* Liang Li (liang.z...@intel.com) wrote:
> Now, multiple thread compression can co-work with xbzrle. when
> xbzrle is on, multiple thread compression will only work at the
> first round of ram data sync.
>
> Signed-off-by: Liang Li <liang.z...@intel.com>
> Signed-off-by: Yang Zhang <yang.z.zh...@intel.com>
> ---
> arch_init.c | 13 +++++++++++--
> 1 file changed, 11 insertions(+), 2 deletions(-)
>
> diff --git a/arch_init.c b/arch_init.c
> index 4109ad7..14bc486 100644
> --- a/arch_init.c
> +++ b/arch_init.c
> @@ -905,8 +905,11 @@ static int ram_save_compressed_page(QEMUFile *f,
> RAMBlock* block,
> * block, and all the pages in last block should have been sent
> * out, keeping this order is important.
> */
> - if (block != last_sent_block) {
> - flush_compressed_data(f);
> + if ((!ram_bulk_stage && migrate_use_xbzrle()) ||
> + block != last_sent_block) {
> + if (block != last_sent_block) {
> + flush_compressed_data(f);
> + }
> bytes_sent = save_zero_and_xbzrle_page(f, block, offset,
> last_stage, NULL);
> if (bytes_sent == -1) {
> @@ -961,6 +964,12 @@ static int ram_find_and_save_block(QEMUFile *f, bool
> last_stage)
> block = QTAILQ_FIRST(&ram_list.blocks);
> complete_round = true;
> ram_bulk_stage = false;
> + if (migrate_use_xbzrle()) {
> + /* if xbzrle is on, we terminate the compression thread
> + * at this point, there is no benefit from muti-thead */
Typo: 'muti-thead' -> 'multi-thread'
Dave
> + flush_compressed_data(f);
> + terminate_compression_threads();
> + }
> }
> } else {
> if (migrate_use_compression()) {
> --
> 1.8.3.1
>
--
Dr. David Alan Gilbert / dgilb...@redhat.com / Manchester, UK
