Current Path : /sys/amd64/compile/hs32/modules/usr/src/sys/modules/ata/atapci/chipsets/atacyrix/@/dev/hatm/ |
FreeBSD hs32.drive.ne.jp 9.1-RELEASE FreeBSD 9.1-RELEASE #1: Wed Jan 14 12:18:08 JST 2015 root@hs32.drive.ne.jp:/sys/amd64/compile/hs32 amd64 |
Current File : //sys/amd64/compile/hs32/modules/usr/src/sys/modules/ata/atapci/chipsets/atacyrix/@/dev/hatm/if_hatm_intr.c |
/*- * Copyright (c) 2001-2003 * Fraunhofer Institute for Open Communication Systems (FhG Fokus). * All rights reserved. * Author: Hartmut Brandt <harti@freebsd.org> * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. */ #include <sys/cdefs.h> __FBSDID("$FreeBSD: release/9.1.0/sys/dev/hatm/if_hatm_intr.c 175872 2008-02-01 19:36:27Z phk $"); /* * ForeHE driver. * * Interrupt handler. */ #include "opt_inet.h" #include "opt_natm.h" #include <sys/types.h> #include <sys/param.h> #include <sys/systm.h> #include <sys/malloc.h> #include <sys/kernel.h> #include <sys/bus.h> #include <sys/errno.h> #include <sys/conf.h> #include <sys/module.h> #include <sys/queue.h> #include <sys/syslog.h> #include <sys/condvar.h> #include <sys/sysctl.h> #include <vm/uma.h> #include <sys/sockio.h> #include <sys/mbuf.h> #include <sys/socket.h> #include <net/if.h> #include <net/if_media.h> #include <net/if_atm.h> #include <net/route.h> #include <netinet/in.h> #include <netinet/if_atm.h> #include <machine/bus.h> #include <machine/resource.h> #include <sys/bus.h> #include <sys/rman.h> #include <dev/pci/pcireg.h> #include <dev/pci/pcivar.h> #include <dev/utopia/utopia.h> #include <dev/hatm/if_hatmconf.h> #include <dev/hatm/if_hatmreg.h> #include <dev/hatm/if_hatmvar.h> CTASSERT(sizeof(struct mbuf_page) == MBUF_ALLOC_SIZE); CTASSERT(sizeof(struct mbuf0_chunk) == MBUF0_CHUNK); CTASSERT(sizeof(struct mbuf1_chunk) == MBUF1_CHUNK); CTASSERT(sizeof(((struct mbuf0_chunk *)NULL)->storage) >= MBUF0_SIZE); CTASSERT(sizeof(((struct mbuf1_chunk *)NULL)->storage) >= MBUF1_SIZE); CTASSERT(sizeof(struct tpd) <= HE_TPD_SIZE); CTASSERT(MBUF0_PER_PAGE <= 256); CTASSERT(MBUF1_PER_PAGE <= 256); static void hatm_mbuf_page_alloc(struct hatm_softc *sc, u_int group); /* * Free an external mbuf to a list. We use atomic functions so that * we don't need a mutex for the list. * * Note that in general this algorithm is not safe when multiple readers * and writers are present. To cite from a mail from David Schultz * <das@freebsd.org>: * * It looks like this is subject to the ABA problem. For instance, * suppose X, Y, and Z are the top things on the freelist and a * thread attempts to make an allocation. You set buf to X and load * buf->link (Y) into a register. Then the thread get preempted, and * another thread allocates both X and Y, then frees X. When the * original thread gets the CPU again, X is still on top of the * freelist, so the atomic operation succeeds. However, the atomic * op places Y on top of the freelist, even though Y is no longer * free. * * We are, however sure that we have only one thread that ever allocates * buffers because the only place we're call from is the interrupt handler. * Under these circumstances the code looks safe. */ void hatm_ext_free(struct mbufx_free **list, struct mbufx_free *buf) { for (;;) { buf->link = *list; if (atomic_cmpset_ptr((uintptr_t *)list, (uintptr_t)buf->link, (uintptr_t)buf)) break; } } static __inline struct mbufx_free * hatm_ext_alloc(struct hatm_softc *sc, u_int g) { struct mbufx_free *buf; for (;;) { if ((buf = sc->mbuf_list[g]) == NULL) break; if (atomic_cmpset_ptr((uintptr_t *)&sc->mbuf_list[g], (uintptr_t)buf, (uintptr_t)buf->link)) break; } if (buf == NULL) { hatm_mbuf_page_alloc(sc, g); for (;;) { if ((buf = sc->mbuf_list[g]) == NULL) break; if (atomic_cmpset_ptr((uintptr_t *)&sc->mbuf_list[g], (uintptr_t)buf, (uintptr_t)buf->link)) break; } } return (buf); } /* * Either the queue treshold was crossed or a TPD with the INTR bit set * was transmitted. */ static void he_intr_tbrq(struct hatm_softc *sc, struct hetbrq *q, u_int group) { uint32_t *tailp = &sc->hsp->group[group].tbrq_tail; u_int no; while (q->head != (*tailp >> 2)) { no = (q->tbrq[q->head].addr & HE_REGM_TBRQ_ADDR) >> HE_REGS_TPD_ADDR; hatm_tx_complete(sc, TPD_ADDR(sc, no), (q->tbrq[q->head].addr & HE_REGM_TBRQ_FLAGS)); if (++q->head == q->size) q->head = 0; } WRITE4(sc, HE_REGO_TBRQ_H(group), q->head << 2); } /* * DMA loader function for external mbuf page. */ static void hatm_extbuf_helper(void *arg, bus_dma_segment_t *segs, int nsegs, int error) { if (error) { printf("%s: mapping error %d\n", __func__, error); return; } KASSERT(nsegs == 1, ("too many segments for DMA: %d", nsegs)); KASSERT(segs[0].ds_addr <= 0xffffffffLU, ("phys addr too large %lx", (u_long)segs[0].ds_addr)); *(uint32_t *)arg = segs[0].ds_addr; } /* * Allocate a page of external mbuf storage for the small pools. * Create a DMA map and load it. Put all the chunks onto the right * free list. */ static void hatm_mbuf_page_alloc(struct hatm_softc *sc, u_int group) { struct mbuf_page *pg; int err; u_int i; if (sc->mbuf_npages == sc->mbuf_max_pages) return; if ((pg = malloc(MBUF_ALLOC_SIZE, M_DEVBUF, M_NOWAIT)) == NULL) return; err = bus_dmamap_create(sc->mbuf_tag, 0, &pg->hdr.map); if (err != 0) { if_printf(sc->ifp, "%s -- bus_dmamap_create: %d\n", __func__, err); free(pg, M_DEVBUF); return; } err = bus_dmamap_load(sc->mbuf_tag, pg->hdr.map, pg, MBUF_ALLOC_SIZE, hatm_extbuf_helper, &pg->hdr.phys, BUS_DMA_NOWAIT); if (err != 0) { if_printf(sc->ifp, "%s -- mbuf mapping failed %d\n", __func__, err); bus_dmamap_destroy(sc->mbuf_tag, pg->hdr.map); free(pg, M_DEVBUF); return; } sc->mbuf_pages[sc->mbuf_npages] = pg; if (group == 0) { struct mbuf0_chunk *c; pg->hdr.pool = 0; pg->hdr.nchunks = MBUF0_PER_PAGE; pg->hdr.chunksize = MBUF0_CHUNK; pg->hdr.hdroff = sizeof(c->storage); c = (struct mbuf0_chunk *)pg; for (i = 0; i < MBUF0_PER_PAGE; i++, c++) { c->hdr.pageno = sc->mbuf_npages; c->hdr.chunkno = i; c->hdr.flags = 0; hatm_ext_free(&sc->mbuf_list[0], (struct mbufx_free *)c); } } else { struct mbuf1_chunk *c; pg->hdr.pool = 1; pg->hdr.nchunks = MBUF1_PER_PAGE; pg->hdr.chunksize = MBUF1_CHUNK; pg->hdr.hdroff = sizeof(c->storage); c = (struct mbuf1_chunk *)pg; for (i = 0; i < MBUF1_PER_PAGE; i++, c++) { c->hdr.pageno = sc->mbuf_npages; c->hdr.chunkno = i; c->hdr.flags = 0; hatm_ext_free(&sc->mbuf_list[1], (struct mbufx_free *)c); } } sc->mbuf_npages++; } /* * Free an mbuf and put it onto the free list. */ static void hatm_mbuf0_free(void *buf, void *args) { struct hatm_softc *sc = args; struct mbuf0_chunk *c = buf; KASSERT((c->hdr.flags & (MBUF_USED | MBUF_CARD)) == MBUF_USED, ("freeing unused mbuf %x", c->hdr.flags)); c->hdr.flags &= ~MBUF_USED; hatm_ext_free(&sc->mbuf_list[0], (struct mbufx_free *)c); } static void hatm_mbuf1_free(void *buf, void *args) { struct hatm_softc *sc = args; struct mbuf1_chunk *c = buf; KASSERT((c->hdr.flags & (MBUF_USED | MBUF_CARD)) == MBUF_USED, ("freeing unused mbuf %x", c->hdr.flags)); c->hdr.flags &= ~MBUF_USED; hatm_ext_free(&sc->mbuf_list[1], (struct mbufx_free *)c); } static void hatm_mbuf_helper(void *arg, bus_dma_segment_t *segs, int nsegs, int error) { uint32_t *ptr = (uint32_t *)arg; if (nsegs == 0) { printf("%s: error=%d\n", __func__, error); return; } KASSERT(nsegs == 1, ("too many segments for mbuf: %d", nsegs)); KASSERT(segs[0].ds_addr <= 0xffffffffLU, ("phys addr too large %lx", (u_long)segs[0].ds_addr)); *ptr = segs[0].ds_addr; } /* * Receive buffer pool interrupt. This means the number of entries in the * queue has dropped below the threshold. Try to supply new buffers. */ static void he_intr_rbp(struct hatm_softc *sc, struct herbp *rbp, u_int large, u_int group) { u_int ntail; struct mbuf *m; int error; struct mbufx_free *cf; struct mbuf_page *pg; struct mbuf0_chunk *buf0; struct mbuf1_chunk *buf1; DBG(sc, INTR, ("%s buffer supply threshold crossed for group %u", large ? "large" : "small", group)); rbp->head = (READ4(sc, HE_REGO_RBP_S(large, group)) >> HE_REGS_RBP_HEAD) & (rbp->size - 1); for (;;) { if ((ntail = rbp->tail + 1) == rbp->size) ntail = 0; if (ntail == rbp->head) break; m = NULL; if (large) { /* allocate the MBUF */ if ((m = m_getcl(M_DONTWAIT, MT_DATA, M_PKTHDR)) == NULL) { if_printf(sc->ifp, "no mbuf clusters\n"); break; } m->m_data += MBUFL_OFFSET; if (sc->lbufs[sc->lbufs_next] != NULL) panic("hatm: lbufs full %u", sc->lbufs_next); sc->lbufs[sc->lbufs_next] = m; if ((error = bus_dmamap_load(sc->mbuf_tag, sc->rmaps[sc->lbufs_next], m->m_data, rbp->bsize, hatm_mbuf_helper, &rbp->rbp[rbp->tail].phys, BUS_DMA_NOWAIT)) != 0) panic("hatm: mbuf mapping failed %d", error); bus_dmamap_sync(sc->mbuf_tag, sc->rmaps[sc->lbufs_next], BUS_DMASYNC_PREREAD); rbp->rbp[rbp->tail].handle = MBUF_MAKE_LHANDLE(sc->lbufs_next); if (++sc->lbufs_next == sc->lbufs_size) sc->lbufs_next = 0; } else if (group == 0) { /* * Allocate small buffer in group 0 */ if ((cf = hatm_ext_alloc(sc, 0)) == NULL) break; buf0 = (struct mbuf0_chunk *)cf; pg = sc->mbuf_pages[buf0->hdr.pageno]; buf0->hdr.flags |= MBUF_CARD; rbp->rbp[rbp->tail].phys = pg->hdr.phys + buf0->hdr.chunkno * MBUF0_CHUNK + MBUF0_OFFSET; rbp->rbp[rbp->tail].handle = MBUF_MAKE_HANDLE(buf0->hdr.pageno, buf0->hdr.chunkno); bus_dmamap_sync(sc->mbuf_tag, pg->hdr.map, BUS_DMASYNC_PREREAD); } else if (group == 1) { /* * Allocate small buffer in group 1 */ if ((cf = hatm_ext_alloc(sc, 1)) == NULL) break; buf1 = (struct mbuf1_chunk *)cf; pg = sc->mbuf_pages[buf1->hdr.pageno]; buf1->hdr.flags |= MBUF_CARD; rbp->rbp[rbp->tail].phys = pg->hdr.phys + buf1->hdr.chunkno * MBUF1_CHUNK + MBUF1_OFFSET; rbp->rbp[rbp->tail].handle = MBUF_MAKE_HANDLE(buf1->hdr.pageno, buf1->hdr.chunkno); bus_dmamap_sync(sc->mbuf_tag, pg->hdr.map, BUS_DMASYNC_PREREAD); } else /* ups */ break; DBG(sc, DMA, ("MBUF loaded: handle=%x m=%p phys=%x", rbp->rbp[rbp->tail].handle, m, rbp->rbp[rbp->tail].phys)); rbp->tail = ntail; } WRITE4(sc, HE_REGO_RBP_T(large, group), (rbp->tail << HE_REGS_RBP_TAIL)); } /* * Extract the buffer and hand it to the receive routine */ static struct mbuf * hatm_rx_buffer(struct hatm_softc *sc, u_int group, u_int handle) { u_int pageno; u_int chunkno; struct mbuf *m; if (handle & MBUF_LARGE_FLAG) { /* large buffer - sync and unload */ MBUF_PARSE_LHANDLE(handle, handle); DBG(sc, RX, ("RX large handle=%x", handle)); bus_dmamap_sync(sc->mbuf_tag, sc->rmaps[handle], BUS_DMASYNC_POSTREAD); bus_dmamap_unload(sc->mbuf_tag, sc->rmaps[handle]); m = sc->lbufs[handle]; sc->lbufs[handle] = NULL; return (m); } MBUF_PARSE_HANDLE(handle, pageno, chunkno); DBG(sc, RX, ("RX group=%u handle=%x page=%u chunk=%u", group, handle, pageno, chunkno)); MGETHDR(m, M_DONTWAIT, MT_DATA); if (group == 0) { struct mbuf0_chunk *c0; c0 = (struct mbuf0_chunk *)sc->mbuf_pages[pageno] + chunkno; KASSERT(c0->hdr.pageno == pageno, ("pageno = %u/%u", c0->hdr.pageno, pageno)); KASSERT(c0->hdr.chunkno == chunkno, ("chunkno = %u/%u", c0->hdr.chunkno, chunkno)); KASSERT(c0->hdr.flags & MBUF_CARD, ("mbuf not on card %u/%u", pageno, chunkno)); KASSERT(!(c0->hdr.flags & MBUF_USED), ("used mbuf %u/%u", pageno, chunkno)); c0->hdr.flags |= MBUF_USED; c0->hdr.flags &= ~MBUF_CARD; if (m != NULL) { m->m_ext.ref_cnt = &c0->hdr.ref_cnt; MEXTADD(m, (void *)c0, MBUF0_SIZE, hatm_mbuf0_free, c0, sc, M_PKTHDR, EXT_EXTREF); m->m_data += MBUF0_OFFSET; } else hatm_mbuf0_free(c0, sc); } else { struct mbuf1_chunk *c1; c1 = (struct mbuf1_chunk *)sc->mbuf_pages[pageno] + chunkno; KASSERT(c1->hdr.pageno == pageno, ("pageno = %u/%u", c1->hdr.pageno, pageno)); KASSERT(c1->hdr.chunkno == chunkno, ("chunkno = %u/%u", c1->hdr.chunkno, chunkno)); KASSERT(c1->hdr.flags & MBUF_CARD, ("mbuf not on card %u/%u", pageno, chunkno)); KASSERT(!(c1->hdr.flags & MBUF_USED), ("used mbuf %u/%u", pageno, chunkno)); c1->hdr.flags |= MBUF_USED; c1->hdr.flags &= ~MBUF_CARD; if (m != NULL) { m->m_ext.ref_cnt = &c1->hdr.ref_cnt; MEXTADD(m, (void *)c1, MBUF1_SIZE, hatm_mbuf1_free, c1, sc, M_PKTHDR, EXT_EXTREF); m->m_data += MBUF1_OFFSET; } else hatm_mbuf1_free(c1, sc); } return (m); } /* * Interrupt because of receive buffer returned. */ static void he_intr_rbrq(struct hatm_softc *sc, struct herbrq *rq, u_int group) { struct he_rbrqen *e; uint32_t flags, tail; u_int cid, len; struct mbuf *m; for (;;) { tail = sc->hsp->group[group].rbrq_tail >> 3; if (rq->head == tail) break; e = &rq->rbrq[rq->head]; flags = e->addr & HE_REGM_RBRQ_FLAGS; if (!(flags & HE_REGM_RBRQ_HBUF_ERROR)) m = hatm_rx_buffer(sc, group, e->addr); else m = NULL; cid = (e->len & HE_REGM_RBRQ_CID) >> HE_REGS_RBRQ_CID; len = 4 * (e->len & HE_REGM_RBRQ_LEN); hatm_rx(sc, cid, flags, m, len); if (++rq->head == rq->size) rq->head = 0; } WRITE4(sc, HE_REGO_RBRQ_H(group), rq->head << 3); } void hatm_intr(void *p) { struct heirq *q = p; struct hatm_softc *sc = q->sc; u_int status; u_int tail; /* if we have a stray interrupt with a non-initialized card, * we cannot even lock before looking at the flag */ if (!(sc->ifp->if_drv_flags & IFF_DRV_RUNNING)) return; mtx_lock(&sc->mtx); (void)READ4(sc, HE_REGO_INT_FIFO); tail = *q->tailp; if (q->head == tail) { /* workaround for tail pointer not updated bug (8.1.1) */ DBG(sc, INTR, ("hatm: intr tailq not updated bug triggered")); /* read the tail pointer from the card */ tail = READ4(sc, HE_REGO_IRQ_BASE(q->group)) & HE_REGM_IRQ_BASE_TAIL; BARRIER_R(sc); sc->istats.bug_no_irq_upd++; } /* clear the interrupt */ WRITE4(sc, HE_REGO_INT_FIFO, HE_REGM_INT_FIFO_CLRA); BARRIER_W(sc); while (q->head != tail) { status = q->irq[q->head]; q->irq[q->head] = HE_REGM_ITYPE_INVALID; if (++q->head == (q->size - 1)) q->head = 0; switch (status & HE_REGM_ITYPE) { case HE_REGM_ITYPE_TBRQ: DBG(sc, INTR, ("TBRQ treshold %u", status & HE_REGM_IGROUP)); sc->istats.itype_tbrq++; he_intr_tbrq(sc, &sc->tbrq, status & HE_REGM_IGROUP); break; case HE_REGM_ITYPE_TPD: DBG(sc, INTR, ("TPD ready %u", status & HE_REGM_IGROUP)); sc->istats.itype_tpd++; he_intr_tbrq(sc, &sc->tbrq, status & HE_REGM_IGROUP); break; case HE_REGM_ITYPE_RBPS: sc->istats.itype_rbps++; switch (status & HE_REGM_IGROUP) { case 0: he_intr_rbp(sc, &sc->rbp_s0, 0, 0); break; case 1: he_intr_rbp(sc, &sc->rbp_s1, 0, 1); break; default: if_printf(sc->ifp, "bad INTR RBPS%u\n", status & HE_REGM_IGROUP); break; } break; case HE_REGM_ITYPE_RBPL: sc->istats.itype_rbpl++; switch (status & HE_REGM_IGROUP) { case 0: he_intr_rbp(sc, &sc->rbp_l0, 1, 0); break; default: if_printf(sc->ifp, "bad INTR RBPL%u\n", status & HE_REGM_IGROUP); break; } break; case HE_REGM_ITYPE_RBRQ: DBG(sc, INTR, ("INTERRUPT RBRQ %u", status & HE_REGM_IGROUP)); sc->istats.itype_rbrq++; switch (status & HE_REGM_IGROUP) { case 0: he_intr_rbrq(sc, &sc->rbrq_0, 0); break; case 1: if (sc->rbrq_1.size > 0) { he_intr_rbrq(sc, &sc->rbrq_1, 1); break; } /* FALLTHRU */ default: if_printf(sc->ifp, "bad INTR RBRQ%u\n", status & HE_REGM_IGROUP); break; } break; case HE_REGM_ITYPE_RBRQT: DBG(sc, INTR, ("INTERRUPT RBRQT %u", status & HE_REGM_IGROUP)); sc->istats.itype_rbrqt++; switch (status & HE_REGM_IGROUP) { case 0: he_intr_rbrq(sc, &sc->rbrq_0, 0); break; case 1: if (sc->rbrq_1.size > 0) { he_intr_rbrq(sc, &sc->rbrq_1, 1); break; } /* FALLTHRU */ default: if_printf(sc->ifp, "bad INTR RBRQT%u\n", status & HE_REGM_IGROUP); break; } break; case HE_REGM_ITYPE_PHYS: sc->istats.itype_phys++; utopia_intr(&sc->utopia); break; #if HE_REGM_ITYPE_UNKNOWN != HE_REGM_ITYPE_INVALID case HE_REGM_ITYPE_UNKNOWN: sc->istats.itype_unknown++; if_printf(sc->ifp, "bad interrupt\n"); break; #endif case HE_REGM_ITYPE_ERR: sc->istats.itype_err++; switch (status) { case HE_REGM_ITYPE_PERR: if_printf(sc->ifp, "parity error\n"); break; case HE_REGM_ITYPE_ABORT: if_printf(sc->ifp, "abort interrupt " "addr=0x%08x\n", READ4(sc, HE_REGO_ABORT_ADDR)); break; default: if_printf(sc->ifp, "bad interrupt type %08x\n", status); break; } break; case HE_REGM_ITYPE_INVALID: /* this is the documented fix for the ISW bug 8.1.1 * Note, that the documented fix is partly wrong: * the ISWs should be intialized to 0xf8 not 0xff */ sc->istats.bug_bad_isw++; DBG(sc, INTR, ("hatm: invalid ISW bug triggered")); he_intr_tbrq(sc, &sc->tbrq, 0); he_intr_rbp(sc, &sc->rbp_s0, 0, 0); he_intr_rbp(sc, &sc->rbp_l0, 1, 0); he_intr_rbp(sc, &sc->rbp_s1, 0, 1); he_intr_rbrq(sc, &sc->rbrq_0, 0); he_intr_rbrq(sc, &sc->rbrq_1, 1); utopia_intr(&sc->utopia); break; default: if_printf(sc->ifp, "bad interrupt type %08x\n", status); break; } } /* write back head to clear queue */ WRITE4(sc, HE_REGO_IRQ_HEAD(0), ((q->size - 1) << HE_REGS_IRQ_HEAD_SIZE) | (q->thresh << HE_REGS_IRQ_HEAD_THRESH) | (q->head << HE_REGS_IRQ_HEAD_HEAD)); BARRIER_W(sc); /* workaround the back-to-back irq access problem (8.1.2) */ (void)READ4(sc, HE_REGO_INT_FIFO); BARRIER_R(sc); mtx_unlock(&sc->mtx); }