2022-01-07 02:51:46

by Xiu Jianfeng

Subject: [PATCH -next] sched: Use struct_size() helper in task_numa_group()

Make use of struct_size() helper instead of an open-coded calculation.
There is no functional change in this patch.

Link: https://github.com/KSPP/linux/issues/160
Signed-off-by: Xiu Jianfeng <[email protected]>
---
kernel/sched/fair.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 095b0aa378df..265e37be0c92 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2437,11 +2437,8 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags,
int i;

if (unlikely(!deref_curr_numa_group(p))) {
- unsigned int size = sizeof(struct numa_group) +
- NR_NUMA_HINT_FAULT_STATS *
- nr_node_ids * sizeof(unsigned long);
-
- grp = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
+ grp = kzalloc(struct_size(grp, faults, NR_NUMA_HINT_FAULT_STATS * nr_node_ids),
+ GFP_KERNEL | __GFP_NOWARN);
if (!grp)
return;

--
2.17.1

2022-01-07 14:36:24

by Steven Rostedt

Subject: Re: [PATCH -next] sched: Use struct_size() helper in task_numa_group()

On Fri, 7 Jan 2022 10:52:12 +0800
Xiu Jianfeng <[email protected]> wrote:

> Make use of struct_size() helper instead of an open-coded calculation.
> There is no functional change in this patch.
>
> Link: https://github.com/KSPP/linux/issues/160
> Signed-off-by: Xiu Jianfeng <[email protected]>
> ---
> kernel/sched/fair.c | 7 ++-----
> 1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 095b0aa378df..265e37be0c92 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -2437,11 +2437,8 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags,
> int i;
>
> if (unlikely(!deref_curr_numa_group(p))) {
> - unsigned int size = sizeof(struct numa_group) +
> - NR_NUMA_HINT_FAULT_STATS *
> - nr_node_ids * sizeof(unsigned long);
> -
> - grp = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
> + grp = kzalloc(struct_size(grp, faults, NR_NUMA_HINT_FAULT_STATS * nr_node_ids),
> + GFP_KERNEL | __GFP_NOWARN);

I just replied yesterday. Keep the size variable. The above is too much for
a kzalloc().

-- Steve




> if (!grp)
> return;
>

2022-01-10 01:21:13

by Xiu Jianfeng

Subject: Re: [PATCH -next] sched: Use struct_size() helper in task_numa_group()

?? 2022/1/7 22:36, Steven Rostedt д??:
> On Fri, 7 Jan 2022 10:52:12 +0800
> Xiu Jianfeng <[email protected]> wrote:
>
>> Make use of struct_size() helper instead of an open-coded calculation.
>> There is no functional change in this patch.
>>
>> Link: https://github.com/KSPP/linux/issues/160
>> Signed-off-by: Xiu Jianfeng <[email protected]>
>> ---
>> kernel/sched/fair.c | 7 ++-----
>> 1 file changed, 2 insertions(+), 5 deletions(-)
>>
>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> index 095b0aa378df..265e37be0c92 100644
>> --- a/kernel/sched/fair.c
>> +++ b/kernel/sched/fair.c
>> @@ -2437,11 +2437,8 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags,
>> int i;
>>
>> if (unlikely(!deref_curr_numa_group(p))) {
>> - unsigned int size = sizeof(struct numa_group) +
>> - NR_NUMA_HINT_FAULT_STATS *
>> - nr_node_ids * sizeof(unsigned long);
>> -
>> - grp = kzalloc(size, GFP_KERNEL | __GFP_NOWARN);
>> + grp = kzalloc(struct_size(grp, faults, NR_NUMA_HINT_FAULT_STATS * nr_node_ids),
>> + GFP_KERNEL | __GFP_NOWARN);
> I just replied yesterday. Keep the size variable. The above is too much for
> a kzalloc().

thanks, I may missed your reply mail, I'll modify and resend this patch.

>
> -- Steve
>
>
>
>
>> if (!grp)
>> return;
>>
> .

2022-01-10 16:07:04

by Peter Zijlstra

Subject: Re: [PATCH -next] sched: Use struct_size() helper in task_numa_group()

On Fri, Jan 07, 2022 at 10:52:12AM +0800, Xiu Jianfeng wrote:
> Make use of struct_size() helper instead of an open-coded calculation.
> There is no functional change in this patch.

Why ?!? This makes something that was trivial to read into something
weird.

2022-01-10 16:11:48

by Steven Rostedt

Subject: Re: [PATCH -next] sched: Use struct_size() helper in task_numa_group()

On Mon, 10 Jan 2022 17:06:30 +0100
Peter Zijlstra <[email protected]> wrote:

> On Fri, Jan 07, 2022 at 10:52:12AM +0800, Xiu Jianfeng wrote:
> > Make use of struct_size() helper instead of an open-coded calculation.
> > There is no functional change in this patch.
>
> Why ?!? This makes something that was trivial to read into something
> weird.

Please take his last patch (which only modifies the update to size) and
makes it more robust.

https://lore.kernel.org/all/[email protected]/

Which is much easier to read.

We should really replace all open coded struct_size() code.

-- Steve