[PULL 29/38] migration: Do chunk page in postcopy_each_ram_send_discard()

Juan Quintela posted 38 patches 4 years ago
Maintainers: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>, Fam Zheng <fam@euphon.net>, Alistair Francis <alistair@alistair23.me>, Andrew Baumann <Andrew.Baumann@microsoft.com>, "Michael S. Tsirkin" <mst@redhat.com>, Vikram Garhwal <fnu.vikram@xilinx.com>, Aurelien Jarno <aurelien@aurel32.net>, "Cédric Le Goater" <clg@kaod.org>, David Gibson <david@gibson.dropbear.id.au>, Paolo Bonzini <pbonzini@redhat.com>, Stafford Horne <shorne@gmail.com>, Juan Quintela <quintela@redhat.com>, "Philippe Mathieu-Daudé" <f4bug@amsat.org>, Pavel Pisa <pisa@cmp.felk.cvut.cz>, Laurent Vivier <laurent@vivier.eu>, "Dr. David Alan Gilbert" <dgilbert@redhat.com>, Artyom Tarasenko <atar4qemu@gmail.com>, Gerd Hoffmann <kraxel@redhat.com>, Ani Sinha <ani@anisinha.ca>, Greg Kurz <groug@kaod.org>, Marcel Apfelbaum <marcel.apfelbaum@gmail.com>, Hannes Reinecke <hare@suse.com>, "Edgar E. Iglesias" <edgar.iglesias@gmail.com>, Jason Wang <jasowang@redhat.com>, Peter Maydell <peter.maydell@linaro.org>, Daniel Henrique Barboza <danielhb413@gmail.com>, Igor Mammedov <imammedo@redhat.com>
[PULL 29/38] migration: Do chunk page in postcopy_each_ram_send_discard()
Posted by Juan Quintela 4 years ago
From: Peter Xu <peterx@redhat.com>

Right now we loop ramblocks for twice, the 1st time chunk the dirty bits with
huge page information; the 2nd time we send the discard ranges.  That's not
necessary - we can do them in a single loop.

Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
---
 migration/ram.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/migration/ram.c b/migration/ram.c
index 91d32002e8..d74a8c04b6 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -2454,6 +2454,8 @@ static int postcopy_send_discard_bm_ram(MigrationState *ms, RAMBlock *block)
     return 0;
 }
 
+static void postcopy_chunk_hostpages_pass(MigrationState *ms, RAMBlock *block);
+
 /**
  * postcopy_each_ram_send_discard: discard all RAMBlocks
  *
@@ -2475,6 +2477,14 @@ static int postcopy_each_ram_send_discard(MigrationState *ms)
     RAMBLOCK_FOREACH_NOT_IGNORED(block) {
         postcopy_discard_send_init(ms, block->idstr);
 
+        /*
+         * Deal with TPS != HPS and huge pages.  It discard any partially sent
+         * host-page size chunks, mark any partially dirty host-page size
+         * chunks as all dirty.  In this case the host-page is the host-page
+         * for the particular RAMBlock, i.e. it might be a huge page.
+         */
+        postcopy_chunk_hostpages_pass(ms, block);
+
         /*
          * Postcopy sends chunks of bitmap over the wire, but it
          * just needs indexes at this point, avoids it having
@@ -2575,7 +2585,6 @@ static void postcopy_chunk_hostpages_pass(MigrationState *ms, RAMBlock *block)
 int ram_postcopy_send_discard_bitmap(MigrationState *ms)
 {
     RAMState *rs = ram_state;
-    RAMBlock *block;
 
     RCU_READ_LOCK_GUARD();
 
@@ -2587,15 +2596,6 @@ int ram_postcopy_send_discard_bitmap(MigrationState *ms)
     rs->last_sent_block = NULL;
     rs->last_page = 0;
 
-    RAMBLOCK_FOREACH_NOT_IGNORED(block) {
-        /*
-         * Deal with TPS != HPS and huge pages.  It discard any partially sent
-         * host-page size chunks, mark any partially dirty host-page size
-         * chunks as all dirty.  In this case the host-page is the host-page
-         * for the particular RAMBlock, i.e. it might be a huge page.
-         */
-        postcopy_chunk_hostpages_pass(ms, block);
-    }
     trace_ram_postcopy_send_discard_bitmap();
 
     return postcopy_each_ram_send_discard(ms);
-- 
2.34.1