From: Joel Fernandes <joel@joelfernandes.org>
To: Peikan Tsai <peikantsai@gmail.com>
Cc: gregkh@linuxfoundation.org, arve@android.com, tkjos@android.com,
maco@android.com, christian@brauner.io,
devel@driverdev.osuosl.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] binder: Use kmem_cache for binder_thread
Date: Thu, 29 Aug 2019 09:43:29 -0400 [thread overview]
Message-ID: <20190829134329.GA63638@google.com> (raw)
In-Reply-To: <20190829054953.GA18328@mark-All-Series>
On Thu, Aug 29, 2019 at 01:49:53PM +0800, Peikan Tsai wrote:
> Hi,
>
> The allocated size for each binder_thread is 512 bytes by kzalloc.
> Because the size of binder_thread is fixed and it's only 304 bytes.
> It will save 208 bytes per binder_thread when use create a kmem_cache
> for the binder_thread.
Awesome change and observation!!!
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
(Another thought: how did you discover this? Are you using some tools to look
into slab fragmentation?).
thanks,
- Joel
> Signed-off-by: Peikan Tsai <peikantsai@gmail.com>
> ---
> drivers/android/binder.c | 16 +++++++++++++---
> 1 file changed, 13 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/android/binder.c b/drivers/android/binder.c
> index dc1c83eafc22..043e0ebd0fe7 100644
> --- a/drivers/android/binder.c
> +++ b/drivers/android/binder.c
> @@ -87,6 +87,8 @@ static struct dentry *binder_debugfs_dir_entry_root;
> static struct dentry *binder_debugfs_dir_entry_proc;
> static atomic_t binder_last_id;
>
> +static struct kmem_cache *binder_thread_cachep;
> +
> static int proc_show(struct seq_file *m, void *unused);
> DEFINE_SHOW_ATTRIBUTE(proc);
>
> @@ -4696,14 +4698,15 @@ static struct binder_thread *binder_get_thread(struct binder_proc *proc)
> thread = binder_get_thread_ilocked(proc, NULL);
> binder_inner_proc_unlock(proc);
> if (!thread) {
> - new_thread = kzalloc(sizeof(*thread), GFP_KERNEL);
> + new_thread = kmem_cache_zalloc(binder_thread_cachep,
> + GFP_KERNEL);
> if (new_thread == NULL)
> return NULL;
> binder_inner_proc_lock(proc);
> thread = binder_get_thread_ilocked(proc, new_thread);
> binder_inner_proc_unlock(proc);
> if (thread != new_thread)
> - kfree(new_thread);
> + kmem_cache_free(binder_thread_cachep, new_thread);
> }
> return thread;
> }
> @@ -4723,7 +4726,7 @@ static void binder_free_thread(struct binder_thread *thread)
> BUG_ON(!list_empty(&thread->todo));
> binder_stats_deleted(BINDER_STAT_THREAD);
> binder_proc_dec_tmpref(thread->proc);
> - kfree(thread);
> + kmem_cache_free(binder_thread_cachep, thread);
> }
>
> static int binder_thread_release(struct binder_proc *proc,
> @@ -6095,6 +6098,12 @@ static int __init binder_init(void)
> if (ret)
> return ret;
>
> + binder_thread_cachep = kmem_cache_create("binder_thread",
> + sizeof(struct binder_thread),
> + 0, 0, NULL);
> + if (!binder_thread_cachep)
> + return -ENOMEM;
> +
> atomic_set(&binder_transaction_log.cur, ~0U);
> atomic_set(&binder_transaction_log_failed.cur, ~0U);
>
> @@ -6167,6 +6176,7 @@ static int __init binder_init(void)
>
> err_alloc_device_names_failed:
> debugfs_remove_recursive(binder_debugfs_dir_entry_root);
> + kmem_cache_destroy(binder_thread_cachep);
>
> return ret;
> }
> --
> 2.17.1
>
prev parent reply other threads:[~2019-08-29 13:43 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-29 5:49 [PATCH] binder: Use kmem_cache for binder_thread Peikan Tsai
2019-08-29 6:42 ` Greg KH
2019-08-29 13:53 ` Joel Fernandes
2019-08-29 15:27 ` Christian Brauner
2019-08-29 18:59 ` Peikan Tsai
2019-08-29 19:30 ` joel
2019-08-30 6:39 ` Greg KH
2019-09-02 14:12 ` Peikan Tsai
2019-08-30 6:38 ` Greg KH
2019-08-30 12:12 ` Christian Brauner
2019-08-29 13:43 ` Joel Fernandes [this message]
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=20190829134329.GA63638@google.com \
--to=joel@joelfernandes.org \
--cc=arve@android.com \
--cc=christian@brauner.io \
--cc=devel@driverdev.osuosl.org \
--cc=gregkh@linuxfoundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=maco@android.com \
--cc=peikantsai@gmail.com \
--cc=tkjos@android.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.