* [RFC PATCH] math.h: Account for 64-bit division on i386
@ 2025-04-19 11:51 cem
2025-04-19 18:28 ` kernel test robot
` (2 more replies)
0 siblings, 3 replies; 8+ messages in thread
From: cem @ 2025-04-19 11:51 UTC (permalink / raw)
To: linux-kernel
Cc: Hans.Holmberg, oe-kbuild-all, hch, lukas,
angelogioacchino.delregno, Jonathan.Cameron
From: Carlos Maiolino <cem@kernel.org>
Building linux on i386 might fail if a 64bit type is passed to
mult_fract(). To prevent the failure, use do_div() for the division
calculation instead of hardcoding a / b.
Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202504181233.F7D9Atra-lkp@intel.com/
---
I'm sending it as a RFC because I didn't to extensive testing on this
patch, also I'm not sure if mult_frac() was intended to work on 32-bit
only types. If that's the case, perhaps, a new mult_frac64() might be a
better idea?!
include/linux/math.h | 19 ++++++++++---------
1 file changed, 10 insertions(+), 9 deletions(-)
diff --git a/include/linux/math.h b/include/linux/math.h
index 0198c92cbe3e..05ea853b75b4 100644
--- a/include/linux/math.h
+++ b/include/linux/math.h
@@ -133,15 +133,16 @@ __STRUCT_FRACT(u32)
#undef __STRUCT_FRACT
/* Calculate "x * n / d" without unnecessary overflow or loss of precision. */
-#define mult_frac(x, n, d) \
-({ \
- typeof(x) x_ = (x); \
- typeof(n) n_ = (n); \
- typeof(d) d_ = (d); \
- \
- typeof(x_) q = x_ / d_; \
- typeof(x_) r = x_ % d_; \
- q * n_ + r * n_ / d_; \
+#define mult_frac(x, n, d) \
+({ \
+ typeof(x) x_ = (x); \
+ typeof(n) n_ = (n); \
+ typeof(d) d_ = (d); \
+ \
+ typeof(x_) r = do_div(x_, d_); \
+ r *= n_; \
+ do_div(r, d_); \
+ x_ * n_ + r; \
})
#define sector_div(a, b) do_div(a, b)
--
2.49.0
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [RFC PATCH] math.h: Account for 64-bit division on i386
2025-04-19 11:51 [RFC PATCH] math.h: Account for 64-bit division on i386 cem
@ 2025-04-19 18:28 ` kernel test robot
2025-04-19 18:59 ` kernel test robot
2025-04-20 15:40 ` Guenter Roeck
2 siblings, 0 replies; 8+ messages in thread
From: kernel test robot @ 2025-04-19 18:28 UTC (permalink / raw)
To: cem; +Cc: oe-kbuild-all
Hi,
[This is a private test report for your RFC patch.]
kernel test robot noticed the following build errors:
[auto build test ERROR on linus/master]
[also build test ERROR on hch-configfs/for-next v6.15-rc2 next-20250417]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/cem-kernel-org/math-h-Account-for-64-bit-division-on-i386/20250419-195242
base: linus/master
patch link: https://lore.kernel.org/r/20250419115157.567249-1-cem%40kernel.org
patch subject: [RFC PATCH] math.h: Account for 64-bit division on i386
config: csky-randconfig-002-20250420 (https://download.01.org/0day-ci/archive/20250420/202504200218.98vwdj3m-lkp@intel.com/config)
compiler: csky-linux-gcc (GCC) 14.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250420/202504200218.98vwdj3m-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202504200218.98vwdj3m-lkp@intel.com/
All error/warnings (new ones prefixed by >>):
In file included from ./arch/csky/include/generated/asm/div64.h:1,
from include/linux/math.h:6,
from include/linux/math64.h:6,
from include/linux/time.h:6,
from include/linux/stat.h:19,
from include/linux/module.h:13,
from drivers/power/supply/ltc4162-l-charger.c:12:
drivers/power/supply/ltc4162-l-charger.c: In function 'ltc4162l_get_ibat':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:360:23: note: in expansion of macro 'mult_frac'
360 | val->intval = mult_frac(ret, chip_info->ibat_resolution_pv, info->rsnsb);
| ^~~~~~~~~
In file included from include/linux/build_bug.h:5,
from include/linux/bitfield.h:10,
from drivers/power/supply/ltc4162-l-charger.c:11:
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:360:23: note: in expansion of macro 'mult_frac'
360 | val->intval = mult_frac(ret, chip_info->ibat_resolution_pv, info->rsnsb);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:360:23: note: in expansion of macro 'mult_frac'
360 | val->intval = mult_frac(ret, chip_info->ibat_resolution_pv, info->rsnsb);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:360:23: note: in expansion of macro 'mult_frac'
360 | val->intval = mult_frac(ret, chip_info->ibat_resolution_pv, info->rsnsb);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:360:23: note: in expansion of macro 'mult_frac'
360 | val->intval = mult_frac(ret, chip_info->ibat_resolution_pv, info->rsnsb);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| int *
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:360:23: note: in expansion of macro 'mult_frac'
360 | val->intval = mult_frac(ret, chip_info->ibat_resolution_pv, info->rsnsb);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
drivers/power/supply/ltc4162-l-charger.c: In function 'ltc4162l_get_icharge':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:417:32: note: in expansion of macro 'mult_frac'
417 | val->intval = 10000u * mult_frac(regval, 100000u, info->rsnsb);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:417:32: note: in expansion of macro 'mult_frac'
417 | val->intval = 10000u * mult_frac(regval, 100000u, info->rsnsb);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:417:32: note: in expansion of macro 'mult_frac'
417 | val->intval = 10000u * mult_frac(regval, 100000u, info->rsnsb);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:417:32: note: in expansion of macro 'mult_frac'
417 | val->intval = 10000u * mult_frac(regval, 100000u, info->rsnsb);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:417:32: note: in expansion of macro 'mult_frac'
417 | val->intval = 10000u * mult_frac(regval, 100000u, info->rsnsb);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:417:32: note: in expansion of macro 'mult_frac'
417 | val->intval = 10000u * mult_frac(regval, 100000u, info->rsnsb);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
drivers/power/supply/ltc4162-l-charger.c: In function 'ltc4162l_set_icharge':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:426:17: note: in expansion of macro 'mult_frac'
426 | value = mult_frac(value, info->rsnsb, 100000u);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:426:17: note: in expansion of macro 'mult_frac'
426 | value = mult_frac(value, info->rsnsb, 100000u);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:426:17: note: in expansion of macro 'mult_frac'
426 | value = mult_frac(value, info->rsnsb, 100000u);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:426:17: note: in expansion of macro 'mult_frac'
426 | value = mult_frac(value, info->rsnsb, 100000u);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:426:17: note: in expansion of macro 'mult_frac'
426 | value = mult_frac(value, info->rsnsb, 100000u);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:426:17: note: in expansion of macro 'mult_frac'
426 | value = mult_frac(value, info->rsnsb, 100000u);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
drivers/power/supply/ltc4162-l-charger.c: In function 'ltc4015_get_vcharge':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:528:37: note: in expansion of macro 'mult_frac'
528 | voltage = 2000000 + mult_frac(regval, 1000000, 105);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:528:37: note: in expansion of macro 'mult_frac'
528 | voltage = 2000000 + mult_frac(regval, 1000000, 105);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:528:37: note: in expansion of macro 'mult_frac'
528 | voltage = 2000000 + mult_frac(regval, 1000000, 105);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:528:37: note: in expansion of macro 'mult_frac'
528 | voltage = 2000000 + mult_frac(regval, 1000000, 105);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:528:37: note: in expansion of macro 'mult_frac'
528 | voltage = 2000000 + mult_frac(regval, 1000000, 105);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:528:37: note: in expansion of macro 'mult_frac'
528 | voltage = 2000000 + mult_frac(regval, 1000000, 105);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
drivers/power/supply/ltc4162-l-charger.c: In function 'ltc4162l_set_iin_limit':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:655:18: note: in expansion of macro 'mult_frac'
655 | regval = mult_frac(value, info->rsnsi, 50000u);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:655:18: note: in expansion of macro 'mult_frac'
655 | regval = mult_frac(value, info->rsnsi, 50000u);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:655:18: note: in expansion of macro 'mult_frac'
655 | regval = mult_frac(value, info->rsnsi, 50000u);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:655:18: note: in expansion of macro 'mult_frac'
655 | regval = mult_frac(value, info->rsnsi, 50000u);
| ^~~~~~~~~
>> include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:655:18: note: in expansion of macro 'mult_frac'
655 | regval = mult_frac(value, info->rsnsi, 50000u);
| ^~~~~~~~~
>> include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:655:18: note: in expansion of macro 'mult_frac'
655 | regval = mult_frac(value, info->rsnsi, 50000u);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
drivers/power/supply/ltc4162-l-charger.c: In function 'ltc4162l_set_term_current':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:746:18: note: in expansion of macro 'mult_frac'
746 | regval = mult_frac(value, info->rsnsb, 14660u);
| ^~~~~~~~~
include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:746:18: note: in expansion of macro 'mult_frac'
746 | regval = mult_frac(value, info->rsnsb, 14660u);
| ^~~~~~~~~
include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:746:18: note: in expansion of macro 'mult_frac'
746 | regval = mult_frac(value, info->rsnsb, 14660u);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast [-Wcompare-distinct-pointer-types]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:746:18: note: in expansion of macro 'mult_frac'
746 | regval = mult_frac(value, info->rsnsb, 14660u);
| ^~~~~~~~~
include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:746:18: note: in expansion of macro 'mult_frac'
746 | regval = mult_frac(value, info->rsnsb, 14660u);
| ^~~~~~~~~
include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Wincompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/power/supply/ltc4162-l-charger.c:746:18: note: in expansion of macro 'mult_frac'
746 | regval = mult_frac(value, info->rsnsb, 14660u);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
vim +/__div64_32 +199 include/asm-generic/div64.h
^1da177e4c3f41 Linus Torvalds 2005-04-16 176
^1da177e4c3f41 Linus Torvalds 2005-04-16 177 /* The unnecessary pointer compare is there
^1da177e4c3f41 Linus Torvalds 2005-04-16 178 * to check for type safety (n must be 64bit)
^1da177e4c3f41 Linus Torvalds 2005-04-16 179 */
^1da177e4c3f41 Linus Torvalds 2005-04-16 180 # define do_div(n,base) ({ \
^1da177e4c3f41 Linus Torvalds 2005-04-16 181 uint32_t __base = (base); \
^1da177e4c3f41 Linus Torvalds 2005-04-16 182 uint32_t __rem; \
^1da177e4c3f41 Linus Torvalds 2005-04-16 183 (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 184 if (__builtin_constant_p(__base) && \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 185 is_power_of_2(__base)) { \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 186 __rem = (n) & (__base - 1); \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 187 (n) >>= ilog2(__base); \
c747ce4706190e Geert Uytterhoeven 2021-08-11 188 } else if (__builtin_constant_p(__base) && \
461a5e51060c93 Nicolas Pitre 2015-10-30 189 __base != 0) { \
461a5e51060c93 Nicolas Pitre 2015-10-30 190 uint32_t __res_lo, __n_lo = (n); \
461a5e51060c93 Nicolas Pitre 2015-10-30 191 (n) = __div64_const32(n, __base); \
461a5e51060c93 Nicolas Pitre 2015-10-30 192 /* the remainder can be computed with 32-bit regs */ \
461a5e51060c93 Nicolas Pitre 2015-10-30 193 __res_lo = (n); \
461a5e51060c93 Nicolas Pitre 2015-10-30 194 __rem = __n_lo - __res_lo * __base; \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 @195 } else if (likely(((n) >> 32) == 0)) { \
^1da177e4c3f41 Linus Torvalds 2005-04-16 196 __rem = (uint32_t)(n) % __base; \
^1da177e4c3f41 Linus Torvalds 2005-04-16 197 (n) = (uint32_t)(n) / __base; \
c747ce4706190e Geert Uytterhoeven 2021-08-11 198 } else { \
^1da177e4c3f41 Linus Torvalds 2005-04-16 @199 __rem = __div64_32(&(n), __base); \
c747ce4706190e Geert Uytterhoeven 2021-08-11 200 } \
^1da177e4c3f41 Linus Torvalds 2005-04-16 201 __rem; \
^1da177e4c3f41 Linus Torvalds 2005-04-16 202 })
^1da177e4c3f41 Linus Torvalds 2005-04-16 203
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [RFC PATCH] math.h: Account for 64-bit division on i386
2025-04-19 11:51 [RFC PATCH] math.h: Account for 64-bit division on i386 cem
2025-04-19 18:28 ` kernel test robot
@ 2025-04-19 18:59 ` kernel test robot
2025-04-20 15:40 ` Guenter Roeck
2 siblings, 0 replies; 8+ messages in thread
From: kernel test robot @ 2025-04-19 18:59 UTC (permalink / raw)
To: cem; +Cc: oe-kbuild-all
Hi,
[This is a private test report for your RFC patch.]
kernel test robot noticed the following build errors:
[auto build test ERROR on linus/master]
[also build test ERROR on v6.15-rc2 next-20250417]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/cem-kernel-org/math-h-Account-for-64-bit-division-on-i386/20250419-195242
base: linus/master
patch link: https://lore.kernel.org/r/20250419115157.567249-1-cem%40kernel.org
patch subject: [RFC PATCH] math.h: Account for 64-bit division on i386
config: arc-randconfig-001-20250420 (https://download.01.org/0day-ci/archive/20250420/202504200217.lAUS8TXX-lkp@intel.com/config)
compiler: arc-linux-gcc (GCC) 12.4.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250420/202504200217.lAUS8TXX-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202504200217.lAUS8TXX-lkp@intel.com/
All errors (new ones prefixed by >>):
In file included from ./arch/arc/include/generated/asm/div64.h:1,
from include/linux/math.h:6,
from include/linux/kernel.h:27,
from include/linux/cpumask.h:11,
from include/linux/smp.h:13,
from include/linux/lockdep.h:14,
from include/linux/mutex.h:17,
from include/linux/interconnect.h:10,
from drivers/soc/qcom/icc-bwmon.c:10:
drivers/soc/qcom/icc-bwmon.c: In function 'bwmon_set_threshold':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:558:17: note: in expansion of macro 'mult_frac'
558 | thres = mult_frac(bwmon_kbps_to_count(bwmon, kbps),
| ^~~~~~~~~
In file included from include/linux/err.h:5,
from drivers/soc/qcom/icc-bwmon.c:9:
include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:558:17: note: in expansion of macro 'mult_frac'
558 | thres = mult_frac(bwmon_kbps_to_count(bwmon, kbps),
| ^~~~~~~~~
include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Werror=incompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:558:17: note: in expansion of macro 'mult_frac'
558 | thres = mult_frac(bwmon_kbps_to_count(bwmon, kbps),
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:558:17: note: in expansion of macro 'mult_frac'
558 | thres = mult_frac(bwmon_kbps_to_count(bwmon, kbps),
| ^~~~~~~~~
include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:558:17: note: in expansion of macro 'mult_frac'
558 | thres = mult_frac(bwmon_kbps_to_count(bwmon, kbps),
| ^~~~~~~~~
include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Werror=incompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:558:17: note: in expansion of macro 'mult_frac'
558 | thres = mult_frac(bwmon_kbps_to_count(bwmon, kbps),
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
drivers/soc/qcom/icc-bwmon.c: In function 'bwmon_start':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
>> include/asm-generic/div64.h:187:21: error: assignment of read-only variable 'x_'
187 | (n) >>= ilog2(__base); \
| ^~~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:191:21: error: assignment of read-only variable 'x_'
191 | (n) = __div64_const32(n, __base); \
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:197:21: error: assignment of read-only variable 'x_'
197 | (n) = (uint32_t)(n) / __base; \
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Werror=incompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| const unsigned int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'const unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
>> include/linux/math.h:143:11: error: assignment of read-only variable 'r'
143 | r *= n_; \
| ^~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
>> include/asm-generic/div64.h:187:21: error: assignment of read-only variable 'r'
187 | (n) >>= ilog2(__base); \
| ^~~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:191:21: error: assignment of read-only variable 'r'
191 | (n) = __div64_const32(n, __base); \
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:197:21: error: assignment of read-only variable 'r'
197 | (n) = (uint32_t)(n) / __base; \
| ^
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Werror=incompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| const unsigned int *
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:574:18: note: in expansion of macro 'mult_frac'
574 | window = mult_frac(bwmon->data->sample_ms, HW_TIMER_HZ, MSEC_PER_SEC);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'const unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
drivers/soc/qcom/icc-bwmon.c: In function 'bwmon_intr':
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:640:30: note: in expansion of macro 'mult_frac'
640 | bwmon->target_kbps = mult_frac(max, MSEC_PER_SEC, bwmon->data->sample_ms);
| ^~~~~~~~~
include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
include/linux/compiler.h:76:45: note: in definition of macro 'likely'
76 | # define likely(x) __builtin_expect(!!(x), 1)
| ^
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:640:30: note: in expansion of macro 'mult_frac'
640 | bwmon->target_kbps = mult_frac(max, MSEC_PER_SEC, bwmon->data->sample_ms);
| ^~~~~~~~~
include/asm-generic/div64.h:199:36: error: passing argument 1 of '__div64_32' from incompatible pointer type [-Werror=incompatible-pointer-types]
199 | __rem = __div64_32(&(n), __base); \
| ^~~~
| |
| unsigned int *
include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:640:30: note: in expansion of macro 'mult_frac'
640 | bwmon->target_kbps = mult_frac(max, MSEC_PER_SEC, bwmon->data->sample_ms);
| ^~~~~~~~~
include/asm-generic/div64.h:174:38: note: expected 'uint64_t *' {aka 'long long unsigned int *'} but argument is of type 'unsigned int *'
174 | extern uint32_t __div64_32(uint64_t *dividend, uint32_t divisor);
| ~~~~~~~~~~^~~~~~~~
include/asm-generic/div64.h:183:35: warning: comparison of distinct pointer types lacks a cast
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
include/linux/math.h:144:9: note: in expansion of macro 'do_div'
144 | do_div(r, d_); \
| ^~~~~~
drivers/soc/qcom/icc-bwmon.c:640:30: note: in expansion of macro 'mult_frac'
640 | bwmon->target_kbps = mult_frac(max, MSEC_PER_SEC, bwmon->data->sample_ms);
| ^~~~~~~~~
include/asm-generic/div64.h:195:32: warning: right shift count >= width of type [-Wshift-count-overflow]
195 | } else if (likely(((n) >> 32) == 0)) { \
| ^~
vim +/x_ +187 include/asm-generic/div64.h
^1da177e4c3f41 Linus Torvalds 2005-04-16 176
^1da177e4c3f41 Linus Torvalds 2005-04-16 177 /* The unnecessary pointer compare is there
^1da177e4c3f41 Linus Torvalds 2005-04-16 178 * to check for type safety (n must be 64bit)
^1da177e4c3f41 Linus Torvalds 2005-04-16 179 */
^1da177e4c3f41 Linus Torvalds 2005-04-16 180 # define do_div(n,base) ({ \
^1da177e4c3f41 Linus Torvalds 2005-04-16 181 uint32_t __base = (base); \
^1da177e4c3f41 Linus Torvalds 2005-04-16 182 uint32_t __rem; \
^1da177e4c3f41 Linus Torvalds 2005-04-16 @183 (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 184 if (__builtin_constant_p(__base) && \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 185 is_power_of_2(__base)) { \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 186 __rem = (n) & (__base - 1); \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 @187 (n) >>= ilog2(__base); \
c747ce4706190e Geert Uytterhoeven 2021-08-11 188 } else if (__builtin_constant_p(__base) && \
461a5e51060c93 Nicolas Pitre 2015-10-30 189 __base != 0) { \
461a5e51060c93 Nicolas Pitre 2015-10-30 190 uint32_t __res_lo, __n_lo = (n); \
461a5e51060c93 Nicolas Pitre 2015-10-30 191 (n) = __div64_const32(n, __base); \
461a5e51060c93 Nicolas Pitre 2015-10-30 192 /* the remainder can be computed with 32-bit regs */ \
461a5e51060c93 Nicolas Pitre 2015-10-30 193 __res_lo = (n); \
461a5e51060c93 Nicolas Pitre 2015-10-30 194 __rem = __n_lo - __res_lo * __base; \
911918aa7ef6f8 Nicolas Pitre 2015-11-02 195 } else if (likely(((n) >> 32) == 0)) { \
^1da177e4c3f41 Linus Torvalds 2005-04-16 196 __rem = (uint32_t)(n) % __base; \
^1da177e4c3f41 Linus Torvalds 2005-04-16 197 (n) = (uint32_t)(n) / __base; \
c747ce4706190e Geert Uytterhoeven 2021-08-11 198 } else { \
^1da177e4c3f41 Linus Torvalds 2005-04-16 199 __rem = __div64_32(&(n), __base); \
c747ce4706190e Geert Uytterhoeven 2021-08-11 200 } \
^1da177e4c3f41 Linus Torvalds 2005-04-16 201 __rem; \
^1da177e4c3f41 Linus Torvalds 2005-04-16 202 })
^1da177e4c3f41 Linus Torvalds 2005-04-16 203
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [RFC PATCH] math.h: Account for 64-bit division on i386
2025-04-19 11:51 [RFC PATCH] math.h: Account for 64-bit division on i386 cem
2025-04-19 18:28 ` kernel test robot
2025-04-19 18:59 ` kernel test robot
@ 2025-04-20 15:40 ` Guenter Roeck
2025-04-20 17:42 ` Guenter Roeck
2025-04-20 18:01 ` Carlos Maiolino
2 siblings, 2 replies; 8+ messages in thread
From: Guenter Roeck @ 2025-04-20 15:40 UTC (permalink / raw)
To: cem
Cc: linux-kernel, Hans.Holmberg, oe-kbuild-all, hch, lukas,
angelogioacchino.delregno, Jonathan.Cameron
On Sat, Apr 19, 2025 at 01:51:46PM +0200, cem@kernel.org wrote:
> From: Carlos Maiolino <cem@kernel.org>
>
> Building linux on i386 might fail if a 64bit type is passed to
i386 actually builds. Its compiler is probably able to convert
the offending mult_frac() without helpers since the divisor is
a constant. I see the problem with openrisc and parisc, with
gcc 13.3.0.
> mult_fract(). To prevent the failure, use do_div() for the division
> calculation instead of hardcoding a / b.
>
> Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202504181233.F7D9Atra-lkp@intel.com/
> ---
>
> I'm sending it as a RFC because I didn't to extensive testing on this
> patch, also I'm not sure if mult_frac() was intended to work on 32-bit
> only types. If that's the case, perhaps, a new mult_frac64() might be a
> better idea?!
>
> include/linux/math.h | 19 ++++++++++---------
> 1 file changed, 10 insertions(+), 9 deletions(-)
>
> diff --git a/include/linux/math.h b/include/linux/math.h
> index 0198c92cbe3e..05ea853b75b4 100644
> --- a/include/linux/math.h
> +++ b/include/linux/math.h
> @@ -133,15 +133,16 @@ __STRUCT_FRACT(u32)
> #undef __STRUCT_FRACT
>
> /* Calculate "x * n / d" without unnecessary overflow or loss of precision. */
> -#define mult_frac(x, n, d) \
> -({ \
> - typeof(x) x_ = (x); \
> - typeof(n) n_ = (n); \
> - typeof(d) d_ = (d); \
> - \
> - typeof(x_) q = x_ / d_; \
> - typeof(x_) r = x_ % d_; \
> - q * n_ + r * n_ / d_; \
> +#define mult_frac(x, n, d) \
> +({ \
> + typeof(x) x_ = (x); \
> + typeof(n) n_ = (n); \
> + typeof(d) d_ = (d); \
> + \
> + typeof(x_) r = do_div(x_, d_); \
> + r *= n_; \
> + do_div(r, d_); \
> + x_ * n_ + r; \
> })
>
Unfortunately that doesn't work. I get build errors on parisc.
In file included from ./arch/parisc/include/generated/asm/div64.h:1,
from ./include/linux/math.h:6,
from ./include/linux/kernel.h:27,
from ./arch/parisc/include/asm/bug.h:5,
from ./include/linux/bug.h:5,
from ./include/linux/mmdebug.h:5,
from ./include/linux/mm.h:6,
from mm/page_alloc.c:19:
mm/page_alloc.c: In function 'boost_watermark':
./include/asm-generic/div64.h:183:35: error: comparison of distinct pointer types lacks a cast [-Werror]
183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
./include/linux/math.h:142:24: note: in expansion of macro 'do_div'
142 | typeof(x_) r = do_div(x_, d_); \
| ^~~~~~
mm/page_alloc.c:2010:21: note: in expansion of macro 'mult_frac'
2010 | max_boost = mult_frac(zone->_watermark[WMARK_HIGH],
That is just one example. It seems to affect all uses of
mult_frac().
Guenter
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [RFC PATCH] math.h: Account for 64-bit division on i386
2025-04-20 15:40 ` Guenter Roeck
@ 2025-04-20 17:42 ` Guenter Roeck
2025-04-20 18:05 ` Carlos Maiolino
2025-04-23 7:02 ` Geert Uytterhoeven
2025-04-20 18:01 ` Carlos Maiolino
1 sibling, 2 replies; 8+ messages in thread
From: Guenter Roeck @ 2025-04-20 17:42 UTC (permalink / raw)
To: cem
Cc: linux-kernel, Hans.Holmberg, oe-kbuild-all, hch, lukas,
angelogioacchino.delregno, Jonathan.Cameron
On Sun, Apr 20, 2025 at 08:40:27AM -0700, Guenter Roeck wrote:
> On Sat, Apr 19, 2025 at 01:51:46PM +0200, cem@kernel.org wrote:
> > From: Carlos Maiolino <cem@kernel.org>
> >
> > Building linux on i386 might fail if a 64bit type is passed to
>
> i386 actually builds. Its compiler is probably able to convert
> the offending mult_frac() without helpers since the divisor is
> a constant. I see the problem with openrisc and parisc, with
> gcc 13.3.0.
>
> > mult_fract(). To prevent the failure, use do_div() for the division
> > calculation instead of hardcoding a / b.
> >
> > Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
> > Reported-by: kernel test robot <lkp@intel.com>
> > Closes: https://lore.kernel.org/oe-kbuild-all/202504181233.F7D9Atra-lkp@intel.com/
> > ---
> >
> > I'm sending it as a RFC because I didn't to extensive testing on this
> > patch, also I'm not sure if mult_frac() was intended to work on 32-bit
> > only types. If that's the case, perhaps, a new mult_frac64() might be a
> > better idea?!
> >
> > include/linux/math.h | 19 ++++++++++---------
> > 1 file changed, 10 insertions(+), 9 deletions(-)
> >
> > diff --git a/include/linux/math.h b/include/linux/math.h
> > index 0198c92cbe3e..05ea853b75b4 100644
> > --- a/include/linux/math.h
> > +++ b/include/linux/math.h
> > @@ -133,15 +133,16 @@ __STRUCT_FRACT(u32)
> > #undef __STRUCT_FRACT
> >
> > /* Calculate "x * n / d" without unnecessary overflow or loss of precision. */
> > -#define mult_frac(x, n, d) \
> > -({ \
> > - typeof(x) x_ = (x); \
> > - typeof(n) n_ = (n); \
> > - typeof(d) d_ = (d); \
> > - \
> > - typeof(x_) q = x_ / d_; \
> > - typeof(x_) r = x_ % d_; \
> > - q * n_ + r * n_ / d_; \
> > +#define mult_frac(x, n, d) \
> > +({ \
> > + typeof(x) x_ = (x); \
> > + typeof(n) n_ = (n); \
> > + typeof(d) d_ = (d); \
> > + \
> > + typeof(x_) r = do_div(x_, d_); \
> > + r *= n_; \
> > + do_div(r, d_); \
> > + x_ * n_ + r; \
> > })
> >
>
> Unfortunately that doesn't work. I get build errors on parisc.
>
Turns out the first parameter of do_div needs to be u64, not s64,
at least on parisc.
Guenter
> In file included from ./arch/parisc/include/generated/asm/div64.h:1,
> from ./include/linux/math.h:6,
> from ./include/linux/kernel.h:27,
> from ./arch/parisc/include/asm/bug.h:5,
> from ./include/linux/bug.h:5,
> from ./include/linux/mmdebug.h:5,
> from ./include/linux/mm.h:6,
> from mm/page_alloc.c:19:
> mm/page_alloc.c: In function 'boost_watermark':
> ./include/asm-generic/div64.h:183:35: error: comparison of distinct pointer types lacks a cast [-Werror]
> 183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
> | ^~
> ./include/linux/math.h:142:24: note: in expansion of macro 'do_div'
> 142 | typeof(x_) r = do_div(x_, d_); \
> | ^~~~~~
> mm/page_alloc.c:2010:21: note: in expansion of macro 'mult_frac'
> 2010 | max_boost = mult_frac(zone->_watermark[WMARK_HIGH],
>
> That is just one example. It seems to affect all uses of
> mult_frac().
>
> Guenter
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [RFC PATCH] math.h: Account for 64-bit division on i386
2025-04-20 15:40 ` Guenter Roeck
2025-04-20 17:42 ` Guenter Roeck
@ 2025-04-20 18:01 ` Carlos Maiolino
1 sibling, 0 replies; 8+ messages in thread
From: Carlos Maiolino @ 2025-04-20 18:01 UTC (permalink / raw)
To: Guenter Roeck
Cc: linux-kernel, Hans.Holmberg, oe-kbuild-all, hch, lukas,
angelogioacchino.delregno, Jonathan.Cameron
On Sun, Apr 20, 2025 at 08:40:25AM -0700, Guenter Roeck wrote:
> On Sat, Apr 19, 2025 at 01:51:46PM +0200, cem@kernel.org wrote:
> > From: Carlos Maiolino <cem@kernel.org>
> >
> > Building linux on i386 might fail if a 64bit type is passed to
>
> i386 actually builds. Its compiler is probably able to convert
> the offending mult_frac() without helpers since the divisor is
> a constant. I see the problem with openrisc and parisc, with
> gcc 13.3.0.
It does fail to build, that's how it got identified in the first place. I
managed to reproduce here according to the reproducer sent by the test robot.
I think it works if you don't enable build checks (in this case W=1) though.
I didn't try other architectures other than x86 though, thanks for the heads up.
>
> > mult_fract(). To prevent the failure, use do_div() for the division
> > calculation instead of hardcoding a / b.
> >
> > Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
> > Reported-by: kernel test robot <lkp@intel.com>
> > Closes: https://lore.kernel.org/oe-kbuild-all/202504181233.F7D9Atra-lkp@intel.com/
> > ---
> >
> > I'm sending it as a RFC because I didn't to extensive testing on this
> > patch, also I'm not sure if mult_frac() was intended to work on 32-bit
> > only types. If that's the case, perhaps, a new mult_frac64() might be a
> > better idea?!
> >
> > include/linux/math.h | 19 ++++++++++---------
> > 1 file changed, 10 insertions(+), 9 deletions(-)
> >
> > diff --git a/include/linux/math.h b/include/linux/math.h
> > index 0198c92cbe3e..05ea853b75b4 100644
> > --- a/include/linux/math.h
> > +++ b/include/linux/math.h
> > @@ -133,15 +133,16 @@ __STRUCT_FRACT(u32)
> > #undef __STRUCT_FRACT
> >
> > /* Calculate "x * n / d" without unnecessary overflow or loss of precision. */
> > -#define mult_frac(x, n, d) \
> > -({ \
> > - typeof(x) x_ = (x); \
> > - typeof(n) n_ = (n); \
> > - typeof(d) d_ = (d); \
> > - \
> > - typeof(x_) q = x_ / d_; \
> > - typeof(x_) r = x_ % d_; \
> > - q * n_ + r * n_ / d_; \
> > +#define mult_frac(x, n, d) \
> > +({ \
> > + typeof(x) x_ = (x); \
> > + typeof(n) n_ = (n); \
> > + typeof(d) d_ = (d); \
> > + \
> > + typeof(x_) r = do_div(x_, d_); \
> > + r *= n_; \
> > + do_div(r, d_); \
> > + x_ * n_ + r; \
> > })
> >
>
> Unfortunately that doesn't work. I get build errors on parisc.
>
> In file included from ./arch/parisc/include/generated/asm/div64.h:1,
> from ./include/linux/math.h:6,
> from ./include/linux/kernel.h:27,
> from ./arch/parisc/include/asm/bug.h:5,
> from ./include/linux/bug.h:5,
> from ./include/linux/mmdebug.h:5,
> from ./include/linux/mm.h:6,
> from mm/page_alloc.c:19:
> mm/page_alloc.c: In function 'boost_watermark':
> ./include/asm-generic/div64.h:183:35: error: comparison of distinct pointer types lacks a cast [-Werror]
> 183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
> | ^~
> ./include/linux/math.h:142:24: note: in expansion of macro 'do_div'
> 142 | typeof(x_) r = do_div(x_, d_); \
> | ^~~~~~
> mm/page_alloc.c:2010:21: note: in expansion of macro 'mult_frac'
> 2010 | max_boost = mult_frac(zone->_watermark[WMARK_HIGH],
>
> That is just one example. It seems to affect all uses of
> mult_frac().
>
> Guenter
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [RFC PATCH] math.h: Account for 64-bit division on i386
2025-04-20 17:42 ` Guenter Roeck
@ 2025-04-20 18:05 ` Carlos Maiolino
2025-04-23 7:02 ` Geert Uytterhoeven
1 sibling, 0 replies; 8+ messages in thread
From: Carlos Maiolino @ 2025-04-20 18:05 UTC (permalink / raw)
To: Guenter Roeck
Cc: linux-kernel, Hans.Holmberg, oe-kbuild-all, hch, lukas,
angelogioacchino.delregno, Jonathan.Cameron
On Sun, Apr 20, 2025 at 10:42:18AM -0700, Guenter Roeck wrote:
> On Sun, Apr 20, 2025 at 08:40:27AM -0700, Guenter Roeck wrote:
> > On Sat, Apr 19, 2025 at 01:51:46PM +0200, cem@kernel.org wrote:
> > > From: Carlos Maiolino <cem@kernel.org>
> > >
> > > Building linux on i386 might fail if a 64bit type is passed to
> >
> > i386 actually builds. Its compiler is probably able to convert
> > the offending mult_frac() without helpers since the divisor is
> > a constant. I see the problem with openrisc and parisc, with
> > gcc 13.3.0.
> >
> > > mult_fract(). To prevent the failure, use do_div() for the division
> > > calculation instead of hardcoding a / b.
> > >
> > > Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
> > > Reported-by: kernel test robot <lkp@intel.com>
> > > Closes: https://lore.kernel.org/oe-kbuild-all/202504181233.F7D9Atra-lkp@intel.com/
> > > ---
> > >
> > > I'm sending it as a RFC because I didn't to extensive testing on this
> > > patch, also I'm not sure if mult_frac() was intended to work on 32-bit
> > > only types. If that's the case, perhaps, a new mult_frac64() might be a
> > > better idea?!
> > >
> > > include/linux/math.h | 19 ++++++++++---------
> > > 1 file changed, 10 insertions(+), 9 deletions(-)
> > >
> > > diff --git a/include/linux/math.h b/include/linux/math.h
> > > index 0198c92cbe3e..05ea853b75b4 100644
> > > --- a/include/linux/math.h
> > > +++ b/include/linux/math.h
> > > @@ -133,15 +133,16 @@ __STRUCT_FRACT(u32)
> > > #undef __STRUCT_FRACT
> > >
> > > /* Calculate "x * n / d" without unnecessary overflow or loss of precision. */
> > > -#define mult_frac(x, n, d) \
> > > -({ \
> > > - typeof(x) x_ = (x); \
> > > - typeof(n) n_ = (n); \
> > > - typeof(d) d_ = (d); \
> > > - \
> > > - typeof(x_) q = x_ / d_; \
> > > - typeof(x_) r = x_ % d_; \
> > > - q * n_ + r * n_ / d_; \
> > > +#define mult_frac(x, n, d) \
> > > +({ \
> > > + typeof(x) x_ = (x); \
> > > + typeof(n) n_ = (n); \
> > > + typeof(d) d_ = (d); \
> > > + \
> > > + typeof(x_) r = do_div(x_, d_); \
> > > + r *= n_; \
> > > + do_div(r, d_); \
> > > + x_ * n_ + r; \
> > > })
> > >
> >
> > Unfortunately that doesn't work. I get build errors on parisc.
> >
>
> Turns out the first parameter of do_div needs to be u64, not s64,
> at least on parisc.
Yup, that's true for x86 too, needs to be fixed once we decide which
path to take to fix it.
>
> Guenter
>
> > In file included from ./arch/parisc/include/generated/asm/div64.h:1,
> > from ./include/linux/math.h:6,
> > from ./include/linux/kernel.h:27,
> > from ./arch/parisc/include/asm/bug.h:5,
> > from ./include/linux/bug.h:5,
> > from ./include/linux/mmdebug.h:5,
> > from ./include/linux/mm.h:6,
> > from mm/page_alloc.c:19:
> > mm/page_alloc.c: In function 'boost_watermark':
> > ./include/asm-generic/div64.h:183:35: error: comparison of distinct pointer types lacks a cast [-Werror]
> > 183 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
> > | ^~
> > ./include/linux/math.h:142:24: note: in expansion of macro 'do_div'
> > 142 | typeof(x_) r = do_div(x_, d_); \
> > | ^~~~~~
> > mm/page_alloc.c:2010:21: note: in expansion of macro 'mult_frac'
> > 2010 | max_boost = mult_frac(zone->_watermark[WMARK_HIGH],
> >
> > That is just one example. It seems to affect all uses of
> > mult_frac().
> >
> > Guenter
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [RFC PATCH] math.h: Account for 64-bit division on i386
2025-04-20 17:42 ` Guenter Roeck
2025-04-20 18:05 ` Carlos Maiolino
@ 2025-04-23 7:02 ` Geert Uytterhoeven
1 sibling, 0 replies; 8+ messages in thread
From: Geert Uytterhoeven @ 2025-04-23 7:02 UTC (permalink / raw)
To: Guenter Roeck
Cc: cem, linux-kernel, Hans.Holmberg, oe-kbuild-all, hch, lukas,
angelogioacchino.delregno, Jonathan.Cameron
On Sun, 20 Apr 2025 at 19:42, Guenter Roeck <linux@roeck-us.net> wrote:
> On Sun, Apr 20, 2025 at 08:40:27AM -0700, Guenter Roeck wrote:
> > On Sat, Apr 19, 2025 at 01:51:46PM +0200, cem@kernel.org wrote:
> > > From: Carlos Maiolino <cem@kernel.org>
> > >
> > > Building linux on i386 might fail if a 64bit type is passed to
> >
> > i386 actually builds. Its compiler is probably able to convert
> > the offending mult_frac() without helpers since the divisor is
> > a constant. I see the problem with openrisc and parisc, with
> > gcc 13.3.0.
> >
> > > mult_fract(). To prevent the failure, use do_div() for the division
> > > calculation instead of hardcoding a / b.
> > >
> > > Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
> > > Reported-by: kernel test robot <lkp@intel.com>
> > > Closes: https://lore.kernel.org/oe-kbuild-all/202504181233.F7D9Atra-lkp@intel.com/
> > > ---
> > >
> > > I'm sending it as a RFC because I didn't to extensive testing on this
> > > patch, also I'm not sure if mult_frac() was intended to work on 32-bit
> > > only types. If that's the case, perhaps, a new mult_frac64() might be a
> > > better idea?!
> > >
> > > include/linux/math.h | 19 ++++++++++---------
> > > 1 file changed, 10 insertions(+), 9 deletions(-)
> > >
> > > diff --git a/include/linux/math.h b/include/linux/math.h
> > > index 0198c92cbe3e..05ea853b75b4 100644
> > > --- a/include/linux/math.h
> > > +++ b/include/linux/math.h
> > > @@ -133,15 +133,16 @@ __STRUCT_FRACT(u32)
> > > #undef __STRUCT_FRACT
> > >
> > > /* Calculate "x * n / d" without unnecessary overflow or loss of precision. */
> > > -#define mult_frac(x, n, d) \
> > > -({ \
> > > - typeof(x) x_ = (x); \
> > > - typeof(n) n_ = (n); \
> > > - typeof(d) d_ = (d); \
> > > - \
> > > - typeof(x_) q = x_ / d_; \
> > > - typeof(x_) r = x_ % d_; \
> > > - q * n_ + r * n_ / d_; \
> > > +#define mult_frac(x, n, d) \
> > > +({ \
> > > + typeof(x) x_ = (x); \
> > > + typeof(n) n_ = (n); \
> > > + typeof(d) d_ = (d); \
> > > + \
> > > + typeof(x_) r = do_div(x_, d_); \
> > > + r *= n_; \
> > > + do_div(r, d_); \
> > > + x_ * n_ + r; \
> > > })
> > >
> >
> > Unfortunately that doesn't work. I get build errors on parisc.
> >
>
> Turns out the first parameter of do_div needs to be u64, not s64,
> at least on parisc.
That is correct: the first parameter must be u64, the second must be u32:
https://elixir.bootlin.com/linux/v6.14.3/source/include/asm-generic/div64.h
As you must never use open-coded 64-bit divisions in the kernel,
simple helpers like mult_frac() can only be used for 32-bit values.
For anything involving 64-bit divisions, you must use the helpers from
<linux/math64.h>, e.g. mul_u64_u32_div().
https://elixir.bootlin.com/linux/v6.14.3/source/include/linux/math64.h#L257
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
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
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2025-04-23 7:02 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-04-19 11:51 [RFC PATCH] math.h: Account for 64-bit division on i386 cem
2025-04-19 18:28 ` kernel test robot
2025-04-19 18:59 ` kernel test robot
2025-04-20 15:40 ` Guenter Roeck
2025-04-20 17:42 ` Guenter Roeck
2025-04-20 18:05 ` Carlos Maiolino
2025-04-23 7:02 ` Geert Uytterhoeven
2025-04-20 18:01 ` Carlos Maiolino
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.