigned long len)
 	for (cnt = i = 0; i < BITMAP_LEN; cnt++)
 		i = find_next_zero_bit(bitmap, len, i) + 1;
 	cycles = get_cycles() - cycles;
-	pr_err("find_next_zero_bit:\t%ld cycles,\t%ld iterations\n",
-								cycles, cnt);
+	pr_err("find_next_zero_bit:\t%llu cycles,\t%ld iterations\n",
+	       (u64)cycles, cnt);
 
 	return 0;
 }
@@ -99,7 +101,8 @@ static int __init test_find_last_bit(const void *bitmap, unsigned long len)
 		len = l;
 	} while (len);
 	cycles = get_cycles() - cycles;
-	pr_err("find_last_bit:\t\t%ld cycles,\t%ld iterations\n", cycles, cnt);
+	pr_err("find_last_bit:\t\t%llu cycles,\t%ld iterations\n",
+	       (u64)cycles, cnt);
 
 	return 0;
 }
-- 
2.9.0

Re: [PATCH] lib: use correct format string for find-bit tests

[Yury Norov]

On Mon, Nov 13, 2017 at 02:55:45PM +0100, Arnd Bergmann wrote:
> The cycles_t definition is architecture specific, which causes
> a link error on all architectures that use a 'long long' or 'int'
> type for it:
> 
> lib/test_find_bit.c: In function 'test_find_last_bit':
> include/linux/kern_levels.h:5:18: error: format '%ld' expects argument of type 'long int', but argument 2 has type 'cycles_t {aka long long unsigned int}' [-Werror=format=]
> 
> This adds an explicit cast to 'u64' for it, which lets us use
> '%llu' everywhere.
> 
> Fixes: 09588b1f1d58 ("lib: test module for find_*_bit() functions")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Hi Arnd,

patch looks OK. Thank you.

Acked-by: Yury Norov <ynorov@caviumnetworks.com>