'\" t
.\"     Title: seqno_fence_init
.\"    Author: [FIXME: author] [see http://docbook.sf.net/el/author]
.\" Generator: DocBook XSL Stylesheets v1.79.1 <http://docbook.sf.net/>
.\"      Date: January 2017
.\"    Manual: Device drivers infrastructure
.\"    Source: Kernel Hackers Manual 4.8.15
.\"  Language: English
.\"
.TH "SEQNO_FENCE_INIT" "9" "January 2017" "Kernel Hackers Manual 4\&.8\&." "Device drivers infrastructure"
.\" -----------------------------------------------------------------
.\" * Define some portability stuff
.\" -----------------------------------------------------------------
.\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.\" http://bugs.debian.org/507673
.\" http://lists.gnu.org/archive/html/groff/2009-02/msg00013.html
.\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.ie \n(.g .ds Aq \(aq
.el       .ds Aq '
.\" -----------------------------------------------------------------
.\" * set default formatting
.\" -----------------------------------------------------------------
.\" disable hyphenation
.nh
.\" disable justification (adjust text to left margin only)
.ad l
.\" -----------------------------------------------------------------
.\" * MAIN CONTENT STARTS HERE *
.\" -----------------------------------------------------------------
.SH "NAME"
seqno_fence_init \- initialize a seqno fence
.SH "SYNOPSIS"
.HP \w'void\ seqno_fence_init('u
.BI "void seqno_fence_init(struct\ seqno_fence\ *\ " "fence" ", spinlock_t\ *\ " "lock" ", struct\ dma_buf\ *\ " "sync_buf" ", uint32_t\ " "context" ", uint32_t\ " "seqno_ofs" ", uint32_t\ " "seqno" ", enum\ seqno_fence_condition\ " "cond" ", const\ struct\ fence_ops\ *\ " "ops" ");"
.SH "ARGUMENTS"
.PP
\fIfence\fR
.RS 4
seqno_fence to initialize
.RE
.PP
\fIlock\fR
.RS 4
pointer to spinlock to use for fence
.RE
.PP
\fIsync_buf\fR
.RS 4
buffer containing the memory location to signal on
.RE
.PP
\fIcontext\fR
.RS 4
the execution context this fence is a part of
.RE
.PP
\fIseqno_ofs\fR
.RS 4
the offset within
\fIsync_buf\fR
.RE
.PP
\fIseqno\fR
.RS 4
the sequence # to signal on
.RE
.PP
\fIcond\fR
.RS 4
fence wait condition
.RE
.PP
\fIops\fR
.RS 4
the fence_ops for operations on this seqno fence
.RE
.SH "DESCRIPTION"
.PP
This function initializes a struct seqno_fence with passed parameters, and takes a reference on sync_buf which is released on fence destruction\&.
.PP
A seqno_fence is a dma_fence which can complete in software when enable_signaling is called, but it also completes when (s32)((sync_buf)[seqno_ofs] \- seqno) >= 0 is true
.PP
The seqno_fence will take a refcount on the sync_buf until it\*(Aqs destroyed, but actual lifetime of sync_buf may be longer if one of the callers take a reference to it\&.
.PP
Certain hardware have instructions to insert this type of wait condition in the command stream, so no intervention from software would be needed\&. This type of fence can be destroyed before completed, however a reference on the sync_buf dma\-buf can be taken\&. It is encouraged to re\-use the same dma\-buf for sync_buf, since mapping or unmapping the sync_buf to the device\*(Aqs vm can be expensive\&.
.PP
It is recommended for creators of seqno_fence to call fence_signal before destruction\&. This will prevent possible issues from wraparound at time of issue vs time of check, since users can check fence_is_signaled before submitting instructions for the hardware to wait on the fence\&. However, when ops\&.enable_signaling is not called, it doesn\*(Aqt have to be done as soon as possible, just before there\*(Aqs any real danger of seqno wraparound\&.
.SH "COPYRIGHT"
.br