2005-01-06 01:43:34 +00:00
|
|
|
/*-
|
2017-11-18 14:26:50 +00:00
|
|
|
* SPDX-License-Identifier: BSD-4-Clause
|
|
|
|
*
|
2004-10-13 05:38:42 +00:00
|
|
|
* Copyright (C) 2003,2004
|
|
|
|
* Hidetoshi Shimokawa. All rights reserved.
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
*
|
2004-10-13 05:38:42 +00:00
|
|
|
* 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.
|
|
|
|
* 3. All advertising materials mentioning features or use of this software
|
|
|
|
* must display the following acknowledgement:
|
|
|
|
*
|
|
|
|
* This product includes software developed by Hidetoshi Shimokawa.
|
|
|
|
*
|
|
|
|
* 4. Neither the name of the author nor the names of its contributors
|
|
|
|
* may be used to endorse or promote products derived from this software
|
|
|
|
* without specific prior written permission.
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
*
|
2004-10-13 05:38:42 +00:00
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``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 REGENTS OR CONTRIBUTORS 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.
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
*
|
2004-10-13 05:38:42 +00:00
|
|
|
* $FreeBSD$
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <sys/param.h>
|
|
|
|
#include <sys/kdb.h>
|
|
|
|
#include <gdb/gdb.h>
|
|
|
|
#include <sys/kernel.h>
|
|
|
|
#include <sys/module.h>
|
|
|
|
#include <sys/systm.h>
|
|
|
|
#include <sys/types.h>
|
|
|
|
#include <sys/conf.h>
|
|
|
|
#include <sys/cons.h>
|
|
|
|
#include <sys/consio.h>
|
|
|
|
#include <sys/tty.h>
|
|
|
|
#include <sys/malloc.h>
|
2006-11-06 13:42:10 +00:00
|
|
|
#include <sys/priv.h>
|
2004-10-13 05:38:42 +00:00
|
|
|
#include <sys/proc.h>
|
|
|
|
#include <sys/ucred.h>
|
|
|
|
|
|
|
|
#include <machine/bus.h>
|
|
|
|
|
|
|
|
#include <dev/dcons/dcons.h>
|
|
|
|
#include <dev/dcons/dcons_os.h>
|
|
|
|
|
|
|
|
#include <ddb/ddb.h>
|
|
|
|
#include <sys/reboot.h>
|
|
|
|
|
|
|
|
#include <sys/sysctl.h>
|
|
|
|
|
|
|
|
#include <vm/vm.h>
|
|
|
|
#include <vm/vm_param.h>
|
|
|
|
#include <vm/pmap.h>
|
|
|
|
|
|
|
|
#include "opt_dcons.h"
|
2007-05-27 13:58:43 +00:00
|
|
|
#include "opt_kdb.h"
|
|
|
|
#include "opt_gdb.h"
|
|
|
|
#include "opt_ddb.h"
|
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
|
|
|
|
#ifndef DCONS_POLL_HZ
|
2009-04-06 13:09:02 +00:00
|
|
|
#define DCONS_POLL_HZ 25
|
2004-10-13 05:38:42 +00:00
|
|
|
#endif
|
|
|
|
|
2013-02-19 18:22:25 +00:00
|
|
|
#ifndef DCONS_POLL_IDLE
|
|
|
|
#define DCONS_POLL_IDLE 256
|
|
|
|
#endif
|
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
#ifndef DCONS_BUF_SIZE
|
|
|
|
#define DCONS_BUF_SIZE (16*1024)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifndef DCONS_FORCE_CONSOLE
|
|
|
|
#define DCONS_FORCE_CONSOLE 0 /* Mostly for FreeBSD-4/DragonFly */
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifndef KLD_MODULE
|
|
|
|
static char bssbuf[DCONS_BUF_SIZE]; /* buf in bss */
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/* global data */
|
|
|
|
static struct dcons_global dg;
|
|
|
|
struct dcons_global *dcons_conf;
|
|
|
|
static int poll_hz = DCONS_POLL_HZ;
|
2013-02-19 18:22:25 +00:00
|
|
|
static u_int poll_idle = DCONS_POLL_HZ * DCONS_POLL_IDLE;
|
2004-10-13 05:38:42 +00:00
|
|
|
|
|
|
|
static struct dcons_softc sc[DCONS_NPORT];
|
|
|
|
|
2011-11-07 15:43:11 +00:00
|
|
|
static SYSCTL_NODE(_kern, OID_AUTO, dcons, CTLFLAG_RD, 0, "Dumb Console");
|
2004-10-13 05:38:42 +00:00
|
|
|
SYSCTL_INT(_kern_dcons, OID_AUTO, poll_hz, CTLFLAG_RW, &poll_hz, 0,
|
|
|
|
"dcons polling rate");
|
|
|
|
|
|
|
|
static int drv_init = 0;
|
|
|
|
static struct callout dcons_callout;
|
|
|
|
struct dcons_buf *dcons_buf; /* for local dconschat */
|
|
|
|
|
|
|
|
static void dcons_timeout(void *);
|
|
|
|
static int dcons_drv_init(int);
|
|
|
|
|
|
|
|
static cn_probe_t dcons_cnprobe;
|
|
|
|
static cn_init_t dcons_cninit;
|
2006-05-26 13:51:38 +00:00
|
|
|
static cn_term_t dcons_cnterm;
|
2004-10-13 05:38:42 +00:00
|
|
|
static cn_getc_t dcons_cngetc;
|
|
|
|
static cn_putc_t dcons_cnputc;
|
2011-12-17 15:08:43 +00:00
|
|
|
static cn_grab_t dcons_cngrab;
|
|
|
|
static cn_ungrab_t dcons_cnungrab;
|
2004-10-13 05:38:42 +00:00
|
|
|
|
2006-05-26 13:51:38 +00:00
|
|
|
CONSOLE_DRIVER(dcons);
|
2004-10-13 05:38:42 +00:00
|
|
|
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
#if defined(GDB)
|
|
|
|
static gdb_probe_f dcons_dbg_probe;
|
|
|
|
static gdb_init_f dcons_dbg_init;
|
|
|
|
static gdb_term_f dcons_dbg_term;
|
|
|
|
static gdb_getc_f dcons_dbg_getc;
|
|
|
|
static gdb_putc_f dcons_dbg_putc;
|
2004-10-13 05:38:42 +00:00
|
|
|
|
|
|
|
GDB_DBGPORT(dcons, dcons_dbg_probe, dcons_dbg_init, dcons_dbg_term,
|
2006-05-26 11:54:32 +00:00
|
|
|
dcons_dbg_getc, dcons_dbg_putc);
|
2004-10-13 05:38:42 +00:00
|
|
|
|
|
|
|
extern struct gdb_dbgport *gdb_cur;
|
|
|
|
#endif
|
|
|
|
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
static tsw_outwakeup_t dcons_outwakeup;
|
|
|
|
|
|
|
|
static struct ttydevsw dcons_ttydevsw = {
|
|
|
|
.tsw_flags = TF_NOPREFIX,
|
|
|
|
.tsw_outwakeup = dcons_outwakeup,
|
|
|
|
};
|
|
|
|
|
Attempt to make break-to-debugger and alternative break-to-debugger more
accessible:
(1) Always compile in support for breaking into the debugger if options
KDB is present in the kernel.
(2) Disable both by default, but allow them to be enabled via tunables
and sysctls debug.kdb.break_to_debugger and
debug.kdb.alt_break_to_debugger.
(3) options BREAK_TO_DEBUGGER and options ALT_BREAK_TO_DEBUGGER continue
to behave as before -- only now instead of compiling in
break-to-debugger support, they change the default values of the
above sysctls to enable those features by default. Current kernel
configurations should, therefore, continue to behave as expected.
(4) Migrate alternative break-to-debugger state machine logic out of
individual device drivers into centralised KDB code. This has a
number of upsides, but also one downside: it's now tricky to release
sio spin locks when entering the debugger, so we don't. However,
similar logic does not exist in other device drivers, including uart.
(5) dcons requires some special handling; unlike other console types, it
allows overriding KDB's own debugger selection, so we need a new
interface to KDB to allow that to work.
GENERIC kernels in -CURRENT will now support break-to-debugger as long as
appropriate boot/run-time options are set, which should improve the
debuggability of BETA kernels significantly.
MFC after: 3 weeks
Reviewed by: kib, nwhitehorn
Approved by: re (bz)
2011-08-26 21:46:36 +00:00
|
|
|
#if (defined(GDB) || defined(DDB))
|
2004-10-13 05:38:42 +00:00
|
|
|
static int
|
|
|
|
dcons_check_break(struct dcons_softc *dc, int c)
|
|
|
|
{
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
if (c < 0)
|
|
|
|
return (c);
|
|
|
|
|
2007-05-28 02:20:40 +00:00
|
|
|
#ifdef GDB
|
Attempt to make break-to-debugger and alternative break-to-debugger more
accessible:
(1) Always compile in support for breaking into the debugger if options
KDB is present in the kernel.
(2) Disable both by default, but allow them to be enabled via tunables
and sysctls debug.kdb.break_to_debugger and
debug.kdb.alt_break_to_debugger.
(3) options BREAK_TO_DEBUGGER and options ALT_BREAK_TO_DEBUGGER continue
to behave as before -- only now instead of compiling in
break-to-debugger support, they change the default values of the
above sysctls to enable those features by default. Current kernel
configurations should, therefore, continue to behave as expected.
(4) Migrate alternative break-to-debugger state machine logic out of
individual device drivers into centralised KDB code. This has a
number of upsides, but also one downside: it's now tricky to release
sio spin locks when entering the debugger, so we don't. However,
similar logic does not exist in other device drivers, including uart.
(5) dcons requires some special handling; unlike other console types, it
allows overriding KDB's own debugger selection, so we need a new
interface to KDB to allow that to work.
GENERIC kernels in -CURRENT will now support break-to-debugger as long as
appropriate boot/run-time options are set, which should improve the
debuggability of BETA kernels significantly.
MFC after: 3 weeks
Reviewed by: kib, nwhitehorn
Approved by: re (bz)
2011-08-26 21:46:36 +00:00
|
|
|
if ((dc->flags & DC_GDB) != 0 && gdb_cur == &dcons_gdb_dbgport)
|
|
|
|
kdb_alt_break_gdb(c, &dc->brk_state);
|
|
|
|
else
|
2007-05-27 13:58:43 +00:00
|
|
|
#endif
|
Attempt to make break-to-debugger and alternative break-to-debugger more
accessible:
(1) Always compile in support for breaking into the debugger if options
KDB is present in the kernel.
(2) Disable both by default, but allow them to be enabled via tunables
and sysctls debug.kdb.break_to_debugger and
debug.kdb.alt_break_to_debugger.
(3) options BREAK_TO_DEBUGGER and options ALT_BREAK_TO_DEBUGGER continue
to behave as before -- only now instead of compiling in
break-to-debugger support, they change the default values of the
above sysctls to enable those features by default. Current kernel
configurations should, therefore, continue to behave as expected.
(4) Migrate alternative break-to-debugger state machine logic out of
individual device drivers into centralised KDB code. This has a
number of upsides, but also one downside: it's now tricky to release
sio spin locks when entering the debugger, so we don't. However,
similar logic does not exist in other device drivers, including uart.
(5) dcons requires some special handling; unlike other console types, it
allows overriding KDB's own debugger selection, so we need a new
interface to KDB to allow that to work.
GENERIC kernels in -CURRENT will now support break-to-debugger as long as
appropriate boot/run-time options are set, which should improve the
debuggability of BETA kernels significantly.
MFC after: 3 weeks
Reviewed by: kib, nwhitehorn
Approved by: re (bz)
2011-08-26 21:46:36 +00:00
|
|
|
kdb_alt_break(c, &dc->brk_state);
|
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
return (c);
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
#define dcons_check_break(dc, c) (c)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
static int
|
2007-08-17 05:32:39 +00:00
|
|
|
dcons_os_checkc_nopoll(struct dcons_softc *dc)
|
2004-10-13 05:38:42 +00:00
|
|
|
{
|
|
|
|
int c;
|
|
|
|
|
|
|
|
if (dg.dma_tag != NULL)
|
|
|
|
bus_dmamap_sync(dg.dma_tag, dg.dma_map, BUS_DMASYNC_POSTREAD);
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
c = dcons_check_break(dc, dcons_checkc(dc));
|
|
|
|
|
|
|
|
if (dg.dma_tag != NULL)
|
|
|
|
bus_dmamap_sync(dg.dma_tag, dg.dma_map, BUS_DMASYNC_PREREAD);
|
|
|
|
|
|
|
|
return (c);
|
|
|
|
}
|
|
|
|
|
2007-08-17 05:32:39 +00:00
|
|
|
static int
|
|
|
|
dcons_os_checkc(struct dcons_softc *dc)
|
|
|
|
{
|
|
|
|
EVENTHANDLER_INVOKE(dcons_poll, 0);
|
|
|
|
return (dcons_os_checkc_nopoll(dc));
|
|
|
|
}
|
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
static void
|
|
|
|
dcons_os_putc(struct dcons_softc *dc, int c)
|
|
|
|
{
|
|
|
|
if (dg.dma_tag != NULL)
|
|
|
|
bus_dmamap_sync(dg.dma_tag, dg.dma_map, BUS_DMASYNC_POSTWRITE);
|
|
|
|
|
|
|
|
dcons_putc(dc, c);
|
|
|
|
|
|
|
|
if (dg.dma_tag != NULL)
|
|
|
|
bus_dmamap_sync(dg.dma_tag, dg.dma_map, BUS_DMASYNC_PREWRITE);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
dcons_outwakeup(struct tty *tp)
|
2004-10-13 05:38:42 +00:00
|
|
|
{
|
|
|
|
struct dcons_softc *dc;
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
char ch;
|
2004-10-13 05:38:42 +00:00
|
|
|
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
dc = tty_softc(tp);
|
2004-10-13 05:38:42 +00:00
|
|
|
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
while (ttydisc_getc(tp, &ch, sizeof ch) != 0)
|
|
|
|
dcons_os_putc(dc, ch);
|
2004-10-13 05:38:42 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
dcons_timeout(void *v)
|
|
|
|
{
|
|
|
|
struct tty *tp;
|
|
|
|
struct dcons_softc *dc;
|
|
|
|
int i, c, polltime;
|
|
|
|
|
|
|
|
for (i = 0; i < DCONS_NPORT; i ++) {
|
|
|
|
dc = &sc[i];
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
tp = dc->tty;
|
|
|
|
|
|
|
|
tty_lock(tp);
|
2013-02-19 18:22:25 +00:00
|
|
|
while ((c = dcons_os_checkc_nopoll(dc)) != -1) {
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
ttydisc_rint(tp, c, 0);
|
2013-02-19 18:22:25 +00:00
|
|
|
poll_idle = 0;
|
|
|
|
}
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
ttydisc_rint_done(tp);
|
|
|
|
tty_unlock(tp);
|
2004-10-13 05:38:42 +00:00
|
|
|
}
|
2013-02-19 18:22:25 +00:00
|
|
|
poll_idle++;
|
|
|
|
polltime = hz;
|
|
|
|
if (poll_idle <= (poll_hz * DCONS_POLL_IDLE))
|
|
|
|
polltime /= poll_hz;
|
2004-10-13 05:38:42 +00:00
|
|
|
callout_reset(&dcons_callout, polltime, dcons_timeout, tp);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
dcons_cnprobe(struct consdev *cp)
|
|
|
|
{
|
|
|
|
sprintf(cp->cn_name, "dcons");
|
|
|
|
#if DCONS_FORCE_CONSOLE
|
|
|
|
cp->cn_pri = CN_REMOTE;
|
|
|
|
#else
|
|
|
|
cp->cn_pri = CN_NORMAL;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
dcons_cninit(struct consdev *cp)
|
|
|
|
{
|
|
|
|
dcons_drv_init(0);
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
cp->cn_arg = (void *)&sc[DCONS_CON]; /* share port0 with unit0 */
|
2004-10-13 05:38:42 +00:00
|
|
|
}
|
|
|
|
|
2006-05-26 13:51:38 +00:00
|
|
|
static void
|
|
|
|
dcons_cnterm(struct consdev *cp)
|
2004-10-13 05:38:42 +00:00
|
|
|
{
|
|
|
|
}
|
2006-05-26 13:51:38 +00:00
|
|
|
|
2011-12-17 15:08:43 +00:00
|
|
|
static void
|
|
|
|
dcons_cngrab(struct consdev *cp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
dcons_cnungrab(struct consdev *cp)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
static int
|
2006-05-26 13:51:38 +00:00
|
|
|
dcons_cngetc(struct consdev *cp)
|
2004-10-13 05:38:42 +00:00
|
|
|
{
|
|
|
|
struct dcons_softc *dc = (struct dcons_softc *)cp->cn_arg;
|
|
|
|
return (dcons_os_checkc(dc));
|
|
|
|
}
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
static void
|
|
|
|
dcons_cnputc(struct consdev *cp, int c)
|
|
|
|
{
|
|
|
|
struct dcons_softc *dc = (struct dcons_softc *)cp->cn_arg;
|
|
|
|
dcons_os_putc(dc, c);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
dcons_drv_init(int stage)
|
|
|
|
{
|
2004-10-28 12:18:22 +00:00
|
|
|
#if defined(__i386__) || defined(__amd64__)
|
2004-10-24 12:41:04 +00:00
|
|
|
quad_t addr, size;
|
2004-10-14 00:21:32 +00:00
|
|
|
#endif
|
2004-10-13 05:38:42 +00:00
|
|
|
|
|
|
|
if (drv_init)
|
|
|
|
return(drv_init);
|
|
|
|
|
|
|
|
drv_init = -1;
|
|
|
|
|
|
|
|
bzero(&dg, sizeof(dg));
|
|
|
|
dcons_conf = &dg;
|
|
|
|
dg.cdev = &dcons_consdev;
|
|
|
|
dg.buf = NULL;
|
|
|
|
dg.size = DCONS_BUF_SIZE;
|
|
|
|
|
2004-10-28 12:18:22 +00:00
|
|
|
#if defined(__i386__) || defined(__amd64__)
|
2004-10-24 12:41:04 +00:00
|
|
|
if (getenv_quad("dcons.addr", &addr) > 0 &&
|
|
|
|
getenv_quad("dcons.size", &size) > 0) {
|
2004-10-28 12:18:22 +00:00
|
|
|
#ifdef __i386__
|
2004-10-24 12:41:04 +00:00
|
|
|
vm_paddr_t pa;
|
|
|
|
/*
|
|
|
|
* Allow read/write access to dcons buffer.
|
|
|
|
*/
|
|
|
|
for (pa = trunc_page(addr); pa < addr + size; pa += PAGE_SIZE)
|
i386: Merge PAE and non-PAE pmaps into same kernel.
Effectively all i386 kernels now have two pmaps compiled in: one
managing PAE pagetables, and another non-PAE. The implementation is
selected at cold time depending on the CPU features. The vm_paddr_t is
always 64bit now. As result, nx bit can be used on all capable CPUs.
Option PAE only affects the bus_addr_t: it is still 32bit for non-PAE
configs, for drivers compatibility. Kernel layout, esp. max kernel
address, low memory PDEs and max user address (same as trampoline
start) are now same for PAE and for non-PAE regardless of the type of
page tables used.
Non-PAE kernel (when using PAE pagetables) can handle physical memory
up to 24G now, larger memory requires re-tuning the KVA consumers and
instead the code caps the maximum at 24G. Unfortunately, a lot of
drivers do not use busdma(9) properly so by default even 4G barrier is
not easy. There are two tunables added: hw.above4g_allow and
hw.above24g_allow, the first one is kept enabled for now to evaluate
the status on HEAD, second is only for dev use.
i386 now creates three freelists if there is any memory above 4G, to
allow proper bounce pages allocation. Also, VM_KMEM_SIZE_SCALE changed
from 3 to 1.
The PAE_TABLES kernel config option is retired.
In collaboarion with: pho
Discussed with: emaste
Reviewed by: markj
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D18894
2019-01-30 02:07:13 +00:00
|
|
|
pmap_ksetrw(PMAP_MAP_LOW + pa);
|
2004-10-24 12:41:04 +00:00
|
|
|
invltlb();
|
2004-10-28 12:18:22 +00:00
|
|
|
#endif
|
2004-10-13 05:38:42 +00:00
|
|
|
/* XXX P to V */
|
i386 4/4G split.
The change makes the user and kernel address spaces on i386
independent, giving each almost the full 4G of usable virtual addresses
except for one PDE at top used for trampoline and per-CPU trampoline
stacks, and system structures that must be always mapped, namely IDT,
GDT, common TSS and LDT, and process-private TSS and LDT if allocated.
By using 1:1 mapping for the kernel text and data, it appeared
possible to eliminate assembler part of the locore.S which bootstraps
initial page table and KPTmap. The code is rewritten in C and moved
into the pmap_cold(). The comment in vmparam.h explains the KVA
layout.
There is no PCID mechanism available in protected mode, so each
kernel/user switch forth and back completely flushes the TLB, except
for the trampoline PTD region. The TLB invalidations for userspace
becomes trivial, because IPI handlers switch page tables. On the other
hand, context switches no longer need to reload %cr3.
copyout(9) was rewritten to use vm_fault_quick_hold(). An issue for
new copyout(9) is compatibility with wiring user buffers around sysctl
handlers. This explains two kind of locks for copyout ptes and
accounting of the vslock() calls. The vm_fault_quick_hold() AKA slow
path, is only tried after the 'fast path' failed, which temporary
changes mapping to the userspace and copies the data to/from small
per-cpu buffer in the trampoline. If a page fault occurs during the
copy, it is short-circuit by exception.s to not even reach C code.
The change was motivated by the need to implement the Meltdown
mitigation, but instead of KPTI the full split is done. The i386
architecture already shows the sizing problems, in particular, it is
impossible to link clang and lld with debugging. I expect that the
issues due to the virtual address space limits would only exaggerate
and the split gives more liveness to the platform.
Tested by: pho
Discussed with: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 month
Differential revision: https://reviews.freebsd.org/D14633
2018-04-13 20:30:49 +00:00
|
|
|
#ifdef __amd64__
|
2004-10-24 12:41:04 +00:00
|
|
|
dg.buf = (struct dcons_buf *)(vm_offset_t)(KERNBASE + addr);
|
i386 4/4G split.
The change makes the user and kernel address spaces on i386
independent, giving each almost the full 4G of usable virtual addresses
except for one PDE at top used for trampoline and per-CPU trampoline
stacks, and system structures that must be always mapped, namely IDT,
GDT, common TSS and LDT, and process-private TSS and LDT if allocated.
By using 1:1 mapping for the kernel text and data, it appeared
possible to eliminate assembler part of the locore.S which bootstraps
initial page table and KPTmap. The code is rewritten in C and moved
into the pmap_cold(). The comment in vmparam.h explains the KVA
layout.
There is no PCID mechanism available in protected mode, so each
kernel/user switch forth and back completely flushes the TLB, except
for the trampoline PTD region. The TLB invalidations for userspace
becomes trivial, because IPI handlers switch page tables. On the other
hand, context switches no longer need to reload %cr3.
copyout(9) was rewritten to use vm_fault_quick_hold(). An issue for
new copyout(9) is compatibility with wiring user buffers around sysctl
handlers. This explains two kind of locks for copyout ptes and
accounting of the vslock() calls. The vm_fault_quick_hold() AKA slow
path, is only tried after the 'fast path' failed, which temporary
changes mapping to the userspace and copies the data to/from small
per-cpu buffer in the trampoline. If a page fault occurs during the
copy, it is short-circuit by exception.s to not even reach C code.
The change was motivated by the need to implement the Meltdown
mitigation, but instead of KPTI the full split is done. The i386
architecture already shows the sizing problems, in particular, it is
impossible to link clang and lld with debugging. I expect that the
issues due to the virtual address space limits would only exaggerate
and the split gives more liveness to the platform.
Tested by: pho
Discussed with: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 month
Differential revision: https://reviews.freebsd.org/D14633
2018-04-13 20:30:49 +00:00
|
|
|
#else /* __i386__ */
|
|
|
|
dg.buf = (struct dcons_buf *)((vm_offset_t)PMAP_MAP_LOW +
|
|
|
|
addr);
|
|
|
|
#endif
|
2004-10-13 05:38:42 +00:00
|
|
|
dg.size = size;
|
|
|
|
if (dcons_load_buffer(dg.buf, dg.size, sc) < 0)
|
|
|
|
dg.buf = NULL;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
if (dg.buf != NULL)
|
|
|
|
goto ok;
|
|
|
|
|
|
|
|
#ifndef KLD_MODULE
|
|
|
|
if (stage == 0) { /* XXX or cold */
|
|
|
|
/*
|
|
|
|
* DCONS_FORCE_CONSOLE == 1 and statically linked.
|
|
|
|
* called from cninit(). can't use contigmalloc yet .
|
|
|
|
*/
|
|
|
|
dg.buf = (struct dcons_buf *) bssbuf;
|
|
|
|
dcons_init(dg.buf, dg.size, sc);
|
|
|
|
} else
|
|
|
|
#endif
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* DCONS_FORCE_CONSOLE == 0 or kernel module case.
|
|
|
|
* if the module is loaded after boot,
|
|
|
|
* bssbuf could be non-continuous.
|
|
|
|
*/
|
|
|
|
dg.buf = (struct dcons_buf *) contigmalloc(dg.size,
|
|
|
|
M_DEVBUF, 0, 0x10000, 0xffffffff, PAGE_SIZE, 0ul);
|
2009-01-07 21:25:44 +00:00
|
|
|
if (dg.buf == NULL)
|
|
|
|
return (-1);
|
2004-10-13 05:38:42 +00:00
|
|
|
dcons_init(dg.buf, dg.size, sc);
|
|
|
|
}
|
|
|
|
|
|
|
|
ok:
|
|
|
|
dcons_buf = dg.buf;
|
|
|
|
|
|
|
|
drv_init = 1;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
dcons_attach_port(int port, char *name, int flags)
|
|
|
|
{
|
|
|
|
struct dcons_softc *dc;
|
|
|
|
struct tty *tp;
|
|
|
|
|
|
|
|
dc = &sc[port];
|
2009-05-29 06:41:23 +00:00
|
|
|
tp = tty_alloc(&dcons_ttydevsw, dc);
|
2004-10-13 05:38:42 +00:00
|
|
|
dc->flags = flags;
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
dc->tty = tp;
|
2009-03-13 02:15:49 +00:00
|
|
|
tty_init_console(tp, 0);
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
tty_makedev(tp, NULL, "%s", name);
|
2004-10-13 05:38:42 +00:00
|
|
|
return(0);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
dcons_attach(void)
|
|
|
|
{
|
|
|
|
int polltime;
|
|
|
|
|
|
|
|
dcons_attach_port(DCONS_CON, "dcons", 0);
|
|
|
|
dcons_attach_port(DCONS_GDB, "dgdb", DC_GDB);
|
2015-05-22 17:05:21 +00:00
|
|
|
callout_init(&dcons_callout, 1);
|
2004-10-13 05:38:42 +00:00
|
|
|
polltime = hz / poll_hz;
|
|
|
|
callout_reset(&dcons_callout, polltime, dcons_timeout, NULL);
|
|
|
|
return(0);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
dcons_detach(int port)
|
|
|
|
{
|
|
|
|
struct tty *tp;
|
|
|
|
struct dcons_softc *dc;
|
|
|
|
|
|
|
|
dc = &sc[port];
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
tp = dc->tty;
|
2004-10-13 05:38:42 +00:00
|
|
|
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
tty_lock(tp);
|
|
|
|
tty_rel_gone(tp);
|
2004-10-13 05:38:42 +00:00
|
|
|
|
|
|
|
return(0);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
dcons_modevent(module_t mode, int type, void *data)
|
|
|
|
{
|
|
|
|
int err = 0, ret;
|
|
|
|
|
|
|
|
switch (type) {
|
|
|
|
case MOD_LOAD:
|
|
|
|
ret = dcons_drv_init(1);
|
2009-04-06 02:29:28 +00:00
|
|
|
if (ret != -1)
|
2009-01-07 21:25:44 +00:00
|
|
|
dcons_attach();
|
2009-04-06 02:29:28 +00:00
|
|
|
if (ret == 0) {
|
2004-10-13 05:38:42 +00:00
|
|
|
dcons_cnprobe(&dcons_consdev);
|
|
|
|
dcons_cninit(&dcons_consdev);
|
|
|
|
cnadd(&dcons_consdev);
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case MOD_UNLOAD:
|
|
|
|
printf("dcons: unload\n");
|
2009-01-07 21:25:44 +00:00
|
|
|
if (drv_init == 1) {
|
|
|
|
callout_stop(&dcons_callout);
|
|
|
|
cnremove(&dcons_consdev);
|
|
|
|
dcons_detach(DCONS_CON);
|
|
|
|
dcons_detach(DCONS_GDB);
|
|
|
|
dg.buf->magic = 0;
|
|
|
|
|
|
|
|
contigfree(dg.buf, DCONS_BUF_SIZE, M_DEVBUF);
|
|
|
|
}
|
2004-10-13 05:38:42 +00:00
|
|
|
|
|
|
|
break;
|
|
|
|
case MOD_SHUTDOWN:
|
2007-06-08 08:23:14 +00:00
|
|
|
#if 0 /* Keep connection after halt */
|
2004-10-13 05:38:42 +00:00
|
|
|
dg.buf->magic = 0;
|
2007-06-08 08:23:14 +00:00
|
|
|
#endif
|
2004-10-13 05:38:42 +00:00
|
|
|
break;
|
|
|
|
default:
|
|
|
|
err = EOPNOTSUPP;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
return(err);
|
|
|
|
}
|
|
|
|
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
#if defined(GDB)
|
2004-10-13 05:38:42 +00:00
|
|
|
/* Debugger interface */
|
|
|
|
|
Integrate the new MPSAFE TTY layer to the FreeBSD operating system.
The last half year I've been working on a replacement TTY layer for the
FreeBSD kernel. The new TTY layer was designed to improve the following:
- Improved driver model:
The old TTY layer has a driver model that is not abstract enough to
make it friendly to use. A good example is the output path, where the
device drivers directly access the output buffers. This means that an
in-kernel PPP implementation must always convert network buffers into
TTY buffers.
If a PPP implementation would be built on top of the new TTY layer
(still needs a hooks layer, though), it would allow the PPP
implementation to directly hand the data to the TTY driver.
- Improved hotplugging:
With the old TTY layer, it isn't entirely safe to destroy TTY's from
the system. This implementation has a two-step destructing design,
where the driver first abandons the TTY. After all threads have left
the TTY, the TTY layer calls a routine in the driver, which can be
used to free resources (unit numbers, etc).
The pts(4) driver also implements this feature, which means
posix_openpt() will now return PTY's that are created on the fly.
- Improved performance:
One of the major improvements is the per-TTY mutex, which is expected
to improve scalability when compared to the old Giant locking.
Another change is the unbuffered copying to userspace, which is both
used on TTY device nodes and PTY masters.
Upgrading should be quite straightforward. Unlike previous versions,
existing kernel configuration files do not need to be changed, except
when they reference device drivers that are listed in UPDATING.
Obtained from: //depot/projects/mpsafetty/...
Approved by: philip (ex-mentor)
Discussed: on the lists, at BSDCan, at the DevSummit
Sponsored by: Snow B.V., the Netherlands
dcons(4) fixed by: kan
2008-08-20 08:31:58 +00:00
|
|
|
static int
|
|
|
|
dcons_os_getc(struct dcons_softc *dc)
|
|
|
|
{
|
|
|
|
int c;
|
|
|
|
|
|
|
|
while ((c = dcons_os_checkc(dc)) == -1);
|
|
|
|
|
|
|
|
return (c & 0xff);
|
|
|
|
}
|
|
|
|
|
2004-10-13 05:38:42 +00:00
|
|
|
static int
|
|
|
|
dcons_dbg_probe(void)
|
|
|
|
{
|
2007-06-11 04:08:50 +00:00
|
|
|
int dcons_gdb;
|
|
|
|
|
|
|
|
if (getenv_int("dcons_gdb", &dcons_gdb) == 0)
|
|
|
|
return (-1);
|
|
|
|
return (dcons_gdb);
|
2004-10-13 05:38:42 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
dcons_dbg_init(void)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
dcons_dbg_term(void)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
dcons_dbg_putc(int c)
|
|
|
|
{
|
|
|
|
struct dcons_softc *dc = &sc[DCONS_GDB];
|
|
|
|
dcons_os_putc(dc, c);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
dcons_dbg_getc(void)
|
|
|
|
{
|
|
|
|
struct dcons_softc *dc = &sc[DCONS_GDB];
|
|
|
|
return (dcons_os_getc(dc));
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
DEV_MODULE(dcons, dcons_modevent, NULL);
|
|
|
|
MODULE_VERSION(dcons, DCONS_VERSION);
|