public inbox for linux-rdma@vger.kernel.org
 help / color / mirror / Atom feed
From: Jason Gunthorpe <jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
To: Bart Van Assche <Bart.VanAssche-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
Cc: "linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
	<linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
	"sean.hefty-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org"
	<sean.hefty-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Subject: Re: [PATCH rdma-core 4/6] rdmacm: Use C11 stdatomic for all atomics
Date: Wed, 15 Mar 2017 17:18:37 -0600	[thread overview]
Message-ID: <20170315231837.GA23082@obsidianresearch.com> (raw)
In-Reply-To: <1489618538.2660.14.camel-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>

On Wed, Mar 15, 2017 at 10:55:52PM +0000, Bart Van Assche wrote:
> On Wed, 2017-03-15 at 16:12 -0600, Jason Gunthorpe wrote:
> > @@ -1013,7 +1012,7 @@ int close(int socket)
> >  			return ret;
> >  	}
> >  
> > -	if (atomic_dec(&fdi->refcnt))
> > +	if (atomic_fetch_sub(&fdi->refcnt, 1))
> >  		return 0;
> >  
> >  	idm_clear(&idm, socket);
> > @@ -898,7 +898,7 @@ static int rs_create_ep(struct rsocket *rs)
> >  
> >  static void rs_release_iomap_mr(struct rs_iomap_mr *iomr)
> >  {
> > -	if (atomic_dec(&iomr->refcnt))
> > +	if (atomic_fetch_sub(&iomr->refcnt, 1))
> >  		return;
> >  
> >  	dlist_remove(&iomr->entry);
> 
> Hello Jason,
> 
> In the gcc documentation I read that __sync_sub_and_fetch() (used to
> implement atomic_dec()) returns the new value. In the C11 standard I read
> that atomic_fetch_sub() returns the old value. Do you agree with this?

Indeed, you are right, I will sqush in this patch, thanks

diff --git a/librdmacm/cma.h b/librdmacm/cma.h
index 645d1e43576279..2f5f86cf498918 100644
--- a/librdmacm/cma.h
+++ b/librdmacm/cma.h
@@ -69,12 +69,12 @@ static inline void fastlock_destroy(fastlock_t *lock)
 }
 static inline void fastlock_acquire(fastlock_t *lock)
 {
-	if (atomic_fetch_add(&lock->cnt, 1) > 1)
+	if (atomic_fetch_add(&lock->cnt, 1) > 0)
 		sem_wait(&lock->sem);
 }
 static inline void fastlock_release(fastlock_t *lock)
 {
-	if (atomic_fetch_sub(&lock->cnt, 1) > 0)
+	if (atomic_fetch_sub(&lock->cnt, 1) > 1)
 		sem_post(&lock->sem);
 }
 
diff --git a/librdmacm/preload.c b/librdmacm/preload.c
index bd1bcb1d701015..82cb0c03850b05 100644
--- a/librdmacm/preload.c
+++ b/librdmacm/preload.c
@@ -1012,7 +1012,7 @@ int close(int socket)
 			return ret;
 	}
 
-	if (atomic_fetch_sub(&fdi->refcnt, 1))
+	if (atomic_fetch_sub(&fdi->refcnt, 1) != 1)
 		return 0;
 
 	idm_clear(&idm, socket);
diff --git a/librdmacm/rsocket.c b/librdmacm/rsocket.c
index 4fbe7aaa938610..9a71035fbeed8b 100644
--- a/librdmacm/rsocket.c
+++ b/librdmacm/rsocket.c
@@ -898,7 +898,7 @@ static int rs_create_ep(struct rsocket *rs)
 
 static void rs_release_iomap_mr(struct rs_iomap_mr *iomr)
 {
-	if (atomic_fetch_sub(&iomr->refcnt, 1))
+	if (atomic_fetch_sub(&iomr->refcnt, 1) != 1)
 		return;
 
 	dlist_remove(&iomr->entry);
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

  parent reply	other threads:[~2017-03-15 23:18 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-15 22:12 [PATCH rdma-core 0/6] Sparse code changes Jason Gunthorpe
     [not found] ` <1489615927-12117-1-git-send-email-jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-03-15 22:12   ` [PATCH rdma-core 1/6] Use NULL instead of 0 to silence sparse Jason Gunthorpe
2017-03-15 22:12   ` [PATCH rdma-core 2/6] INADDR_* needs to be byteswapped before being used Jason Gunthorpe
2017-03-15 22:12   ` [PATCH rdma-core 3/6] rstream: Use waitpid instead of wait Jason Gunthorpe
2017-03-15 22:12   ` [PATCH rdma-core 4/6] rdmacm: Use C11 stdatomic for all atomics Jason Gunthorpe
     [not found]     ` <1489615927-12117-5-git-send-email-jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-03-15 22:55       ` Bart Van Assche
     [not found]         ` <1489618538.2660.14.camel-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2017-03-15 23:18           ` Jason Gunthorpe [this message]
     [not found]             ` <20170315231837.GA23082-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
2017-03-16  1:04               ` Bart Van Assche
     [not found]                 ` <1489626275.3542.1.camel-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2017-03-16 16:16                   ` Jason Gunthorpe
2017-03-15 22:12   ` [PATCH rdma-core 5/6] rsocket: Fix byte swapping when constructing path records Jason Gunthorpe
2017-03-15 22:12   ` [PATCH rdma-core 6/6] verbs: Consistently apply __attribute_const Jason Gunthorpe
2017-03-21 17:24   ` [PATCH rdma-core 0/6] Sparse code changes Doug Ledford

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=20170315231837.GA23082@obsidianresearch.com \
    --to=jgunthorpe-epgobjl8dl3ta4ec/59zmfatqe2ktcn/@public.gmane.org \
    --cc=Bart.VanAssche-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org \
    --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=sean.hefty-ral2JQCrhuEAvxtiuMwx3w@public.gmane.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