Date: Mon, 30 Jun 2014 09:54:08 +0000 (UTC) From: Konstantin Belousov <kib@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-stable@freebsd.org, svn-src-stable-10@freebsd.org Subject: svn commit: r268035 - stable/10/share/man/man9 Message-ID: <201406300954.s5U9s8MA095993@svn.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: kib Date: Mon Jun 30 09:54:08 2014 New Revision: 268035 URL: http://svnweb.freebsd.org/changeset/base/268035 Log: MFC r267768: Add documentation for the fpu_kern(9) interfaces. MFC r267785 (by brueffer): Several small fixes. Added: stable/10/share/man/man9/fpu_kern.9 - copied, changed from r267768, head/share/man/man9/fpu_kern.9 Modified: Directory Properties: stable/10/ (props changed) Copied and modified: stable/10/share/man/man9/fpu_kern.9 (from r267768, head/share/man/man9/fpu_kern.9) ============================================================================== --- head/share/man/man9/fpu_kern.9 Mon Jun 23 07:45:44 2014 (r267768, copy source) +++ stable/10/share/man/man9/fpu_kern.9 Mon Jun 30 09:54:08 2014 (r268035) @@ -23,8 +23,8 @@ .\" .\" $FreeBSD$ .\" -.Dd June 21, 2014 -.Dt KERN_FPU 9 +.Dd June 23, 2014 +.Dt FPU_KERN 9 .Os .Sh NAME .Nm fpu_kern @@ -47,11 +47,11 @@ 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 +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. +Typical kernel code does not need 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 @@ -80,6 +80,7 @@ without sleep. .It 0 No special handling is required. .El +.Pp The function returns the allocated context area, or .Va NULL if the allocation failed. @@ -121,6 +122,7 @@ The function correctly handles such contexts. .El .El +.Pp The function does not sleep or block. It could cause the .Nm Device Not Available @@ -173,7 +175,7 @@ and false otherwise. .Sh NOTES The .Nm -is currently implemented only for i386 and amd64 architectures. +is currently implemented only for the i386 and amd64 architectures. .Pp There is no way to handle floating point exceptions raised from kernel mode.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201406300954.s5U9s8MA095993>