qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Liang Li <liang.z.li@intel.com>
To: qemu-devel@nongnu.org
Cc: quintela@redhat.com, Liang Li <liang.z.li@intel.com>,
	armbru@redhat.com, lcapitulino@redhat.com,
	Yang Zhang <yang.z.zhang@intel.com>,
	amit.shah@redhat.com, dgilbert@redhat.com
Subject: [Qemu-devel] [v5 10/12] migration: Add the core code for decompression
Date: Wed, 11 Feb 2015 11:06:24 +0800	[thread overview]
Message-ID: <1423623986-590-11-git-send-email-liang.z.li@intel.com> (raw)
In-Reply-To: <1423623986-590-1-git-send-email-liang.z.li@intel.com>

Implement the core logic of multiple thread decompression,
the decompression can work now.

Signed-off-by: Liang Li <liang.z.li@intel.com>
Signed-off-by: Yang Zhang <yang.z.zhang@intel.com>
---
 arch_init.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 48 insertions(+), 2 deletions(-)

diff --git a/arch_init.c b/arch_init.c
index 12dfa34..f9e6a95 100644
--- a/arch_init.c
+++ b/arch_init.c
@@ -833,6 +833,13 @@ static inline void start_compression(CompressParam *param)
     qemu_mutex_unlock(&param->mutex);
 }
 
+static inline void start_decompression(DecompressParam *param)
+{
+    qemu_mutex_lock(&param->mutex);
+    param->busy = true;
+    qemu_cond_signal(&param->cond);
+    qemu_mutex_unlock(&param->mutex);
+}
 
 static uint64_t bytes_transferred;
 
@@ -1379,8 +1386,26 @@ void ram_handle_compressed(void *host, uint8_t ch, uint64_t size)
 
 static void *do_data_decompress(void *opaque)
 {
+    DecompressParam *param = opaque;
+    size_t pagesize;
+
     while (!quit_decomp_thread) {
-        /* To be done */
+        qemu_mutex_lock(&param->mutex);
+        while (!param->busy && !quit_decomp_thread) {
+            qemu_cond_wait(&param->cond, &param->mutex);
+            pagesize = TARGET_PAGE_SIZE;
+            if (!quit_decomp_thread) {
+                /* uncompress() will return failed in some case, especially
+                 * when the page is dirted when doing the compression, it's
+                 * not a problem because the dirty page will be retransferred
+                 * and uncompress() won't break the data in other pages.
+                 */
+                uncompress((Bytef *)param->des, &pagesize,
+                           (const Bytef *)param->compbuf, param->len);
+            }
+            param->busy = false;
+        }
+        qemu_mutex_unlock(&param->mutex);
     }
 
     return NULL;
@@ -1411,6 +1436,11 @@ void migrate_decompress_threads_join(void)
     quit_decomp_thread = true;
     thread_count = migrate_decompress_threads();
     for (i = 0; i < thread_count; i++) {
+        qemu_mutex_lock(&decomp_param[i].mutex);
+        qemu_cond_signal(&decomp_param[i].cond);
+        qemu_mutex_unlock(&decomp_param[i].mutex);
+    }
+    for (i = 0; i < thread_count; i++) {
         qemu_thread_join(decompress_threads + i);
         qemu_mutex_destroy(&decomp_param[i].mutex);
         qemu_cond_destroy(&decomp_param[i].cond);
@@ -1427,7 +1457,23 @@ void migrate_decompress_threads_join(void)
 static void decompress_data_with_multi_threads(uint8_t *compbuf,
                                                void *host, int len)
 {
-    /* To be done */
+    int idx, thread_count;
+
+    thread_count = migrate_decompress_threads();
+    while (true) {
+        for (idx = 0; idx < thread_count; idx++) {
+            if (!decomp_param[idx].busy) {
+                memcpy(decomp_param[idx].compbuf, compbuf, len);
+                decomp_param[idx].des = host;
+                decomp_param[idx].len = len;
+                start_decompression(&decomp_param[idx]);
+                break;
+            }
+        }
+        if (idx < thread_count) {
+            break;
+        }
+    }
 }
 
 static int ram_load(QEMUFile *f, void *opaque, int version_id)
-- 
1.9.1

  parent reply	other threads:[~2015-02-11  3:15 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-11  3:06 [Qemu-devel] [PATCH v5 0/12] migration: Add a new feature to do live migration Liang Li
2015-02-11  3:06 ` [Qemu-devel] [v5 01/12] docs: Add a doc about multiple thread compression Liang Li
2015-02-11  8:46   ` Juan Quintela
2015-02-11  3:06 ` [Qemu-devel] [v5 02/12] migration: Add the framework of multi-thread compression Liang Li
2015-02-11  8:52   ` Juan Quintela
2015-02-11  8:55   ` Juan Quintela
2015-02-11 11:10   ` Juan Quintela
2015-02-12  7:24     ` Li, Liang Z
2015-02-12 12:31       ` Juan Quintela
2015-02-11  3:06 ` [Qemu-devel] [v5 03/12] migration: Add the framework of multi-thread decompression Liang Li
2015-02-11  8:52   ` Juan Quintela
2015-02-11  3:06 ` [Qemu-devel] [v5 04/12] qemu-file: Add compression functions to QEMUFile Liang Li
2015-02-12 12:06   ` Juan Quintela
2015-02-11  3:06 ` [Qemu-devel] [v5 05/12] arch_init: Alloc and free data struct for compression Liang Li
2015-02-11  9:03   ` Juan Quintela
2015-02-12  5:32     ` Li, Liang Z
2015-02-12 12:05       ` Juan Quintela
2015-02-11  3:06 ` [Qemu-devel] [v5 06/12] arch_init: Add and free data struct for decompression Liang Li
2015-02-11  9:04   ` Juan Quintela
2015-02-11  3:06 ` [Qemu-devel] [v5 07/12] migration: Split the function ram_save_page Liang Li
2015-02-11  9:08   ` Juan Quintela
2015-02-11 11:02   ` Juan Quintela
2015-02-11  3:06 ` [Qemu-devel] [v5 08/12] migration: Add the core code of multi-thread compression Liang Li
2015-02-11 11:44   ` Juan Quintela
2015-02-12  7:43     ` Li, Liang Z
2015-03-02  2:46     ` Li, Liang Z
2015-02-11  3:06 ` [Qemu-devel] [v5 09/12] migration: Make compression co-work with xbzrle Liang Li
2015-02-11 11:46   ` Juan Quintela
2015-02-12  2:24     ` Li, Liang Z
2015-02-12 12:22       ` Juan Quintela
2015-02-12 15:10         ` Li, Liang Z
2015-02-11  3:06 ` Liang Li [this message]
2015-02-11 11:48   ` [Qemu-devel] [v5 10/12] migration: Add the core code for decompression Juan Quintela
2015-02-11  3:06 ` [Qemu-devel] [v5 11/12] migration: Add interface to control compression Liang Li
2015-02-11  3:06 ` [Qemu-devel] [v5 12/12] migration: Add commands to set and query parameter Liang Li
2015-02-11 11:53   ` Juan Quintela
2015-03-11 16:57   ` Dr. David Alan Gilbert
2015-03-12 10:30   ` Markus Armbruster
2015-03-19  2:30     ` Li, Liang Z
2015-03-19  7:49       ` Markus Armbruster
2015-03-19 14:21         ` Li, Liang Z
2015-03-20  5:16           ` Li, Liang Z
2015-03-19 14:33       ` Eric Blake

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1423623986-590-11-git-send-email-liang.z.li@intel.com \
    --to=liang.z.li@intel.com \
    --cc=amit.shah@redhat.com \
    --cc=armbru@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=lcapitulino@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@redhat.com \
    --cc=yang.z.zhang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).