qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Pavel Dovgalyuk" <dovgaluk@ispras.ru>
To: "'Alex Bennée'" <alex.bennee@linaro.org>, qemu-devel@nongnu.org
Cc: Pavel.Dovgaluk@ispras.ru, vilanova@ac.upc.edu, cota@braap.org,
	'Stefan Hajnoczi' <stefanha@redhat.com>
Subject: Re: [Qemu-devel] [RFC PATCH 19/21] plugins: add an example hotblocks plugin
Date: Mon, 8 Oct 2018 15:59:22 +0300	[thread overview]
Message-ID: <005c01d45f06$bbd7c6b0$33875410$@ru> (raw)
In-Reply-To: <20181005154910.3099-20-alex.bennee@linaro.org>

I guess this one is too tcg-dependent.
It count TB's, but breaking code into TBs may depend on many things,
like breakpoints, record/replay, ...

I mean that this measuring approach may be used only in some specific
cases, and not ok as an example.

Pavel Dovgalyuk

> -----Original Message-----
> From: Alex Bennée [mailto:alex.bennee@linaro.org]
> Sent: Friday, October 05, 2018 6:49 PM
> To: qemu-devel@nongnu.org
> Cc: Pavel.Dovgaluk@ispras.ru; vilanova@ac.upc.edu; cota@braap.org; Alex Bennée; Stefan
> Hajnoczi
> Subject: [RFC PATCH 19/21] plugins: add an example hotblocks plugin
> 
> This plugin attempts to track hotblocks by total execution count and
> average time to return to block. It is intended to be lower impact
> than saving a long log file and post-processing it.
> 
> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
> ---
>  trace/plugins/hotblocks/hotblocks.c | 69 +++++++++++++++++++++++++++++
>  1 file changed, 69 insertions(+)
>  create mode 100644 trace/plugins/hotblocks/hotblocks.c
> 
> diff --git a/trace/plugins/hotblocks/hotblocks.c b/trace/plugins/hotblocks/hotblocks.c
> new file mode 100644
> index 0000000000..e9166ad1a5
> --- /dev/null
> +++ b/trace/plugins/hotblocks/hotblocks.c
> @@ -0,0 +1,69 @@
> +/*
> + * Execution Hotblocks Plugin
> + *
> + * Copyright (c) 2018
> + * Written by Alex Bennée <alex.bennee@linaro.org>
> + *
> + * This code is licensed under the GNU GPL v2.
> + */
> +
> +#include <stdint.h>
> +#include <stdio.h>
> +#include <glib.h>
> +#include <time.h>
> +#include "plugins.h"
> +
> +/* Plugins need to take care of their own locking */
> +GMutex lock;
> +GHashTable *hotblocks;
> +
> +typedef struct {
> +    uintptr_t pc;
> +    unsigned int hits;
> +    struct timespec last;
> +    unsigned long total_time;
> +} ExecCount;
> +
> +bool plugin_init(const char *args)
> +{
> +    hotblocks = g_hash_table_new(NULL, g_direct_equal);
> +    return true;
> +}
> +
> +char *plugin_status(void)
> +{
> +    GString *report = g_string_new("We have ");
> +    char *r;
> +    g_mutex_lock(&lock);
> +    g_string_append_printf(report, "%ud entries in the hash table\n",
> +                           g_hash_table_size(hotblocks));
> +    g_mutex_unlock(&lock);
> +    r = report->str;
> +    g_string_free(report, FALSE);
> +    return r;
> +}
> +
> +bool exec_tb(void *tb, uintptr_t pc)
> +{
> +    ExecCount *cnt;
> +    struct timespec current;
> +    clock_gettime(CLOCK_MONOTONIC, &current);
> +
> +    g_mutex_lock(&lock);
> +    cnt = (ExecCount *) g_hash_table_lookup(hotblocks, (gconstpointer) pc);
> +    if (cnt) {
> +        cnt->hits++;
> +        cnt->total_time += current.tv_nsec - cnt->last.tv_nsec;
> +        cnt->last = current;
> +    } else {
> +        cnt = g_new0(ExecCount, 1);
> +        cnt->pc = pc;
> +        cnt->last = current;
> +        cnt->hits = 1;
> +        g_hash_table_insert(hotblocks, (gpointer) pc, (gpointer) cnt);
> +    }
> +    g_mutex_unlock(&lock);
> +
> +    /* As we are collecting up samples no reason to continue tracing */
> +    return false;
> +}
> --
> 2.17.1

  reply	other threads:[~2018-10-08 12:59 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-05 15:48 [Qemu-devel] [RFC PATCH 00/21] Trace updates and plugin RFC Alex Bennée
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 01/21] util/log: allow -dfilter to stack Alex Bennée
2018-10-06 16:57   ` Richard Henderson
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 02/21] util/log: add qemu_dfilter_append_range() Alex Bennée
2018-10-06 17:00   ` Richard Henderson
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 03/21] linux-user: add -dfilter progtext shortcut Alex Bennée
2018-10-06 17:12   ` Richard Henderson
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 04/21] trace: enable the exec_tb trace events Alex Bennée
2018-10-06 17:15   ` Richard Henderson
2018-10-07  1:42   ` Emilio G. Cota
2018-10-08  9:41     ` Alex Bennée
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 05/21] trace: keep a count of trace-point hits Alex Bennée
2018-10-06 18:26   ` Richard Henderson
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 06/21] trace: show trace point counts in the monitor Alex Bennée
2018-10-06 18:27   ` Richard Henderson
2018-10-08 12:51   ` Markus Armbruster
2018-10-17 11:52   ` Dr. David Alan Gilbert
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 07/21] accel/tcg/cputlb: convert tlb_flush debugging into trace events Alex Bennée
2018-10-15 16:33   ` Richard Henderson
2018-10-15 18:23     ` Alex Bennée
2018-10-15 18:37       ` Richard Henderson
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 08/21] accel/tcg/cputlb: convert remaining tlb_debug() to " Alex Bennée
2018-10-15 16:38   ` Richard Henderson
2018-10-15 18:17     ` Alex Bennée
2018-10-15 18:35       ` Richard Henderson
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 09/21] trace: suppress log output of trace points Alex Bennée
2018-10-15 16:47   ` Richard Henderson
2018-10-05 15:48 ` [Qemu-devel] [RFC PATCH 10/21] qom/cpu: add a cpu_exit trace event Alex Bennée
2018-10-06 18:51   ` Richard Henderson
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 11/21] trace: expose a plugin fn pointer in TraceEvent Alex Bennée
2018-10-15 16:52   ` Richard Henderson
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 12/21] configure: expose a plugin to the trace-backends Alex Bennée
2018-10-15 17:14   ` Richard Henderson
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 13/21] tracetool: generate plugin snippets Alex Bennée
2018-10-15 17:02   ` Richard Henderson
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 14/21] trace: add support for plugin infrastructure Alex Bennée
2018-10-07  1:29   ` Emilio G. Cota
2018-10-15 17:11   ` Richard Henderson
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 15/21] trace: add linux-user plugin support Alex Bennée
2018-10-15 17:13   ` Richard Henderson
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 16/21] trace: add infrastructure for building plugins Alex Bennée
2018-10-15 17:24   ` Richard Henderson
2018-10-15 18:16     ` Alex Bennée
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 17/21] hmp: expose status of plugins to the monitor Alex Bennée
2018-10-15 17:27   ` Richard Henderson
2018-10-17 12:04   ` Dr. David Alan Gilbert
2018-10-17 17:36     ` Markus Armbruster
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 18/21] linux-user: allow dumping of plugin status at end of run Alex Bennée
2018-10-15 17:28   ` Richard Henderson
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 19/21] plugins: add an example hotblocks plugin Alex Bennée
2018-10-08 12:59   ` Pavel Dovgalyuk [this message]
2018-10-08 14:05     ` Alex Bennée
2018-10-15 17:33   ` Richard Henderson
2018-10-15 18:15     ` Alex Bennée
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 20/21] plugins: add hotness summary to hotblocks Alex Bennée
2018-10-15 17:34   ` Richard Henderson
2018-10-05 15:49 ` [Qemu-devel] [RFC PATCH 21/21] plugin: add tlbflush stats plugin Alex Bennée
2018-10-07  1:23 ` [Qemu-devel] [RFC PATCH 00/21] Trace updates and plugin RFC Emilio G. Cota
2018-10-08 10:28   ` Alex Bennée
2018-10-08 12:51     ` Philippe Mathieu-Daudé
2018-10-08 13:59     ` Emilio G. Cota
2018-10-08 14:15       ` Alex Bennée
2018-10-09  6:28 ` Pavel Dovgalyuk
2018-10-09  8:31   ` Alex Bennée
2018-10-09  8:38     ` Pavel Dovgalyuk
2018-10-09  9:26       ` Alex Bennée
2018-10-29  7:46         ` Pavel Dovgalyuk
2018-10-29 11:30           ` Alex Bennée
2018-10-29 12:24             ` Pavel Dovgalyuk
2018-10-29 15:03               ` 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='005c01d45f06$bbd7c6b0$33875410$@ru' \
    --to=dovgaluk@ispras.ru \
    --cc=Pavel.Dovgaluk@ispras.ru \
    --cc=alex.bennee@linaro.org \
    --cc=cota@braap.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    --cc=vilanova@ac.upc.edu \
    /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).