[PATCH] ntb_perf: Fix 64-bit division on 32-bit architectures

[PATCH] ntb_perf: Fix 64-bit division on 32-bit architectures

[Nathan Chancellor]

When compiling for a 32-bit architecture, such as arm, an error occurs
during modpost:

  ERROR: modpost: "__aeabi_uldivmod" [drivers/ntb/test/ntb_perf.ko] undefined!

The tries member of struct perf_thread is u64 so a 64-bit division
helper is needed. Use div_u64_rem() to get the remainder of the division
so that it can be checked against zero.

Fixes: dc150dfb081f ("ntb_perf: extend with burst latency measurement")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
---
 drivers/ntb/test/ntb_perf.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/ntb/test/ntb_perf.c b/drivers/ntb/test/ntb_perf.c
index 23e154bd41b9..4e05c7aa070d 100644
--- a/drivers/ntb/test/ntb_perf.c
+++ b/drivers/ntb/test/ntb_perf.c
@@ -1118,6 +1118,7 @@ static int perf_run_latency(struct perf_thread *pthr)
 	void __iomem *flt_dst, *bnd_dst;
 	void *flt_src;
 	u64 stop_at;
+	u32 rem;
 	int ret;
 
 	pthr->tries = 0;
@@ -1146,7 +1147,8 @@ static int perf_run_latency(struct perf_thread *pthr)
 		}
 
 		/* Avoid processor soft lock-ups */
-		if (!(pthr->tries % RESCHEDULE_RATIO))
+		div_u64_rem(pthr->tries, RESCHEDULE_RATIO, &rem);
+		if (!rem)
 			schedule();
 	}
 

base-commit: 0c4b285d9636cc850f55151fa6a250dd131e5192
-- 
2.37.0.rc1

Re: [PATCH] ntb_perf: Fix 64-bit division on 32-bit architectures

[Jon Mason]

On Wed, Jun 22, 2022 at 10:43:26AM -0700, Nathan Chancellor wrote:
> When compiling for a 32-bit architecture, such as arm, an error occurs
> during modpost:
> 
>   ERROR: modpost: "__aeabi_uldivmod" [drivers/ntb/test/ntb_perf.ko] undefined!
> 
> The tries member of struct perf_thread is u64 so a 64-bit division
> helper is needed. Use div_u64_rem() to get the remainder of the division
> so that it can be checked against zero.
> 
> Fixes: dc150dfb081f ("ntb_perf: extend with burst latency measurement")
> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> ---
>  drivers/ntb/test/ntb_perf.c | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/ntb/test/ntb_perf.c b/drivers/ntb/test/ntb_perf.c
> index 23e154bd41b9..4e05c7aa070d 100644
> --- a/drivers/ntb/test/ntb_perf.c
> +++ b/drivers/ntb/test/ntb_perf.c
> @@ -1118,6 +1118,7 @@ static int perf_run_latency(struct perf_thread *pthr)
>  	void __iomem *flt_dst, *bnd_dst;
>  	void *flt_src;
>  	u64 stop_at;
> +	u32 rem;
>  	int ret;
>  
>  	pthr->tries = 0;
> @@ -1146,7 +1147,8 @@ static int perf_run_latency(struct perf_thread *pthr)
>  		}
>  
>  		/* Avoid processor soft lock-ups */
> -		if (!(pthr->tries % RESCHEDULE_RATIO))
> +		div_u64_rem(pthr->tries, RESCHEDULE_RATIO, &rem);
> +		if (!rem)
>  			schedule();
>  	}
>  
> 
> base-commit: 0c4b285d9636cc850f55151fa6a250dd131e5192
> -- 
> 2.37.0.rc1
> 

Sorry for the extremely long delay in response.  This patch is in my
ntb branch and will be in my pull request for v5.20 which should be
going out later today.

Thanks,
Jon

Re: [PATCH] ntb_perf: Fix 64-bit division on 32-bit architectures

[Nathan Chancellor]

Hi Jon,

On Fri, Aug 12, 2022 at 10:04:59AM -0400, Jon Mason wrote:
> On Wed, Jun 22, 2022 at 10:43:26AM -0700, Nathan Chancellor wrote:
> > When compiling for a 32-bit architecture, such as arm, an error occurs
> > during modpost:
> > 
> >   ERROR: modpost: "__aeabi_uldivmod" [drivers/ntb/test/ntb_perf.ko] undefined!
> > 
> > The tries member of struct perf_thread is u64 so a 64-bit division
> > helper is needed. Use div_u64_rem() to get the remainder of the division
> > so that it can be checked against zero.
> > 
> > Fixes: dc150dfb081f ("ntb_perf: extend with burst latency measurement")
> > Reported-by: kernel test robot <lkp@intel.com>
> > Signed-off-by: Nathan Chancellor <nathan@kernel.org>
> > ---
> >  drivers/ntb/test/ntb_perf.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> > 
> > diff --git a/drivers/ntb/test/ntb_perf.c b/drivers/ntb/test/ntb_perf.c
> > index 23e154bd41b9..4e05c7aa070d 100644
> > --- a/drivers/ntb/test/ntb_perf.c
> > +++ b/drivers/ntb/test/ntb_perf.c
> > @@ -1118,6 +1118,7 @@ static int perf_run_latency(struct perf_thread *pthr)
> >  	void __iomem *flt_dst, *bnd_dst;
> >  	void *flt_src;
> >  	u64 stop_at;
> > +	u32 rem;
> >  	int ret;
> >  
> >  	pthr->tries = 0;
> > @@ -1146,7 +1147,8 @@ static int perf_run_latency(struct perf_thread *pthr)
> >  		}
> >  
> >  		/* Avoid processor soft lock-ups */
> > -		if (!(pthr->tries % RESCHEDULE_RATIO))
> > +		div_u64_rem(pthr->tries, RESCHEDULE_RATIO, &rem);
> > +		if (!rem)
> >  			schedule();
> >  	}
> >  
> > 
> > base-commit: 0c4b285d9636cc850f55151fa6a250dd131e5192
> > -- 
> > 2.37.0.rc1
> > 
> 
> Sorry for the extremely long delay in response.  This patch is in my
> ntb branch and will be in my pull request for v5.20 which should be
> going out later today.

No worries, we all get busy :)

For the record, I believe this was addressed by Max's patch [1], which I
did see in -next for a time. It doesn't matter to me which one goes in,
just so long as the build doesn't break :)

[1]: https://lore.kernel.org/20220627155710.2067032-1-jcmvbkbc@gmail.com/

Cheers,
Nathan