#include "u.h" #include "../port/lib.h" #include "mem.h" #include "dat.h" #include "fns.h" #include "../port/error.h" struct Mntrpc { Chan* c; /* Channel for whom we are working */ Mntrpc* list; /* Free/pending list */ Fcall request; /* Outgoing file system protocol message */ Fcall reply; /* Incoming reply */ Mnt* m; /* Mount device during rpc */ Rendez r; /* Place to hang out */ char* rpc; /* I/O Data buffer */ char done; /* Rpc completed */ char flushed; /* Flush was sent */ ushort flushtag; /* Tag flush sent on */ char flush[MAXMSG]; /* Somewhere to build flush */ uvlong stime; /* start time for mnt statistics */ ulong reqlen; /* request length for mnt statistics */ ulong replen; /* reply length for mnt statistics */ }; struct Mntalloc { Lock; Mnt* list; /* Mount devices in use */ Mnt* mntfree; /* Free list */ Mntrpc* rpcfree; int nrpcfree; int nrpcused; ulong id; int rpctag; }mntalloc; #define MAXRPC (16*1024+MAXMSG) #define limit(n, max) (n > max ? max : n) void mattach(Mnt*, Chan*, char*); void mntauth(Mnt*, Mntrpc*, char*, ushort); Mnt* mntchk(Chan*); void mntdirfix(uchar*, Chan*); int mntflush(Mnt*, Mntrpc*); void mntfree(Mntrpc*); void mntgate(Mnt*); void mntpntfree(Mnt*); void mntqrm(Mnt*, Mntrpc*); Mntrpc* mntralloc(Chan*); long mntrdwr(int, Chan*, void*, long, vlong); long mnt9prdwr(int, Chan*, void*, long, vlong); void mntrpcread(Mnt*, Mntrpc*); void mountio(Mnt*, Mntrpc*); void mountmux(Mnt*, Mntrpc*); void mountrpc(Mnt*, Mntrpc*); int rpcattn(void*); void mclose(Mnt*, Chan*); void mntrecover(Mnt*, Mntrpc*); Chan* mntchan(void); int defmaxmsg = MAXFDATA; void (*mntstats)(int, Chan*, uvlong, ulong); enum { Tagspace = 1, Tagfls = 0x8000, Tagend = 0xfffe, }; static void mntreset(void) { mntalloc.id = 1; mntalloc.rpctag = Tagspace; cinit(); } static Chan* mntattach(char *muxattach) { Mnt *m; Chan *c, *mc; char buf[NAMELEN]; struct bogus{ Chan *chan; char *spec; int flags; }bogus; bogus = *((struct bogus *)muxattach); c = bogus.chan; lock(&mntalloc); for(m = mntalloc.list; m; m = m->list) { if(m->c == c && m->id) { lock(m); if(m->id && m->ref > 0 && m->c == c) { m->ref++; unlock(m); unlock(&mntalloc); c = mntchan(); if(waserror()) { chanfree(c); nexterror(); } mattach(m, c, bogus.spec); poperror(); if(bogus.flags&MCACHE) c->flag |= CCACHE; return c; } unlock(m); } } m = mntalloc.mntfree; if(m != 0) mntalloc.mntfree = m->list; else { m = malloc(sizeof(Mnt)); if(m == 0) { unlock(&mntalloc); exhausted("mount devices"); } m->flushbase = Tagfls; m->flushtag = Tagfls; } m->list = mntalloc.list; mntalloc.list = m; m->id = mntalloc.id++; unlock(&mntalloc); lock(m); m->ref = 1; m->queue = 0; m->rip = 0; m->c = c; m->c->flag |= CMSG; if(strncmp(bogus.spec, "mntblk=", 7) == 0) { m->blocksize = strtoul(bogus.spec+7, 0, 0); if(m->blocksize > MAXFDATA) m->blocksize = MAXFDATA; print("mount blk %d\n", m->blocksize); bogus.spec = ""; } else m->blocksize = defmaxmsg; m->flags = bogus.flags & ~MCACHE; incref(m->c); sprint(buf, "#M%lud", m->id); m->tree.root = ptenter(&m->tree, 0, buf); unlock(m); c = mntchan(); if(waserror()) { mclose(m, c); /* Close must not be called since it will * call mnt recursively */ chanfree(c); nexterror(); } mattach(m, c, bogus.spec); poperror(); /* * Detect a recursive mount for a mount point served by exportfs. * If CHDIR is clear in the returned qid, the foreign server is * requesting the mount point be folded into the connection * to the exportfs. In this case the remote mount driver does * the multiplexing. */ mc = m->c; if(mc->type == devno('M', 0) && (c->qid.path&CHDIR) == 0) { mclose(m, c); c->qid.path |= CHDIR; c->mntptr = mc->mntptr; c->mchan = c->mntptr->c; c->mqid = c->qid; c->path = c->mntptr->tree.root; incref(c->path); incref(c->mntptr); } if(bogus.flags & MCACHE) c->flag |= CCACHE; return c; } Chan* mntchan(void) { Chan *c; c = devattach('M', 0); lock(&mntalloc); c->dev = mntalloc.id++; unlock(&mntalloc); return c; } void mattach(Mnt *m, Chan *c, char *spec) { ulong id; Mntrpc *r; r = mntralloc(0); c->mntptr = m; if(waserror()) { mntfree(r); nexterror(); } r->request.type = Tattach; r->request.fid = c->fid; memmove(r->request.uname, up->user, NAMELEN); strncpy(r->request.aname, spec, NAMELEN); id = authrequest(m->c->session, &r->request); mountrpc(m, r); authreply(m->c->session, id, &r->reply); c->qid = r->reply.qid; c->mchan = m->c; c->mqid = c->qid; c->path = m->tree.root; incref(c->path); poperror(); mntfree(r); } static Chan* mntclone(Chan *c, Chan *nc) { Mnt *m; Mntrpc *r; int alloc = 0; m = mntchk(c); r = mntralloc(c); if(nc == 0) { nc = newchan(); alloc = 1; } if(waserror()) { mntfree(r); if(alloc) cclose(nc); nexterror(); } r->request.type = Tclone; r->request.fid = c->fid; r->request.newfid = nc->fid; mountrpc(m, r); devclone(c, nc); nc->mqid = c->qid; incref(m); USED(alloc); poperror(); mntfree(r); return nc; } static int mntwalk(Chan *c, char *name) { Mnt *m; Path *op; Mntrpc *r; m = mntchk(c); r = mntralloc(c); if(waserror()) { mntfree(r); return 0; } r->request.type = Twalk; r->request.fid = c->fid; strncpy(r->request.name, name, NAMELEN); mountrpc(m, r); c->qid = r->reply.qid; op = c->path; c->path = ptenter(&m->tree, op, name); decref(op); poperror(); mntfree(r); return 1; } static void mntstat(Chan *c, char *dp) { Mnt *m; Mntrpc *r; m = mntchk(c); r = mntralloc(c); if(waserror()) { mntfree(r); nexterror(); } r->request.type = Tstat; r->request.fid = c->fid; mountrpc(m, r); memmove(dp, r->reply.stat, DIRLEN); mntdirfix((uchar*)dp, c); poperror(); mntfree(r); } static Chan* mntopen(Chan *c, int omode) { Mnt *m; Mntrpc *r; m = mntchk(c); r = mntralloc(c); if(waserror()) { mntfree(r); nexterror(); } r->request.type = Topen; r->request.fid = c->fid; r->request.mode = omode; mountrpc(m, r); c->qid = r->reply.qid; c->offset = 0; c->mode = openmode(omode); c->flag |= COPEN; poperror(); mntfree(r); if(c->flag & CCACHE) copen(c); return c; } static void mntcreate(Chan *c, char *name, int omode, ulong perm) { Mnt *m; Mntrpc *r; m = mntchk(c); r = mntralloc(c); if(waserror()) { mntfree(r); nexterror(); } r->request.type = Tcreate; r->request.fid = c->fid; r->request.mode = omode; r->request.perm = perm; strncpy(r->request.name, name, NAMELEN); mountrpc(m, r); c->qid = r->reply.qid; c->flag |= COPEN; c->mode = openmode(omode); poperror(); mntfree(r); if(c->flag & CCACHE) copen(c); } static void mntclunk(Chan *c, int t) { Mnt *m; Mntrpc *r; m = mntchk(c); r = mntralloc(c); if(waserror()){ mntfree(r); mclose(m, c); nexterror(); } r->request.type = t; r->request.fid = c->fid; mountrpc(m, r); mntfree(r); mclose(m, c); poperror(); } void mclose(Mnt *m, Chan *c) { Mntrpc *q, *r; if(decref(m) != 0) return; c->path = 0; ptclose(&m->tree); for(q = m->queue; q; q = r) { r = q->list; q->flushed = 0; mntfree(q); } m->id = 0; cclose(m->c); mntpntfree(m); } void mntpntfree(Mnt *m) { Mnt *f, **l; lock(&mntalloc); l = &mntalloc.list; for(f = *l; f; f = f->list) { if(f == m) { *l = m->list; break; } l = &f->list; } m->list = mntalloc.mntfree; mntalloc.mntfree = m; unlock(&mntalloc); } static void mntclose(Chan *c) { mntclunk(c, Tclunk); } static void mntremove(Chan *c) { mntclunk(c, Tremove); } static void mntwstat(Chan *c, char *dp) { Mnt *m; Mntrpc *r; m = mntchk(c); r = mntralloc(c); if(waserror()) { mntfree(r); nexterror(); } r->request.type = Twstat; r->request.fid = c->fid; memmove(r->request.stat, dp, DIRLEN); mountrpc(m, r); poperror(); mntfree(r); } long mntread9p(Chan *c, void *buf, long n, vlong off) { return mnt9prdwr(Tread, c, buf, n, off); } static long mntread(Chan *c, void *buf, long n, vlong off) { uchar *p, *e; int nc, cache, isdir; isdir = 0; cache = c->flag & CCACHE; if(c->qid.path & CHDIR) { cache = 0; isdir = 1; } p = buf; if(cache) { nc = cread(c, buf, n, off); if(nc > 0) { n -= nc; if(n == 0) return nc; p += nc; off += nc; } n = mntrdwr(Tread, c, p, n, off); cupdate(c, p, n, off); return n + nc; } n = mntrdwr(Tread, c, buf, n, off); if(isdir) { for(e = &p[n]; p < e; p += DIRLEN) mntdirfix(p, c); } return n; } long mntwrite9p(Chan *c, void *buf, long n, vlong off) { return mnt9prdwr(Twrite, c, buf, n, off); } static long mntwrite(Chan *c, void *buf, long n, vlong off) { return mntrdwr(Twrite, c, buf, n, off); } long mnt9prdwr(int type, Chan *c, void *buf, long n, vlong off) { Mnt *m; ulong nr; Mntrpc *r; if(n > MAXRPC-32) { if(type == Twrite) error("write9p too long"); n = MAXRPC-32; } m = mntchk(c); r = mntralloc(c); if(waserror()) { mntfree(r); nexterror(); } r->request.type = type; r->request.fid = c->fid; r->request.offset = off; r->request.data = buf; r->request.count = n; mountrpc(m, r); nr = r->reply.count; if(nr > r->request.count) nr = r->request.count; if(type == Tread) memmove(buf, r->reply.data, nr); poperror(); mntfree(r); return nr; } long mntrdwr(int type, Chan *c, void *buf, long n, vlong off) { Mnt *m; Mntrpc *r; char *uba; int cache; ulong cnt, nr, nreq; m = mntchk(c); uba = buf; cnt = 0; cache = c->flag & CCACHE; if(c->qid.path & CHDIR) cache = 0; for(;;) { r = mntralloc(c); if(waserror()) { mntfree(r); nexterror(); } r->request.type = type; r->request.fid = c->fid; r->request.offset = off; r->request.data = uba; r->request.count = limit(n, m->blocksize); mountrpc(m, r); nreq = r->request.count; nr = r->reply.count; if(nr > nreq) nr = nreq; if(type == Tread) memmove(uba, r->reply.data, nr); else if(cache) cwrite(c, (uchar*)uba, nr, off); poperror(); mntfree(r); off += nr; uba += nr; cnt += nr; n -= nr; if(nr != nreq || n == 0 || up->nnote) break; } return cnt; } void mountrpc(Mnt *m, Mntrpc *r) { int t; r->reply.tag = 0; r->reply.type = 4; while(waserror()) { if((m->flags&MRECOV) == 0) nexterror(); mntrecover(m, r); } mountio(m, r); poperror(); t = r->reply.type; switch(t) { case Rerror: error(r->reply.ename); case Rflush: error(Eintr); default: if(t == r->request.type+1) break; print("mnt: proc %s %lud: mismatch rep 0x%lux T%d R%d rq %d fls %d rp %d\n", up->text, up->pid, r, r->request.type, r->reply.type, r->request.tag, r->flushtag, r->reply.tag); error(Emountrpc); } } void mountio(Mnt *m, Mntrpc *r) { int n; lock(m); r->flushed = 0; r->m = m; r->list = m->queue; m->queue = r; unlock(m); /* Transmit a file system rpc */ n = convS2M(&r->request, r->rpc); if(n < 0) panic("bad message type in mountio"); if(waserror()) { if(mntflush(m, r) == 0) nexterror(); } else { if(devtab[m->c->type]->dc == L'M'){ if(mnt9prdwr(Twrite, m->c, r->rpc, n, 0) != n) error(Emountrpc); }else{ if(devtab[m->c->type]->write(m->c, r->rpc, n, 0) != n) error(Emountrpc); } r->stime = fastticks(nil); r->reqlen = n; poperror(); } /* Gate readers onto the mount point one at a time */ for(;;) { lock(m); if(m->rip == 0) break; unlock(m); if(waserror()) { if(mntflush(m, r) == 0) nexterror(); continue; } sleep(&r->r, rpcattn, r); poperror(); if(r->done) return; } m->rip = up; unlock(m); while(r->done == 0) { mntrpcread(m, r); mountmux(m, r); } mntgate(m); } void mntrpcread(Mnt *m, Mntrpc *r) { int n; for(;;) { if(waserror()) { if(mntflush(m, r) == 0) { mntgate(m); nexterror(); } continue; } r->reply.type = 0; r->reply.tag = 0; if(devtab[m->c->type]->dc == L'M') n = mnt9prdwr(Tread, m->c, r->rpc, MAXRPC, 0); else n = devtab[m->c->type]->read(m->c, r->rpc, MAXRPC, 0); poperror(); if(n == 0) continue; r->replen = n; if(convM2S(r->rpc, &r->reply, n) != 0) return; } } void mntgate(Mnt *m) { Mntrpc *q; lock(m); m->rip = 0; for(q = m->queue; q; q = q->list) { if(q->done == 0) if(wakeup(&q->r)) break; } unlock(m); } void mountmux(Mnt *m, Mntrpc *r) { char *dp; Mntrpc **l, *q; lock(m); l = &m->queue; for(q = *l; q; q = q->list) { if(q->request.tag == r->reply.tag || q->flushed && q->flushtag == r->reply.tag) { *l = q->list; unlock(m); if(q != r) { /* * Completed someone else. * Trade pointers to receive buffer. */ dp = q->rpc; q->rpc = r->rpc; r->rpc = dp; q->reply = r->reply; q->done = 1; if(mntstats != nil) (*mntstats)(q->request.type, m->c, q->stime, q->reqlen + r->replen); wakeup(&q->r); }else { if(mntstats != nil) (*mntstats)(r->request.type, m->c, r->stime, r->reqlen + r->replen); q->done = 1; } return; } l = &q->list; } unlock(m); } int mntflush(Mnt *m, Mntrpc *r) { int n, l; Fcall flush; lock(m); r->flushtag = m->flushtag++; if(m->flushtag == Tagend) m->flushtag = m->flushbase; r->flushed = 1; unlock(m); flush.type = Tflush; flush.tag = r->flushtag; flush.oldtag = r->request.tag; n = convS2M(&flush, r->flush); if(n < 0) panic("bad message type in mntflush"); if(waserror()) { if(strcmp(up->error, Eintr) == 0) return 1; mntqrm(m, r); return 0; } l = devtab[m->c->type]->write(m->c, r->flush, n, 0); if(l != n) error(Ehungup); poperror(); return 1; } Mntrpc* mntralloc(Chan *c) { Mntrpc *new; lock(&mntalloc); new = mntalloc.rpcfree; if(new == nil){ new = malloc(sizeof(Mntrpc)); if(new == nil) { unlock(&mntalloc); exhausted("mount rpc header"); } /* * The header is split from the data buffer as * mountmux may swap the buffer with another header. */ new->rpc = mallocz(MAXRPC, 0); if(new->rpc == nil){ free(new); unlock(&mntalloc); exhausted("mount rpc buffer"); } new->request.tag = mntalloc.rpctag++; } else { mntalloc.rpcfree = new->list; mntalloc.nrpcfree--; } mntalloc.nrpcused++; unlock(&mntalloc); new->c = c; new->done = 0; new->flushed = 0; new->flushtag = 0; return new; } void mntfree(Mntrpc *r) { lock(&mntalloc); if(mntalloc.nrpcfree >= 10){ free(r->rpc); free(r); } else{ r->list = mntalloc.rpcfree; mntalloc.rpcfree = r; mntalloc.nrpcfree++; } mntalloc.nrpcused--; unlock(&mntalloc); } void mntqrm(Mnt *m, Mntrpc *r) { Mntrpc **l, *f; lock(m); r->done = 1; r->flushed = 0; l = &m->queue; for(f = *l; f; f = f->list) { if(f == r) { *l = r->list; break; } l = &f->list; } unlock(m); } void recoverchan(Mnt *m, Chan *c) { int i, n, flg; Path *safe, *p, **pav; if(m->c == 0) error(Eshutdown); flg = c->flag; /* Don't recursively recover */ c->flag &= ~(COPEN|CRECOV); n = 0; for(p = c->path; p; p = p->parent) n++; pav = smalloc(sizeof(Path*)*n); i = n; for(p = c->path; p; p = p->parent) pav[--i] = p; safe = c->path; if(waserror()) { c->flag = flg; free(pav); nexterror(); } /* Attach the fid onto the file server (sets c->path to #Mxxx) */ mattach(m, c, c->xmh->mount->spec); poperror(); /* * c is now at the root so we free where * the chan was before the server connection was lost */ decref(safe); for(i = 1; i < n; i++) { if(mntwalk(c, pav[i]->elem) == 0) { free(pav); /* Shut down the channel */ c->dev = m->id-1; error(Erecover); } } free(pav); if(flg&COPEN) mntopen(c, c->mode); } Mnt* mntchk(Chan *c) { Mnt *m; m = c->mntptr; /* * Was it closed and reused */ if(m->id == 0 || m->id >= c->dev) error(Eshutdown); /* * Try and get the channel back after a crash */ if((c->flag&CRECOV) && m->recprog == 0) recoverchan(m, c); return m; } void mntdirfix(uchar *dirbuf, Chan *c) { int r; r = devtab[c->type]->dc; dirbuf[DIRLEN-4] = r>>0; dirbuf[DIRLEN-3] = r>>8; dirbuf[DIRLEN-2] = c->dev; dirbuf[DIRLEN-1] = c->dev>>8; } int rpcattn(void *v) { Mntrpc *r; r = v; return r->done || r->m->rip == 0; } int recdone(void *v) { Mnt *m; m = v; return m->recprog == 0; } void mntrecdel(Mnt *m, Mntrpc *r) { Mntrpc *f, **l; lock(m); l = &m->recwait; for(f = *l; f; f = f->list) { if(f == r) { *l = r->list; break; } } unlock(m); } void mntrecover(Mnt *m, Mntrpc *r) { char *ps; lock(m); if(m->recprog == 0) { m->recprog = 1; unlock(m); chanrec(m); /* * Send a message to boot via #/recover */ rootrecover(m->c->path, m->tree.root->elem); lock(m); } r->list = m->recwait; m->recwait = r; unlock(m); pprint("lost server connection, wait...\n"); ps = up->psstate; up->psstate = "Recover"; if(waserror()) { up->psstate = ps; mntrecdel(m, r); nexterror(); } sleep(&r->r, recdone, m); poperror(); r->done = 0; mntrecdel(m, r); if(r->c != 0) recoverchan(m, r->c); up->psstate = ps; } void mntrepl(char *buf) { int fd; Mnt *m; char *p; Chan *c1, *c2; Mntrpc *r; /* reply from boot is 'fd #M23' */ fd = strtoul(buf, &p, 0); p++; lock(&mntalloc); for(m = mntalloc.list; m; m = m->list) { if(strcmp(p, m->tree.root->elem) == 0) break; } unlock(&mntalloc); if(m == 0) error(Eunmount); c1 = fdtochan(fd, ORDWR, 0, 1); /* error check and inc ref */ /* If the channel was posted fix it up */ srvrecover(m->c, c1); lock(m); c2 = m->c; m->c = c1; m->recprog = 0; /* Wakeup partially complete rpc */ for(r = m->recwait; r; r = r->list) wakeup(&r->r); unlock(m); cclose(c2); } Dev mntdevtab = { 'M', "mnt", mntreset, devinit, mntattach, mntclone, mntwalk, mntstat, mntopen, mntcreate, mntclose, mntread, devbread, mntwrite, devbwrite, mntremove, mntwstat, };