freebsd-nq/sys/kern/tty_inq.c

484 lines
12 KiB
C
Raw Normal View History

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
/*-
* Copyright (c) 2008 Ed Schouten <ed@FreeBSD.org>
* All rights reserved.
*
* Portions of this software were developed under sponsorship from Snow
* B.V., the Netherlands.
*
* 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 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 AUTHOR 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.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/queue.h>
#include <sys/sysctl.h>
#include <sys/systm.h>
#include <sys/tty.h>
#include <sys/uio.h>
#include <vm/uma.h>
/*
* TTY input queue buffering.
*
* Unlike the output queue, the input queue has more features that are
* needed to properly implement various features offered by the TTY
* interface:
*
* - Data can be removed from the tail of the queue, which is used to
* implement backspace.
* - Once in a while, input has to be `canonicalized'. When ICANON is
* turned on, this will be done after a CR has been inserted.
* Otherwise, it should be done after any character has been inserted.
* - The input queue can store one bit per byte, called the quoting bit.
* This bit is used by TTYDISC to make backspace work on quoted
* characters.
*
* In most cases, there is probably less input than output, so unlike
* the outq, we'll stick to 128 byte blocks here.
*/
/* Statistics. */
static long ttyinq_nfast = 0;
SYSCTL_LONG(_kern, OID_AUTO, tty_inq_nfast, CTLFLAG_RD,
&ttyinq_nfast, 0, "Unbuffered reads to userspace on input");
static long ttyinq_nslow = 0;
SYSCTL_LONG(_kern, OID_AUTO, tty_inq_nslow, CTLFLAG_RD,
&ttyinq_nslow, 0, "Buffered reads to userspace on input");
#define TTYINQ_QUOTESIZE (TTYINQ_DATASIZE / BMSIZE)
#define BMSIZE 32
#define GETBIT(tib,boff) \
((tib)->tib_quotes[(boff) / BMSIZE] & (1 << ((boff) % BMSIZE)))
#define SETBIT(tib,boff) \
((tib)->tib_quotes[(boff) / BMSIZE] |= (1 << ((boff) % BMSIZE)))
#define CLRBIT(tib,boff) \
((tib)->tib_quotes[(boff) / BMSIZE] &= ~(1 << ((boff) % BMSIZE)))
struct ttyinq_block {
TAILQ_ENTRY(ttyinq_block) tib_list;
uint32_t tib_quotes[TTYINQ_QUOTESIZE];
char tib_data[TTYINQ_DATASIZE];
};
static uma_zone_t ttyinq_zone;
void
ttyinq_setsize(struct ttyinq *ti, struct tty *tp, size_t size)
{
struct ttyinq_block *tib;
ti->ti_quota = howmany(size, TTYINQ_DATASIZE);
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 (ti->ti_quota > ti->ti_nblocks) {
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
/*
* List is getting bigger.
* Add new blocks to the tail of the list.
*
* We must unlock the TTY temporarily, because we need
* to allocate memory. This won't be a problem, because
* in the worst case, another thread ends up here, which
* may cause us to allocate too many blocks, but this
* will be caught by the loop below.
*/
tty_unlock(tp);
tib = uma_zalloc(ttyinq_zone, M_WAITOK);
tty_lock(tp);
TAILQ_INSERT_TAIL(&ti->ti_list, tib, tib_list);
ti->ti_nblocks++;
}
}
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
void
ttyinq_free(struct ttyinq *ti)
{
struct ttyinq_block *tib;
ttyinq_flush(ti);
ti->ti_quota = 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
while ((tib = TAILQ_FIRST(&ti->ti_list)) != NULL) {
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
TAILQ_REMOVE(&ti->ti_list, tib, tib_list);
uma_zfree(ttyinq_zone, tib);
ti->ti_nblocks--;
}
MPASS(ti->ti_nblocks == 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
}
int
ttyinq_read_uio(struct ttyinq *ti, struct tty *tp, struct uio *uio,
size_t rlen, size_t flen)
{
MPASS(rlen <= uio->uio_resid);
while (rlen > 0) {
int error;
struct ttyinq_block *tib;
size_t cbegin, cend, clen;
/* See if there still is data. */
if (ti->ti_begin == ti->ti_linestart)
return (0);
tib = TAILQ_FIRST(&ti->ti_list);
if (tib == NULL)
return (0);
/*
* The end address should be the lowest of these three:
* - The write pointer
* - The blocksize - we can't read beyond the block
* - The end address if we could perform the full read
*/
cbegin = ti->ti_begin;
cend = MIN(MIN(ti->ti_linestart, ti->ti_begin + rlen),
TTYINQ_DATASIZE);
clen = cend - cbegin;
MPASS(clen >= flen);
rlen -= clen;
/*
* We can prevent buffering in some cases:
* - We need to read the block until the end.
* - We don't need to read the block until the end, but
* there is no data beyond it, which allows us to move
* the write pointer to a new block.
*/
if (cend == TTYINQ_DATASIZE || cend == ti->ti_end) {
atomic_add_long(&ttyinq_nfast, 1);
/*
* Fast path: zero copy. Remove the first block,
* so we can unlock the TTY temporarily.
*/
TAILQ_REMOVE(&ti->ti_list, tib, tib_list);
ti->ti_nblocks--;
ti->ti_begin = 0;
/*
* Because we remove the first block, we must
* fix up the block offsets.
*/
#define CORRECT_BLOCK(t) do { \
if (t <= TTYINQ_DATASIZE) { \
t = 0; \
} else { \
t -= TTYINQ_DATASIZE; \
} \
} while (0)
CORRECT_BLOCK(ti->ti_linestart);
CORRECT_BLOCK(ti->ti_reprint);
CORRECT_BLOCK(ti->ti_end);
#undef CORRECT_BLOCK
/*
* Temporary unlock and copy the data to
* userspace. We may need to flush trailing
* bytes, like EOF characters.
*/
tty_unlock(tp);
error = uiomove(tib->tib_data + cbegin,
clen - flen, uio);
tty_lock(tp);
/* Block can now be readded to the list. */
if (ti->ti_quota <= ti->ti_nblocks) {
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
uma_zfree(ttyinq_zone, tib);
} else {
TAILQ_INSERT_TAIL(&ti->ti_list, tib, tib_list);
ti->ti_nblocks++;
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
}
} else {
char ob[TTYINQ_DATASIZE - 1];
atomic_add_long(&ttyinq_nslow, 1);
/*
* Slow path: store data in a temporary buffer.
*/
memcpy(ob, tib->tib_data + cbegin, clen - flen);
ti->ti_begin += clen;
MPASS(ti->ti_begin < TTYINQ_DATASIZE);
/* Temporary unlock and copy the data to userspace. */
tty_unlock(tp);
error = uiomove(ob, clen - flen, uio);
tty_lock(tp);
}
if (error != 0)
return (error);
if (tty_gone(tp))
return (ENXIO);
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
}
return (0);
}
static __inline void
ttyinq_set_quotes(struct ttyinq_block *tib, size_t offset,
size_t length, int value)
{
if (value) {
/* Set the bits. */
for (; length > 0; length--, offset++)
SETBIT(tib, offset);
} else {
/* Unset the bits. */
for (; length > 0; length--, offset++)
CLRBIT(tib, offset);
}
}
size_t
ttyinq_write(struct ttyinq *ti, const void *buf, size_t nbytes, int quote)
{
const char *cbuf = buf;
struct ttyinq_block *tib;
unsigned int boff;
size_t l;
while (nbytes > 0) {
tib = ti->ti_lastblock;
boff = ti->ti_end % TTYINQ_DATASIZE;
if (ti->ti_end == 0) {
/* First time we're being used or drained. */
MPASS(ti->ti_begin == 0);
tib = ti->ti_lastblock = TAILQ_FIRST(&ti->ti_list);
if (tib == NULL) {
/* Queue has no blocks. */
break;
}
} else if (boff == 0) {
/* We reached the end of this block on last write. */
tib = TAILQ_NEXT(tib, tib_list);
if (tib == NULL) {
/* We've reached the watermark. */
break;
}
ti->ti_lastblock = tib;
}
/* Don't copy more than was requested. */
l = MIN(nbytes, TTYINQ_DATASIZE - boff);
MPASS(l > 0);
memcpy(tib->tib_data + boff, cbuf, l);
/* Set the quoting bits for the proper region. */
ttyinq_set_quotes(tib, boff, l, quote);
cbuf += l;
nbytes -= l;
ti->ti_end += l;
}
return (cbuf - (const char *)buf);
}
int
ttyinq_write_nofrag(struct ttyinq *ti, const void *buf, size_t nbytes, int quote)
{
size_t ret;
if (ttyinq_bytesleft(ti) < nbytes)
return (-1);
/* We should always be able to write it back. */
ret = ttyinq_write(ti, buf, nbytes, quote);
MPASS(ret == nbytes);
return (0);
}
void
ttyinq_canonicalize(struct ttyinq *ti)
{
ti->ti_linestart = ti->ti_reprint = ti->ti_end;
ti->ti_startblock = ti->ti_reprintblock = ti->ti_lastblock;
}
size_t
ttyinq_findchar(struct ttyinq *ti, const char *breakc, size_t maxlen,
char *lastc)
{
struct ttyinq_block *tib = TAILQ_FIRST(&ti->ti_list);
unsigned int boff = ti->ti_begin;
unsigned int bend = MIN(MIN(TTYINQ_DATASIZE, ti->ti_linestart),
ti->ti_begin + maxlen);
MPASS(maxlen > 0);
if (tib == NULL)
return (0);
while (boff < bend) {
if (index(breakc, tib->tib_data[boff]) && !GETBIT(tib, boff)) {
*lastc = tib->tib_data[boff];
return (boff - ti->ti_begin + 1);
}
boff++;
}
/* Not found - just process the entire block. */
return (bend - ti->ti_begin);
}
void
ttyinq_flush(struct ttyinq *ti)
{
ti->ti_begin = 0;
ti->ti_linestart = 0;
ti->ti_reprint = 0;
ti->ti_end = 0;
}
#if 0
void
ttyinq_flush_safe(struct ttyinq *ti)
{
struct ttyinq_block *tib;
ttyinq_flush(ti);
/* Zero all data in the input queue to make it more safe */
TAILQ_FOREACH(tib, &ti->ti_list, tib_list) {
bzero(&tib->tib_quotes, sizeof tib->tib_quotes);
bzero(&tib->tib_data, sizeof tib->tib_data);
}
}
#endif
int
ttyinq_peekchar(struct ttyinq *ti, char *c, int *quote)
{
unsigned int boff;
struct ttyinq_block *tib = ti->ti_lastblock;
if (ti->ti_linestart == ti->ti_end)
return (-1);
MPASS(ti->ti_end > 0);
boff = (ti->ti_end - 1) % TTYINQ_DATASIZE;
*c = tib->tib_data[boff];
*quote = GETBIT(tib, boff);
return (0);
}
void
ttyinq_unputchar(struct ttyinq *ti)
{
MPASS(ti->ti_linestart < ti->ti_end);
if (--ti->ti_end % TTYINQ_DATASIZE == 0) {
/* Roll back to the previous block. */
ti->ti_lastblock = TAILQ_PREV(ti->ti_lastblock,
ttyinq_bhead, tib_list);
/*
* This can only fail if we are unputchar()'ing the
* first character in the queue.
*/
MPASS((ti->ti_lastblock == NULL) == (ti->ti_end == 0));
}
}
void
ttyinq_reprintpos_set(struct ttyinq *ti)
{
ti->ti_reprint = ti->ti_end;
ti->ti_reprintblock = ti->ti_lastblock;
}
void
ttyinq_reprintpos_reset(struct ttyinq *ti)
{
ti->ti_reprint = ti->ti_linestart;
ti->ti_reprintblock = ti->ti_startblock;
}
static void
ttyinq_line_iterate(struct ttyinq *ti,
ttyinq_line_iterator_t *iterator, void *data,
unsigned int offset, struct ttyinq_block *tib)
{
unsigned int boff;
/* Use the proper block when we're at the queue head. */
if (offset == 0)
tib = TAILQ_FIRST(&ti->ti_list);
/* Iterate all characters and call the iterator function. */
for (; offset < ti->ti_end; offset++) {
boff = offset % TTYINQ_DATASIZE;
MPASS(tib != NULL);
/* Call back the iterator function. */
iterator(data, tib->tib_data[boff], GETBIT(tib, boff));
/* Last byte iterated - go to the next block. */
if (boff == TTYINQ_DATASIZE - 1)
tib = TAILQ_NEXT(tib, tib_list);
MPASS(tib != NULL);
}
}
void
ttyinq_line_iterate_from_linestart(struct ttyinq *ti,
ttyinq_line_iterator_t *iterator, void *data)
{
ttyinq_line_iterate(ti, iterator, data,
ti->ti_linestart, ti->ti_startblock);
}
void
ttyinq_line_iterate_from_reprintpos(struct ttyinq *ti,
ttyinq_line_iterator_t *iterator, void *data)
{
ttyinq_line_iterate(ti, iterator, data,
ti->ti_reprint, ti->ti_reprintblock);
}
static void
ttyinq_startup(void *dummy)
{
ttyinq_zone = uma_zcreate("ttyinq", sizeof(struct ttyinq_block),
NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, 0);
}
SYSINIT(ttyinq, SI_SUB_DRIVERS, SI_ORDER_FIRST, ttyinq_startup, NULL);