From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: David Hildenbrand <david@redhat.com>
Cc: "Pankaj Gupta" <pankaj.gupta.linux@gmail.com>,
"Daniel P . Berrangé" <berrange@redhat.com>,
"Eduardo Habkost" <ehabkost@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
qemu-devel@nongnu.org, "Igor Mammedov" <imammedo@redhat.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Marek Kedzierski" <mkedzier@redhat.com>
Subject: Re: [PATCH v2 3/6] util/oslib-posix: Don't create too many threads with small memory or little pages
Date: Tue, 27 Jul 2021 20:01:15 +0100 [thread overview]
Message-ID: <YQBX+3teXvfsZaXJ@work-vm> (raw)
In-Reply-To: <20210722123635.60608-4-david@redhat.com>
* David Hildenbrand (david@redhat.com) wrote:
> Let's limit the number of threads to something sane, especially that
> - We don't have more threads than the number of pages we have
> - We don't have threads that initialize small (< 64 MiB) memory
>
> Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> ---
> util/oslib-posix.c | 12 ++++++++++--
> 1 file changed, 10 insertions(+), 2 deletions(-)
>
> diff --git a/util/oslib-posix.c b/util/oslib-posix.c
> index 2ae6c3aaab..a1d309d495 100644
> --- a/util/oslib-posix.c
> +++ b/util/oslib-posix.c
> @@ -40,6 +40,7 @@
> #include <libgen.h>
> #include "qemu/cutils.h"
> #include "qemu/compiler.h"
> +#include "qemu/units.h"
>
> #ifdef CONFIG_LINUX
> #include <sys/syscall.h>
> @@ -529,7 +530,8 @@ static void *do_madv_populate_write_pages(void *arg)
> return NULL;
> }
>
> -static inline int get_memset_num_threads(int smp_cpus)
> +static inline int get_memset_num_threads(size_t hpagesize, size_t numpages,
> + int smp_cpus)
> {
> long host_procs = sysconf(_SC_NPROCESSORS_ONLN);
> int ret = 1;
> @@ -537,6 +539,12 @@ static inline int get_memset_num_threads(int smp_cpus)
> if (host_procs > 0) {
> ret = MIN(MIN(host_procs, MAX_MEM_PREALLOC_THREAD_COUNT), smp_cpus);
> }
> +
> + /* Especially with gigantic pages, don't create more threads than pages. */
> + ret = MIN(ret, numpages);
> + /* Don't start threads to prealloc comparatively little memory. */
> + ret = MIN(ret, MAX(1, hpagesize * numpages / (64 * MiB)));
> +
> /* In case sysconf() fails, we fall back to single threaded */
> return ret;
> }
> @@ -546,7 +554,7 @@ static bool touch_all_pages(char *area, size_t hpagesize, size_t numpages,
> {
> static gsize initialized = 0;
> MemsetContext context = {
> - .num_threads = get_memset_num_threads(smp_cpus),
> + .num_threads = get_memset_num_threads(hpagesize, numpages, smp_cpus),
> };
> size_t numpages_per_thread, leftover;
> void *(*touch_fn)(void *);
> --
> 2.31.1
>
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
next prev parent reply other threads:[~2021-07-27 19:02 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-22 12:36 [PATCH v2 0/6] util/oslib-posix: Support MADV_POPULATE_WRITE for os_mem_prealloc() David Hildenbrand
2021-07-22 12:36 ` [PATCH v2 1/6] " David Hildenbrand
2021-07-22 13:31 ` Daniel P. Berrangé
2021-07-22 13:39 ` David Hildenbrand
2021-07-22 13:47 ` Daniel P. Berrangé
2021-07-22 14:13 ` David Hildenbrand
2021-07-22 12:36 ` [PATCH v2 2/6] util/oslib-posix: Introduce and use MemsetContext for touch_all_pages() David Hildenbrand
2021-07-22 12:36 ` [PATCH v2 3/6] util/oslib-posix: Don't create too many threads with small memory or little pages David Hildenbrand
2021-07-27 19:01 ` Dr. David Alan Gilbert [this message]
2021-07-28 11:23 ` Pankaj Gupta
2021-07-22 12:36 ` [PATCH v2 4/6] util/oslib-posix: Avoid creating a single thread with MADV_POPULATE_WRITE David Hildenbrand
2021-07-27 19:04 ` Dr. David Alan Gilbert
2021-07-30 15:13 ` David Hildenbrand
2021-07-28 11:32 ` Pankaj Gupta
2021-07-22 12:36 ` [PATCH v2 5/6] util/oslib-posix: Support concurrent os_mem_prealloc() invocation David Hildenbrand
2021-07-22 12:36 ` [PATCH v2 6/6] util/oslib-posix: Forward SIGBUS to MCE handler under Linux David Hildenbrand
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=YQBX+3teXvfsZaXJ@work-vm \
--to=dgilbert@redhat.com \
--cc=berrange@redhat.com \
--cc=david@redhat.com \
--cc=ehabkost@redhat.com \
--cc=imammedo@redhat.com \
--cc=mkedzier@redhat.com \
--cc=mst@redhat.com \
--cc=pankaj.gupta.linux@gmail.com \
--cc=pbonzini@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).