mdoc(7) police: markup overhaul.
Approved by: re
This commit is contained in:
parent
3e0fbd7bc4
commit
e479f313a6
@ -40,38 +40,26 @@
|
||||
.Sh SYNOPSIS
|
||||
.In sys/jumbo.h
|
||||
.Ft int
|
||||
.Fo jumbo_vm_init
|
||||
.Fa "void"
|
||||
.Fc
|
||||
.Fn jumbo_vm_init void
|
||||
.Ft vm_page_t
|
||||
.Fo jumbo_pg_alloc
|
||||
.Fa "void"
|
||||
.Fc
|
||||
.Fn jumbo_pg_alloc void
|
||||
.Ft void
|
||||
.Fo jumbo_pg_free
|
||||
.Fa "vm_offset_t addr"
|
||||
.Fc
|
||||
.Fn jumbo_pg_free "vm_offset_t addr"
|
||||
.Ft void
|
||||
.Fo jumbo_freem
|
||||
.Fa "caddr_t addr"
|
||||
.Fa "void *args"
|
||||
.Fc
|
||||
.Fn jumbo_freem "caddr_t addr" "void *args"
|
||||
.Ft void
|
||||
.Fo jumbo_pg_steal
|
||||
.Fa "vm_page_t pg"
|
||||
.Fc
|
||||
.Fn jumbo_pg_steal "vm_page_t pg"
|
||||
.Ft caddr_t
|
||||
.Fo jumbo_phys_to_kva
|
||||
.Fa "vm_offset_t pa"
|
||||
.Fc
|
||||
.Fn jumbo_phys_to_kva "vm_offset_t pa"
|
||||
.Sh DESCRIPTION
|
||||
The
|
||||
.Nm
|
||||
buffer facility is designed for allocating disposable page-sized
|
||||
buffers. Buffers allocated via this facility can either be returned or
|
||||
not. This facility is primarily intended for use with network adapters
|
||||
that have MTUs of a page or greater size. The buffers will normally be
|
||||
disposed of by the
|
||||
buffers.
|
||||
Buffers allocated via this facility can either be returned or not.
|
||||
This facility is primarily intended for use with network adapters
|
||||
that have MTUs of a page or greater size.
|
||||
The buffers will normally be disposed of by the
|
||||
.Xr zero_copy 9
|
||||
receive code.
|
||||
.Pp
|
||||
@ -79,39 +67,47 @@ receive code.
|
||||
initializes the pool of KVA the
|
||||
.Nm
|
||||
code needs to operate and does some
|
||||
other initialization to prepare the subsystem for operation. This routine
|
||||
only needs to be called once. Calling it multiple times will have no
|
||||
effect. It is recommended that this initialization routine be called in a
|
||||
other initialization to prepare the subsystem for operation.
|
||||
This routine only needs to be called once.
|
||||
Calling it multiple times will have no effect.
|
||||
It is recommended that this initialization routine be called in a
|
||||
device driver attach routine, so that resources are not allocated if the
|
||||
.Nm
|
||||
subsystem won't end up being used.
|
||||
subsystem will not end up being used.
|
||||
.Fn jumbo_vm_init
|
||||
returns 1 upon successful completion, and 0 upon failure.
|
||||
.Pp
|
||||
.Fn jumbo_pg_alloc
|
||||
allocates a physical page and assigns a piece of KVA from the
|
||||
.Nm
|
||||
KVA pool. It returns the allocated page if successful, and NULL in the case of
|
||||
failure.
|
||||
KVA pool.
|
||||
It returns the allocated page if successful, and
|
||||
.Dv NULL
|
||||
in the case of failure.
|
||||
.Pp
|
||||
.Fn jumbo_pg_free
|
||||
frees a page allocated by
|
||||
.Fn jumbo_pg_alloc .
|
||||
It takes the address of the memory in question as an argument. This
|
||||
routine will normally be used in cases where the allocated
|
||||
.Nm jumbo
|
||||
page cannot be used for some reason. The normal free path is via
|
||||
It takes the address of the memory in question as an argument.
|
||||
This routine will normally be used in cases where the allocated
|
||||
.Nm
|
||||
page cannot be used for some reason.
|
||||
The normal free path is via
|
||||
.Fn jumbo_freem .
|
||||
.Pp
|
||||
.Fn jumbo_freem
|
||||
is the routine that should be given as the external free routine when an
|
||||
external mbuf is allocated using pages from the
|
||||
external
|
||||
.Vt mbuf
|
||||
is allocated using pages from the
|
||||
.Nm
|
||||
allocator. It takes the virtual address of the page in question, and
|
||||
allocator.
|
||||
It takes the virtual address of the page in question, and
|
||||
ignores the second argument.
|
||||
.Pp
|
||||
.Fn jumbo_pg_steal
|
||||
"steals" a page and recycles its KVA space.
|
||||
.Dq steals
|
||||
a page and recycles its KVA space.
|
||||
.Pp
|
||||
.Fn jumbo_phys_to_kva
|
||||
translates the physical address of a
|
||||
@ -128,9 +124,12 @@ by
|
||||
.An Andrew Gallatin Aq gallatin@FreeBSD.org
|
||||
as part of a set of zero copy patches for the
|
||||
.Xr ti 4
|
||||
driver. The allocator was taken out of the
|
||||
driver.
|
||||
The allocator was taken out of the
|
||||
.Xr ti 4
|
||||
driver, cleaned up and ported to the new mutex interface by
|
||||
driver, cleaned up and ported to the new
|
||||
.Xr mutex 9
|
||||
interface by
|
||||
.An Kenneth Merry Aq ken@FreeBSD.org .
|
||||
.Pp
|
||||
The
|
||||
@ -140,8 +139,7 @@ allocator first appeared in
|
||||
and has existed in patch form since at least 1999.
|
||||
.Sh AUTHORS
|
||||
.An Andrew Gallatin Aq gallatin@FreeBSD.org
|
||||
and
|
||||
.An Kenneth Merry Aq ken@FreeBSD.org .
|
||||
.An Kenneth Merry Aq ken@FreeBSD.org
|
||||
.Sh BUGS
|
||||
There is currently a static number of KVA pages allocated by the
|
||||
.Nm
|
||||
@ -150,4 +148,4 @@ allocated should demand exceed supply.
|
||||
.Pp
|
||||
The
|
||||
.Fn jumbo_pg_steal
|
||||
function isn't currently used anywhere.
|
||||
function is not currently used anywhere.
|
||||
|
Loading…
Reference in New Issue
Block a user