All of lore.kernel.org
 help / color / mirror / Atom feed
From: Amit Shah <amit.shah@redhat.com>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: Juan Quintela <quintela@redhat.com>,
	"Dr. David Alan Gilbert" <dgilbert@redhat.com>,
	liang.z.li@intel.com, den@openvz.org,
	qemu list <qemu-devel@nongnu.org>,
	Amit Shah <amit.shah@redhat.com>
Subject: [Qemu-devel] [PULL 04/13] migration: Fix multi-thread compression bug
Date: Fri, 17 Jun 2016 18:36:43 +0530	[thread overview]
Message-ID: <73a8912b8aeed1c992e3f9cb4880e9d1edb935de.1466168448.git.amit.shah@redhat.com> (raw)
In-Reply-To: <cover.1466168448.git.amit.shah@redhat.com>
In-Reply-To: <cover.1466168448.git.amit.shah@redhat.com>

From: Liang Li <liang.z.li@intel.com>

Recently, a bug related to multiple thread compression feature for
live migration is reported. The destination side will be blocked
during live migration if there are heavy workload in host and
memory intensive workload in guest, this is most likely to happen
when there is one decompression thread.

Some parts of the decompression code are incorrect:
1. The main thread receives data from source side will enter a busy
loop to wait for a free decompression thread.
2. A lock is needed to protect the decomp_param[idx]->start, because
it is checked in the main thread and is updated in the decompression
thread.

Fix these two issues by following the code pattern for compression.

Signed-off-by: Liang Li <liang.z.li@intel.com>
Reported-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Tested-by: Daniel P. Berrange <berrange@redhat.com>

Signed-off-by: Liang Li <liang.z.li@intel.com>
Message-Id: <1462433579-13691-2-git-send-email-liang.z.li@intel.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
---
 migration/ram.c | 38 +++++++++++++++++++++++++++-----------
 1 file changed, 27 insertions(+), 11 deletions(-)

diff --git a/migration/ram.c b/migration/ram.c
index 42fb8ac..f3fe6c7 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -265,6 +265,7 @@ typedef struct CompressParam CompressParam;
 
 struct DecompressParam {
     bool start;
+    bool done;
     QemuMutex mutex;
     QemuCond cond;
     void *des;
@@ -289,6 +290,8 @@ static bool quit_comp_thread;
 static bool quit_decomp_thread;
 static DecompressParam *decomp_param;
 static QemuThread *decompress_threads;
+static QemuMutex decomp_done_lock;
+static QemuCond decomp_done_cond;
 
 static int do_compress_ram_page(CompressParam *param);
 
@@ -834,6 +837,7 @@ static inline void start_compression(CompressParam *param)
 
 static inline void start_decompression(DecompressParam *param)
 {
+    param->done = false;
     qemu_mutex_lock(&param->mutex);
     param->start = true;
     qemu_cond_signal(&param->cond);
@@ -2196,19 +2200,24 @@ static void *do_data_decompress(void *opaque)
         qemu_mutex_lock(&param->mutex);
         while (!param->start && !quit_decomp_thread) {
             qemu_cond_wait(&param->cond, &param->mutex);
+        }
+        if (!quit_decomp_thread) {
             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->start = false;
+            /* 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->start = false;
         qemu_mutex_unlock(&param->mutex);
+
+        qemu_mutex_lock(&decomp_done_lock);
+        param->done = true;
+        qemu_cond_signal(&decomp_done_cond);
+        qemu_mutex_unlock(&decomp_done_lock);
     }
 
     return NULL;
@@ -2222,10 +2231,13 @@ void migrate_decompress_threads_create(void)
     decompress_threads = g_new0(QemuThread, thread_count);
     decomp_param = g_new0(DecompressParam, thread_count);
     quit_decomp_thread = false;
+    qemu_mutex_init(&decomp_done_lock);
+    qemu_cond_init(&decomp_done_cond);
     for (i = 0; i < thread_count; i++) {
         qemu_mutex_init(&decomp_param[i].mutex);
         qemu_cond_init(&decomp_param[i].cond);
         decomp_param[i].compbuf = g_malloc0(compressBound(TARGET_PAGE_SIZE));
+        decomp_param[i].done = true;
         qemu_thread_create(decompress_threads + i, "decompress",
                            do_data_decompress, decomp_param + i,
                            QEMU_THREAD_JOINABLE);
@@ -2261,9 +2273,10 @@ static void decompress_data_with_multi_threads(QEMUFile *f,
     int idx, thread_count;
 
     thread_count = migrate_decompress_threads();
+    qemu_mutex_lock(&decomp_done_lock);
     while (true) {
         for (idx = 0; idx < thread_count; idx++) {
-            if (!decomp_param[idx].start) {
+            if (decomp_param[idx].done) {
                 qemu_get_buffer(f, decomp_param[idx].compbuf, len);
                 decomp_param[idx].des = host;
                 decomp_param[idx].len = len;
@@ -2273,8 +2286,11 @@ static void decompress_data_with_multi_threads(QEMUFile *f,
         }
         if (idx < thread_count) {
             break;
+        } else {
+            qemu_cond_wait(&decomp_done_cond, &decomp_done_lock);
         }
     }
+    qemu_mutex_unlock(&decomp_done_lock);
 }
 
 /*
-- 
2.7.4

  parent reply	other threads:[~2016-06-17 13:07 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-17 13:06 [Qemu-devel] [PULL 00/13] migration: many fixes Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 01/13] migration: Don't use *_to_cpup() and cpu_to_*w() Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 02/13] migration: Trace improvements Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 03/13] migration: fix inability to save VM after snapshot Amit Shah
2016-06-17 13:06 ` Amit Shah [this message]
2016-06-17 13:06 ` [Qemu-devel] [PULL 05/13] migration: Fix a potential issue Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 06/13] migration: remove useless code Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 07/13] qemu-file: Fix qemu_put_compression_data flaw Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 08/13] migration: refine ram_save_compressed_page Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 09/13] migration: protect the quit flag by lock Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 10/13] migration: refine the compression code Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 11/13] migration: refine the decompression code Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 12/13] migration: code clean up Amit Shah
2016-06-17 13:06 ` [Qemu-devel] [PULL 13/13] vmstate-static-checker: fix size mismatch detection in unused fields Amit Shah
2016-06-17 13:58 ` [Qemu-devel] [PULL 00/13] migration: many fixes Peter Maydell

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=73a8912b8aeed1c992e3f9cb4880e9d1edb935de.1466168448.git.amit.shah@redhat.com \
    --to=amit.shah@redhat.com \
    --cc=den@openvz.org \
    --cc=dgilbert@redhat.com \
    --cc=liang.z.li@intel.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=quintela@redhat.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.