ipvs: use div_s64 for signed division
Subsystems (4)
PATCH 1/1 ipvs: use div_s64 for signed division
From: Arnd Bergmann <[email protected]>
do_div() is only well-behaved for positive numbers, and now warns when the first argument is a an s64:
net/netfilter/ipvs/ip_vs_est.c: In function 'ip_vs_est_calc_limits':include/asm-generic/div64.h:222:35: error: comparison of distinct pointer types lacks a cast [-Werror]
222 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~net/netfilter/ipvs/ip_vs_est.c:694:17: note: in expansion of macro 'do_div'
694 | do_div(val, loops);
Convert to using the more appropriate div_s64(), which also simplifies the code a bit.
--- net/netfilter/ipvs/ip_vs_est.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
net/netfilter/ipvs/ip_vs_est.c +2 -4
@@ -691,15 +691,13 @@ static int ip_vs_est_calc_limits(struct netns_ipvs *ipvs, int *chain_max) }if (diff >=NSEC_PER_SEC )continue ; - val = diff; -do_div (val, loops); + val =div_s64 (diff, loops);if (!min_est || val < min_est) { min_est = val;/* goal: 95usec per chain */ val =95 *NSEC_PER_USEC ;if (val >= min_est) { -do_div (val, min_est); - max = (int )val; + max =div_s64 (val, min_est); }else { max =1 ; }
Signature · 1 line
2.35.1
Julian Anastasov
maintainerHello,
Quoted 11 lines · Arnd Bergmann
From: Arnd Bergmann <[email protected]>
do_div()is only well-behaved for positive numbers, and now warns when the first argument is a an s64:
net/netfilter/ipvs/ip_vs_est.c: In function 'ip_vs_est_calc_limits':include/asm-generic/div64.h:222:35: error: comparison of distinct pointer types lacks a cast [-Werror]
222 | (void)(((typeof((n)) *)0) == ((uint64_t*)0)); \
| ^~net/netfilter/ipvs/ip_vs_est.c:694:17: note: in expansion of macro 'do_div'
694 |do_div(val, loops);
net-next already contains fix for this warning
and changes val to u64.
Quoted 14 lines
Convert to using the more appropriate
div_s64(), which also simplifies the code a bit.Fixes 705dd3444081 ("ipvs: use kthreads for stats estimation")Signed-off-by Arnd Bergmann <[email protected]>--- net/netfilter/ipvs/ip_vs_est.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)net/netfilter/ipvs/ip_vs_est.c +0
-0
net/netfilter/ipvs/ip_vs_est.c static int ip_vs_est_calc_limits(struct netns_ipvs *ipvs, int *chain_max) · line 694
Quoted 6 lines
net/netfilter/ipvs/ip_vs_est.c +1
-2@@ -691,15 +691,13 @@ static int ip_vs_est_calc_limits(struct netns_ipvs *ipvs, int *chain_max) }if (diff >=NSEC_PER_SEC )continue ; - val = diff; -do_div (val, loops); + val =div_s64 (diff, loops);
On CONFIG_X86_32 both versions execute single divl
for our case but div_s64 is not inlined. I'm not expert in this area but if you think div_u64 is more appropriate then post another patch. Note that now val is u64 and min_est is still s32 (can be u32).
Quoted 13 lines
net/netfilter/ipvs/ip_vs_est.c +1
-2if (!min_est || val < min_est) { min_est = val;/* goal: 95usec per chain */ val =95 *NSEC_PER_USEC ;if (val >= min_est) { -do_div (val, min_est); - max = (int )val; + max =div_s64 (val, min_est); }else { max =1 ; }Signature · 1 line
2.35.1
Regards
--
Julian Anastasov <[email protected]>
Pablo Neira Ayuso
maintainerHi Julian,
Quoted 17 lines · Julian Anastasov
Hello,
Quoted 11 lines · Arnd Bergmann
From: Arnd Bergmann <[email protected]>
do_div()is only well-behaved for positive numbers, and now warns when the first argument is a an s64:
net/netfilter/ipvs/ip_vs_est.c: In function 'ip_vs_est_calc_limits':include/asm-generic/div64.h:222:35: error: comparison of distinct pointer types lacks a cast [-Werror]
222 | (void)(((typeof((n)) *)0) == ((uint64_t*)0)); \
| ^~net/netfilter/ipvs/ip_vs_est.c:694:17: note: in expansion of macro 'do_div'
694 |do_div(val, loops);
Quoted 2 lines
net-next already contains fix for this warning
and changes val to u64.
Arnd's patch applies fine on top of net-next, maybe he is addressing something else?
Quoted 46 lines
Quoted 14 lines
Convert to using the more appropriate
div_s64(), which also simplifies the code a bit.Fixes 705dd3444081 ("ipvs: use kthreads for stats estimation")Signed-off-by Arnd Bergmann <[email protected]>--- net/netfilter/ipvs/ip_vs_est.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)net/netfilter/ipvs/ip_vs_est.c +0
-0net/netfilter/ipvs/ip_vs_est.c static int ip_vs_est_calc_limits(struct netns_ipvs *ipvs, int *chain_max) · line 694
Quoted 6 lines
net/netfilter/ipvs/ip_vs_est.c +1
-2@@ -691,15 +691,13 @@ static int ip_vs_est_calc_limits(struct netns_ipvs *ipvs, int *chain_max) }if (diff >=NSEC_PER_SEC )continue ; - val = diff; -do_div (val, loops); + val =div_s64 (diff, loops);On
CONFIG_X86_32both versions execute single divl
for our case butdiv_s64is not inlined. I'm not expert in this area but if you thinkdiv_u64is more appropriate then post another patch. Note that now val is u64 andmin_estis still s32 (can be u32).Quoted 13 lines
net/netfilter/ipvs/ip_vs_est.c +1
-2if (!min_est || val < min_est) { min_est = val;/* goal: 95usec per chain */ val =95 *NSEC_PER_USEC ;if (val >= min_est) { -do_div (val, min_est); - max = (int )val; + max =div_s64 (val, min_est); }else { max =1 ; }Signature · 1 line
2.35.1Regards
--
Julian Anastasov <[email protected]>
Quoted 20 lines · Pablo Neira Ayuso
Hi Julian,
Quoted 17 lines · Julian Anastasov
Hello,
Quoted 11 lines · Arnd Bergmann
From: Arnd Bergmann <[email protected]>
do_div()is only well-behaved for positive numbers, and now warns when the first argument is a an s64:
net/netfilter/ipvs/ip_vs_est.c: In function 'ip_vs_est_calc_limits':include/asm-generic/div64.h:222:35: error: comparison of distinct pointer types lacks a cast [-Werror]
222 | (void)(((typeof((n)) *)0) == ((uint64_t*)0)); \
| ^~net/netfilter/ipvs/ip_vs_est.c:694:17: note: in expansion of macro 'do_div'
694 |do_div(val, loops);
Quoted 5 lines
Quoted 2 lines
net-next already contains fix for this warning
and changes val to u64.Arnd's patch applies fine on top of net-next, maybe he is addressing something else?
No, it's the same bug. I had prepared my patch before the other fix went in, and only one of the two is needed.
FWIW, I find my version slightly more readable, but Jakub's fix is probably more efficient, because the unsigned 64-bit division is better optimized on 32-bit, while div_s64() goes through an extern function.
Arnd