Re: [PATCH v2 4/5] blk-iocost: fix sleeping in atomic context warnning

[Jens Axboe]

On 11/22/22 2:10 PM, Tejun Heo wrote:
> On Thu, Nov 17, 2022 at 07:28:50PM +0800, Yu Kuai wrote:
>> Hi, Tejun!
>>
>> 在 2022/11/15 6:07, Tejun Heo 写道:
>>
>>>
>>> Any chance I can persuade you into updating match_NUMBER() helpers to not
>>> use match_strdup()? They can easily disable irq/preemption and use percpu
>>> buffers and we won't need most of this patchset.
>>
>> Does the following patch match your proposal?
>>
>> diff --git a/lib/parser.c b/lib/parser.c
>> index bcb23484100e..ded652471919 100644
>> --- a/lib/parser.c
>> +++ b/lib/parser.c
>> @@ -11,6 +11,24 @@
>>  #include <linux/slab.h>
>>  #include <linux/string.h>
>>
>> +#define U64_MAX_SIZE 20
>> +
>> +static DEFINE_PER_CPU(char, buffer[U64_MAX_SIZE]);
>> +
>> +static char *get_buffer(void)
>> +{
>> +       preempt_disable();
>> +       local_irq_disable();
>> +
>> +       return this_cpu_ptr(buffer);
>> +}
>> +
>> +static void put_buffer(void)
>> +{
>> +       local_irq_enable();
>> +       preempt_enable();
>> +}
>> +
>>
>> Then match_strdup() and kfree() in match_NUMBER() can be replaced with
>> get_buffer() and put_buffer().
> 
> Sorry about the late reply. Yeah, something like this.

Doesn't local_irq_disable() imply preemption disable as well?

-- 
Jens Axboe