d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
06d6750e0d
freebsd-skq/stand/libsa/netif.c

317 lines
7.3 KiB
C
Raw Normal View History

This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
/* $NetBSD: netif.c,v 1.10 1997/09/06 13:57:14 drochner Exp $ */
/*
* Copyright (c) 1993 Adam Glass
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by Adam Glass.
* 4. The name of the Author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY Adam Glass ``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.
*/
Add __FBSDID()s to libstand
2001-09-30 22:28:01 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
#include <sys/param.h>
#include <sys/types.h>
#include <sys/cdefs.h>
#include <sys/mount.h>
#include <string.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include "stand.h"
#include "net.h"
#include "netif.h"
struct iodesc sockets[SOPEN_MAX];
#ifdef NETIF_DEBUG
int netif_debug = 0;
#endif
/*
* netif_init:
*
* initialize the generic network interface layer
*/
void
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_init(void)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
struct netif_driver *drv;
int d, i;
#ifdef NETIF_DEBUG
if (netif_debug)
printf("netif_init: called\n");
#endif
for (d = 0; netif_drivers[d]; d++) {
drv = netif_drivers[d];
for (i = 0; i < drv->netif_nifs; i++)
drv->netif_ifs[i].dif_used = 0;
}
}
int
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_match(struct netif *nif, void *machdep_hint)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
struct netif_driver *drv = nif->nif_driver;
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
#if NETIF_DEBUG
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
if (netif_debug)
printf("%s%d: netif_match (%d)\n", drv->netif_bname,
nif->nif_unit, nif->nif_sel);
#endif
return drv->netif_match(nif, machdep_hint);
}
struct netif *
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_select(void *machdep_hint)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
int d, u, unit_done, s;
struct netif_driver *drv;
struct netif cur_if;
static struct netif best_if;
int best_val;
int val;
best_val = 0;
best_if.nif_driver = NULL;
for (d = 0; netif_drivers[d] != NULL; d++) {
cur_if.nif_driver = netif_drivers[d];
drv = cur_if.nif_driver;
for (u = 0; u < drv->netif_nifs; u++) {
cur_if.nif_unit = u;
unit_done = 0;
#ifdef NETIF_DEBUG
if (netif_debug)
printf("\t%s%d:", drv->netif_bname,
cur_if.nif_unit);
#endif
for (s = 0; s < drv->netif_ifs[u].dif_nsel; s++) {
cur_if.nif_sel = s;
if (drv->netif_ifs[u].dif_used & (1 << s)) {
#ifdef NETIF_DEBUG
if (netif_debug)
printf(" [%d used]", s);
#endif
continue;
}
val = netif_match(&cur_if, machdep_hint);
#ifdef NETIF_DEBUG
if (netif_debug)
printf(" [%d -> %d]", s, val);
#endif
if (val > best_val) {
best_val = val;
best_if = cur_if;
}
}
#ifdef NETIF_DEBUG
if (netif_debug)
printf("\n");
#endif
}
}
if (best_if.nif_driver == NULL)
return NULL;
best_if.nif_driver->
netif_ifs[best_if.nif_unit].dif_used |= (1 << best_if.nif_sel);
#ifdef NETIF_DEBUG
if (netif_debug)
printf("netif_select: %s%d(%d) wins\n",
best_if.nif_driver->netif_bname,
best_if.nif_unit, best_if.nif_sel);
#endif
return &best_if;
}
int
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_probe(struct netif *nif, void *machdep_hint)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
struct netif_driver *drv = nif->nif_driver;
#ifdef NETIF_DEBUG
if (netif_debug)
printf("%s%d: netif_probe\n", drv->netif_bname, nif->nif_unit);
#endif
return drv->netif_probe(nif, machdep_hint);
}
void
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_attach(struct netif *nif, struct iodesc *desc, void *machdep_hint)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
struct netif_driver *drv = nif->nif_driver;
#ifdef NETIF_DEBUG
if (netif_debug)
printf("%s%d: netif_attach\n", drv->netif_bname, nif->nif_unit);
#endif
desc->io_netif = nif;
#ifdef PARANOID
if (drv->netif_init == NULL)
panic("%s%d: no netif_init support\n", drv->netif_bname,
nif->nif_unit);
#endif
drv->netif_init(desc, machdep_hint);
bzero(drv->netif_ifs[nif->nif_unit].dif_stats,
sizeof(struct netif_stats));
}
void
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_detach(struct netif *nif)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
struct netif_driver *drv = nif->nif_driver;
#ifdef NETIF_DEBUG
if (netif_debug)
printf("%s%d: netif_detach\n", drv->netif_bname, nif->nif_unit);
#endif
#ifdef PARANOID
if (drv->netif_end == NULL)
panic("%s%d: no netif_end support\n", drv->netif_bname,
nif->nif_unit);
#endif
drv->netif_end(nif);
}
ssize_t
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_get(struct iodesc *desc, void **pkt, time_t timo)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
#ifdef NETIF_DEBUG
struct netif *nif = desc->io_netif;
#endif
struct netif_driver *drv = desc->io_netif->nif_driver;
ssize_t rv;
#ifdef NETIF_DEBUG
if (netif_debug)
printf("%s%d: netif_get\n", drv->netif_bname, nif->nif_unit);
#endif
#ifdef PARANOID
if (drv->netif_get == NULL)
panic("%s%d: no netif_get support\n", drv->netif_bname,
nif->nif_unit);
#endif
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
rv = drv->netif_get(desc, pkt, timo);
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
#ifdef NETIF_DEBUG
if (netif_debug)
printf("%s%d: netif_get returning %d\n", drv->netif_bname,
nif->nif_unit, (int)rv);
#endif
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
return (rv);
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
}
ssize_t
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_put(struct iodesc *desc, void *pkt, size_t len)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
#ifdef NETIF_DEBUG
struct netif *nif = desc->io_netif;
#endif
struct netif_driver *drv = desc->io_netif->nif_driver;
ssize_t rv;
#ifdef NETIF_DEBUG
if (netif_debug)
printf("%s%d: netif_put\n", drv->netif_bname, nif->nif_unit);
#endif
#ifdef PARANOID
if (drv->netif_put == NULL)
panic("%s%d: no netif_put support\n", drv->netif_bname,
nif->nif_unit);
#endif
rv = drv->netif_put(desc, pkt, len);
#ifdef NETIF_DEBUG
if (netif_debug)
printf("%s%d: netif_put returning %d\n", drv->netif_bname,
nif->nif_unit, (int)rv);
#endif
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
return (rv);
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
}
struct iodesc *
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
socktodesc(int sock)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
if (sock >= SOPEN_MAX) {
errno = EBADF;
return (NULL);
}
return (&sockets[sock]);
}
int
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_open(void *machdep_hint)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
int fd;
Remove 'register' keyword.
2002-03-21 23:39:28 +00:00
struct iodesc *s;
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
struct netif *nif;
/* find a free socket */
for (fd = 0, s = sockets; fd < SOPEN_MAX; fd++, s++)
if (s->io_netif == (struct netif *)0)
goto fnd;
errno = EMFILE;
return (-1);
fnd:
bzero(s, sizeof(*s));
netif_init();
nif = netif_select(machdep_hint);
if (!nif)
panic("netboot: no interfaces left untried");
if (netif_probe(nif, machdep_hint)) {
printf("netboot: couldn't probe %s%d\n",
nif->nif_driver->netif_bname, nif->nif_unit);
errno = EINVAL;
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
return (-1);
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
}
netif_attach(nif, s, machdep_hint);
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
return (fd);
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
}
int
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
netif_close(int sock)
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
{
if (sock >= SOPEN_MAX) {
errno = EBADF;
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
return (-1);
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
}
netif_detach(sockets[sock].io_netif);
sockets[sock].io_netif = (struct netif *)0;
loader: network read rework The current read from network is working from up to down - we have some protocol needing the data from the network, so we build the buffer space for that protocol, add the extra space for headers and pass this buffer down to be filled by nif get call in hope, we have guessed the incoming packet size right. Amazingly enough this approach mostly does work, but not always... So, this update does work from down to up - we allocate buffer (based on MTU or frame size info), fill it up, and pass on for upper layers. The obvious problem is that when we should free the buffer - if at all. In the current implementation the upper layer will free the packet on error or when the packet is no longer needed. While working on the issue, the additional issue did pop up - the bios implementation does not have generic get/put interface but is using pxe udpsend/udpreceive instead. So the udp calls are gone and undi interface is implemented instead. Which in turn means slight other changes as we do not need to have duplicated pxe implementation and can just use dev_net. To align packet content, the actual read from nic is using shifted buffer by ETHER_ALIGN (2). Reviewed by: bapt Differential Revision: https://reviews.freebsd.org/D10232
2017-05-06 20:32:27 +00:00
return (0);
This is libstand; a support library for standalone executables (eg. bootstrap modules). Obtained from: NetBSD, with some architectural changes and many additions.
1998-08-20 08:19:55 +00:00
}
Reference in New Issue Copy Permalink