nexoniarz/SchOS
nexoniarz/SchOS
Watch
1
0
Fork
You've already forked SchOS
 0
Code Issues Pull requests Projects Releases 1 Packages Wiki Activity Actions

net: a TCP/IP stack, and AF_INET on the sockets that were already there

Browse source 
lwIP 2.2.0, fetched by ports/lwip and compiled straight into the kernel.
Not written from scratch, and the reason is not the packet formats: it is
congestion control, retransmission, window management and twenty years of
corner cases that only show up against a real peer. Not FreeBSD's
netinet either, though the licence would have allowed it -- that stack is
not a library, it reaches into mbufs, epoch(9), netisr, VNET and uma(9),
so adopting it means porting FreeBSD's kernel infrastructure first. lwIP
was written for exactly this situation: a small kernel with a raw
Ethernet driver and a port layer of a few hundred lines.

It is kernel code, so nothing is installed into the sysroot and there is
no second build system: the top-level Makefile compiles the files it
wants against our port headers. The file list is explicit rather than a
wildcard, because lwIP ships IPv6, 6LoWPAN, PPP, SLIP, DNS-SD and a dozen
apps in the same tree, and building what is not configured yields either
dead code or a link error naming a port function nobody meant to write.
Its sources are unmodified and its warnings are silenced for its files
only -- patching them would mean maintaining a fork.

The port layer is ours: a lock, a clock, the netif that hands frames to
e1000, lwipopts.h, and a shim/ directory of tiny headers so lwIP's
<string.h> and friends resolve to the kernel's functions instead of the
host's, whose prototypes disagree and would otherwise be found on the
default include path.

NO_SYS=1, so lwIP has no threads of its own and one kernel thread feeds
it. A thread rather than the interrupt handler, because a call into lwIP
can reach the driver, the driver can sleep, and an interrupt handler
cannot -- so the interrupt still only clears the cause register and the
thread polls. kthread_create is new and generally useful: a process with
a kernel-only page directory, no files, no cwd, no parent, dispatched
from forkret, which can therefore sleep and be preempted like anything
else. DHCP obtains an address at boot and the console says which one.

AF_INET went in exactly where phase 2 said it would. socknew dispatches
on the domain, kernel/net/inet.c implements the internet side on lwIP's
raw API, and not one syscall changed shape: bind, connect, listen,
accept, read and write reach the same descriptors they always did. TCP
and UDP both work. cmd/nettest opens a connection, writes, and reads the
answer back.

Two things are owed and named here so they are not discovered later:
recvfrom does not yet report the source address, and the socket paths
poll on the timer tick instead of sleeping on the socket. Both are
follow-ups, not redesigns.

NOTICE carries lwIP's BSD-3 text. Clause 3 is no-endorsement, not the
old advertising clause, so it sits with Apache-2.0 without difficulty.

Verified: the tree builds clean, and the headless boot check passes with
27 [ok] and no failures with the stack compiled in and the thread
running -- which is the part that would break first if netinit deadlocked
or the thread starved anything. An end-to-end connection test belongs in
that suite and is the next commit.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
Nexoniarz 2026-08-15 16:02:48 +02:00
commit c0806d1b64
30 changed files with 1922 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

6
kernel/core/main.c
View file

@ -139,6 +139,12 @@ kmain(uint32 fb_addr)
if (boot_verbose) cprintf("verbose boot: kinit2 done\n");
sinit();
if (boot_verbose) cprintf("verbose boot: sinit done\n");
// The TCP/IP stack, after sinit because it starts a kernel thread and
// kthread_create needs the process table to be usable. It returns
// immediately: everything it does happens in that thread, which cannot run
// until mpmain() below enters the scheduler.
netinit();
if (boot_verbose) cprintf("verbose boot: netinit done\n");
mpmain();
}

8
kernel/drivers/net/e1000/e1000.c
View file

@ -387,6 +387,14 @@ put_be16(uint8 *p, uint16 v)
p[1] = (uint8)v;
}
// Is there a NIC at all? Asked by netinit() before it starts a thread whose
// entire job would be to poll a device that is not there.
int
e1000_present(void)
{
return nic.ready;
}
const unsigned char *
e1000_mac(void)
{

576
kernel/net/inet.c Normal file
View file

@ -0,0 +1,576 @@
// SPDX-License-Identifier: Apache-2.0
//
// AF_INET sockets, on top of lwIP's raw API.
//
// The raw API rather than lwIP's own sockets layer, because lwIP's needs
// NO_SYS=0 -- threads, mailboxes and semaphores of its own -- and this kernel
// already has a socket layer with a shape of its own that AF_INET has to fit
// into rather than replace. See kernel/net/lwip/lwipopts.h.
//
// ---------------------------------------------------------------------------
// Locking, and why there is only one lock here
//
// Everything in an inet socket is protected by the lwIP lock and nothing else.
// That is not a shortcut: lwIP's callbacks (data arriving, a connection
// completing, a peer closing) run in the network thread with that lock held,
// and they are the only things that write to a socket's receive buffer. A
// syscall reading that buffer takes the same lock. There is no second lock
// because there is no second writer.
//
// ---------------------------------------------------------------------------
// Blocking, and what it costs today
//
// A blocking read waits by polling on the timer tick rather than by sleeping on
// a channel a callback wakes. The reason is a lock mismatch: sleep() wants a
// spinlock, the lwIP lock is a sleeplock (it must be, because a call into lwIP
// can reach the driver and the driver can sleep), and interleaving the two
// correctly is a genuinely error-prone piece of code.
//
// The cost is up to 10 ms of latency on an operation that has to wait -- and
// none at all on one that does not, which is the case that matters for
// throughput: a bulk transfer only blocks when the buffer is genuinely empty.
// Replacing this with a proper sleep/wakeup is worth doing and is deliberately
// not the first version.
#include "types.h"
#include "defs.h"
#include "param.h"
#include "spinlock.h"
#include "sleeplock.h"
#include "proc.h"
#include "fs.h" /* NDIRECT, which file.h's struct inode needs */
#include "file.h"
#define _KERNEL_SOCKET_INTERNALS
#include "socket.h"
#include "errno.h"
#include "lwip/opt.h"
#include "lwip/tcp.h"
#include "lwip/udp.h"
#include "lwip/ip_addr.h"
#include "lwip/pbuf.h"
#define NINET 16 // AF_INET sockets in the system
#define INET_RXBUF 8192 // per socket; one lwIP window plus slack
struct inet_sock {
int inuse;
int type; // SOCK_STREAM or SOCK_DGRAM
struct tcp_pcb *tpcb;
struct udp_pcb *upcb;
// Receive ring. lwIP hands over pbufs from the network thread; a reader
// drains bytes. Full means the sender is told to back off, which for TCP is
// the window closing rather than data being dropped.
char rx[INET_RXBUF];
uint rhead, rtail; // rhead == rtail is empty
int connecting; // connect() issued, no answer yet
int connected;
int conn_failed;
int peer_closed; // FIN seen, or the connection died
int listening;
// Connections lwIP has completed but nobody has accept()ed yet.
struct inet_sock *acceptq[NBACKLOG];
int naccept;
};
static struct inet_sock itab[NINET];
/* ------------------------------------------------------------------ */
/* The receive ring. Callers hold the lwIP lock. */
/* ------------------------------------------------------------------ */
static uint
rx_used(struct inet_sock *is)
{
return is->rhead - is->rtail;
}
static uint
rx_space(struct inet_sock *is)
{
return INET_RXBUF - rx_used(is);
}
static void
rx_put(struct inet_sock *is, const char *p, uint n)
{
for (uint i = 0; i < n; i++)
is->rx[(is->rhead + i) % INET_RXBUF] = p[i];
is->rhead += n;
}
static uint
rx_take(struct inet_sock *is, char *p, uint n)
{
uint have = rx_used(is);
if (n > have)
n = have;
for (uint i = 0; i < n; i++)
p[i] = is->rx[(is->rtail + i) % INET_RXBUF];
is->rtail += n;
return n;
}
/* ------------------------------------------------------------------ */
/* lwIP callbacks. All of these run in the network thread, holding the */
/* lwIP lock. */
/* ------------------------------------------------------------------ */
static void
tcp_fail(struct inet_sock *is)
{
is->peer_closed = 1;
is->conn_failed = 1;
is->connecting = 0;
// The pcb is already freed by lwIP when err_cb fires; keeping the pointer
// would leave a dangling handle that close() would then free a second time.
is->tpcb = 0;
}
static void
inet_err_cb(void *arg, err_t err)
{
(void)err;
if (arg)
tcp_fail((struct inet_sock *)arg);
}
static err_t
inet_recv_cb(void *arg, struct tcp_pcb *pcb, struct pbuf *p, err_t err)
{
struct inet_sock *is = arg;
if (is == 0) {
if (p) pbuf_free(p);
return ERR_OK;
}
if (err != ERR_OK) {
if (p) pbuf_free(p);
return err;
}
// A null pbuf is the peer's FIN, not an error: the connection stays readable
// until the buffer drains, and only then does read() report end of file.
if (p == NULL) {
is->peer_closed = 1;
return ERR_OK;
}
// Refusing what will not fit leaves it queued in lwIP, which closes the
// window and makes the sender wait -- which is the whole point of a window.
// Accepting and dropping would look like a working receive that loses data.
if (p->tot_len > rx_space(is))
return ERR_MEM;
for (struct pbuf *q = p; q != NULL; q = q->next)
rx_put(is, (const char *)q->payload, q->len);
// Only now is the data ours, so only now may the window reopen.
tcp_recved(pcb, p->tot_len);
pbuf_free(p);
return ERR_OK;
}
static err_t
inet_connected_cb(void *arg, struct tcp_pcb *pcb, err_t err)
{
struct inet_sock *is = arg;
(void)pcb;
if (is == 0)
return ERR_OK;
if (err != ERR_OK) {
tcp_fail(is);
return ERR_OK;
}
is->connecting = 0;
is->connected = 1;
return ERR_OK;
}
static struct inet_sock *inet_alloc(int type);
static err_t
inet_accept_cb(void *arg, struct tcp_pcb *newpcb, err_t err)
{
struct inet_sock *ls = arg;
if (ls == 0 || err != ERR_OK)
return ERR_VAL;
if (ls->naccept >= NBACKLOG)
return ERR_MEM; // backlog full: lwIP refuses the connection
struct inet_sock *is = inet_alloc(SOCK_STREAM);
if (is == 0)
return ERR_MEM;
is->tpcb = newpcb;
is->connected = 1;
tcp_arg(newpcb, is);
tcp_recv(newpcb, inet_recv_cb);
tcp_err(newpcb, inet_err_cb);
ls->acceptq[ls->naccept++] = is;
return ERR_OK;
}
// UDP delivers datagrams with no connection. The source is dropped on the
// floor for now: recvfrom is not wired, and inventing a per-datagram source
// queue that nothing reads would be storage with no reader.
static void
inet_udp_recv_cb(void *arg, struct udp_pcb *pcb, struct pbuf *p,
const ip_addr_t *addr, u16_t port)
{
struct inet_sock *is = arg;
(void)pcb; (void)addr; (void)port;
if (is == 0 || p == NULL) {
if (p) pbuf_free(p);
return;
}
if (p->tot_len <= rx_space(is))
for (struct pbuf *q = p; q != NULL; q = q->next)
rx_put(is, (const char *)q->payload, q->len);
pbuf_free(p);
}
/* ------------------------------------------------------------------ */
/* Allocation */
/* ------------------------------------------------------------------ */
static struct inet_sock *
inet_alloc(int type)
{
for (int i = 0; i < NINET; i++)
if (!itab[i].inuse) {
struct inet_sock *is = &itab[i];
memset(is, 0, sizeof(*is));
is->inuse = 1;
is->type = type;
return is;
}
return 0;
}
// Wait one timer tick. See the note at the top on why waiting is a poll.
static void
inet_wait(void)
{
acquire(&tickslock);
uint start = ticks;
while (ticks == start)
sleep(&ticks, &tickslock);
release(&tickslock);
}
/* ------------------------------------------------------------------ */
/* The interface kernel/net/unix.c dispatches to */
/* ------------------------------------------------------------------ */
void *
inet_new(int type)
{
if (type != SOCK_STREAM && type != SOCK_DGRAM)
return 0;
if (!net_is_up())
return 0;
lwip_sys_lock();
struct inet_sock *is = inet_alloc(type);
if (is == 0) {
lwip_sys_unlock();
return 0;
}
if (type == SOCK_STREAM) {
is->tpcb = tcp_new();
if (is->tpcb == 0) {
is->inuse = 0;
lwip_sys_unlock();
return 0;
}
tcp_arg(is->tpcb, is);
tcp_err(is->tpcb, inet_err_cb);
tcp_recv(is->tpcb, inet_recv_cb);
} else {
is->upcb = udp_new();
if (is->upcb == 0) {
is->inuse = 0;
lwip_sys_unlock();
return 0;
}
udp_recv(is->upcb, inet_udp_recv_cb, is);
}
lwip_sys_unlock();
return is;
}
int
inet_bind(void *h, uint32 addr, uint16 port)
{
struct inet_sock *is = h;
ip_addr_t ip;
err_t e;
ip_addr_set_ip4_u32(&ip, addr); // already network byte order
lwip_sys_lock();
e = (is->type == SOCK_STREAM) ? tcp_bind(is->tpcb, &ip, port)
: udp_bind(is->upcb, &ip, port);
lwip_sys_unlock();
if (e == ERR_USE)
return -EADDRINUSE;
return e == ERR_OK ? 0 : -EINVAL;
}
int
inet_listen(void *h, int backlog)
{
struct inet_sock *is = h;
if (is->type != SOCK_STREAM)
return -EOPNOTSUPP;
lwip_sys_lock();
// tcp_listen frees the old pcb and returns a smaller listening one, so the
// handle has to be replaced rather than kept. Losing this is a use-after-free
// that only shows up on the second connection.
struct tcp_pcb *lp = tcp_listen_with_backlog(is->tpcb,
backlog > 0 && backlog < NBACKLOG ? (u8_t)backlog
: NBACKLOG);
if (lp == 0) {
lwip_sys_unlock();
return -ENOMEM;
}
is->tpcb = lp;
is->listening = 1;
tcp_arg(lp, is);
tcp_accept(lp, inet_accept_cb);
lwip_sys_unlock();
return 0;
}
void *
inet_accept(void *h)
{
struct inet_sock *ls = h;
if (!ls->listening)
return 0;
for (;;) {
lwip_sys_lock();
if (ls->naccept > 0) {
struct inet_sock *is = ls->acceptq[0];
for (int i = 1; i < ls->naccept; i++)
ls->acceptq[i - 1] = ls->acceptq[i];
ls->naccept--;
lwip_sys_unlock();
return is;
}
lwip_sys_unlock();
if (myproc()->killed)
return 0;
inet_wait();
}
}
int
inet_connect(void *h, uint32 addr, uint16 port)
{
struct inet_sock *is = h;
ip_addr_t ip;
ip_addr_set_ip4_u32(&ip, addr);
if (is->type == SOCK_DGRAM) {
// A connected UDP socket just remembers where to send; there is no
// handshake to wait for.
lwip_sys_lock();
err_t e = udp_connect(is->upcb, &ip, port);
lwip_sys_unlock();
return e == ERR_OK ? 0 : -EINVAL;
}
lwip_sys_lock();
is->connecting = 1;
err_t e = tcp_connect(is->tpcb, &ip, port, inet_connected_cb);
lwip_sys_unlock();
if (e != ERR_OK) {
is->connecting = 0;
return -ECONNREFUSED;
}
for (;;) {
lwip_sys_lock();
int done = is->connected, failed = is->conn_failed;
lwip_sys_unlock();
if (done)
return 0;
if (failed)
return -ECONNREFUSED;
if (myproc()->killed)
return -EINTR;
inet_wait();
}
}
int
inet_read(void *h, char *buf, int n)
{
struct inet_sock *is = h;
if (n <= 0)
return 0;
for (;;) {
lwip_sys_lock();
uint got = rx_take(is, buf, (uint)n);
int eof = is->peer_closed;
if (got > 0 && is->type == SOCK_STREAM && is->tpcb)
; // window already reopened in the receive callback
lwip_sys_unlock();
if (got > 0)
return (int)got;
if (eof)
return 0; // end of file, not an error
if (myproc()->killed)
return -EINTR;
inet_wait();
}
}
int
inet_write(void *h, char *buf, int n)
{
struct inet_sock *is = h;
int done = 0;
if (n <= 0)
return 0;
if (is->type == SOCK_DGRAM) {
lwip_sys_lock();
struct pbuf *p = pbuf_alloc(PBUF_TRANSPORT, (u16_t)n, PBUF_RAM);
if (p == 0) {
lwip_sys_unlock();
return -ENOBUFS;
}
pbuf_take(p, buf, (u16_t)n);
err_t e = udp_send(is->upcb, p);
pbuf_free(p);
lwip_sys_unlock();
return e == ERR_OK ? n : -EIO;
}
// TCP: the send buffer is finite, so a large write becomes several passes.
while (done < n) {
lwip_sys_lock();
if (is->peer_closed || is->tpcb == 0) {
lwip_sys_unlock();
// Same contract as a pipe with no reader, which is what the AF_UNIX side
// does: the signal first, then the error.
return done > 0 ? done : -EPIPE;
}
int room = tcp_sndbuf(is->tpcb);
int want = n - done;
if (want > room)
want = room;
if (want > 0) {
// TCP_WRITE_FLAG_COPY because `buf` is a kernel bounce buffer the caller
// reuses the moment this returns; lwIP would otherwise hold a pointer
// into memory that has changed by the time the segment goes out.
err_t e = tcp_write(is->tpcb, buf + done, (u16_t)want,
TCP_WRITE_FLAG_COPY);
if (e == ERR_OK) {
tcp_output(is->tpcb);
done += want;
} else if (e != ERR_MEM) {
lwip_sys_unlock();
return done > 0 ? done : -EIO;
}
}
lwip_sys_unlock();
if (done < n) {
if (myproc()->killed)
return done > 0 ? done : -EINTR;
inet_wait(); // wait for the window or the buffer
}
}
return done;
}
int
inet_readable(void *h)
{
struct inet_sock *is = h;
int r;
lwip_sys_lock();
r = rx_used(is) > 0 || is->peer_closed || is->naccept > 0;
lwip_sys_unlock();
return r;
}
int
inet_hup(void *h)
{
struct inet_sock *is = h;
int r;
lwip_sys_lock();
r = is->peer_closed && rx_used(is) == 0;
lwip_sys_unlock();
return r;
}
void
inet_close(void *h)
{
struct inet_sock *is = h;
if (is == 0)
return;
lwip_sys_lock();
// Anything queued but never accepted is closed too, or those connections
// would sit half-open until the peer gave up on them.
for (int i = 0; i < is->naccept; i++) {
struct inet_sock *q = is->acceptq[i];
if (q->tpcb) {
tcp_arg(q->tpcb, 0);
tcp_close(q->tpcb);
}
q->inuse = 0;
}
is->naccept = 0;
if (is->tpcb) {
// Clearing the argument first: lwIP may still deliver a callback for this
// pcb after tcp_close, and it would otherwise arrive with a pointer to a
// socket that has just been freed.
tcp_arg(is->tpcb, 0);
tcp_recv(is->tpcb, 0);
tcp_err(is->tpcb, 0);
if (tcp_close(is->tpcb) != ERR_OK)
tcp_abort(is->tpcb);
is->tpcb = 0;
}
if (is->upcb) {
udp_recv(is->upcb, 0, 0);
udp_remove(is->upcb);
is->upcb = 0;
}
is->inuse = 0;
lwip_sys_unlock();
}

98
kernel/net/lwip/arch/cc.h Normal file
View file

@ -0,0 +1,98 @@
// SPDX-License-Identifier: Apache-2.0
//
// lwIP's compiler/platform header for SchOS.
//
// lwIP includes this before anything else and expects it to supply the fixed
// width types, the byte order, the diagnostic macros and the packing
// attributes. Everything here is either a typedef onto the kernel's own types
// or a one-line macro; there is no logic, by design -- this file is the seam
// between two codebases and the fewer decisions it makes, the better.
#ifndef LWIP_ARCH_CC_H
#define LWIP_ARCH_CC_H
#include "types.h"
#include "defs.h"
// There is no libc under the kernel, so every header lwIP would normally reach
// for has to be switched off and its contents supplied here instead. That is
// the contract arch.h states: set LWIP_NO_STDINT_H and you own the types, set
// LWIP_NO_LIMITS_H and you own the limits, and so on.
#define LWIP_NO_STDINT_H 1
#define LWIP_NO_INTTYPES_H 1
#define LWIP_NO_LIMITS_H 1
#define LWIP_NO_CTYPE_H 1
#define LWIP_NO_UNISTD_H 1
// lwIP's own type names. Written against the kernel's types rather than C99
// spellings, because the kernel already has names for these and introducing a
// second set that must agree is one more thing to get wrong.
typedef uchar u8_t;
typedef char s8_t;
typedef ushort u16_t;
typedef short s16_t;
typedef uint u32_t;
typedef int s32_t;
typedef uint64 u64_t;
typedef int64 s64_t;
#define LWIP_HAVE_INT64 1
typedef unsigned long mem_ptr_t; // must be able to hold a pointer
// ssize_t is the kernel's, not lwIP's. Defining SSIZE_MAX is how arch.h is
// told that: it only declares its own `typedef int ssize_t` when this is
// absent, and that would collide with the kernel's 64-bit one.
#define SSIZE_MAX 0x7fffffffffffffffL
#define INT_MAX 0x7fffffff
// Format specifiers, since <inttypes.h> is switched off above. The kernel's
// cprintf understands %d and %x and nothing wider, which is why these are the
// plain forms rather than the length-qualified ones a hosted printf would use.
#define X8_F "x"
#define U16_F "d"
#define S16_F "d"
#define X16_F "x"
#define U32_F "d"
#define S32_F "d"
#define X32_F "x"
#define SZT_F "d"
// x86 is little-endian. Getting this wrong does not fail to compile; it fails
// on the wire, as every checksum and every port number arriving byte-swapped.
#define BYTE_ORDER LITTLE_ENDIAN
// On-the-wire structures must not be padded. lwIP wraps every such struct in
// these, and a compiler that ignored them would silently move every field in
// every header lwIP parses.
#define PACK_STRUCT_BEGIN
#define PACK_STRUCT_STRUCT __attribute__((packed))
#define PACK_STRUCT_END
#define PACK_STRUCT_FIELD(x) x
// Diagnostics. cprintf is the kernel's console printf.
#define LWIP_PLATFORM_DIAG(x) do { cprintf x; } while (0)
#define LWIP_PLATFORM_ASSERT(x) \
do { cprintf("lwip assert: %s at %s:%d\n", (x), __FILE__, __LINE__); \
panic("lwip assertion"); } while (0)
// lwIP is single-threaded here (NO_SYS=1) but is still entered from two places:
// the network thread and any syscall that touches a socket. Both take the same
// lock, and these are what lwIP itself uses to bracket its critical sections.
// See kernel/net/lwip/sys_arch.c.
void lwip_sys_lock(void);
void lwip_sys_unlock(void);
typedef int sys_prot_t;
#define SYS_ARCH_DECL_PROTECT(lev)
#define SYS_ARCH_PROTECT(lev) lwip_sys_lock()
#define SYS_ARCH_UNPROTECT(lev) lwip_sys_unlock()
// lwIP calls these for its own buffers. The kernel has no malloc, so lwIP is
// configured (MEM_LIBC_MALLOC 0) to use its own pool allocator out of a static
// heap instead -- these exist only because a few optional modules reference
// them, and reaching one is a bug rather than a fallback.
// Spelled with the underlying type rather than lwIP's u32_t: this header is
// included *before* lwIP's arch.h defines u32_t, so naming it here would be a
// forward reference to a typedef that does not exist yet.
#define LWIP_RAND() (lwip_rand_seed())
u32_t lwip_rand_seed(void);
#endif

94
kernel/net/lwip/lwipopts.h Normal file
View file

@ -0,0 +1,94 @@
// SPDX-License-Identifier: Apache-2.0
//
// lwIP configuration for SchOS.
//
// The shape of this file is one decision: NO_SYS=1.
//
// lwIP can run either as a set of threads of its own (NO_SYS=0), which needs a
// port layer of semaphores, mailboxes and thread creation and then hands you a
// blocking sockets API for free; or bare, with no threading assumptions
// (NO_SYS=1), where every call must come from one context and the blocking is
// yours to write. This kernel takes the second, for two reasons: the sys_arch
// a threaded lwIP wants is more code than the socket layer it would save, and
// the kernel already has a socket layer with its own shape (kernel/net/unix.c)
// that AF_INET has to fit into rather than replace.
//
// "One context" is enforced by a single lock rather than by literally one
// thread: kernel/net/net.c takes it around the network thread's work, and the
// socket syscalls take it around theirs. That is the entire concurrency model.
#ifndef LWIP_LWIPOPTS_H
#define LWIP_LWIPOPTS_H
#define NO_SYS 1
#define SYS_LIGHTWEIGHT_PROT 1
#define LWIP_NETCONN 0 // needs NO_SYS=0
#define LWIP_SOCKET 0 // ours lives in kernel/net/inet.c
// Memory. All of it is static: the kernel's allocator hands out whole pages
// and nothing else, so lwIP's own pools are a better fit than wrapping kalloc.
// These numbers are sized for a handful of connections, not for a server.
#define MEM_LIBC_MALLOC 0
#define MEMP_MEM_MALLOC 0
#define MEM_ALIGNMENT 4
#define MEM_SIZE (192 * 1024)
#define MEMP_NUM_PBUF 32
#define MEMP_NUM_UDP_PCB 8
#define MEMP_NUM_TCP_PCB 16
#define MEMP_NUM_TCP_PCB_LISTEN 8
#define MEMP_NUM_TCP_SEG 64
#define MEMP_NUM_ARP_QUEUE 16
#define MEMP_NUM_SYS_TIMEOUT 16
#define PBUF_POOL_SIZE 48
#define PBUF_POOL_BUFSIZE 1536 // one Ethernet frame plus room
// Protocols. IPv4 only: the driver is proven for it, the QEMU test network is
// v4, and every line of v6 here would be untested.
#define LWIP_IPV4 1
#define LWIP_IPV6 0
#define LWIP_ARP 1
#define LWIP_ETHERNET 1
#define LWIP_ICMP 1
#define LWIP_RAW 1
#define LWIP_UDP 1
#define LWIP_TCP 1
#define LWIP_DHCP 1
#define LWIP_DNS 1
#define LWIP_AUTOIP 0
#define LWIP_IGMP 0
#define LWIP_NETIF_HOSTNAME 1
// DHCP needs UDP checksums and a broadcast path; both are on. Requesting an
// address rather than hard-coding one matters because QEMU's user-mode network
// hands out 10.0.2.15 through DHCP and nothing else will route.
#define LWIP_NETIF_STATUS_CALLBACK 1
#define LWIP_NETIF_LINK_CALLBACK 1
#define TCP_MSS 1460
#define TCP_SND_BUF (8 * TCP_MSS)
#define TCP_WND (8 * TCP_MSS)
#define TCP_SND_QUEUELEN ((4 * TCP_SND_BUF) / TCP_MSS)
#define TCP_LISTEN_BACKLOG 1
// Checksums in software. The e1000 can offload them, but the driver does not
// ask it to, and a checksum computed by a device the stack has not been told
// about is the sort of mismatch that shows up only against a real peer.
#define CHECKSUM_GEN_IP 1
#define CHECKSUM_GEN_UDP 1
#define CHECKSUM_GEN_TCP 1
#define CHECKSUM_CHECK_IP 1
#define CHECKSUM_CHECK_UDP 1
#define CHECKSUM_CHECK_TCP 1
// Statistics are what /proc/net reports; they cost a few counters.
#define LWIP_STATS 1
#define LWIP_STATS_DISPLAY 0
// Debug output off by default -- turning LWIP_DEBUG on here makes the stack
// extremely chatty on the console, which is occasionally exactly what you want.
#define LWIP_DEBUG 0
// lwIP would otherwise provide its own errno, which would then disagree with
// the kernel's for every value a syscall returns.
#define LWIP_PROVIDE_ERRNO 1
#endif

137
kernel/net/lwip/netif_e1000.c Normal file
View file

@ -0,0 +1,137 @@
// SPDX-License-Identifier: Apache-2.0
//
// The lwIP network interface backed by the e1000 driver.
//
// Two directions and almost nothing else: a linkoutput that flattens a pbuf
// chain into a frame and hands it to the driver, and a poll that takes frames
// off the driver and hands them up. Everything about IP, ARP and TCP lives in
// lwIP; everything about descriptors and registers lives in the driver; this
// file is the seam.
#include "types.h"
#include "defs.h"
#include "param.h"
#include "lwip/opt.h"
#include "lwip/netif.h"
#include "lwip/pbuf.h"
#include "lwip/etharp.h"
#include "lwip/ethip6.h"
#include "netif/ethernet.h"
#define ETH_MTU 1500
#define ETH_FRAMESZ 1518 // 1500 payload + 14 header + 4 FCS allowance
static struct netif e1000_netif;
// One staging buffer, not one per call: a frame is 1518 bytes and the kernel
// stack is 4096 in total, so a local of this size on a path that is several
// frames deep is exactly the overflow the kstack canary exists to catch.
// Safe as a single static because every caller holds the lwIP lock.
static char txframe[ETH_FRAMESZ];
static char rxframe[ETH_FRAMESZ];
// Send one packet. lwIP hands over a pbuf chain; the driver wants one flat
// buffer, so this walks the chain into txframe.
static err_t
e1000_linkoutput(struct netif *netif, struct pbuf *p)
{
(void)netif;
if (p->tot_len > sizeof(txframe))
return ERR_IF;
u16_t off = 0;
for (struct pbuf *q = p; q != NULL; q = q->next) {
if (off + q->len > sizeof(txframe))
return ERR_IF;
memmove(txframe + off, q->payload, q->len);
off += q->len;
}
// A full transmit ring is congestion, not failure. Reporting ERR_IF would
// make lwIP drop the segment and wait for a retransmission timer; ERR_WOULDBLOCK
// lets it try again, which for TCP is the difference between a stall of
// milliseconds and one of seconds.
if (e1000_transmit(txframe, off) != 0)
return ERR_WOULDBLOCK;
return ERR_OK;
}
// Bring the interface up. Called once, by the network thread.
static err_t
e1000_netif_init(struct netif *netif)
{
const unsigned char *mac = e1000_mac();
netif->name[0] = 'e';
netif->name[1] = 'n';
netif->output = etharp_output; // IP -> ARP -> linkoutput
netif->linkoutput = e1000_linkoutput;
netif->mtu = ETH_MTU;
netif->hwaddr_len = 6;
for (int i = 0; i < 6; i++)
netif->hwaddr[i] = mac[i];
netif->flags = NETIF_FLAG_BROADCAST | NETIF_FLAG_ETHARP | NETIF_FLAG_LINK_UP;
return ERR_OK;
}
struct netif *
net_e1000_netif(void)
{
return &e1000_netif;
}
// Attach the interface with no address yet -- DHCP fills it in.
int
net_e1000_attach(void)
{
ip4_addr_t any;
IP4_ADDR(&any, 0, 0, 0, 0);
if (netif_add(&e1000_netif, &any, &any, &any, NULL,
e1000_netif_init, netif_input) == NULL)
return -1;
netif_set_default(&e1000_netif);
netif_set_up(&e1000_netif);
netif_set_link_up(&e1000_netif);
return 0;
}
// Take whatever the driver has and push it into lwIP. Returns how many frames
// moved, so the caller can tell a busy poll from an idle one.
//
// Bounded rather than draining: a receive loop with no limit starves everything
// else on a busy network, and the caller comes straight back anyway.
int
net_e1000_poll(void)
{
int moved = 0;
for (int i = 0; i < 32; i++) {
int len = e1000_receive(rxframe, sizeof(rxframe));
if (len <= 0)
break;
// PBUF_POOL, so this comes from lwIP's own pool rather than a heap it
// does not have. A null return means the pool is exhausted, which is a
// dropped packet -- entirely normal under load, and what the protocol
// above is designed to survive.
struct pbuf *p = pbuf_alloc(PBUF_RAW, (u16_t)len, PBUF_POOL);
if (p == NULL)
break;
pbuf_take(p, rxframe, (u16_t)len);
// netif->input is netif_input, set at netif_add: it dispatches on the
// Ethernet type. Ownership of the pbuf passes with the call, including
// when it fails, so there is nothing to free here.
if (e1000_netif.input(p, &e1000_netif) != ERR_OK)
pbuf_free(p);
moved++;
}
return moved;
}

11
kernel/net/lwip/shim/ctype.h Normal file
View file

@ -0,0 +1,11 @@
// SPDX-License-Identifier: Apache-2.0
// <ctype.h> for lwIP, shadowing the host's.
//
// Empty on purpose. Everything lwIP would take from this header is supplied by
// arch/cc.h instead -- see the LWIP_NO_*_H switches there -- and the file exists
// only so the #include resolves to something harmless rather than to the host's
// copy, which declares types the kernel already has and pulls half of glibc in
// behind them. See shim/string.h for the full reasoning.
#ifndef LWIP_SHIM_ctype_H
#define LWIP_SHIM_ctype_H
#endif

11
kernel/net/lwip/shim/errno.h Normal file
View file

@ -0,0 +1,11 @@
// SPDX-License-Identifier: Apache-2.0
// <errno.h> for lwIP, shadowing the host's.
//
// Empty on purpose. Everything lwIP would take from this header is supplied by
// arch/cc.h instead -- see the LWIP_NO_*_H switches there -- and the file exists
// only so the #include resolves to something harmless rather than to the host's
// copy, which declares types the kernel already has and pulls half of glibc in
// behind them. See shim/string.h for the full reasoning.
#ifndef LWIP_SHIM_errno_H
#define LWIP_SHIM_errno_H
#endif

11
kernel/net/lwip/shim/inttypes.h Normal file
View file

@ -0,0 +1,11 @@
// SPDX-License-Identifier: Apache-2.0
// <inttypes.h> for lwIP, shadowing the host's.
//
// Empty on purpose. Everything lwIP would take from this header is supplied by
// arch/cc.h instead -- see the LWIP_NO_*_H switches there -- and the file exists
// only so the #include resolves to something harmless rather than to the host's
// copy, which declares types the kernel already has and pulls half of glibc in
// behind them. See shim/string.h for the full reasoning.
#ifndef LWIP_SHIM_inttypes_H
#define LWIP_SHIM_inttypes_H
#endif

11
kernel/net/lwip/shim/limits.h Normal file
View file

@ -0,0 +1,11 @@
// SPDX-License-Identifier: Apache-2.0
// <limits.h> for lwIP, shadowing the host's.
//
// Empty on purpose. Everything lwIP would take from this header is supplied by
// arch/cc.h instead -- see the LWIP_NO_*_H switches there -- and the file exists
// only so the #include resolves to something harmless rather than to the host's
// copy, which declares types the kernel already has and pulls half of glibc in
// behind them. See shim/string.h for the full reasoning.
#ifndef LWIP_SHIM_limits_H
#define LWIP_SHIM_limits_H
#endif

11
kernel/net/lwip/shim/stddef.h Normal file
View file

@ -0,0 +1,11 @@
// SPDX-License-Identifier: Apache-2.0
// <stddef.h> for lwIP, shadowing the host's.
//
// Empty on purpose. Everything lwIP would take from this header is supplied by
// arch/cc.h instead -- see the LWIP_NO_*_H switches there -- and the file exists
// only so the #include resolves to something harmless rather than to the host's
// copy, which declares types the kernel already has and pulls half of glibc in
// behind them. See shim/string.h for the full reasoning.
#ifndef LWIP_SHIM_stddef_H
#define LWIP_SHIM_stddef_H
#endif

11
kernel/net/lwip/shim/stdint.h Normal file
View file

@ -0,0 +1,11 @@
// SPDX-License-Identifier: Apache-2.0
// <stdint.h> for lwIP, shadowing the host's.
//
// Empty on purpose. Everything lwIP would take from this header is supplied by
// arch/cc.h instead -- see the LWIP_NO_*_H switches there -- and the file exists
// only so the #include resolves to something harmless rather than to the host's
// copy, which declares types the kernel already has and pulls half of glibc in
// behind them. See shim/string.h for the full reasoning.
#ifndef LWIP_SHIM_stdint_H
#define LWIP_SHIM_stdint_H
#endif

11
kernel/net/lwip/shim/stdio.h Normal file
View file

@ -0,0 +1,11 @@
// SPDX-License-Identifier: Apache-2.0
// <stdio.h> for lwIP, shadowing the host's.
//
// Empty on purpose. Everything lwIP would take from this header is supplied by
// arch/cc.h instead -- see the LWIP_NO_*_H switches there -- and the file exists
// only so the #include resolves to something harmless rather than to the host's
// copy, which declares types the kernel already has and pulls half of glibc in
// behind them. See shim/string.h for the full reasoning.
#ifndef LWIP_SHIM_stdio_H
#define LWIP_SHIM_stdio_H
#endif

28
kernel/net/lwip/shim/stdlib.h Normal file
View file

@ -0,0 +1,28 @@
// SPDX-License-Identifier: Apache-2.0
//
// <stdlib.h> for lwIP, shadowing the host's. See shim/string.h for why these
// files exist at all.
//
// lwIP does not need an allocator from here: MEM_LIBC_MALLOC is off, so it uses
// its own pools out of a static heap. That is not a workaround -- the kernel's
// allocator hands out whole pages and lwIP wants byte granularity, so its pools
// are the better fit either way.
//
// atoi is the one thing it genuinely calls, in netif_name_to_index, to turn the
// digits at the end of an interface name into a number.
#ifndef LWIP_SHIM_STDLIB_H
#define LWIP_SHIM_STDLIB_H
// Deliberately not a general atoi: no sign, no whitespace, no overflow
// handling. It is fed the tail of an interface name ("en0" -> "0"), and
// anything cleverer would be untested code answering a question nobody asks.
static inline int
atoi(const char *s)
{
int n = 0;
while (*s >= '0' && *s <= '9')
n = n * 10 + (*s++ - '0');
return n;
}
#endif

37
kernel/net/lwip/shim/string.h Normal file
View file

@ -0,0 +1,37 @@
// SPDX-License-Identifier: Apache-2.0
//
// <string.h> for lwIP, shadowing the host's.
//
// lwIP includes <string.h> in thirty-odd files. The kernel is built
// -ffreestanding and has its own string functions, but the compiler's default
// include path still finds /usr/include/string.h -- which declares strlen
// returning size_t against a kernel that returns int, and drags sigset_t and
// __float128 in behind it. So the lwIP include path puts this directory first.
//
// The prototypes are repeated here rather than taken from defs.h on purpose.
// lwIP includes <string.h> at the very top of a translation unit, before any
// kernel header has been seen, so pulling defs.h in from here would drop the
// whole kernel's declarations into a context where its own types are not
// defined yet -- which shows up as errors in unrelated headers, like file.h
// not knowing what NDIRECT is.
//
// They must match kernel/core/string.c exactly. Only what lwIP calls is here;
// anything else should be added deliberately rather than by growing this into
// a libc.
#ifndef LWIP_SHIM_STRING_H
#define LWIP_SHIM_STRING_H
int memcmp(const void *, const void *, unsigned long);
void *memmove(void *, const void *, unsigned long);
void *memset(void *, int, unsigned long);
int strlen(const char *);
int strncmp(const char *, const char *, unsigned long);
char *strncpy(char *, const char *, int);
// The kernel deliberately provides only memmove, on the grounds that a copy
// which happens to overlap should not be undefined behaviour. lwIP calls
// memcpy in a handful of places, always on buffers it has just allocated, so
// mapping it onto memmove is correct and costs nothing.
#define memcpy(d, s, n) memmove((d), (s), (n))
#endif

11
kernel/net/lwip/shim/unistd.h Normal file
View file

@ -0,0 +1,11 @@
// SPDX-License-Identifier: Apache-2.0
// <unistd.h> for lwIP, shadowing the host's.
//
// Empty on purpose. Everything lwIP would take from this header is supplied by
// arch/cc.h instead -- see the LWIP_NO_*_H switches there -- and the file exists
// only so the #include resolves to something harmless rather than to the host's
// copy, which declares types the kernel already has and pulls half of glibc in
// behind them. See shim/string.h for the full reasoning.
#ifndef LWIP_SHIM_unistd_H
#define LWIP_SHIM_unistd_H
#endif

121
kernel/net/lwip/sys_arch.c Normal file
View file

@ -0,0 +1,121 @@
// SPDX-License-Identifier: Apache-2.0
//
// The lwIP port layer: everything lwIP needs from the system it runs on.
//
// With NO_SYS=1 that is a short list -- a clock, a source of randomness, and a
// lock -- because lwIP is not creating threads or mailboxes of its own. The
// lock is the interesting one.
//
// ---------------------------------------------------------------------------
// The concurrency model, in full
//
// lwIP with NO_SYS=1 is not re-entrant. It is entered from exactly two kinds of
// place here: the network thread (kernel/net/net.c), which feeds it received
// frames and runs its timers, and the socket syscalls, which call into it on
// behalf of a process. Those can run on different CPUs at the same time, so
// something has to serialise them, and lwIP's own SYS_ARCH_PROTECT hooks are
// the sanctioned place to put it.
//
// A sleeplock rather than a spinlock, because a call into lwIP can reach the
// driver and the driver can sleep. That makes this lock unusable from an
// interrupt handler -- which is why the e1000 interrupt does not feed lwIP
// directly and the network thread polls instead.
//
// It is recursive by necessity: lwIP brackets small critical sections inside
// larger ones and expects nesting to be free.
#include "types.h"
#include "defs.h"
#include "param.h"
#include "spinlock.h"
#include "sleeplock.h"
#include "proc.h"
#include "lwip/opt.h"
#include "lwip/sys.h"
static struct sleeplock lwip_lock;
static struct proc *lwip_owner;
static int lwip_depth;
static int lwip_lock_ready;
void
lwip_sys_init(void)
{
initsleeplock(&lwip_lock, "lwip");
lwip_owner = 0;
lwip_depth = 0;
lwip_lock_ready = 1;
}
// Reading `lwip_owner` without holding anything is safe in the one direction it
// is asked: only the process inside the lock ever writes its own proc pointer
// there, and it holds the lock the whole time. No other process can see its own
// pointer here, so a stale read cannot produce a false "I hold it". The null
// check matters because myproc() is 0 in early boot, and 0 == 0 would otherwise
// claim the lock for a process that does not exist.
static int
lwip_lock_mine(void)
{
return lwip_owner != 0 && lwip_owner == myproc();
}
void
lwip_sys_lock(void)
{
// Before lwip_sys_init runs there is exactly one context in existence, so
// there is nothing to serialise against and the sleeplock is not yet valid.
if (!lwip_lock_ready)
return;
if (lwip_lock_mine()) {
lwip_depth++;
return;
}
acquiresleep(&lwip_lock);
lwip_owner = myproc();
lwip_depth = 1;
}
void
lwip_sys_unlock(void)
{
if (!lwip_lock_ready)
return;
if (!lwip_lock_mine())
return; // not ours; releasing would free someone else's
if (--lwip_depth > 0)
return;
lwip_owner = 0;
releasesleep(&lwip_lock);
}
// Milliseconds since boot. lwIP drives every timer off this: retransmission,
// TIME_WAIT, ARP expiry, DHCP renewal. It need not match wall time, but it must
// advance monotonically -- a clock that goes backwards makes lwIP believe every
// timer is due at once.
//
// `ticks` is the timer interrupt count at 100 Hz, so one tick is 10 ms. Taken
// without the lock: it is a single aligned word, and a torn read would cost
// lwIP one early or late timer rather than anything worse.
u32_t
sys_now(void)
{
return (u32_t)ticks * 10u;
}
// lwIP wants randomness for initial TCP sequence numbers and for the DHCP
// transaction id. Sequence numbers that are predictable are a real weakness on
// a network with anyone hostile on it; this is not a CSPRNG and does not claim
// to be, but it is seeded from a clock that an outsider cannot observe.
u32_t
lwip_rand_seed(void)
{
static u32_t s;
if (s == 0)
s = (u32_t)ticks ^ 0x9E3779B9u;
// xorshift32
s ^= s << 13;
s ^= s >> 17;
s ^= s << 5;
return s;
}

130
kernel/net/net.c Normal file
View file

@ -0,0 +1,130 @@
// SPDX-License-Identifier: Apache-2.0
//
// The network thread, and the one place the TCP/IP stack is started.
//
// lwIP with NO_SYS=1 has no threads of its own, so something has to feed it
// received frames and run its timers. That something is a kernel thread rather
// than the e1000 interrupt handler, and the reason is the lock: a call into
// lwIP can reach the driver, the driver can sleep, and an interrupt handler
// cannot. So the interrupt keeps doing what it always did -- clear the cause
// register -- and this thread polls.
//
// Polling costs a wakeup every few milliseconds when the link is idle. The
// alternative, waking this thread from the interrupt, is a worthwhile change
// later; it is deliberately not the first version, because a receive path that
// is woken from an interrupt is much harder to debug than one that is asked.
#include "types.h"
#include "defs.h"
#include "param.h"
#include "spinlock.h"
#include "sleeplock.h"
#include "proc.h"
#include "lwip/opt.h"
#include "lwip/init.h"
#include "lwip/timeouts.h"
#include "lwip/netif.h"
#include "lwip/dhcp.h"
#include "lwip/ip4_addr.h"
// Wait until the next timer tick. `ticks` advances at 100 Hz, so this is a
// sleep of up to 10 ms -- and, more importantly, it is the kernel's only timed
// sleep: there is no higher-resolution timer to wait on. The lwIP lock must not
// be held across it, or nothing else could touch the stack while the link is
// idle, which is most of the time.
static void
wait_tick(void)
{
acquire(&tickslock);
uint start = ticks;
while (ticks == start)
sleep(&ticks, &tickslock);
release(&tickslock);
}
static int net_started;
// Set once DHCP has bound, so `ifconfig`-ish output and /proc/net have
// something true to say rather than reporting 0.0.0.0 forever.
static int net_bound;
int
net_is_up(void)
{
return net_bound;
}
static void
net_thread(void)
{
// lwIP's own initialisation has to happen in this thread, not in main():
// it allocates from lwIP's pools and registers timers, and doing that
// before there is a process context means doing it somewhere that cannot
// sleep if the allocator ever needs to.
lwip_sys_init();
lwip_sys_lock();
lwip_init();
if (net_e1000_attach() != 0) {
lwip_sys_unlock();
cprintf("net: no e1000; the network thread has nothing to drive\n");
// Not a panic and not an exit: a kernel thread must never return, and
// a machine with no NIC is an ordinary machine.
for (;;)
wait_tick();
}
struct netif *nif = net_e1000_netif();
dhcp_start(nif);
lwip_sys_unlock();
cprintf("net: lwIP %d.%d.%d up, requesting an address\n",
LWIP_VERSION_MAJOR, LWIP_VERSION_MINOR, LWIP_VERSION_REVISION);
int announced = 0;
for (;;) {
int moved;
lwip_sys_lock();
moved = net_e1000_poll();
sys_check_timeouts();
// Report the address once, when it first arrives. Anyone who has spent
// an evening wondering whether the guest has an address at all wants
// this on the console rather than in a file they have to know about.
if (!announced && !ip4_addr_isany_val(*netif_ip4_addr(nif))) {
const ip4_addr_t *a = netif_ip4_addr(nif);
const ip4_addr_t *m = netif_ip4_netmask(nif);
const ip4_addr_t *gw = netif_ip4_gw(nif);
cprintf("net: %d.%d.%d.%d/%d.%d.%d.%d via %d.%d.%d.%d\n",
ip4_addr1(a), ip4_addr2(a), ip4_addr3(a), ip4_addr4(a),
ip4_addr1(m), ip4_addr2(m), ip4_addr3(m), ip4_addr4(m),
ip4_addr1(gw), ip4_addr2(gw), ip4_addr3(gw), ip4_addr4(gw));
announced = 1;
net_bound = 1;
}
lwip_sys_unlock();
// Busy while there is traffic, idle otherwise. Sleeping even 1 ms with
// frames waiting would cap throughput at one burst per tick.
if (moved == 0)
wait_tick();
}
}
// Start the stack. Called from main() once processes exist.
void
netinit(void)
{
if (net_started)
return;
if (!e1000_present()) {
// Silent: a machine without an e1000 is not misconfigured, and this
// runs on every boot.
return;
}
net_started = 1;
if (kthread_create(net_thread, "netd") < 0)
cprintf("net: could not start the network thread\n");
}

117
kernel/net/unix.c
View file

@ -108,6 +108,21 @@ sockclose(struct socket *s)
if(s == 0)
return;
// An AF_INET socket's state lives in inet.c behind the lwIP lock, so its
// teardown happens there. The table entry below is freed either way.
if(s->domain == AF_INET){
acquire(&stab.lock);
int last = (--s->ref <= 0);
if(last)
s->state = SS_FREE;
release(&stab.lock);
if(last && s->inet){
inet_close(s->inet);
s->inet = 0;
}
return;
}
acquire(&stab.lock);
if(--s->ref > 0){
release(&stab.lock);
@ -161,6 +176,8 @@ sockread(struct socket *s, char *addr, int n)
if(n < 0)
return -EINVAL;
if(s->domain == AF_INET)
return inet_read(s->inet, addr, n);
acquire(&stab.lock);
while(s->nread == s->nwrite && !s->peer_closed){
@ -192,6 +209,8 @@ sockwrite(struct socket *s, char *addr, int n)
if(n < 0)
return -EINVAL;
if(s->domain == AF_INET)
return inet_write(s->inet, addr, n);
acquire(&stab.lock);
@ -239,14 +258,30 @@ socknew(int domain, int type)
{
struct socket *s;
if(domain != AF_UNIX)
if(domain != AF_UNIX && domain != AF_INET)
return 0;
if(type != SOCK_STREAM)
// AF_UNIX is stream-only here; AF_INET has datagrams too, and inet_new is
// what decides whether it can serve the type asked for.
if(domain == AF_UNIX && type != SOCK_STREAM)
return 0;
acquire(&stab.lock);
s = alloc_locked(type);
if(s)
s->domain = domain;
release(&stab.lock);
if(s == 0)
return 0;
if(domain == AF_INET){
// Outside the table lock: inet_new takes the lwIP lock, which is a
// sleeplock, and sleeping while holding a spinlock is a panic.
s->inet = inet_new(type);
if(s->inet == 0){
sockclose(s);
return 0;
}
}
return s;
}
@ -305,6 +340,9 @@ sockbind(struct socket *s, const char *path)
int
socklisten(struct socket *s, int backlog)
{
if(s->domain == AF_INET)
return inet_listen(s->inet, backlog);
(void)backlog; // the queue is a fixed NBACKLOG either way
acquire(&stab.lock);
@ -325,6 +363,28 @@ sockaccept(struct socket *s)
struct socket *c;
int i;
if(s->domain == AF_INET){
// The connection lwIP hands back needs a table entry of its own, because
// everything above this layer works in `struct socket` and knows nothing
// about lwIP.
void *h = inet_accept(s->inet);
if(h == 0)
return 0;
acquire(&stab.lock);
c = alloc_locked(s->type);
if(c){
c->domain = AF_INET;
c->state = SS_CONNECTED;
}
release(&stab.lock);
if(c == 0){
inet_close(h);
return 0;
}
c->inet = h;
return c;
}
acquire(&stab.lock);
if(s->state != SS_LISTENING){
release(&stab.lock);
@ -426,6 +486,8 @@ sock_shutdown(struct socket *s, int how)
int
sockreadable(struct socket *s)
{
if(s->domain == AF_INET)
return inet_readable(s->inet);
if(s->state == SS_LISTENING)
return s->nqueue > 0;
return s->nread != s->nwrite || s->peer_closed;
@ -435,7 +497,58 @@ sockreadable(struct socket *s)
int
sock_hup(struct socket *s)
{
if(s->domain == AF_INET)
return inet_hup(s->inet);
if(s->state == SS_LISTENING)
return 0;
return s->peer == 0 && s->nread == s->nwrite;
}
/* ------------------------------------------------------------------ */
/* AF_INET addressing */
/* */
/* Separate entry points rather than an overloaded sockbind, because */
/* the two families do not name the same kind of thing: one takes a */
/* path, the other an address and a port. The syscall layer picks by */
/* the family in the sockaddr the caller passed. */
/* ------------------------------------------------------------------ */
int
sockbind_in(struct socket *s, uint addr, ushort port)
{
if(s->domain != AF_INET)
return -EAFNOSUPPORT;
int r = inet_bind(s->inet, addr, port);
if(r == 0){
acquire(&stab.lock);
s->state = SS_BOUND;
release(&stab.lock);
}
return r;
}
int
sockconnect_in(struct socket *s, uint addr, ushort port)
{
if(s->domain != AF_INET)
return -EAFNOSUPPORT;
int r = inet_connect(s->inet, addr, port);
if(r == 0){
acquire(&stab.lock);
s->state = SS_CONNECTED;
release(&stab.lock);
}
return r;
}
// Which family a socket belongs to. An accessor because kernel/sys/sys_net.c is
// the translation layer and deliberately does not see inside struct socket --
// it works in descriptors and user addresses, not in the socket's guts.
int
sock_domain(struct socket *s)
{
return s ? s->domain : 0;
}

51
kernel/proc/proc.c
View file

@ -118,6 +118,7 @@ found:
p->syscall_map = 0;
p->map_size = 0;
p->mmap_next = 0;
p->kfn = 0;
memset(p->ofile_flags, 0, sizeof(p->ofile_flags));
@ -459,6 +460,19 @@ forkret(void) {
static int first = 1;
release(&ptable.lock);
// A kernel thread starts here and never leaves. It shares the scheduler's
// rotation with everything else, so it can hold sleeplocks and block on the
// disk exactly as a syscall would -- which is the whole reason it is a
// process rather than a timer callback.
//
// Ahead of the first-boot block below on purpose: a kernel thread is only
// ever created long after the root filesystem is mounted, so it must not be
// the one that runs that setup.
if (myproc()->kfn) {
myproc()->kfn();
panic("kthread returned");
}
if (first) {
first = 0;
iinit();
@ -500,6 +514,43 @@ forkret(void) {
}
}
// Start a kernel thread: a process that runs `fn` in kernel mode and never
// returns to userland.
//
// It gets a page directory of its own with only the kernel mapped, because the
// scheduler calls switchuvm() for every process it runs and that panics on a
// null pgdir. Sharing kpgdir directly would work equally well; a private one
// costs a page and keeps the "every proc owns its pgdir" invariant intact.
//
// No files, no cwd, no parent: it is not in the process tree, cannot be waited
// for and must never exit. Returns the pid, or -1.
int
kthread_create(void (*fn)(void), const char *name)
{
struct proc *p;
if ((p = allocproc()) == 0)
return -1;
if ((p->pgdir = setupkvm()) == 0) {
kfree(p->kstack);
p->kstack = 0;
p->state = UNUSED;
return -1;
}
p->sz = PGSIZE; // never used; 0 upsets code that treats it as a size
p->parent = 0;
p->cwd = 0;
p->kfn = fn;
safestrcpy(p->name, name, sizeof(p->name));
acquire(&ptable.lock);
p->state = RUNNABLE;
release(&ptable.lock);
return p->pid;
}
void
sleep(void *chan, struct spinlock *lk)
{

60
kernel/sys/sys_net.c
View file

@ -4,10 +4,12 @@
// between a syscall's user-supplied `struct sockaddr` and the path string the
// socket layer actually wants, plus the file-descriptor bookkeeping.
//
// Only AF_UNIX is answered. AF_INET is recognised and refused with
// Two families now. AF_UNIX is answered by kernel/net/unix.c, AF_INET by
// kernel/net/inet.c on top of lwIP; the socket layer dispatches on the domain
// the socket was created with, so this file's only extra job is to read the
// right shape of sockaddr out of user memory. AF_INET6 is still refused with
// EAFNOSUPPORT rather than EINVAL, because a program that probes for IPv6 and
// falls back to IPv4 needs to be able to tell "this address family does not
// exist here" from "you passed nonsense".
// falls back to IPv4 needs to tell "not here" from "you passed nonsense".
#include "types.h"
#include "defs.h"
@ -112,14 +114,16 @@ SYSCALL_DEFINE(socket)
argint(1, &type);
argint(2, &protocol);
if (domain != AF_UNIX)
if (domain != AF_UNIX && domain != AF_INET)
return -EAFNOSUPPORT;
// SOCK_CLOEXEC / SOCK_NONBLOCK ride in the high bits of `type`. Neither
// is honoured yet, but masking them off means a musl program that always
// sets SOCK_CLOEXEC still gets a socket instead of EINVAL.
type &= 0xFF;
if (type != SOCK_STREAM)
if (domain == AF_UNIX && type != SOCK_STREAM)
return -EPROTOTYPE;
if (domain == AF_INET && type != SOCK_STREAM && type != SOCK_DGRAM)
return -EPROTOTYPE;
if (protocol != 0)
return -EPROTONOSUPPORT;
@ -134,6 +138,36 @@ SYSCALL_DEFINE(socket)
return fd;
}
// Pull an address and port out of a user `struct sockaddr_in`.
//
// Both come back in network byte order and stay that way: lwIP wants them that
// way, so converting here and back again would be two swaps that cancel.
static int
fetch_sin(addr_t uaddr, int ulen, uint *addr, ushort *port)
{
struct sockaddr_in sa;
if (ulen < (int)sizeof(sa))
return -EINVAL;
if (copy_from_user(&sa, uaddr, sizeof(sa)) < 0)
return -EFAULT;
if (sa.sin_family != AF_INET)
return -EAFNOSUPPORT;
*addr = sa.sin_addr;
*port = sa.sin_port;
return 0;
}
// Network byte order to host, for a port. The kernel is little-endian and the
// wire is big-endian; lwIP's own ntohs is not reachable from here without
// dragging its headers into the syscall layer.
static ushort
port_from_net(ushort p)
{
return (ushort)(((p & 0xFF) << 8) | ((p >> 8) & 0xFF));
}
SYSCALL_DEFINE(bind)
{
struct socket *s;
@ -146,6 +180,14 @@ SYSCALL_DEFINE(bind)
argaddr(1, &uaddr);
argint(2, &ulen);
if (sock_domain(s) == AF_INET) {
uint ip;
ushort port;
if ((r = fetch_sin(uaddr, ulen, &ip, &port)) < 0)
return r;
return sockbind_in(s, ip, port_from_net(port));
}
if ((r = fetch_sun_path(uaddr, ulen, path)) < 0)
return r;
return sockbind(s, path);
@ -174,6 +216,14 @@ SYSCALL_DEFINE(connect)
argaddr(1, &uaddr);
argint(2, &ulen);
if (sock_domain(s) == AF_INET) {
uint ip;
ushort port;
if ((r = fetch_sin(uaddr, ulen, &ip, &port)) < 0)
return r;
return sockconnect_in(s, ip, port_from_net(port));
}
if ((r = fetch_sun_path(uaddr, ulen, path)) < 0)
return r;
return sockconnect(s, path);