From: Ben Peart <peartben@gmail.com>
To: Duy Nguyen <pclouds@gmail.com>
Cc: Git Mailing List <git@vger.kernel.org>,
Junio C Hamano <gitster@pobox.com>, Jeff King <peff@peff.net>
Subject: Re: [PATCH v2 06/10] preload-index.c: remove #ifdef NO_PTHREADS
Date: Mon, 29 Oct 2018 14:05:58 -0400 [thread overview]
Message-ID: <2935e3e8-c529-4292-baa3-fc3c9074028b@gmail.com> (raw)
In-Reply-To: <CACsJy8CYnngeA8JcFy83qYxKj3saB21Y32VV3R5o1PY_tkAq5Q@mail.gmail.com>
On 10/29/2018 1:26 PM, Duy Nguyen wrote:
> On Mon, Oct 29, 2018 at 6:21 PM Ben Peart <peartben@gmail.com> wrote:
>> @@ -114,6 +104,9 @@ static void preload_index(struct index_state *index,
>> threads = index->cache_nr / THREAD_COST;
>> if ((index->cache_nr > 1) && (threads < 2) &&
>> git_env_bool("GIT_TEST_PRELOAD_INDEX", 0))
>> threads = 2;
>> + cpus = online_cpus();
>> + if (threads > cpus)
>> + threads = cpus;
>> if (threads < 2)
>> return;
>> trace_performance_enter();
>
> Capping the number of threads to online_cpus() does not always make
> sense. In this case (or at least the original use case where we stat()
> over nfs) we want to issue as many requests to nfs server as possible
> to reduce latency and it has nothing to do with how many cores we
> have. Using more threads than cores might make sense since threads are
> blocked by nfs client, but we still want to send more to the server.
>
That makes sense. Some test will be necessary then.
I guess HAVE_THREADS is functionally the same as online_cpus() == 1.
For some reason, I prefer the latter - probably because I'm typically
already calling it and so it doesn't feel like a 'special' value/test I
have to remember. I just know I need to make sure the logic works
correctly with any number of cps greater than zero. :-)
next prev parent reply other threads:[~2018-10-29 18:06 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-27 7:09 [PATCH 00/10] Reduce #ifdef NO_PTHREADS Nguyễn Thái Ngọc Duy
2018-10-27 7:09 ` [PATCH 01/10] thread-utils: macros to unconditionally compile pthreads API Nguyễn Thái Ngọc Duy
2018-10-27 7:31 ` Jeff King
2018-10-27 7:40 ` Duy Nguyen
2018-10-27 8:15 ` Jeff King
2018-10-27 14:43 ` Duy Nguyen
2018-10-27 7:09 ` [PATCH 02/10] index-pack: remove #ifdef NO_PTHREADS Nguyễn Thái Ngọc Duy
2018-10-27 7:34 ` Jeff King
2018-10-27 7:09 ` [PATCH 03/10] name-hash.c: " Nguyễn Thái Ngọc Duy
2018-10-27 7:09 ` [PATCH 04/10] attr.c: " Nguyễn Thái Ngọc Duy
2018-10-27 7:09 ` [PATCH 05/10] send-pack.c: " Nguyễn Thái Ngọc Duy
2018-10-27 7:39 ` Jeff King
2018-10-27 7:09 ` [PATCH 06/10] grep: " Nguyễn Thái Ngọc Duy
2018-10-27 7:44 ` Jeff King
2018-10-29 2:16 ` Junio C Hamano
2018-10-29 14:25 ` Jeff King
2018-10-29 16:01 ` Duy Nguyen
2018-10-29 16:20 ` Jeff King
2018-10-30 1:27 ` Junio C Hamano
2018-10-27 7:10 ` [PATCH 07/10] preload-index.c: " Nguyễn Thái Ngọc Duy
2018-10-29 16:52 ` Ben Peart
2018-10-27 7:10 ` [PATCH 08/10] pack-objects: " Nguyễn Thái Ngọc Duy
2018-10-27 7:10 ` [PATCH 09/10] read-cache.c: " Nguyễn Thái Ngọc Duy
2018-10-29 17:05 ` Ben Peart
2018-10-29 17:21 ` Duy Nguyen
2018-10-29 17:58 ` Ben Peart
2018-10-30 1:44 ` Junio C Hamano
2018-10-27 7:10 ` [PATCH 10/10] Clean up pthread_create() error handling Nguyễn Thái Ngọc Duy
2018-10-27 7:24 ` [PATCH 00/10] Reduce #ifdef NO_PTHREADS Jeff King
2018-10-27 8:13 ` Jeff King
2018-10-27 17:07 ` Duy Nguyen
2018-10-27 17:29 ` [PATCH v2 " Nguyễn Thái Ngọc Duy
2018-10-27 17:29 ` [PATCH v2 01/10] thread-utils: macros to unconditionally compile pthreads API Nguyễn Thái Ngọc Duy
2018-10-27 17:30 ` [PATCH v2 02/10] index-pack: remove #ifdef NO_PTHREADS Nguyễn Thái Ngọc Duy
2018-10-27 17:30 ` [PATCH v2 03/10] name-hash.c: " Nguyễn Thái Ngọc Duy
2018-10-27 17:30 ` [PATCH v2 04/10] attr.c: " Nguyễn Thái Ngọc Duy
2018-10-27 17:30 ` [PATCH v2 05/10] grep: " Nguyễn Thái Ngọc Duy
2018-10-27 17:30 ` [PATCH v2 06/10] preload-index.c: " Nguyễn Thái Ngọc Duy
2018-10-29 17:21 ` Ben Peart
2018-10-29 17:26 ` Duy Nguyen
2018-10-29 18:05 ` Ben Peart [this message]
2018-10-29 20:11 ` Jeff King
2018-10-27 17:30 ` [PATCH v2 07/10] pack-objects: " Nguyễn Thái Ngọc Duy
2018-10-27 17:30 ` [PATCH v2 08/10] read-cache.c: " Nguyễn Thái Ngọc Duy
2018-10-29 14:30 ` Jeff King
2018-10-29 17:07 ` Ben Peart
2018-10-29 17:23 ` Ben Peart
2018-10-27 17:30 ` [PATCH v2 09/10] Clean up pthread_create() error handling Nguyễn Thái Ngọc Duy
2018-10-27 17:30 ` [PATCH v2 10/10] read-cache.c: initialize copy_len to shut up gcc 8 Nguyễn Thái Ngọc Duy
2018-10-29 14:31 ` Jeff King
2018-11-03 8:48 ` [PATCH v3 00/14] Reduce #ifdef NO_PTHREADS Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 01/14] thread-utils: macros to unconditionally compile pthreads API Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 02/14] run-command.h: include thread-utils.h instead of pthread.h Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 03/14] send-pack.c: move async's #ifdef NO_PTHREADS back to run-command.c Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 04/14] index-pack: remove #ifdef NO_PTHREADS Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 05/14] name-hash.c: " Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 06/14] attr.c: " Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 07/14] grep: " Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 08/14] grep: clean up num_threads handling Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 09/14] preload-index.c: remove #ifdef NO_PTHREADS Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 10/14] pack-objects: " Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 11/14] read-cache.c: " Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 12/14] read-cache.c: reduce branching based on HAVE_THREADS Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 13/14] read-cache.c: initialize copy_len to shut up gcc 8 Nguyễn Thái Ngọc Duy
2018-11-03 8:48 ` [PATCH v3 14/14] Clean up pthread_create() error handling Nguyễn Thái Ngọc Duy
2018-11-06 4:51 ` [PATCH v3 00/14] Reduce #ifdef NO_PTHREADS Junio C Hamano
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
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2935e3e8-c529-4292-baa3-fc3c9074028b@gmail.com \
--to=peartben@gmail.com \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=pclouds@gmail.com \
--cc=peff@peff.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.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
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 read-only IMAP folder(s) and NNTP newsgroup(s).