[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <f39081c8-cbcc-4478-5f9b-4f73651bc6a7@linux.ibm.com>
Date: Mon, 18 May 2020 15:18:29 +0200
From: Laurent Dufour <ldufour@...ux.ibm.com>
To: Michel Lespinasse <walken@...gle.com>,
Andrew Morton <akpm@...ux-foundation.org>,
linux-mm <linux-mm@...ck.org>
Cc: LKML <linux-kernel@...r.kernel.org>,
Peter Zijlstra <peterz@...radead.org>,
Vlastimil Babka <vbabka@...e.cz>,
Matthew Wilcox <willy@...radead.org>,
Liam Howlett <Liam.Howlett@...cle.com>,
Jerome Glisse <jglisse@...hat.com>,
Davidlohr Bueso <dave@...olabs.net>,
David Rientjes <rientjes@...gle.com>,
Hugh Dickins <hughd@...gle.com>, Ying Han <yinghan@...gle.com>,
Jason Gunthorpe <jgg@...pe.ca>,
Daniel Jordan <daniel.m.jordan@...cle.com>,
Davidlohr Bueso <dbueso@...e.de>
Subject: Re: [PATCH v5 01/10] mmap locking API: initial implementation as
rwsem wrappers
Le 22/04/2020 à 02:14, Michel Lespinasse a écrit :
> This change wraps the existing mmap_sem related rwsem calls into a new
> mmap locking API. There are two justifications for the new API:
>
> - At first, it provides an easy hooking point to instrument mmap_sem
> locking latencies independently of any other rwsems.
>
> - In the future, it may be a starting point for replacing the rwsem
> implementation with a different one, such as range locks.
>
> Signed-off-by: Michel Lespinasse <walken@...gle.com>
> Reviewed-by: Daniel Jordan <daniel.m.jordan@...cle.com>
> Reviewed-by: Davidlohr Bueso <dbueso@...e.de>
Reviewed-by: Laurent Dufour <ldufour@...ux.ibm.com>
> ---
> include/linux/mm.h | 1 +
> include/linux/mmap_lock.h | 54 +++++++++++++++++++++++++++++++++++++++
> 2 files changed, 55 insertions(+)
> create mode 100644 include/linux/mmap_lock.h
>
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index 5a323422d783..051ec782bdbb 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -15,6 +15,7 @@
> #include <linux/atomic.h>
> #include <linux/debug_locks.h>
> #include <linux/mm_types.h>
> +#include <linux/mmap_lock.h>
> #include <linux/range.h>
> #include <linux/pfn.h>
> #include <linux/percpu-refcount.h>
> diff --git a/include/linux/mmap_lock.h b/include/linux/mmap_lock.h
> new file mode 100644
> index 000000000000..97ac53b66052
> --- /dev/null
> +++ b/include/linux/mmap_lock.h
> @@ -0,0 +1,54 @@
> +#ifndef _LINUX_MMAP_LOCK_H
> +#define _LINUX_MMAP_LOCK_H
> +
> +static inline void mmap_init_lock(struct mm_struct *mm)
> +{
> + init_rwsem(&mm->mmap_sem);
> +}
> +
> +static inline void mmap_write_lock(struct mm_struct *mm)
> +{
> + down_write(&mm->mmap_sem);
> +}
> +
> +static inline int mmap_write_lock_killable(struct mm_struct *mm)
> +{
> + return down_write_killable(&mm->mmap_sem);
> +}
> +
> +static inline bool mmap_write_trylock(struct mm_struct *mm)
> +{
> + return down_write_trylock(&mm->mmap_sem) != 0;
> +}
> +
> +static inline void mmap_write_unlock(struct mm_struct *mm)
> +{
> + up_write(&mm->mmap_sem);
> +}
> +
> +static inline void mmap_write_downgrade(struct mm_struct *mm)
> +{
> + downgrade_write(&mm->mmap_sem);
> +}
> +
> +static inline void mmap_read_lock(struct mm_struct *mm)
> +{
> + down_read(&mm->mmap_sem);
> +}
> +
> +static inline int mmap_read_lock_killable(struct mm_struct *mm)
> +{
> + return down_read_killable(&mm->mmap_sem);
> +}
> +
> +static inline bool mmap_read_trylock(struct mm_struct *mm)
> +{
> + return down_read_trylock(&mm->mmap_sem) != 0;
> +}
> +
> +static inline void mmap_read_unlock(struct mm_struct *mm)
> +{
> + up_read(&mm->mmap_sem);
> +}
> +
> +#endif /* _LINUX_MMAP_LOCK_H */
>
Powered by blists - more mailing lists