Re: [PATCH v2 1/2]: perf util: map data buffer for preserving collected data

[Alexey Budankov]

Hi,

On 28.08.2018 11:45, Jiri Olsa wrote:
> On Mon, Aug 27, 2018 at 12:02:35PM +0300, Alexey Budankov wrote:
>> Hi,
>>
>> On 27.08.2018 11:33, Jiri Olsa wrote:
>>> On Thu, Aug 23, 2018 at 07:42:09PM +0300, Alexey Budankov wrote:
>>>
>>> SNIP
>>>
>>>> diff --git a/tools/perf/util/mmap.c b/tools/perf/util/mmap.c
>>>> index fc832676a798..e71d46cb01cc 100644
>>>> --- a/tools/perf/util/mmap.c
>>>> +++ b/tools/perf/util/mmap.c
>>>> @@ -155,6 +155,10 @@ void __weak auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp __mayb
>>>>  
>>>>  void perf_mmap__munmap(struct perf_mmap *map)
>>>>  {
>>>> +	if (map->data != NULL) {
>>>> +		munmap(map->data, perf_mmap__mmap_len(map));
>>>> +		map->data = NULL;
>>>> +	}
>>>>  	if (map->base != NULL) {
>>>>  		munmap(map->base, perf_mmap__mmap_len(map));
>>>>  		map->base = NULL;
>>>> @@ -190,6 +194,14 @@ int perf_mmap__mmap(struct perf_mmap *map, struct mmap_params *mp, int fd)
>>>>  		map->base = NULL;
>>>>  		return -1;
>>>>  	}
>>>> +	map->data = mmap(NULL, perf_mmap__mmap_len(map), PROT_READ | PROT_WRITE,
>>>> +			MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
>>>
>>> hum, why does map->data need to be mmap-ed?
>>
>> The same way as for kernel buffers. If you see better alternatives it could be applied.
> 
> I meant why not just allocate them with mmaloc?

Yep. Using malloc()/free() makes the implementation shorter. Included into [PATCH v4 1/2].

> 
> jirka
>