Add documentation for the fpu_kern(9) interfaces.

Many thanks to jmg for reviewing the (previous version) of the text
and providing grammar and content fixes.

Sponsored by:	The FreeBSD Foundation
MFC after:	1 week

This commit is contained in:

Konstantin Belousov

2014-06-23 07:45:44 +00:00

parent 633034fe0e

commit 56388ef94b

1 changed files with 193 additions and 0 deletions

									
										share/man/man9/fpu_kern.9
									
		+193
		
												View File
												
				@@ -0,0 +1,193 @@

				.\" Copyright (c) 2014

				.\"	Konstantin Belousov <kib@FreeBSD.org>.  All rights reserved.

				.\"

				.\" Redistribution and use in source and binary forms, with or without

				.\" modification, are permitted provided that the following conditions

				.\" are met:

				.\" 1. Redistributions of source code must retain the above copyright

				.\"    notice, this list of conditions and the following disclaimer.

				.\" 2. Redistributions in binary form must reproduce the above copyright

				.\"    notice, this list of conditions and the following disclaimer in the

				.\"    documentation and/or other materials provided with the distribution.

				.\"

				.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR

				.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES

				.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.

				.\" IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,

				.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT

				.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,

				.\" DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY

				.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT

				.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF

				.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

				.\"

				.\" $FreeBSD$

				.\"

				.Dd June 21, 2014

				.Dt KERN_FPU 9

				.Os

				.Sh NAME

				.Nm fpu_kern

				.Nd "facility to use the FPU in the kernel"

				.Sh SYNOPSIS

				.Ft struct fpu_kern_ctx *

				.Fn fpu_kern_alloc_ctx "u_int flags"

				.Ft void

				.Fn fpu_kern_free_ctx "struct fpu_kern_ctx *ctx"

				.Ft int

				.Fn fpu_kern_enter "struct thread *td" "struct fpu_kern_ctx *ctx" "u_int flags"

				.Ft int

				.Fn fpu_kern_leave "struct thread *td" "struct fpu_kern_ctx *ctx"

				.Ft int

				.Fn fpu_kern_thread "u_int flags"

				.Ft int

				.Fn is_fpu_kern_thread "u_int flags"

				.Sh DESCRIPTION

				The

				.Nm

				family of functions allows the use of FPU hardware in kernel code.

				Modern FPUs are not limited to providing hardware implementation for

				floating point arithmetic, they offer advanced accelerators for cryptography

				and other computational-intensive algorithms.

				These facilities share registers with the FPU hardware.

				.Pp

				Typical kernel code does not need to access to the FPU.

				Saving a large register file on each entry to the kernel would waste

				time.

				When kernel code uses the FPU, the current FPU state must be saved to

				avoid corrupting the user-mode state, and vice versa.

				.Pp

				The management of the save and restore is automatic.

				The processor catches accesses to the FPU registers

				when the non-current context tries to access them.

				Explicit calls are required for the allocation of the save area and

				the notification of the start and end of the code using the FPU.

				.Pp

				The

				.Fn fpu_kern_alloc_ctx

				function allocates the memory used by

				.Nm

				to track the use of the FPU hardware state and the related software state.

				The

				.Fn fpu_kern_alloc_ctx

				function requires the

				.Fa flags

				argument, which currently accepts the following flags:

				.Bl -tag -width ".Dv FPU_KERN_NOWAIT" -offset indent

				.It Dv FPU_KERN_NOWAIT

				Do not wait for the available memory if the request could not be satisfied

				without sleep.

				.It 0

				No special handling is required.

				.El

				The function returns the allocated context area, or

				.Va NULL

				if the allocation failed.

				.Pp

				The

				.Fn fpu_kern_free_ctx

				function frees the context previously allocated by

				.Fn fpu_kern_alloc_ctx .

				.Pp

				The

				.Fn fpu_kern_enter

				function designates the start of the region of kernel code where the

				use of the FPU is allowed.

				Its arguments are:

				.Bl -tag -width ".Fa ctx" -offset indent

				.It Fa td

				Currently must be

				.Va curthread .

				.It Fa ctx

				The context save area previously allocated by

				.Fn fpu_kern_alloc_ctx

				and not currently in use by another call to

				.Fn fpu_kern_enter .

				.It Fa flags

				This argument currently accepts the following flags:

				.Bl -tag -width ".Dv FPU_KERN_NORMAL" -offset indent

				.It Dv FPU_KERN_NORMAL

				Indicates that the caller intends to access the full FPU state.

				Must be specified currently.

				.It Dv FPU_KERN_KTHR

				Indicates that no saving of the current FPU state should be performed,

				if the thread called

				.Xr fpu_kern_thread 9

				function.

				This is intended to minimize code duplication in callers which

				could be used from both kernel thread and syscall contexts.

				The

				.Fn fpu_kern_leave

				function correctly handles such contexts.

				.El

				.El

				The function does not sleep or block.

				It could cause the

				.Nm Device Not Available

				exception during execution, and on the first FPU access after the

				function returns, as well as after each context switch

				(see Intel Software Developer Manual for the reference).

				Currently, no errors are defined which can be returned by

				.Fn fpu_kern_enter

				to the caller.

				.Pp

				The

				.Fn fpu_kern_leave

				function ends the region started by

				.Fn fpu_kern_enter .

				The uses of FPU in the kernel after the call to

				.Fn fpu_kern_leave

				are erronous until the next call to

				.Fn fpu_kern_enter

				is performed.

				The function takes the

				.Fa td

				thread argument, which currently must be

				.Va curthread ,

				and the

				.Fa ctx

				context pointer, previously passed to

				.Fn fpu_kern_enter .

				After the function returns, the context may be freed or reused

				by other invocation of

				.Fn fpu_kern_enter .

				There are no errors defined for the function, it always returns 0.

				.Pp

				The

				.Fn fpu_kern_thread

				function provides an optimization for threads which never leave to

				the usermode.

				Such thread can reuse the usermode save area for the FPU state,

				which is allowed by the function call.

				There is no flags defined for the function, and no error states

				that the function returns.

				.Pp

				The

				.Fn is_fpu_kern_thread

				function returns the boolean indicating whether the current thread

				entered the mode enabled by

				.Fn fpu_kern_thread .

				There is currently no flags defined for the function, the return

				value is true if the current thread have the permanent FPU save area,

				and false otherwise.

				.Sh NOTES

				The

				.Nm

				is currently implemented only for i386 and amd64 architectures.

				.Pp

				There is no way to handle floating point exceptions raised from

				kernel mode.

				.Pp

				The unused

				.Fa flags

				arguments

				to the

				.Nm

				functions are to be extended to allow specification of the

				set of the FPU hardware state used by the code region.

				This would allow optimizations of saving and restoring the state.

				.Sh AUTHORS

				The

				.Nm

				facitily and this manual page were written by

				.An Konstantin Belousov Aq Mt kib@FreeBSD.org .