From: Maarten Lankhorst <maarten.lankhorst@canonical.com>
To: Francesco Lavra <francescolavra.fl@gmail.com>
Cc: Maarten Lankhorst <m.b.lankhorst@gmail.com>,
dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org,
linux-media@vger.kernel.org, linaro-mm-sig@lists.linaro.org
Subject: Re: [Linaro-mm-sig] [PATCH 6/7] reservation: cross-device reservation support
Date: Tue, 22 Jan 2013 18:04:00 +0100 [thread overview]
Message-ID: <50FEC680.7030602@canonical.com> (raw)
In-Reply-To: <50FEC28F.3090100@gmail.com>
Op 22-01-13 17:47, Francesco Lavra schreef:
> On 01/15/2013 01:34 PM, Maarten Lankhorst wrote:
>> This adds support for a generic reservations framework that can be
>> hooked up to ttm and dma-buf and allows easy sharing of reservations
>> across devices.
>>
>> The idea is that a dma-buf and ttm object both will get a pointer
>> to a struct reservation_object, which has to be reserved before
>> anything is done with the contents of the dma-buf.
>>
>> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
>> ---
>> Documentation/DocBook/device-drivers.tmpl | 2 +
>> drivers/base/Makefile | 2 +-
>> drivers/base/reservation.c | 251 ++++++++++++++++++++++++++++++
>> include/linux/reservation.h | 182 ++++++++++++++++++++++
>> 4 files changed, 436 insertions(+), 1 deletion(-)
>> create mode 100644 drivers/base/reservation.c
>> create mode 100644 include/linux/reservation.h
> [...]
>> diff --git a/include/linux/reservation.h b/include/linux/reservation.h
>> new file mode 100644
>> index 0000000..fc2349d
>> --- /dev/null
>> +++ b/include/linux/reservation.h
>> @@ -0,0 +1,182 @@
>> +/*
>> + * Header file for reservations for dma-buf and ttm
>> + *
>> + * Copyright(C) 2011 Linaro Limited. All rights reserved.
>> + * Copyright (C) 2012 Canonical Ltd
>> + * Copyright (C) 2012 Texas Instruments
>> + *
>> + * Authors:
>> + * Rob Clark <rob.clark@linaro.org>
>> + * Maarten Lankhorst <maarten.lankhorst@canonical.com>
>> + * Thomas Hellstrom <thellstrom-at-vmware-dot-com>
>> + *
>> + * Based on bo.c which bears the following copyright notice,
>> + * but is dual licensed:
>> + *
>> + * Copyright (c) 2006-2009 VMware, Inc., Palo Alto, CA., USA
>> + * All Rights Reserved.
>> + *
>> + * Permission is hereby granted, free of charge, to any person obtaining a
>> + * copy of this software and associated documentation files (the
>> + * "Software"), to deal in the Software without restriction, including
>> + * without limitation the rights to use, copy, modify, merge, publish,
>> + * distribute, sub license, and/or sell copies of the Software, and to
>> + * permit persons to whom the Software is furnished to do so, subject to
>> + * the following conditions:
>> + *
>> + * The above copyright notice and this permission notice (including the
>> + * next paragraph) shall be included in all copies or substantial portions
>> + * of the Software.
>> + *
>> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
>> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
>> + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
>> + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
>> + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
>> + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
>> + * USE OR OTHER DEALINGS IN THE SOFTWARE.
>> + */
>> +#ifndef _LINUX_RESERVATION_H
>> +#define _LINUX_RESERVATION_H
>> +
>> +#include <linux/spinlock.h>
>> +#include <linux/mutex.h>
>> +#include <linux/fence.h>
>> +
>> +#define BUF_MAX_SHARED_FENCE 8
>> +
>> +struct fence;
>> +
>> +extern atomic_long_t reservation_counter;
>> +extern const char reservation_object_name[];
>> +extern struct lock_class_key reservation_object_class;
>> +extern const char reservation_ticket_name[];
>> +extern struct lock_class_key reservation_ticket_class;
>> +
>> +struct reservation_object {
>> + struct ticket_mutex lock;
>> +
>> + u32 fence_shared_count;
>> + struct fence *fence_excl;
>> + struct fence *fence_shared[BUF_MAX_SHARED_FENCE];
>> +};
>> +
>> +struct reservation_ticket {
>> + unsigned long seqno;
>> +#ifdef CONFIG_DEBUG_LOCK_ALLOC
>> + struct lockdep_map dep_map;
>> +#endif
>> +};
>> +
>> +/**
>> + * struct reservation_entry - reservation structure for a
>> + * reservation_object
>> + * @head: list entry
>> + * @obj_shared: pointer to a reservation_object to reserve
>> + *
>> + * Bit 0 of obj_shared is set to bool shared, as such pointer has to be
>> + * converted back, which can be done with reservation_entry_get.
>> + */
>> +struct reservation_entry {
>> + struct list_head head;
>> + unsigned long obj_shared;
>> +};
>> +
>> +
>> +static inline void
>> +reservation_object_init(struct reservation_object *obj)
>> +{
>> + obj->fence_shared_count = 0;
>> + obj->fence_excl = NULL;
>> +
>> + __ticket_mutex_init(&obj->lock, reservation_object_name,
>> + &reservation_object_class);
>> +}
>> +
>> +static inline void
>> +reservation_object_fini(struct reservation_object *obj)
>> +{
>> + int i;
>> +
>> + if (obj->fence_excl)
>> + fence_put(obj->fence_excl);
>> + for (i = 0; i < obj->fence_shared_count; ++i)
>> + fence_put(obj->fence_shared[i]);
>> +
>> + mutex_destroy(&obj->lock.base);
>> +}
>> +
>> +static inline void
>> +reservation_ticket_init(struct reservation_ticket *t)
>> +{
>> +#ifdef CONFIG_DEBUG_LOCK_ALLOC
>> + /*
>> + * Make sure we are not reinitializing a held ticket:
>> + */
>> +
>> + debug_check_no_locks_freed((void *)t, sizeof(*t));
>> + lockdep_init_map(&t->dep_map, reservation_ticket_name,
>> + &reservation_ticket_class, 0);
>> +#endif
>> + mutex_acquire(&t->dep_map, 0, 0, _THIS_IP_);
> If CONFIG_DEBUG_LOCK_ALLOC is not defined, t->dep_map is not there.
And mutex_acquire will not expand either, so it's harmless. :-)
>> + do {
>> + t->seqno = atomic_long_inc_return(&reservation_counter);
>> + } while (unlikely(!t->seqno));
>> +}
> --
> Francesco
>
next prev parent reply other threads:[~2013-01-22 17:04 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-15 12:33 [PATCH 0/7] cross-device reservation for dma-buf support Maarten Lankhorst
2013-01-15 12:33 ` [PATCH 1/7] arch: add __mutex_fastpath_lock_retval_arg to generic/sh/x86/powerpc/ia64 Maarten Lankhorst
2013-01-15 13:49 ` Maarten Lankhorst
2013-01-15 12:33 ` [PATCH 2/7] mutex: add support for reservation style locks Maarten Lankhorst
2013-01-15 13:43 ` Maarten Lankhorst
2013-01-30 1:07 ` Rob Clark
2013-01-30 11:08 ` Daniel Vetter
2013-01-30 11:52 ` Rob Clark
2013-01-31 13:38 ` Rob Clark
2013-01-30 11:16 ` Maarten Lankhorst
2013-01-15 12:34 ` [PATCH 3/7] sched: allow try_to_wake_up to be used internally outside of core.c Maarten Lankhorst
2013-01-15 12:34 ` [PATCH 4/7] fence: dma-buf cross-device synchronization (v11) Maarten Lankhorst
2013-01-22 15:13 ` [Linaro-mm-sig] " Francesco Lavra
2013-01-23 14:56 ` Maarten Lankhorst
2013-01-23 17:14 ` Francesco Lavra
2013-01-31 9:32 ` Inki Dae
2013-01-31 9:53 ` Maarten Lankhorst
2013-01-31 9:57 ` Daniel Vetter
2013-01-31 14:38 ` Inki Dae
2013-01-31 14:49 ` Daniel Vetter
2013-01-15 12:34 ` [PATCH 5/7] seqno-fence: Hardware dma-buf implementation of fencing (v4) Maarten Lankhorst
2013-01-16 6:28 ` [Linaro-mm-sig] " Inki Dae
2013-01-16 10:36 ` Maarten Lankhorst
2013-01-16 12:00 ` Inki Dae
2013-01-24 14:52 ` Inki Dae
2013-01-15 12:34 ` [PATCH 6/7] reservation: cross-device reservation support Maarten Lankhorst
2013-01-22 16:47 ` [Linaro-mm-sig] " Francesco Lavra
2013-01-22 17:04 ` Maarten Lankhorst [this message]
2013-02-04 7:06 ` Inki Dae
2013-02-04 9:57 ` Maarten Lankhorst
2013-02-04 14:51 ` Daniel Vetter
2013-01-15 12:34 ` [PATCH 7/7] reservation: Add lockdep annotation and selftests Maarten Lankhorst
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=50FEC680.7030602@canonical.com \
--to=maarten.lankhorst@canonical.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=francescolavra.fl@gmail.com \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=m.b.lankhorst@gmail.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 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 NNTP newsgroup(s).