Re: [PATCH v4 08/10] Reduce the PVM stop time during Checkpoint

["Dr. David Alan Gilbert" <dgilbert@redhat.com>]

* leirao (lei.rao@intel.com) wrote:
> From: "Rao, Lei" <lei.rao@intel.com>
> 
> When flushing memory from ram cache to ram during every checkpoint
> on secondary VM, we can copy continuous chunks of memory instead of
> 4096 bytes per time to reduce the time of VM stop during checkpoint.
> 
> Signed-off-by: Lei Rao <lei.rao@intel.com>

A minor comment below, but :

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

> ---
>  migration/ram.c | 45 ++++++++++++++++++++++++++++++++++++++++++---
>  1 file changed, 42 insertions(+), 3 deletions(-)
> 
> diff --git a/migration/ram.c b/migration/ram.c
> index c69a8e0..a258466 100644
> --- a/migration/ram.c
> +++ b/migration/ram.c
> @@ -822,6 +822,39 @@ unsigned long migration_bitmap_find_dirty(RAMState *rs, RAMBlock *rb,
>      return next;
>  }
>  
> +/*
> + * colo_bitmap_find_diry:find contiguous dirty pages from start
> + *
> + * Returns the page offset within memory region of the start of the contiguout
> + * dirty page
> + *
> + * @rs: current RAM state
> + * @rb: RAMBlock where to search for dirty pages
> + * @start: page where we start the search
> + * @num: the number of contiguous dirty pages
> + */
> +static inline
> +unsigned long colo_bitmap_find_dirty(RAMState *rs, RAMBlock *rb,
> +                                     unsigned long start, unsigned long *num)
> +{
> +    unsigned long size = rb->used_length >> TARGET_PAGE_BITS;
> +    unsigned long *bitmap = rb->bmap;
> +    unsigned long first, next;

It might be better to add the 
       *num = 0

here, which means this function always writes num

> +    if (ramblock_is_ignored(rb)) {
> +        return size;
> +    }
> +
> +    first = find_next_bit(bitmap, size, start);
> +    if (first >= size) {
> +        return first;
> +    }
> +    next = find_next_zero_bit(bitmap, size, first + 1);
> +    assert(next >= first);
> +    *num = next - first;
> +    return first;
> +}
> +
>  static inline bool migration_bitmap_clear_dirty(RAMState *rs,
>                                                  RAMBlock *rb,
>                                                  unsigned long page)
> @@ -3666,6 +3699,8 @@ void colo_flush_ram_cache(void)
>      void *dst_host;
>      void *src_host;
>      unsigned long offset = 0;
> +    unsigned long num = 0;

that could move inside the while loop.

> +    unsigned long i = 0;

This line could move inside the 'else' clause below that uses it.

>      memory_global_dirty_log_sync();
>      WITH_RCU_READ_LOCK_GUARD() {
> @@ -3679,19 +3714,23 @@ void colo_flush_ram_cache(void)
>          block = QLIST_FIRST_RCU(&ram_list.blocks);
>  
>          while (block) {
> -            offset = migration_bitmap_find_dirty(ram_state, block, offset);
> +            offset = colo_bitmap_find_dirty(ram_state, block, offset, &num);
>  
>              if (((ram_addr_t)offset) << TARGET_PAGE_BITS
>                  >= block->used_length) {
>                  offset = 0;
> +                num = 0;
>                  block = QLIST_NEXT_RCU(block, next);
>              } else {
> -                migration_bitmap_clear_dirty(ram_state, block, offset);
> +                for (i = 0; i < num; i++) {
> +                    migration_bitmap_clear_dirty(ram_state, block, offset + i);
> +                }
>                  dst_host = block->host
>                           + (((ram_addr_t)offset) << TARGET_PAGE_BITS);
>                  src_host = block->colo_cache
>                           + (((ram_addr_t)offset) << TARGET_PAGE_BITS);
> -                memcpy(dst_host, src_host, TARGET_PAGE_SIZE);
> +                memcpy(dst_host, src_host, TARGET_PAGE_SIZE * num);
> +                offset += num;

I was initially confused as to why the old code didn't have an offset++
but I guess that means it just checked the bit a second time that was
just cleared.

Dave


>              }
>          }
>      }
> -- 
> 1.8.3.1
> 
-- 
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK