From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Yang Zhong <yang.zhong@intel.com>
Subject: [Qemu-devel] [PULL 41/46] rcu: reduce more than 7MB heap memory by malloc_trim()
Date: Wed, 20 Dec 2017 18:14:53 +0100 [thread overview]
Message-ID: <1513790098-9815-42-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1513790098-9815-1-git-send-email-pbonzini@redhat.com>
From: Yang Zhong <yang.zhong@intel.com>
Since there are some issues in memory alloc/free machenism
in glibc for little chunk memory, if Qemu frequently
alloc/free little chunk memory, the glibc doesn't alloc
little chunk memory from free list of glibc and still
allocate from OS, which make the heap size bigger and bigger.
This patch introduce malloc_trim(), which will free heap
memory when there is no rcu call during rcu thread loop.
malloc_trim() can be enabled/disabled by --enable-malloc-trim/
--disable-malloc-trim in the Qemu configure command. The
default malloc_trim() is enabled for libc.
Below are test results from smaps file.
(1)without patch
55f0783e1000-55f07992a000 rw-p 00000000 00:00 0 [heap]
Size: 21796 kB
Rss: 14260 kB
Pss: 14260 kB
(2)with patch
55cc5fadf000-55cc61008000 rw-p 00000000 00:00 0 [heap]
Size: 21668 kB
Rss: 6940 kB
Pss: 6940 kB
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <1513775806-19779-1-git-send-email-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
configure | 35 +++++++++++++++++++++++++++++++++++
util/rcu.c | 6 ++++++
2 files changed, 41 insertions(+)
diff --git a/configure b/configure
index 99ccc17..100309c 100755
--- a/configure
+++ b/configure
@@ -426,6 +426,7 @@ vxhs=""
supported_cpu="no"
supported_os="no"
bogus_os="no"
+malloc_trim=""
# parse CC options first
for opt do
@@ -1047,6 +1048,10 @@ for opt do
;;
--enable-tcg) tcg="yes"
;;
+ --disable-malloc-trim) malloc_trim="no"
+ ;;
+ --enable-malloc-trim) malloc_trim="yes"
+ ;;
--disable-spice) spice="no"
;;
--enable-spice) spice="yes"
@@ -1466,6 +1471,7 @@ Advanced options (experts only):
Default:trace-<pid>
--disable-slirp disable SLIRP userspace network connectivity
--enable-tcg-interpreter enable TCG with bytecode interpreter (TCI)
+ --enable-malloc-trim enable libc malloc_trim() for memory optimization
--oss-lib path to OSS library
--cpu=CPU Build for host CPU [$cpu]
--with-coroutine=BACKEND coroutine backend. Supported options:
@@ -3860,6 +3866,30 @@ if test "$tcmalloc" = "yes" && test "$jemalloc" = "yes" ; then
exit 1
fi
+# Even if malloc_trim() is available, these non-libc memory allocators
+# do not support it.
+if test "$tcmalloc" = "yes" || test "$jemalloc" = "yes" ; then
+ if test "$malloc_trim" = "yes" ; then
+ echo "Disabling malloc_trim with non-libc memory allocator"
+ fi
+ malloc_trim="no"
+fi
+
+#######################################
+# malloc_trim
+
+if test "$malloc_trim" != "no" ; then
+ cat > $TMPC << EOF
+#include <malloc.h>
+int main(void) { malloc_trim(0); return 0; }
+EOF
+ if compile_prog "" "" ; then
+ malloc_trim="yes"
+ else
+ malloc_trim="no"
+ fi
+fi
+
##########################################
# tcmalloc probe
@@ -5505,6 +5535,7 @@ if test "$tcg" = "yes" ; then
echo "TCG debug enabled $debug_tcg"
echo "TCG interpreter $tcg_interpreter"
fi
+echo "malloc trim support $malloc_trim"
echo "RDMA support $rdma"
echo "fdt support $fdt"
echo "preadv support $preadv"
@@ -6015,6 +6046,10 @@ if test "$opengl" = "yes" ; then
fi
fi
+if test "$malloc_trim" = "yes" ; then
+ echo "CONFIG_MALLOC_TRIM=y" >> $config_host_mak
+fi
+
if test "$avx2_opt" = "yes" ; then
echo "CONFIG_AVX2_OPT=y" >> $config_host_mak
fi
diff --git a/util/rcu.c b/util/rcu.c
index ca5a63e..f4d09c8 100644
--- a/util/rcu.c
+++ b/util/rcu.c
@@ -32,6 +32,9 @@
#include "qemu/atomic.h"
#include "qemu/thread.h"
#include "qemu/main-loop.h"
+#if defined(CONFIG_MALLOC_TRIM)
+#include <malloc.h>
+#endif
/*
* Global grace period counter. Bit 0 is always one in rcu_gp_ctr.
@@ -246,6 +249,9 @@ static void *call_rcu_thread(void *opaque)
qemu_event_reset(&rcu_call_ready_event);
n = atomic_read(&rcu_call_count);
if (n == 0) {
+#if defined(CONFIG_MALLOC_TRIM)
+ malloc_trim(4 * 1024 * 1024);
+#endif
qemu_event_wait(&rcu_call_ready_event);
}
}
--
1.8.3.1
next prev parent reply other threads:[~2017-12-20 17:16 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-20 17:14 [Qemu-devel] [PULL 00/46] First batch of misc patches for QEMU 2.12 Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 01/46] memfd: fix configure test Paolo Bonzini
2018-04-30 15:49 ` Greg Kurz
2017-12-20 17:14 ` [Qemu-devel] [PULL 02/46] qemu-thread: fix races on threads that exit very quickly Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 03/46] qemu-pr-helper: miscellaneous fixes Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 04/46] contrib: add systemd unit files Paolo Bonzini
2018-01-10 14:44 ` Daniel P. Berrange
2017-12-20 17:14 ` [Qemu-devel] [PULL 05/46] Revert "docker: Enable features explicitly in test-full" Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 06/46] scsi-block: Add share-rw option Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 07/46] MAITAINERS: List Fam Zheng as reviewer for SCSI patches Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 08/46] x86/cpu: Enable new SSE/AVX/AVX512 cpu features Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 09/46] hyperv: set partition-wide MSRs only on first vcpu Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 10/46] hyperv: ensure SINTx msrs are reset properly Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 11/46] hyperv: make SynIC version msr constant Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 12/46] cpus: make pause_all_cpus() play with SMP on single threaded TCG Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 13/46] cpu-exec: fix missed CPU kick during interrupt injection Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 14/46] target/i386: Fix compiler warnings Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 15/46] baum: Truncate braille device size to 84x1 Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 16/46] sockets: remove obsolete code that updated listen address Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 17/46] target/i386: Fix handling of VEX prefixes Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 18/46] tests/boot-serial-test: Make sure that we check the timeout regularly Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 19/46] tests/boot-serial-test: Add code to allow to specify our own kernel or bios Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 20/46] tests/boot-serial-test: Add support for the mcf5208evb board Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 21/46] tests/boot-serial-test: Add tests for microblaze boards Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 22/46] hw/moxie/moxiesim: Add support for loading a BIOS on moxiesim Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 23/46] tests/boot-serial-test: Add a test for the moxiesim machine Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 24/46] tests/boot-serial-test: Add support for the raspi2 machine Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 25/46] cpu: refactor cpu_address_space_init() Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 26/46] cpu: suffix cpu address spaces with cpu index Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 27/46] block/iscsi: dont leave allocmap in an invalid state on UNMAP failure Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 28/46] block/iscsi: only report an iSCSI Failure if we don't handle it gracefully Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 29/46] exec: Don't reuse unassigned_mem_ops for io_mem_rom Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 30/46] hw/mips/boston: Remove workaround for writes to ROM aborting Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 31/46] hw/i386/vmport: replace fprintf() by trace events or LOG_UNIMP Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 32/46] scsi: provide general-purpose functions to manage sense data Paolo Bonzini
2017-12-22 15:25 ` Roman Kagan
2017-12-20 17:14 ` [Qemu-devel] [PULL 33/46] scsi: replace hex constants with #defines Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 34/46] Remove legacy -no-kvm-pit option Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 35/46] i8259: convert DPRINTFs into trace Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 36/46] i8259: use DEBUG_IRQ_COUNT always Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 37/46] i8259: generalize statistics into common code Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 38/46] kvm-i8259: support "info pic" and "info irq" Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 39/46] i8259: move TYPE_INTERRUPT_STATS_PROVIDER upper Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 40/46] checkpatch: volatile with a comment or sig_atomic_t is okay Paolo Bonzini
2017-12-20 17:14 ` Paolo Bonzini [this message]
2017-12-20 17:14 ` [Qemu-devel] [PULL 42/46] chardev: fix backend events regression with mux chardev Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 43/46] test: add some chardev mux event tests Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 44/46] blockdev: convert internal NBD server to QIONetListener Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 45/46] blockdev: convert qemu-nbd " Paolo Bonzini
2017-12-20 17:14 ` [Qemu-devel] [PULL 46/46] chardev: convert the socket " Paolo Bonzini
2017-12-20 18:21 ` [Qemu-devel] [PULL 00/46] First batch of misc patches for QEMU 2.12 no-reply
2017-12-20 19:20 ` Peter Maydell
2017-12-20 21:56 ` Paolo Bonzini
2017-12-21 12:49 ` [Qemu-devel] out of bounds in set_cc_op() (was: [PULL 00/46] First batch of misc patches for QEMU 2.12) Thomas Huth
2017-12-21 13:07 ` [Qemu-devel] out of bounds in set_cc_op() Laurent Vivier
2017-12-21 13:32 ` Laurent Vivier
2017-12-21 14:10 ` Paolo Bonzini
2017-12-21 14:13 ` Laurent Vivier
2017-12-21 14:14 ` Paolo Bonzini
2017-12-21 14:36 ` Laurent Vivier
2017-12-21 19:20 ` Laurent Vivier
2017-12-21 19:30 ` Paolo Bonzini
2018-01-02 16:57 ` Thomas Huth
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=1513790098-9815-42-git-send-email-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=yang.zhong@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).