[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <27975756-0602-6a5b-f5cf-ba9a33e1863a@huawei.com>
Date: Thu, 8 Mar 2018 09:05:40 +0800
From: Yisheng Xie <xieyisheng1@...wei.com>
To: Arnaldo Carvalho de Melo <acme@...hat.com>,
Jiri Olsa <jolsa@...hat.com>
CC: <mingo@...hat.com>, <peterz@...radead.org>, <namhyung@...nel.org>,
<acme@...nel.org>, <alexander.shishkin@...ux.intel.com>,
<linux-kernel@...r.kernel.org>,
Kan Liang <kan.liang@...ux.intel.com>
Subject: Re: [PATCH] perf mmap: Discard head in overwrite_rb_find_range
Hi Arnaldo,
On 2018/3/7 21:58, Arnaldo Carvalho de Melo wrote:
> Em Wed, Mar 07, 2018 at 11:21:57AM +0100, Jiri Olsa escreveu:
>> On Tue, Mar 06, 2018 at 07:44:23PM +0800, Yisheng Xie wrote:
>>> In overwrite mode, start will be set to head in perf_mmap__read_init.
>>> Therefore, it is no need to set the start one more in
>>> overwrite_rb_find_range and *start can be used as head instead of
>>> passing head to overwrite_rb_find_range.
>>
>> it conflicts a little with recent cleanup from Kan Liang,,
>> https://marc.info/?t=152035082800001&r=1&w=2
>>
>> otherwise it looks ok to me
>
> yeah, wait a while till I push Kan's simplifications and then resubmit,
> ok?
>
Sure, I will wait and resubmit.
Thanks
Yisheng
> - Arnaldo
>
>> jirka
>>
>>>
>>> Signed-off-by: Yisheng Xie <xieyisheng1@...wei.com>
>>> ---
>>> tools/perf/util/mmap.c | 15 +++++++--------
>>> 1 file changed, 7 insertions(+), 8 deletions(-)
>>>
>>> diff --git a/tools/perf/util/mmap.c b/tools/perf/util/mmap.c
>>> index 91531a7..6751ac9 100644
>>> --- a/tools/perf/util/mmap.c
>>> +++ b/tools/perf/util/mmap.c
>>> @@ -223,19 +223,18 @@ int perf_mmap__mmap(struct perf_mmap *map, struct mmap_params *mp, int fd)
>>> return 0;
>>> }
>>>
>>> -static int overwrite_rb_find_range(void *buf, int mask, u64 head, u64 *start, u64 *end)
>>> +static int overwrite_rb_find_range(void *buf, int mask, u64 *start, u64 *end)
>>> {
>>> struct perf_event_header *pheader;
>>> - u64 evt_head = head;
>>> + u64 evt_head = *start;
>>> int size = mask + 1;
>>>
>>> - pr_debug2("overwrite_rb_find_range: buf=%p, head=%"PRIx64"\n", buf, head);
>>> - pheader = (struct perf_event_header *)(buf + (head & mask));
>>> - *start = head;
>>> + pr_debug2("%s: buf=%p, start=%"PRIx64"\n", __func__, buf, *start);
>>> + pheader = (struct perf_event_header *)(buf + (*start & mask));
>>> while (true) {
>>> - if (evt_head - head >= (unsigned int)size) {
>>> + if (evt_head - *start >= (unsigned int)size) {
>>> pr_debug("Finished reading overwrite ring buffer: rewind\n");
>>> - if (evt_head - head > (unsigned int)size)
>>> + if (evt_head - *start > (unsigned int)size)
>>> evt_head -= pheader->size;
>>> *end = evt_head;
>>> return 0;
>>> @@ -287,7 +286,7 @@ int perf_mmap__read_init(struct perf_mmap *md, bool overwrite,
>>> * Backward ring buffer is full. We still have a chance to read
>>> * most of data from it.
>>> */
>>> - if (overwrite_rb_find_range(data, md->mask, head, startp, endp))
>>> + if (overwrite_rb_find_range(data, md->mask, startp, endp))
>>> return -EINVAL;
>>> }
>>>
>>> --
>>> 1.7.12.4
>>>
>
> .
>
Powered by blists - more mailing lists