From: Sergey Fedorov <serge.fdrv@gmail.com>
To: "Emilio G. Cota" <cota@braap.org>,
QEMU Developers <qemu-devel@nongnu.org>,
MTTCG Devel <mttcg@listserver.greensocs.com>
Cc: "Alex Bennée" <alex.bennee@linaro.org>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Richard Henderson" <rth@twiddle.net>
Subject: Re: [Qemu-devel] [PATCH v6 12/15] qht: add qht-bench, a performance benchmark
Date: Sun, 29 May 2016 23:45:23 +0300 [thread overview]
Message-ID: <574B54E3.3010908@gmail.com> (raw)
In-Reply-To: <1464138802-23503-13-git-send-email-cota@braap.org>
On 25/05/16 04:13, Emilio G. Cota wrote:
> diff --git a/tests/qht-bench.c b/tests/qht-bench.c
> new file mode 100644
> index 0000000..30d27c8
> --- /dev/null
> +++ b/tests/qht-bench.c
> @@ -0,0 +1,474 @@
(snip)
> +static void do_rw(struct thread_info *info)
> +{
> + struct thread_stats *stats = &info->stats;
> + uint32_t hash;
> + long *p;
> +
> + if (info->r >= update_threshold) {
> + bool read;
> +
> + p = &keys[info->r & (lookup_range - 1)];
> + hash = h(*p);
> + read = qht_lookup(&ht, is_equal, p, hash);
> + if (read) {
> + stats->rd++;
> + } else {
> + stats->not_rd++;
> + }
> + } else {
> + p = &keys[info->r & (update_range - 1)];
> + hash = h(*p);
The previous two lines are common for the both "if" branches. Lets move
it above the "if".
> + if (info->write_op) {
> + bool written = false;
> +
> + if (qht_lookup(&ht, is_equal, p, hash) == NULL) {
> + written = qht_insert(&ht, p, hash);
> + }
> + if (written) {
> + stats->in++;
> + } else {
> + stats->not_in++;
> + }
> + } else {
> + bool removed = false;
> +
> + if (qht_lookup(&ht, is_equal, p, hash)) {
> + removed = qht_remove(&ht, p, hash);
> + }
> + if (removed) {
> + stats->rm++;
> + } else {
> + stats->not_rm++;
> + }
> + }
> + info->write_op = !info->write_op;
> + }
> +}
> +
> +static void *thread_func(void *p)
> +{
> + struct thread_info *info = p;
> +
> + while (!atomic_mb_read(&test_start)) {
> + cpu_relax();
> + }
> +
> + rcu_register_thread();
Shouldn't we do this before checking for 'test_start'?
> +
> + rcu_read_lock();
Why don't we do rcu_read_lock()/rcu_read_unlock() inside the loop?
> + while (!atomic_read(&test_stop)) {
> + info->r = xorshift64star(info->r);
> + info->func(info);
> + }
> + rcu_read_unlock();
> +
> + rcu_unregister_thread();
> + return NULL;
> +}
> +
> +/* sets everything except info->func */
> +static void prepare_thread_info(struct thread_info *info, int i)
> +{
> + /* seed for the RNG; each thread should have a different one */
> + info->r = (i + 1) ^ time(NULL);
> + /* the first update will be a write */
> + info->write_op = true;
> + /* the first resize will be down */
> + info->resize_down = true;
> +
> + memset(&info->stats, 0, sizeof(info->stats));
> +}
> +
> +static void
> +th_create_n(QemuThread **threads, struct thread_info **infos, const char *name,
> + void (*func)(struct thread_info *), int offset, int n)
'offset' is not used in this function.
> +{
> + struct thread_info *info;
> + QemuThread *th;
> + int i;
> +
> + th = g_malloc(sizeof(*th) * n);
> + *threads = th;
> +
> + info = qemu_memalign(64, sizeof(*info) * n);
> + *infos = info;
> +
> + for (i = 0; i < n; i++) {
> + prepare_thread_info(&info[i], i);
> + info[i].func = func;
> + qemu_thread_create(&th[i], name, thread_func, &info[i],
> + QEMU_THREAD_JOINABLE);
> + }
> +}
> +
(snip)
> +
> +static void run_test(void)
> +{
> + unsigned int remaining;
> + int i;
> +
Are we sure all the threads are ready at this point? Otherwise why
bother with 'test_start' flag?
> + atomic_mb_set(&test_start, true);
> + do {
> + remaining = sleep(duration);
> + } while (remaining);
> + atomic_mb_set(&test_stop, true);
> +
> + for (i = 0; i < n_rw_threads; i++) {
> + qemu_thread_join(&rw_threads[i]);
> + }
> + for (i = 0; i < n_rz_threads; i++) {
> + qemu_thread_join(&rz_threads[i]);
> + }
> +}
> +
>
Kind regards,
Sergey
next prev parent reply other threads:[~2016-05-29 20:45 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-25 1:13 [Qemu-devel] [PATCH v6 00/15] tb hash improvements Emilio G. Cota
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 01/15] compiler.h: add QEMU_ALIGNED() to enforce struct alignment Emilio G. Cota
2016-05-27 19:54 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 02/15] seqlock: remove optional mutex Emilio G. Cota
2016-05-27 19:55 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 03/15] seqlock: rename write_lock/unlock to write_begin/end Emilio G. Cota
2016-05-27 19:59 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 04/15] include/processor.h: define cpu_relax() Emilio G. Cota
2016-05-27 20:53 ` Sergey Fedorov
2016-05-27 21:10 ` Emilio G. Cota
2016-05-28 12:35 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 05/15] qemu-thread: add simple test-and-set spinlock Emilio G. Cota
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 06/15] exec: add tb_hash_func5, derived from xxhash Emilio G. Cota
2016-05-28 12:36 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 07/15] tb hash: hash phys_pc, pc, and flags with xxhash Emilio G. Cota
2016-05-28 12:39 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 08/15] qdist: add module to represent frequency distributions of data Emilio G. Cota
2016-05-28 18:15 ` Sergey Fedorov
2016-06-03 17:22 ` Emilio G. Cota
2016-06-03 17:29 ` Sergey Fedorov
2016-06-03 17:46 ` Sergey Fedorov
2016-06-06 23:40 ` Emilio G. Cota
2016-06-07 14:06 ` Sergey Fedorov
2016-06-07 22:53 ` Emilio G. Cota
2016-06-08 13:09 ` Sergey Fedorov
2016-06-07 1:05 ` Emilio G. Cota
2016-06-07 15:56 ` Sergey Fedorov
2016-06-08 0:02 ` Emilio G. Cota
2016-06-08 14:10 ` Sergey Fedorov
2016-06-08 18:06 ` Emilio G. Cota
2016-06-08 18:18 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 09/15] qdist: add test program Emilio G. Cota
2016-05-28 18:56 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 10/15] qht: QEMU's fast, resizable and scalable Hash Table Emilio G. Cota
2016-05-29 19:52 ` Sergey Fedorov
2016-05-29 19:55 ` Sergey Fedorov
2016-05-31 7:46 ` Alex Bennée
2016-06-01 20:53 ` Sergey Fedorov
2016-06-03 9:18 ` Emilio G. Cota
2016-06-03 15:19 ` Sergey Fedorov
2016-06-03 11:01 ` Emilio G. Cota
2016-06-03 15:34 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 11/15] qht: add test program Emilio G. Cota
2016-05-29 20:15 ` Sergey Fedorov
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 12/15] qht: add qht-bench, a performance benchmark Emilio G. Cota
2016-05-29 20:45 ` Sergey Fedorov [this message]
2016-06-03 11:41 ` Emilio G. Cota
2016-06-03 15:41 ` Sergey Fedorov
2016-05-31 15:12 ` Alex Bennée
2016-05-31 16:44 ` Emilio G. Cota
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 13/15] qht: add test-qht-par to invoke qht-bench from 'check' target Emilio G. Cota
2016-05-29 20:53 ` Sergey Fedorov
2016-06-03 11:07 ` Emilio G. Cota
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 14/15] tb hash: track translated blocks with qht Emilio G. Cota
2016-05-29 21:09 ` Sergey Fedorov
2016-05-31 8:39 ` Alex Bennée
2016-05-25 1:13 ` [Qemu-devel] [PATCH v6 15/15] translate-all: add tb hash bucket info to 'info jit' dump Emilio G. Cota
2016-05-29 21:14 ` Sergey Fedorov
2016-06-08 6:25 ` [Qemu-devel] [PATCH v6 00/15] tb hash improvements Alex Bennée
2016-06-08 15:16 ` Emilio G. Cota
2016-06-08 15:35 ` Richard Henderson
2016-06-08 15:37 ` Sergey Fedorov
2016-06-08 16:45 ` Alex Bennée
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=574B54E3.3010908@gmail.com \
--to=serge.fdrv@gmail.com \
--cc=alex.bennee@linaro.org \
--cc=cota@braap.org \
--cc=mttcg@listserver.greensocs.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/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).