From patchwork Fri Feb 23 12:14:02 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Nuno Sa X-Patchwork-Id: 13568977 Received: from mx0b-00128a01.pphosted.com (mx0a-00128a01.pphosted.com [148.163.135.77]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 2D26D7AE78; Fri, 23 Feb 2024 12:11:40 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=148.163.135.77 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708690303; cv=none; b=UDRFwMs3d92Myr2ph8BSo36gUiJOnQfftF9cNQwZnkJUd5tYie6bRjsY6iIOdU2nTvNdRLmmboiQjW4pXs0MSmtyYRies3AEIJ82xKWBs/a7nk9wCTp6pNh7VqYz8mSESXbP+vsDWTM4ZAkOvra1MoJVyBh/RgXER9llf2dOO/c= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708690303; c=relaxed/simple; bh=MMA/LIpuAFXsZIP2g4Qp61L2mQ+IqF27lslq2bL1Ac8=; h=From:Date:Subject:MIME-Version:Content-Type:Message-ID:References: In-Reply-To:To:CC; b=SOor8NNLMfrdfbZkx5KfKjKKHbU90McPe2znmioMKK93jjfa2IWK7DVgPR39aRHvR7ZTBaql4ZQnUCtOY5TUBaXouV3QJo5PCM9h27GJnM+YIIPlO4astyCVUMlL727jK0zlnvKHiwBS5yJoPvemc7ICCIUg0+OpSukX4aI2A6k= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=analog.com; spf=pass smtp.mailfrom=analog.com; dkim=pass (2048-bit key) header.d=analog.com header.i=@analog.com header.b=SYZz4sp/; arc=none smtp.client-ip=148.163.135.77 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=analog.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=analog.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=analog.com header.i=@analog.com header.b="SYZz4sp/" Received: from pps.filterd (m0375855.ppops.net [127.0.0.1]) by mx0b-00128a01.pphosted.com (8.17.1.24/8.17.1.24) with ESMTP id 41N6rd0k030350; Fri, 23 Feb 2024 07:11:21 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=analog.com; h= from:date:subject:mime-version:content-type :content-transfer-encoding:message-id:references:in-reply-to:to :cc; s=DKIM; bh=caKbzzOgQcSL0nwDkNESBmj8xoOV5cNJB/71r9s4Wmk=; b= SYZz4sp/8RVCaAdkGjr2rxCs0UK9OMFMKVsp8dT8rsKGS7Pqiz6/gq2tNZquk2lR WpY4vaLzMxXxuIb6U4ET+9+hlv8RqpNe/j0TZvErH6r30y6W9XzxT98H8p0gjY3G vGuT0V2GRs6PAcqxxIS+ZivFxFZ1WOcwNbQYBNx8yfr0g0lCqzyEp3B/693wa/Qb CL5t1GN7ErWDFwBoBsK4y30PHfi/Ie7zRwwjskzwW6NhhCWCr4RXIydWHedXbwp9 I6OxASyYbzzRdVnPjztRK5vbZYiFJwh3JeyDIuI9YwqG1D4QhBM9CbstQFN/3iAv 7+53Wf5kHJRROMC9B2X23Q== Received: from nwd2mta3.analog.com ([137.71.173.56]) by mx0b-00128a01.pphosted.com (PPS) with ESMTPS id 3wepgc11p7-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Fri, 23 Feb 2024 07:11:21 -0500 (EST) Received: from ASHBMBX8.ad.analog.com (ASHBMBX8.ad.analog.com [10.64.17.5]) by nwd2mta3.analog.com (8.14.7/8.14.7) with ESMTP id 41NCB5rB033127 (version=TLSv1/SSLv3 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=FAIL); Fri, 23 Feb 2024 07:11:05 -0500 Received: from ASHBMBX9.ad.analog.com (10.64.17.10) by ASHBMBX8.ad.analog.com (10.64.17.5) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.986.14; Fri, 23 Feb 2024 07:11:04 -0500 Received: from zeus.spd.analog.com (10.66.68.11) by ashbmbx9.ad.analog.com (10.64.17.10) with Microsoft SMTP Server id 15.2.986.14 via Frontend Transport; Fri, 23 Feb 2024 07:11:04 -0500 Received: from [127.0.0.1] ([10.44.3.55]) by zeus.spd.analog.com (8.15.1/8.15.1) with ESMTP id 41NCAb2V032246; Fri, 23 Feb 2024 07:10:58 -0500 From: Nuno Sa Date: Fri, 23 Feb 2024 13:14:02 +0100 Subject: [PATCH v7 4/6] iio: buffer-dma: Enable support for DMABUFs Precedence: bulk X-Mailing-List: dmaengine@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Message-ID: <20240223-iio-dmabuf-v7-4-78cfaad117b9@analog.com> References: <20240223-iio-dmabuf-v7-0-78cfaad117b9@analog.com> In-Reply-To: <20240223-iio-dmabuf-v7-0-78cfaad117b9@analog.com> To: Vinod Koul , Lars-Peter Clausen , Jonathan Cameron , Sumit Semwal , =?utf-8?q?Christian_K=C3=B6nig?= , Jonathan Corbet , Paul Cercueil CC: Daniel Vetter , Michael Hennerich , , , , , , X-Mailer: b4 0.13.0 X-Developer-Signature: v=1; a=ed25519-sha256; t=1708690439; l=12470; i=nuno.sa@analog.com; s=20231116; h=from:subject:message-id; bh=qkzQZP5LbzNx55b8OghG1F1wB4XBJJdReDWUW43Sz9Y=; b=PohMGitn+TnJk45lhJVfHh+SJFwKB0vQTdIIBn3pkVfEZ2bPv8UA9axgwWdrFFbxUmD+DuleR kqo8nNaZBb3DOhgBpWgZkMMk2J+jGbE7rpLDYRrsqyTObKl0nzPvFH3 X-Developer-Key: i=nuno.sa@analog.com; a=ed25519; pk=3NQwYA013OUYZsmDFBf8rmyyr5iQlxV/9H4/Df83o1E= X-ADIRuleOP-NewSCL: Rule Triggered X-Proofpoint-GUID: QSyGBBlqywm1203IQ7EfbA5W64oHMtpw X-Proofpoint-ORIG-GUID: QSyGBBlqywm1203IQ7EfbA5W64oHMtpw X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.272,Aquarius:18.0.1011,Hydra:6.0.619,FMLib:17.11.176.26 definitions=2024-02-22_15,2024-02-23_01,2023-05-22_02 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 bulkscore=0 impostorscore=0 mlxlogscore=999 clxscore=1015 priorityscore=1501 phishscore=0 malwarescore=0 adultscore=0 mlxscore=0 spamscore=0 lowpriorityscore=0 suspectscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.19.0-2402120000 definitions=main-2402230088 From: Paul Cercueil Implement iio_dma_buffer_attach_dmabuf(), iio_dma_buffer_detach_dmabuf() and iio_dma_buffer_transfer_dmabuf(), which can then be used by the IIO DMA buffer implementations. Signed-off-by: Paul Cercueil Signed-off-by: Nuno Sa --- drivers/iio/buffer/industrialio-buffer-dma.c | 181 +++++++++++++++++++++++++-- include/linux/iio/buffer-dma.h | 31 +++++ 2 files changed, 201 insertions(+), 11 deletions(-) diff --git a/drivers/iio/buffer/industrialio-buffer-dma.c b/drivers/iio/buffer/industrialio-buffer-dma.c index 5610ba67925e..c0f539af98f9 100644 --- a/drivers/iio/buffer/industrialio-buffer-dma.c +++ b/drivers/iio/buffer/industrialio-buffer-dma.c @@ -4,6 +4,8 @@ * Author: Lars-Peter Clausen */ +#include +#include #include #include #include @@ -14,6 +16,8 @@ #include #include #include +#include +#include #include #include @@ -94,13 +98,18 @@ static void iio_buffer_block_release(struct kref *kref) { struct iio_dma_buffer_block *block = container_of(kref, struct iio_dma_buffer_block, kref); + struct iio_dma_buffer_queue *queue = block->queue; - WARN_ON(block->state != IIO_BLOCK_STATE_DEAD); + WARN_ON(block->fileio && block->state != IIO_BLOCK_STATE_DEAD); - dma_free_coherent(block->queue->dev, PAGE_ALIGN(block->size), - block->vaddr, block->phys_addr); + if (block->fileio) { + dma_free_coherent(queue->dev, PAGE_ALIGN(block->size), + block->vaddr, block->phys_addr); + } else { + atomic_dec(&queue->num_dmabufs); + } - iio_buffer_put(&block->queue->buffer); + iio_buffer_put(&queue->buffer); kfree(block); } @@ -163,7 +172,7 @@ static struct iio_dma_buffer_queue *iio_buffer_to_queue(struct iio_buffer *buf) } static struct iio_dma_buffer_block *iio_dma_buffer_alloc_block( - struct iio_dma_buffer_queue *queue, size_t size) + struct iio_dma_buffer_queue *queue, size_t size, bool fileio) { struct iio_dma_buffer_block *block; @@ -171,13 +180,16 @@ static struct iio_dma_buffer_block *iio_dma_buffer_alloc_block( if (!block) return NULL; - block->vaddr = dma_alloc_coherent(queue->dev, PAGE_ALIGN(size), - &block->phys_addr, GFP_KERNEL); - if (!block->vaddr) { - kfree(block); - return NULL; + if (fileio) { + block->vaddr = dma_alloc_coherent(queue->dev, PAGE_ALIGN(size), + &block->phys_addr, GFP_KERNEL); + if (!block->vaddr) { + kfree(block); + return NULL; + } } + block->fileio = fileio; block->size = size; block->state = IIO_BLOCK_STATE_DONE; block->queue = queue; @@ -186,6 +198,9 @@ static struct iio_dma_buffer_block *iio_dma_buffer_alloc_block( iio_buffer_get(&queue->buffer); + if (!fileio) + atomic_inc(&queue->num_dmabufs); + return block; } @@ -206,13 +221,21 @@ void iio_dma_buffer_block_done(struct iio_dma_buffer_block *block) { struct iio_dma_buffer_queue *queue = block->queue; unsigned long flags; + bool cookie; + + cookie = dma_fence_begin_signalling(); spin_lock_irqsave(&queue->list_lock, flags); _iio_dma_buffer_block_done(block); spin_unlock_irqrestore(&queue->list_lock, flags); + if (!block->fileio) + iio_buffer_signal_dmabuf_done(block->fence, 0); + iio_buffer_block_put_atomic(block); wake_up_interruptible_poll(&queue->buffer.pollq, EPOLLIN | EPOLLRDNORM); + + dma_fence_end_signalling(cookie); } EXPORT_SYMBOL_GPL(iio_dma_buffer_block_done); @@ -231,17 +254,27 @@ void iio_dma_buffer_block_list_abort(struct iio_dma_buffer_queue *queue, { struct iio_dma_buffer_block *block, *_block; unsigned long flags; + bool cookie; + + cookie = dma_fence_begin_signalling(); spin_lock_irqsave(&queue->list_lock, flags); list_for_each_entry_safe(block, _block, list, head) { list_del(&block->head); block->bytes_used = 0; _iio_dma_buffer_block_done(block); + + if (!block->fileio) + iio_buffer_signal_dmabuf_done(block->fence, -EINTR); iio_buffer_block_put_atomic(block); } spin_unlock_irqrestore(&queue->list_lock, flags); + if (queue->fileio.enabled) + queue->fileio.enabled = false; + wake_up_interruptible_poll(&queue->buffer.pollq, EPOLLIN | EPOLLRDNORM); + dma_fence_end_signalling(cookie); } EXPORT_SYMBOL_GPL(iio_dma_buffer_block_list_abort); @@ -261,6 +294,16 @@ static bool iio_dma_block_reusable(struct iio_dma_buffer_block *block) } } +static bool iio_dma_buffer_can_use_fileio(struct iio_dma_buffer_queue *queue) +{ + /* + * Note that queue->num_dmabufs cannot increase while the queue is + * locked, it can only decrease, so it does not race against + * iio_dma_buffer_alloc_block(). + */ + return queue->fileio.enabled || !atomic_read(&queue->num_dmabufs); +} + /** * iio_dma_buffer_request_update() - DMA buffer request_update callback * @buffer: The buffer which to request an update @@ -287,6 +330,12 @@ int iio_dma_buffer_request_update(struct iio_buffer *buffer) mutex_lock(&queue->lock); + queue->fileio.enabled = iio_dma_buffer_can_use_fileio(queue); + + /* If DMABUFs were created, disable fileio interface */ + if (!queue->fileio.enabled) + goto out_unlock; + /* Allocations are page aligned */ if (PAGE_ALIGN(queue->fileio.block_size) == PAGE_ALIGN(size)) try_reuse = true; @@ -327,7 +376,7 @@ int iio_dma_buffer_request_update(struct iio_buffer *buffer) } if (!block) { - block = iio_dma_buffer_alloc_block(queue, size); + block = iio_dma_buffer_alloc_block(queue, size, true); if (!block) { ret = -ENOMEM; goto out_unlock; @@ -384,8 +433,12 @@ static void iio_dma_buffer_submit_block(struct iio_dma_buffer_queue *queue, block->state = IIO_BLOCK_STATE_ACTIVE; iio_buffer_block_get(block); + ret = queue->ops->submit(queue, block); if (ret) { + if (!block->fileio) + iio_buffer_signal_dmabuf_done(block->fence, ret); + /* * This is a bit of a problem and there is not much we can do * other then wait for the buffer to be disabled and re-enabled @@ -588,6 +641,112 @@ size_t iio_dma_buffer_data_available(struct iio_buffer *buf) } EXPORT_SYMBOL_GPL(iio_dma_buffer_data_available); +struct iio_dma_buffer_block * +iio_dma_buffer_attach_dmabuf(struct iio_buffer *buffer, + struct dma_buf_attachment *attach) +{ + struct iio_dma_buffer_queue *queue = iio_buffer_to_queue(buffer); + struct iio_dma_buffer_block *block; + + guard(mutex)(&queue->lock); + + /* + * If the buffer is enabled and in fileio mode new blocks can't be + * allocated. + */ + if (queue->fileio.enabled) + return ERR_PTR(-EBUSY); + + block = iio_dma_buffer_alloc_block(queue, attach->dmabuf->size, false); + if (!block) + return ERR_PTR(-ENOMEM); + + block->attach = attach; + + /* Free memory that might be in use for fileio mode */ + iio_dma_buffer_fileio_free(queue); + + return block; +} +EXPORT_SYMBOL_GPL(iio_dma_buffer_attach_dmabuf); + +void iio_dma_buffer_detach_dmabuf(struct iio_buffer *buffer, + struct iio_dma_buffer_block *block) +{ + block->state = IIO_BLOCK_STATE_DEAD; + iio_buffer_block_put_atomic(block); +} +EXPORT_SYMBOL_GPL(iio_dma_buffer_detach_dmabuf); + +static int iio_dma_can_enqueue_block(struct iio_dma_buffer_block *block) +{ + struct iio_dma_buffer_queue *queue = block->queue; + + /* If in fileio mode buffers can't be enqueued. */ + if (queue->fileio.enabled) + return -EBUSY; + + switch (block->state) { + case IIO_BLOCK_STATE_QUEUED: + return -EPERM; + case IIO_BLOCK_STATE_ACTIVE: + case IIO_BLOCK_STATE_DEAD: + return -EBUSY; + case IIO_BLOCK_STATE_DONE: + break; + } + + return 0; +} + +int iio_dma_buffer_enqueue_dmabuf(struct iio_buffer *buffer, + struct iio_dma_buffer_block *block, + struct dma_fence *fence, + struct sg_table *sgt, + size_t size, bool cyclic) +{ + struct iio_dma_buffer_queue *queue = iio_buffer_to_queue(buffer); + bool cookie; + int ret; + + WARN_ON(!mutex_is_locked(&queue->lock)); + + cookie = dma_fence_begin_signalling(); + + ret = iio_dma_can_enqueue_block(block); + if (ret < 0) + goto out_end_signalling; + + block->bytes_used = size; + block->cyclic = cyclic; + block->sg_table = sgt; + block->fence = fence; + + iio_dma_buffer_enqueue(queue, block); + +out_end_signalling: + dma_fence_end_signalling(cookie); + + return ret; +} +EXPORT_SYMBOL_GPL(iio_dma_buffer_enqueue_dmabuf); + +void iio_dma_buffer_lock_queue(struct iio_buffer *buffer) +{ + struct iio_dma_buffer_queue *queue = iio_buffer_to_queue(buffer); + + mutex_lock(&queue->lock); +} +EXPORT_SYMBOL_GPL(iio_dma_buffer_lock_queue); + +void iio_dma_buffer_unlock_queue(struct iio_buffer *buffer) +{ + struct iio_dma_buffer_queue *queue = iio_buffer_to_queue(buffer); + + mutex_unlock(&queue->lock); +} +EXPORT_SYMBOL_GPL(iio_dma_buffer_unlock_queue); + /** * iio_dma_buffer_set_bytes_per_datum() - DMA buffer set_bytes_per_datum callback * @buffer: Buffer to set the bytes-per-datum for diff --git a/include/linux/iio/buffer-dma.h b/include/linux/iio/buffer-dma.h index 18d3702fa95d..b62ff2a3f554 100644 --- a/include/linux/iio/buffer-dma.h +++ b/include/linux/iio/buffer-dma.h @@ -7,6 +7,7 @@ #ifndef __INDUSTRIALIO_DMA_BUFFER_H__ #define __INDUSTRIALIO_DMA_BUFFER_H__ +#include #include #include #include @@ -16,6 +17,9 @@ struct iio_dma_buffer_queue; struct iio_dma_buffer_ops; struct device; +struct dma_buf_attachment; +struct dma_fence; +struct sg_table; /** * enum iio_block_state - State of a struct iio_dma_buffer_block @@ -41,6 +45,8 @@ enum iio_block_state { * @queue: Parent DMA buffer queue * @kref: kref used to manage the lifetime of block * @state: Current state of the block + * @cyclic: True if this is a cyclic buffer + * @fileio: True if this buffer is used for fileio mode */ struct iio_dma_buffer_block { /* May only be accessed by the owner of the block */ @@ -63,6 +69,14 @@ struct iio_dma_buffer_block { * queue->list_lock if the block is not owned by the core. */ enum iio_block_state state; + + bool cyclic; + bool fileio; + + struct dma_buf_attachment *attach; + struct sg_table *sg_table; + + struct dma_fence *fence; }; /** @@ -72,6 +86,7 @@ struct iio_dma_buffer_block { * @pos: Read offset in the active block * @block_size: Size of each block * @next_dequeue: index of next block that will be dequeued + * @enabled: Whether the buffer is operating in fileio mode */ struct iio_dma_buffer_queue_fileio { struct iio_dma_buffer_block *blocks[2]; @@ -80,6 +95,7 @@ struct iio_dma_buffer_queue_fileio { size_t block_size; unsigned int next_dequeue; + bool enabled; }; /** @@ -95,6 +111,7 @@ struct iio_dma_buffer_queue_fileio { * the DMA controller * @incoming: List of buffers on the incoming queue * @active: Whether the buffer is currently active + * @num_dmabufs: Total number of DMABUFs attached to this queue * @fileio: FileIO state */ struct iio_dma_buffer_queue { @@ -107,6 +124,7 @@ struct iio_dma_buffer_queue { struct list_head incoming; bool active; + atomic_t num_dmabufs; struct iio_dma_buffer_queue_fileio fileio; }; @@ -142,4 +160,17 @@ int iio_dma_buffer_init(struct iio_dma_buffer_queue *queue, void iio_dma_buffer_exit(struct iio_dma_buffer_queue *queue); void iio_dma_buffer_release(struct iio_dma_buffer_queue *queue); +struct iio_dma_buffer_block * +iio_dma_buffer_attach_dmabuf(struct iio_buffer *buffer, + struct dma_buf_attachment *attach); +void iio_dma_buffer_detach_dmabuf(struct iio_buffer *buffer, + struct iio_dma_buffer_block *block); +int iio_dma_buffer_enqueue_dmabuf(struct iio_buffer *buffer, + struct iio_dma_buffer_block *block, + struct dma_fence *fence, + struct sg_table *sgt, + size_t size, bool cyclic); +void iio_dma_buffer_lock_queue(struct iio_buffer *buffer); +void iio_dma_buffer_unlock_queue(struct iio_buffer *buffer); + #endif