All of lore.kernel.org
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: famz@redhat.com, stefanha@redhat.com, qemu-block@nongnu.org
Subject: [Qemu-devel] [PATCH 17/18] block: introduce block_account_one_io
Date: Thu, 11 May 2017 16:42:07 +0200	[thread overview]
Message-ID: <20170511144208.24075-18-pbonzini@redhat.com> (raw)
In-Reply-To: <20170511144208.24075-1-pbonzini@redhat.com>

This is the common code to account operations that produced actual I/O.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
        v1->v2: new, to hoist qemu_clock_get_ns out of the
                "if (!stats->account_failed)" case.  Failed operations
                are not a fast path, so getting the current time
                unconditionally is not an issue. [Stefan]

 block/accounting.c | 50 +++++++++++++++++++++-----------------------------
 1 file changed, 21 insertions(+), 29 deletions(-)

diff --git a/block/accounting.c b/block/accounting.c
index 3f457c4e73..0429e31c9f 100644
--- a/block/accounting.c
+++ b/block/accounting.c
@@ -86,7 +86,7 @@ void block_acct_start(BlockAcctStats *stats, BlockAcctCookie *cookie,
     cookie->type = type;
 }
 
-void block_acct_done(BlockAcctStats *stats, BlockAcctCookie *cookie)
+static void block_account_one_io(BlockAcctStats *stats, BlockAcctCookie *cookie, bool failed)
 {
     BlockAcctTimedStats *s;
     int64_t time_ns = qemu_clock_get_ns(clock_type);
@@ -98,31 +98,14 @@ void block_acct_done(BlockAcctStats *stats, BlockAcctCookie *cookie)
 
     assert(cookie->type < BLOCK_MAX_IOTYPE);
 
-    stats->nr_bytes[cookie->type] += cookie->bytes;
-    stats->nr_ops[cookie->type]++;
-    stats->total_time_ns[cookie->type] += latency_ns;
-    stats->last_access_time_ns = time_ns;
-
-    QSLIST_FOREACH(s, &stats->intervals, entries) {
-        timed_average_account(&s->latency[cookie->type], latency_ns);
+    if (failed) {
+        stats->failed_ops[cookie->type]++;
+    } else {
+        stats->nr_bytes[cookie->type] += cookie->bytes;
+        stats->nr_ops[cookie->type]++;
     }
-}
-
-void block_acct_failed(BlockAcctStats *stats, BlockAcctCookie *cookie)
-{
-    assert(cookie->type < BLOCK_MAX_IOTYPE);
-
-    stats->failed_ops[cookie->type]++;
-
-    if (stats->account_failed) {
-        BlockAcctTimedStats *s;
-        int64_t time_ns = qemu_clock_get_ns(clock_type);
-        int64_t latency_ns = time_ns - cookie->start_time_ns;
-
-        if (qtest_enabled()) {
-            latency_ns = qtest_latency_ns;
-        }
 
+    if (!failed || stats->account_failed) {
         stats->total_time_ns[cookie->type] += latency_ns;
         stats->last_access_time_ns = time_ns;
 
@@ -132,15 +115,24 @@ void block_acct_failed(BlockAcctStats *stats, BlockAcctCookie *cookie)
     }
 }
 
+void block_acct_done(BlockAcctStats *stats, BlockAcctCookie *cookie)
+{
+    block_account_one_io(stats, cookie, false);
+}
+
+void block_acct_failed(BlockAcctStats *stats, BlockAcctCookie *cookie)
+{
+    block_account_one_io(stats, cookie, true);
+}
+
 void block_acct_invalid(BlockAcctStats *stats, enum BlockAcctType type)
 {
     assert(type < BLOCK_MAX_IOTYPE);
 
-    /* block_acct_done() and block_acct_failed() update
-     * total_time_ns[], but this one does not. The reason is that
-     * invalid requests are accounted during their submission,
-     * therefore there's no actual I/O involved. */
-
+    /* block_account_one_io() updates total_time_ns[], but this one does
+     * not.  The reason is that invalid requests are accounted during their
+     * submission, therefore there's no actual I/O involved.
+     */
     stats->invalid_ops[type]++;
 
     if (stats->account_invalid) {
-- 
2.12.2

  parent reply	other threads:[~2017-05-11 14:42 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-11 14:41 [Qemu-devel] [PATCH v2 00/18] Block layer thread safety, part 1 Paolo Bonzini
2017-05-11 14:41 ` [Qemu-devel] [PATCH 01/18] block: access copy_on_read with atomic ops Paolo Bonzini
2017-05-16 13:34   ` Stefan Hajnoczi
2017-05-11 14:41 ` [Qemu-devel] [PATCH 02/18] block: access quiesce_counter " Paolo Bonzini
2017-05-12  7:56   ` [Qemu-devel] [Qemu-block] " Alberto Garcia
2017-05-11 14:41 ` [Qemu-devel] [PATCH 03/18] block: access io_limits_disabled " Paolo Bonzini
2017-05-18 12:25   ` [Qemu-devel] [Qemu-block] " Alberto Garcia
2017-05-11 14:41 ` [Qemu-devel] [PATCH 04/18] block: access serialising_in_flight " Paolo Bonzini
2017-05-11 14:41 ` [Qemu-devel] [PATCH 05/18] block: access wakeup " Paolo Bonzini
2017-05-16 13:48   ` Stefan Hajnoczi
2017-05-11 14:41 ` [Qemu-devel] [PATCH 06/18] block: access io_plugged " Paolo Bonzini
2017-05-11 16:10   ` Eric Blake
2017-05-11 14:41 ` [Qemu-devel] [PATCH 07/18] throttle-groups: only start one coroutine from drained_begin Paolo Bonzini
2017-05-16 13:54   ` Stefan Hajnoczi
2017-05-17 21:50   ` [Qemu-devel] [Qemu-block] " Alberto Garcia
2017-05-11 14:41 ` [Qemu-devel] [PATCH 08/18] throttle-groups: do not use qemu_co_enter_next Paolo Bonzini
2017-05-16 14:44   ` Stefan Hajnoczi
2017-05-18 11:56   ` [Qemu-devel] [Qemu-block] " Alberto Garcia
2017-05-11 14:41 ` [Qemu-devel] [PATCH 09/18] throttle-groups: protect throttled requests with a CoMutex Paolo Bonzini
2017-05-16 14:47   ` Stefan Hajnoczi
2017-05-18 12:06   ` [Qemu-devel] [Qemu-block] " Alberto Garcia
2017-05-18 12:08     ` Paolo Bonzini
2017-05-18 12:19       ` Alberto Garcia
2017-05-11 14:42 ` [Qemu-devel] [PATCH 10/18] util: add stats64 module Paolo Bonzini
2017-05-16 14:47   ` Stefan Hajnoczi
2017-05-11 14:42 ` [Qemu-devel] [PATCH 11/18] block: use Stat64 for wr_highest_offset Paolo Bonzini
2017-05-11 14:42 ` [Qemu-devel] [PATCH 12/18] block: access write_gen with atomics Paolo Bonzini
2017-05-11 14:42 ` [Qemu-devel] [PATCH 13/18] block: protect tracked_requests and flush_queue with reqs_lock Paolo Bonzini
2017-05-16 14:50   ` Stefan Hajnoczi
2017-05-11 14:42 ` [Qemu-devel] [PATCH 14/18] block: introduce dirty_bitmap_mutex Paolo Bonzini
2017-05-16 14:55   ` Stefan Hajnoczi
2017-05-11 14:42 ` [Qemu-devel] [PATCH 15/18] migration/block: reset dirty bitmap before reading Paolo Bonzini
2017-05-16 15:03   ` Stefan Hajnoczi
2017-05-11 14:42 ` [Qemu-devel] [PATCH 16/18] block: protect modification of dirty bitmaps with a mutex Paolo Bonzini
2017-05-16 15:05   ` Stefan Hajnoczi
2017-05-11 14:42 ` Paolo Bonzini [this message]
2017-05-16 15:07   ` [Qemu-devel] [PATCH 17/18] block: introduce block_account_one_io Stefan Hajnoczi
2017-05-18 12:09   ` [Qemu-devel] [Qemu-block] " Alberto Garcia
2017-05-25  7:28   ` [Qemu-devel] " Fam Zheng
2017-05-11 14:42 ` [Qemu-devel] [PATCH 18/18] block: make accounting thread-safe Paolo Bonzini
2017-05-16 15:08   ` Stefan Hajnoczi
2017-05-18 12:16   ` [Qemu-devel] [Qemu-block] " Alberto Garcia
2017-05-11 18:21 ` [Qemu-devel] [PATCH v2 00/18] Block layer thread safety, part 1 no-reply
2017-05-11 19:24 ` no-reply
2017-05-16 15:08 ` Stefan Hajnoczi
2017-05-24  8:32 ` Paolo Bonzini
2017-05-25  7:40   ` Fam Zheng
2017-05-25  9:14     ` [Qemu-devel] [Qemu-block] " Paolo Bonzini
2017-05-25 16:17     ` [Qemu-devel] " Paolo Bonzini
2017-05-25 16:32 [Qemu-devel] [PATCH v3 " Paolo Bonzini
2017-05-25 16:32 ` [Qemu-devel] [PATCH 17/18] block: introduce block_account_one_io Paolo Bonzini

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=20170511144208.24075-18-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=famz@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@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.