All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gao Xiang <gaoxiang25@huawei.com>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	<devel@driverdev.osuosl.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	<linux-erofs@lists.ozlabs.org>, "Chao Yu" <yuchao0@huawei.com>,
	Miao Xie <miaoxie@huawei.com>, <weidu.du@huawei.com>,
	Gao Xiang <gaoxiang25@huawei.com>
Subject: [PATCH 03/10] staging: erofs: atomic_cond_read_relaxed on ref-locked workgroup
Date: Tue, 20 Nov 2018 22:34:18 +0800	[thread overview]
Message-ID: <20181120143425.43637-4-gaoxiang25@huawei.com> (raw)
In-Reply-To: <20181120143425.43637-1-gaoxiang25@huawei.com>

It's better to use atomic_cond_read_relaxed, which is implemented
in hardware instructions to monitor a variable changes currently
for ARM64, instead of open-coded busy waiting.

Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Gao Xiang <gaoxiang25@huawei.com>
---
 drivers/staging/erofs/internal.h | 30 ++++++++++++++++++------------
 1 file changed, 18 insertions(+), 12 deletions(-)

diff --git a/drivers/staging/erofs/internal.h b/drivers/staging/erofs/internal.h
index 89dbd0888e53..eb80ba44d072 100644
--- a/drivers/staging/erofs/internal.h
+++ b/drivers/staging/erofs/internal.h
@@ -221,23 +221,29 @@ static inline void erofs_workgroup_unfreeze(
 	preempt_enable();
 }
 
+#if defined(CONFIG_SMP)
+static inline int erofs_wait_on_workgroup_freezed(struct erofs_workgroup *grp)
+{
+	return atomic_cond_read_relaxed(&grp->refcount,
+					VAL != EROFS_LOCKED_MAGIC);
+}
+#else
+static inline int erofs_wait_on_workgroup_freezed(struct erofs_workgroup *grp)
+{
+	int v = atomic_read(&grp->refcount);
+
+	/* workgroup is never freezed on uniprocessor systems */
+	DBG_BUGON(v == EROFS_LOCKED_MAGIC);
+	return v;
+}
+#endif
+
 static inline bool erofs_workgroup_get(struct erofs_workgroup *grp, int *ocnt)
 {
-	const int locked = (int)EROFS_LOCKED_MAGIC;
 	int o;
 
 repeat:
-	o = atomic_read(&grp->refcount);
-
-	/* spin if it is temporarily locked at the reclaim path */
-	if (unlikely(o == locked)) {
-#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
-		do
-			cpu_relax();
-		while (atomic_read(&grp->refcount) == locked);
-#endif
-		goto repeat;
-	}
+	o = erofs_wait_on_workgroup_freezed(grp);
 
 	if (unlikely(o <= 0))
 		return -1;
-- 
2.14.4


WARNING: multiple messages have this Message-ID (diff)
From: gaoxiang25@huawei.com (Gao Xiang)
Subject: [PATCH 03/10] staging: erofs: atomic_cond_read_relaxed on ref-locked workgroup
Date: Tue, 20 Nov 2018 22:34:18 +0800	[thread overview]
Message-ID: <20181120143425.43637-4-gaoxiang25@huawei.com> (raw)
In-Reply-To: <20181120143425.43637-1-gaoxiang25@huawei.com>

It's better to use atomic_cond_read_relaxed, which is implemented
in hardware instructions to monitor a variable changes currently
for ARM64, instead of open-coded busy waiting.

Reviewed-by: Chao Yu <yuchao0 at huawei.com>
Signed-off-by: Gao Xiang <gaoxiang25 at huawei.com>
---
 drivers/staging/erofs/internal.h | 30 ++++++++++++++++++------------
 1 file changed, 18 insertions(+), 12 deletions(-)

diff --git a/drivers/staging/erofs/internal.h b/drivers/staging/erofs/internal.h
index 89dbd0888e53..eb80ba44d072 100644
--- a/drivers/staging/erofs/internal.h
+++ b/drivers/staging/erofs/internal.h
@@ -221,23 +221,29 @@ static inline void erofs_workgroup_unfreeze(
 	preempt_enable();
 }
 
+#if defined(CONFIG_SMP)
+static inline int erofs_wait_on_workgroup_freezed(struct erofs_workgroup *grp)
+{
+	return atomic_cond_read_relaxed(&grp->refcount,
+					VAL != EROFS_LOCKED_MAGIC);
+}
+#else
+static inline int erofs_wait_on_workgroup_freezed(struct erofs_workgroup *grp)
+{
+	int v = atomic_read(&grp->refcount);
+
+	/* workgroup is never freezed on uniprocessor systems */
+	DBG_BUGON(v == EROFS_LOCKED_MAGIC);
+	return v;
+}
+#endif
+
 static inline bool erofs_workgroup_get(struct erofs_workgroup *grp, int *ocnt)
 {
-	const int locked = (int)EROFS_LOCKED_MAGIC;
 	int o;
 
 repeat:
-	o = atomic_read(&grp->refcount);
-
-	/* spin if it is temporarily locked at the reclaim path */
-	if (unlikely(o == locked)) {
-#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
-		do
-			cpu_relax();
-		while (atomic_read(&grp->refcount) == locked);
-#endif
-		goto repeat;
-	}
+	o = erofs_wait_on_workgroup_freezed(grp);
 
 	if (unlikely(o <= 0))
 		return -1;
-- 
2.14.4

  parent reply	other threads:[~2018-11-20 14:21 UTC|newest]

Thread overview: 92+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-11-20 14:34 [PATCH 00/10] staging: erofs: decompression stability enhancement Gao Xiang
2018-11-20 14:34 ` Gao Xiang
2018-11-20 14:34 ` [PATCH 01/10] staging: erofs: fix `trace_erofs_readpage' position Gao Xiang
2018-11-20 14:34   ` Gao Xiang
2018-11-22 10:19   ` Greg Kroah-Hartman
2018-11-22 10:19     ` Greg Kroah-Hartman
2018-11-22 10:49     ` Gao Xiang
2018-11-22 10:49       ` Gao Xiang
2018-11-22 11:00       ` Greg Kroah-Hartman
2018-11-22 11:00         ` Greg Kroah-Hartman
2018-11-20 14:34 ` [PATCH 02/10] staging: erofs: fix race when the managed cache is enabled Gao Xiang
2018-11-20 14:34   ` Gao Xiang
2018-11-22 10:17   ` Greg Kroah-Hartman
2018-11-22 10:17     ` Greg Kroah-Hartman
2018-11-22 10:42     ` Gao Xiang
2018-11-22 10:42       ` Gao Xiang
2018-11-22 11:06       ` Greg Kroah-Hartman
2018-11-22 11:06         ` Greg Kroah-Hartman
2018-11-22 11:43         ` Gao Xiang
2018-11-22 11:43           ` Gao Xiang
2018-11-22 12:26           ` Greg Kroah-Hartman
2018-11-22 12:26             ` Greg Kroah-Hartman
2018-11-22 12:41             ` Gao Xiang
2018-11-22 12:41               ` Gao Xiang
2018-11-22 13:20               ` Greg Kroah-Hartman
2018-11-22 13:20                 ` Greg Kroah-Hartman
2018-11-22 10:19   ` Greg Kroah-Hartman
2018-11-22 10:19     ` Greg Kroah-Hartman
2018-11-20 14:34 ` Gao Xiang [this message]
2018-11-20 14:34   ` [PATCH 03/10] staging: erofs: atomic_cond_read_relaxed on ref-locked workgroup Gao Xiang
2018-11-22 10:20   ` Greg Kroah-Hartman
2018-11-22 10:20     ` Greg Kroah-Hartman
2018-11-20 14:34 ` [PATCH 04/10] staging: erofs: fix `erofs_workgroup_{try_to_freeze, unfreeze}' Gao Xiang
2018-11-20 14:34   ` Gao Xiang
2018-11-22 10:21   ` Greg Kroah-Hartman
2018-11-22 10:21     ` Greg Kroah-Hartman
2018-11-22 10:29     ` Gao Xiang
2018-11-22 10:29       ` Gao Xiang
2018-11-22 11:03       ` Greg Kroah-Hartman
2018-11-22 11:03         ` Greg Kroah-Hartman
2018-11-22 11:05       ` Greg Kroah-Hartman
2018-11-22 11:05         ` Greg Kroah-Hartman
2018-11-22 11:22         ` Gao Xiang
2018-11-22 11:22           ` Gao Xiang
2018-11-20 14:34 ` [PATCH 05/10] staging: erofs: add a full barrier in erofs_workgroup_unfreeze Gao Xiang
2018-11-20 14:34   ` Gao Xiang
2018-11-22 10:22   ` Greg Kroah-Hartman
2018-11-22 10:22     ` Greg Kroah-Hartman
2018-11-22 10:56     ` Gao Xiang
2018-11-22 10:56       ` Gao Xiang
2018-11-22 18:50       ` Andrea Parri
2018-11-22 18:50         ` Andrea Parri
2018-11-23  2:51         ` Gao Xiang
2018-11-23  2:51           ` Gao Xiang
2018-11-23  9:51           ` Andrea Parri
2018-11-23  9:51             ` Andrea Parri
2018-11-23 10:00             ` Gao Xiang
2018-11-23 10:00               ` Gao Xiang
2018-11-20 14:34 ` [PATCH 06/10] staging: erofs: fix the definition of DBG_BUGON Gao Xiang
2018-11-20 14:34   ` Gao Xiang
2018-11-20 14:34 ` [PATCH 07/10] staging: erofs: separate into init_once / always Gao Xiang
2018-11-20 14:34   ` Gao Xiang
2018-11-22 10:23   ` Greg Kroah-Hartman
2018-11-22 10:23     ` Greg Kroah-Hartman
2018-11-22 10:34     ` Gao Xiang
2018-11-22 10:34       ` Gao Xiang
2018-11-22 11:05       ` Greg Kroah-Hartman
2018-11-22 11:05         ` Greg Kroah-Hartman
2018-11-22 11:11         ` Gao Xiang
2018-11-22 11:11           ` Gao Xiang
2018-11-22 11:26           ` Greg Kroah-Hartman
2018-11-22 11:26             ` Greg Kroah-Hartman
2018-11-22 11:37             ` Gao Xiang
2018-11-22 11:37               ` Gao Xiang
2018-11-22 12:00             ` Gao Xiang
2018-11-22 12:00               ` Gao Xiang
2018-11-22 13:01               ` Gao Xiang
2018-11-22 13:01                 ` Gao Xiang
2018-11-22 13:23                 ` Greg Kroah-Hartman
2018-11-22 13:23                   ` Greg Kroah-Hartman
2018-11-22 13:59                   ` Gao Xiang
2018-11-22 13:59                     ` Gao Xiang
2018-11-20 14:34 ` [PATCH 08/10] staging: erofs: locked before registering for all new workgroups Gao Xiang
2018-11-20 14:34   ` Gao Xiang
2018-11-22 10:24   ` Greg Kroah-Hartman
2018-11-22 10:24     ` Greg Kroah-Hartman
2018-11-22 10:35     ` Gao Xiang
2018-11-22 10:35       ` Gao Xiang
2018-11-20 14:34 ` [PATCH 09/10] staging: erofs: decompress asynchronously if PG_readahead page at first Gao Xiang
2018-11-20 14:34   ` Gao Xiang
2018-11-20 14:34 ` [PATCH 10/10] staging: erofs: rename strange variable names in z_erofs_vle_frontend Gao Xiang
2018-11-20 14:34   ` Gao Xiang

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=20181120143425.43637-4-gaoxiang25@huawei.com \
    --to=gaoxiang25@huawei.com \
    --cc=devel@driverdev.osuosl.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miaoxie@huawei.com \
    --cc=weidu.du@huawei.com \
    --cc=yuchao0@huawei.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.