'\" te .\" Copyright (c) 2002 Sun Microsystems, Inc. All Rights Reserved. .\" The contents of this file are subject to the terms of the Common Development and Distribution License (the "License"). You may not use this file except in compliance with the License. .\" You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE or http://www.opensolaris.org/os/licensing. See the License for the specific language governing permissions and limitations under the License. .\" When distributing Covered Code, include this CDDL HEADER in each file and include the License file at usr/src/OPENSOLARIS.LICENSE. If applicable, add the following below this CDDL HEADER, with the fields enclosed by brackets "[]" replaced with your own identifying information: Portions Copyright [yyyy] [name of copyright owner] .TH DDI_DMA_BUF_BIND_HANDLE 9F "Jul 27, 1996" .SH NAME ddi_dma_buf_bind_handle \- binds a system buffer to a DMA handle .SH SYNOPSIS .LP .nf #include #include \fBint\fR \fBddi_dma_buf_bind_handle\fR(\fBddi_dma_handle_t\fR \fIhandle\fR, \fBstruct buf *\fR\fIbp\fR, \fBuint_t\fR \fIflags\fR, \fBint (*\fR\fIcallback\fR)(caddr_t), \fBcaddr_t\fR \fIarg\fR, \fBddi_dma_cookie_t *\fR\fIcookiep\fR, \fBuint_t *\fR\fIccountp\fR); .fi .SH INTERFACE LEVEL .sp .LP Solaris DDI specific (Solaris DDI). .SH PARAMETERS .sp .ne 2 .na \fB\fIhandle\fR\fR .ad .RS 12n The \fBDMA\fR handle previously allocated by a call to \fBddi_dma_alloc_handle\fR(9F). .RE .sp .ne 2 .na \fB\fIbp\fR\fR .ad .RS 12n A pointer to a system buffer structure (see \fBbuf\fR(9S)). .RE .sp .ne 2 .na \fB\fIflags\fR\fR .ad .RS 12n Valid flags include: .sp .ne 2 .na \fB\fBDDI_DMA_WRITE\fR\fR .ad .RS 22n Transfer direction is from memory to I/O .RE .sp .ne 2 .na \fB\fBDDI_DMA_READ\fR\fR .ad .RS 22n Transfer direction is from I/O to memory .RE .sp .ne 2 .na \fB\fBDDI_DMA_RDWR\fR\fR .ad .RS 22n Both read and write .RE .sp .ne 2 .na \fB\fBDDI_DMA_REDZONE\fR\fR .ad .RS 22n Establish an \fBMMU\fR redzone at end of the object. .RE .sp .ne 2 .na \fB\fBDDI_DMA_PARTIAL\fR\fR .ad .RS 22n Partial resource allocation .RE .sp .ne 2 .na \fB\fBDDI_DMA_CONSISTENT\fR\fR .ad .RS 22n Nonsequential, random, and small block transfers. .RE .sp .ne 2 .na \fB\fBDDI_DMA_STREAMING\fR\fR .ad .RS 22n Sequential, unidirectional, block-sized, and block-aligned transfers. .RE .RE .sp .ne 2 .na \fB\fIcallback\fR\fR .ad .RS 12n The address of a function to call back later if resources are not available now. The following special function addresses may also be used. .sp .ne 2 .na \fB\fBDDI_DMA_SLEEP\fR\fR .ad .RS 20n Wait until resources are available. .RE .sp .ne 2 .na \fB\fBDDI_DMA_DONTWAIT\fR\fR .ad .RS 20n Do not wait until resources are available and do not schedule a callback. .RE .RE .sp .ne 2 .na \fB\fIarg\fR\fR .ad .RS 12n Argument to be passed to the callback function, \fIcallback\fR, if such a function is specified. .RE .sp .ne 2 .na \fB\fIcookiep\fR\fR .ad .RS 12n A pointer to the first \fBddi_dma_cookie\fR(9S) structure. .RE .sp .ne 2 .na \fB\fIccountp\fR\fR .ad .RS 12n Upon a successful return, \fIccountp\fR points to a value representing the number of cookies for this \fBDMA\fR object. .RE .SH DESCRIPTION .sp .LP \fBddi_dma_buf_bind_handle()\fR allocates \fBDMA\fR resources for a system buffer such that a device can perform \fBDMA\fR to or from the buffer. \fBDMA\fR resources are allocated considering the device's \fBDMA\fR attributes as expressed by \fBddi_dma_attr\fR(9S) (see \fBddi_dma_alloc_handle\fR(9F)). .sp .LP \fBddi_dma_buf_bind_handle()\fR fills in the first \fBDMA\fR cookie pointed to by \fIcookiep\fR with the appropriate address, length, and bus type. \fB*\fR\fIccountp\fR is set to the number of \fBDMA\fR cookies representing this \fBDMA\fR object. Subsequent \fBDMA\fR cookies must be retrieved by calling \fBddi_dma_nextcookie\fR(9F) \fB*\fR\fIcountp\fR-1 times. .sp .LP When a \fBDMA\fR transfer completes, the driver should free up system \fBDMA\fR resources by calling \fBddi_dma_unbind_handle\fR(9F). .sp .LP The \fIflags\fR argument contains information for mapping routines. .sp .ne 2 .na \fB\fBDDI_DMA_WRITE\fR, \fBDDI_DMA_READ\fR, \fBDDI_DMA_RDWR\fR\fR .ad .sp .6 .RS 4n These flags describe the intended direction of the \fBDMA\fR transfer. .RE .sp .ne 2 .na \fB\fBDDI_DMA_STREAMING\fR \fR .ad .sp .6 .RS 4n This flag should be set if the device is doing sequential, unidirectional, block-sized, and block-aligned transfers to or from memory. The alignment and padding constraints specified by the \fBminxfer\fR and \fBburstsizes\fR fields in the \fBDMA\fR attribute structure, \fBddi_dma_attr\fR(9S) (see \fBddi_dma_alloc_handle\fR(9F)) is used to allocate the most effective hardware support for large transfers. .RE .sp .ne 2 .na \fB\fBDDI_DMA_CONSISTENT\fR \fR .ad .sp .6 .RS 4n This flag should be set if the device accesses memory randomly, or if synchronization steps using \fBddi_dma_sync\fR(9F) need to be as efficient as possible. I/O parameter blocks used for communication between a device and a driver should be allocated using \fBDDI_DMA_CONSISTENT\fR. .RE .sp .ne 2 .na \fB\fBDDI_DMA_REDZONE\fR \fR .ad .sp .6 .RS 4n If this flag is set, the system attempts to establish a protected red zone after the object. The \fBDMA\fR resource allocation functions do not guarantee the success of this request as some implementations may not have the hardware ability to support a red zone. .RE .sp .ne 2 .na \fB\fBDDI_DMA_PARTIAL\fR \fR .ad .sp .6 .RS 4n Setting this flag indicates the caller can accept resources for part of the object. That is, if the size of the object exceeds the resources available, only resources for a portion of the object are allocated. The system indicates this condition returning status \fBDDI_DMA_PARTIAL_MAP\fR. At a later point, the caller can use \fBddi_dma_getwin\fR(9F) to change the valid portion of the object for which resources are allocated. If resources were allocated for only part of the object, \fBddi_dma_addr_bind_handle()\fR returns resources for the first \fBDMA\fR window. Even when \fBDDI_DMA_PARTIAL\fR is set, the system may decide to allocate resources for the entire object (less overhead) in which case \fBDDI_DMA_MAPPED\fR is returned. .RE .sp .LP The callback function, \fIcallback\fR, indicates how a caller wants to handle the possibility of resources not being available. If \fIcallback\fR is set to \fBDDI_DMA_DONTWAIT\fR, the caller does not care if the allocation fails, and can handle an allocation failure appropriately. If \fIcallback\fR is set to \fBDDI_DMA_SLEEP\fR, the caller wishes to have the allocation routines wait for resources to become available. If any other value is set, and a \fBDMA\fR resource allocation fails, this value is assumed to be the address of a function to call at a later time when resources may become available. When the specified function is called, it is passed \fIarg\fR as an argument. The specified callback function must return either \fBDDI_DMA_CALLBACK_RUNOUT\fR or \fBDDI_DMA_CALLBACK_DONE\fR. \fBDDI_DMA_CALLBACK_RUNOUT\fR indicates that the callback function attempted to allocate \fBDMA\fR resources but failed to do so. In this case the callback function is put back on a list to be called again later. \fBDDI_DMA_CALLBACK_DONE\fR indicates either a successful allocation of \fBDMA\fR resources or that the driver no longer wishes to retry. .sp .LP The callback function is called in interrupt context. Therefore, only system functions accessible from interrupt context are be available. The callback function must take whatever steps necessary to protect its critical resources, data structures, queues, etc. .SH RETURN VALUES .sp .LP \fBddi_dma_buf_bind_handle()\fR returns: .sp .ne 2 .na \fB\fBDDI_DMA_MAPPED\fR\fR .ad .RS 23n Successfully allocated resources for the entire object. .RE .sp .ne 2 .na \fB\fBDDI_DMA_PARTIAL_MAP\fR\fR .ad .RS 23n Successfully allocated resources for a part of the object. This is acceptable when partial transfers are permitted by setting the \fBDDI_DMA_PARTIAL\fR flag in \fIflags\fR. .RE .sp .ne 2 .na \fB\fBDDI_DMA_INUSE\fR\fR .ad .RS 23n Another I/O transaction is using the \fBDMA\fR handle. .RE .sp .ne 2 .na \fB\fBDDI_DMA_NORESOURCES\fR\fR .ad .RS 23n No resources are available at the present time. .RE .sp .ne 2 .na \fB\fBDDI_DMA_NOMAPPING\fR\fR .ad .RS 23n The object cannot be reached by the device requesting the resources. .RE .sp .ne 2 .na \fB\fBDDI_DMA_TOOBIG\fR\fR .ad .RS 23n The object is too big. A request of this size can never be satisfied on this particular system. The maximum size varies depending on machine and configuration. .RE .SH CONTEXT .sp .LP \fBddi_dma_buf_bind_handle()\fR can be called from user, kernel, or interrupt context, except when \fIcallback\fR is set to \fBDDI_DMA_SLEEP\fR, in which case it can be called from user or kernel context only. .SH SEE ALSO .sp .LP \fBddi_dma_addr_bind_handle\fR(9F), \fBddi_dma_alloc_handle\fR(9F), \fBddi_dma_free_handle\fR(9F), \fBddi_dma_getwin\fR(9F), \fBddi_dma_nextcookie\fR(9F), \fBddi_dma_sync\fR(9F), \fBddi_dma_unbind_handle\fR(9F), \fBbuf\fR(9S), \fBddi_dma_attr\fR(9S), \fBddi_dma_cookie\fR(9S) .sp .LP \fIWriting Device Drivers\fR .SH NOTES .sp .LP If the driver permits partial mapping with the \fBDDI_DMA_PARTIAL\fR flag, the number of cookies in each window may exceed the size of the device's scatter/gather list as specified in the \fBdma_attr_sgllen\fR field in the \fBddi_dma_attr\fR(9S) structure. In this case, each set of cookies comprising a \fBDMA\fR window will satisfy the \fBDMA\fR attributes as described in the \fBddi_dma_attr\fR(9S) structure in all aspects. The driver should set up its \fBDMA\fR engine and perform one transfer for each set of cookies sufficient for its scatter/gather list, up to the number of cookies for this window, before advancing to the next window using \fBddi_dma_getwin\fR(9F).