qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Fam Zheng <famz@redhat.com>
Subject: [Qemu-devel] [PULL 17/22] configure: Add support for tcmalloc
Date: Tue, 28 Apr 2015 16:40:24 +0200	[thread overview]
Message-ID: <1430232029-9457-18-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1430232029-9457-1-git-send-email-pbonzini@redhat.com>

From: Fam Zheng <famz@redhat.com>

This adds "--enable-tcmalloc" and "--disable-tcmalloc" to allow linking
to libtcmalloc from gperftools.

tcmalloc is a malloc implementation that works well with threads and is
fast, so it is good for performance.

It is disabled by default, because the MALLOC_PERTURB_ flag we use in
tests doesn't work with tcmalloc. However we can enable tcmalloc
specific heap checker and profilers later.

An IOPS gain can be observed with virtio-blk-dataplane, other parts of
QEMU will directly benefit from it as well:

==========================================================
                       glibc malloc
----------------------------------------------------------
rw         bs         iodepth    bw     iops       latency
read       4k         1          150    38511      24
----------------------------------------------------------

==========================================================
                         tcmalloc
----------------------------------------------------------
rw         bs         iodepth    bw     iops       latency
read       4k         1          156    39969      23
----------------------------------------------------------

Signed-off-by: Fam Zheng <famz@redhat.com>
Message-Id: <1427338992-27057-1-git-send-email-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 configure | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/configure b/configure
index 6969f6f..5cf90d3 100755
--- a/configure
+++ b/configure
@@ -336,6 +336,7 @@ libssh2=""
 vhdx=""
 quorum=""
 numa=""
+tcmalloc="no"
 
 # parse CC options first
 for opt do
@@ -1134,6 +1135,10 @@ for opt do
   ;;
   --enable-numa) numa="yes"
   ;;
+  --disable-tcmalloc) tcmalloc="no"
+  ;;
+  --enable-tcmalloc) tcmalloc="yes"
+  ;;
   *)
       echo "ERROR: unknown option $opt"
       echo "Try '$0 --help' for more information"
@@ -1407,6 +1412,8 @@ Advanced options (experts only):
   --enable-quorum          enable quorum block filter support
   --disable-numa           disable libnuma support
   --enable-numa            enable libnuma support
+  --disable-tcmalloc       disable tcmalloc support
+  --enable-tcmalloc        enable tcmalloc support
 
 NOTE: The object files are built at the place where configure is launched
 EOF
@@ -3331,6 +3338,22 @@ EOF
 fi
 
 ##########################################
+# tcmalloc probe
+
+if test "$tcmalloc" == "yes" ; then
+  cat > $TMPC << EOF
+#include <stdlib.h>
+int main(void) { malloc(1); return 0; }
+EOF
+
+  if compile_prog "" "-ltcmalloc" ; then
+    LIBS="-ltcmalloc $LIBS"
+  else
+    feature_not_found "tcmalloc" "install gperftools devel"
+  fi
+fi
+
+##########################################
 # signalfd probe
 signalfd="no"
 cat > $TMPC << EOF
@@ -4441,6 +4464,7 @@ echo "lzo support       $lzo"
 echo "snappy support    $snappy"
 echo "bzip2 support     $bzip2"
 echo "NUMA host support $numa"
+echo "tcmalloc support  $tcmalloc"
 
 if test "$sdl_too_old" = "yes"; then
 echo "-> Your SDL version is too old - please upgrade to have SDL support"
-- 
2.3.5

  parent reply	other threads:[~2015-04-28 14:41 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-04-28 14:40 [Qemu-devel] [PULL 00/22] Memory, TCG, NBD, build system changes for 2015-04-27 Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 01/22] translate-all: use glib for all page descriptor allocations Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 02/22] exec: Atomic access to bounce buffer Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 03/22] linux-user, bsd-user: Remove two calls to cpu_exec_init_all Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 04/22] exec: Protect map_client_list with mutex Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 05/22] exec: Notify cpu_register_map_client caller if the bounce buffer is available Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 06/22] dma-helpers: Fix race condition of continue_after_map_failure and dma_aio_cancel Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 07/22] memory: add memory_region_ram_resize Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 08/22] acpi-build: remove dependency from ram_addr.h Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 09/22] sun4m: fix slavio sysctrl and led register sizes Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 10/22] sb16: remove useless mixer_write_indexw Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 11/22] gus: clean up MemoryRegionPortio Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 12/22] ide: there is only one data port Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 13/22] ioport: remove wrong comment Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 14/22] ioport: loosen assertions on emulation of 16-bit ports Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 15/22] ioport: reserve the whole range of an I/O port in the AddressSpace Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 16/22] exec: Respect as_translate_internal length clamp Paolo Bonzini
2015-04-28 14:40 ` Paolo Bonzini [this message]
2015-04-28 14:40 ` [Qemu-devel] [PULL 18/22] milkymist: do not modify libs-softmmu Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 19/22] Makefile.target: prepend $libs_softmmu to $LIBS Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 20/22] target-i386: disable LINT0 after reset Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 21/22] translate-all: use bitmap helpers for PageDesc's bitmap Paolo Bonzini
2015-04-28 14:40 ` [Qemu-devel] [PULL 22/22] nbd/trivial: fix type cast for ioctl Paolo Bonzini
2015-04-28 15:54 ` [Qemu-devel] [PULL 00/22] Memory, TCG, NBD, build system changes for 2015-04-27 Peter Maydell
2015-04-29  1:53   ` Fam Zheng
2015-04-29  8:15     ` 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=1430232029-9457-18-git-send-email-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=famz@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /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).