Re: [PATCH v5 03/12] migration/dirtyrate: Add RamlockDirtyInfo to store sampled page info

[David Edmondson <dme@dme.org>]

On Wednesday, 2020-08-26 at 11:33:30 +01, Dr. David Alan Gilbert wrote:

> * David Edmondson (dme@dme.org) wrote:
>> On Monday, 2020-08-24 at 17:14:31 +08, Chuan Zheng wrote:
>> 
>> > Add RamlockDirtyInfo to store sampled page info of each ramblock.
>> >
>> > Signed-off-by: Chuan Zheng <zhengchuan@huawei.com>
>> > ---
>> >  migration/dirtyrate.h | 18 ++++++++++++++++++
>> >  1 file changed, 18 insertions(+)
>> >
>> > diff --git a/migration/dirtyrate.h b/migration/dirtyrate.h
>> > index 33669b7..70000da 100644
>> > --- a/migration/dirtyrate.h
>> > +++ b/migration/dirtyrate.h
>> > @@ -19,6 +19,11 @@
>> >   */
>> >  #define DIRTYRATE_DEFAULT_SAMPLE_PAGES            512
>> >  
>> > +/*
>> > + * Record ramblock idstr
>> > + */
>> > +#define RAMBLOCK_INFO_MAX_LEN                     256
>> > +
>> >  /* Take 1s as default for calculation duration */
>> >  #define DEFAULT_FETCH_DIRTYRATE_TIME_SEC          1
>> >  
>> > @@ -27,6 +32,19 @@ struct DirtyRateConfig {
>> >      int64_t sample_period_seconds; /* time duration between two sampling */
>> >  };
>> >  
>> > +/*
>> > + * Store dirtypage info for each ramblock.
>> > + */
>> > +struct RamblockDirtyInfo {
>> > +    char idstr[RAMBLOCK_INFO_MAX_LEN]; /* idstr for each ramblock */
>> > +    uint8_t *ramblock_addr; /* base address of ramblock we measure */
>> > +    uint64_t ramblock_pages; /* ramblock size in 4K-page */
>> 
>> It's probably a stupid question, but why not store a pointer to the
>> RAMBlock rather than copying some of the details?
>
> I think I figured that out in the last round;  this code runs as:
>
>     rcu lock {
>        calculate initial CRCs
>     }
>
>     <sleep 1 second ish>
>     rcu lock {
>        calculate new CRCs
>     }
>
> A RAMBlock might get deleted between the two.

Makes sense, thanks.

dme.
-- 
Why does it have to be like this? I can never tell.