diff options
Diffstat (limited to 'hw')
| -rw-r--r-- | hw/nvme/ctrl.c | 377 | ||||
| -rw-r--r-- | hw/nvme/ns.c | 55 | ||||
| -rw-r--r-- | hw/nvme/nvme.h | 18 | ||||
| -rw-r--r-- | hw/nvme/subsys.c | 9 | ||||
| -rw-r--r-- | hw/nvme/trace-events | 1 |
5 files changed, 271 insertions, 189 deletions
diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c index 629b0d3..43dfaea 100644 --- a/hw/nvme/ctrl.c +++ b/hw/nvme/ctrl.c @@ -439,10 +439,12 @@ static uint8_t nvme_sq_empty(NvmeSQueue *sq) static void nvme_irq_check(NvmeCtrl *n) { + uint32_t intms = ldl_le_p(&n->bar.intms); + if (msix_enabled(&(n->parent_obj))) { return; } - if (~n->bar.intms & n->irq_status) { + if (~intms & n->irq_status) { pci_irq_assert(&n->parent_obj); } else { pci_irq_deassert(&n->parent_obj); @@ -623,6 +625,10 @@ static uint16_t nvme_map_addr(NvmeCtrl *n, NvmeSg *sg, hwaddr addr, size_t len) return NVME_INVALID_USE_OF_CMB | NVME_DNR; } + if (sg->iov.niov + 1 > IOV_MAX) { + goto max_mappings_exceeded; + } + if (cmb) { return nvme_map_addr_cmb(n, &sg->iov, addr, len); } else { @@ -634,9 +640,18 @@ static uint16_t nvme_map_addr(NvmeCtrl *n, NvmeSg *sg, hwaddr addr, size_t len) return NVME_INVALID_USE_OF_CMB | NVME_DNR; } + if (sg->qsg.nsg + 1 > IOV_MAX) { + goto max_mappings_exceeded; + } + qemu_sglist_add(&sg->qsg, addr, len); return NVME_SUCCESS; + +max_mappings_exceeded: + NVME_GUEST_ERR(pci_nvme_ub_too_many_mappings, + "number of mappings exceed 1024"); + return NVME_INTERNAL_DEV_ERROR | NVME_DNR; } static inline bool nvme_addr_is_dma(NvmeCtrl *n, hwaddr addr) @@ -1276,7 +1291,7 @@ static void nvme_post_cqes(void *opaque) if (ret) { trace_pci_nvme_err_addr_write(addr); trace_pci_nvme_err_cfs(); - n->bar.csts = NVME_CSTS_FAILED; + stl_le_p(&n->bar.csts, NVME_CSTS_FAILED); break; } QTAILQ_REMOVE(&cq->req_list, req, entry); @@ -4009,7 +4024,7 @@ static uint16_t nvme_create_sq(NvmeCtrl *n, NvmeRequest *req) trace_pci_nvme_err_invalid_create_sq_sqid(sqid); return NVME_INVALID_QID | NVME_DNR; } - if (unlikely(!qsize || qsize > NVME_CAP_MQES(n->bar.cap))) { + if (unlikely(!qsize || qsize > NVME_CAP_MQES(ldq_le_p(&n->bar.cap)))) { trace_pci_nvme_err_invalid_create_sq_size(qsize); return NVME_MAX_QSIZE_EXCEEDED | NVME_DNR; } @@ -4195,7 +4210,7 @@ static uint16_t nvme_cmd_effects(NvmeCtrl *n, uint8_t csi, uint32_t buf_len, return NVME_INVALID_FIELD | NVME_DNR; } - switch (NVME_CC_CSS(n->bar.cc)) { + switch (NVME_CC_CSS(ldl_le_p(&n->bar.cc))) { case NVME_CC_CSS_NVM: src_iocs = nvme_cse_iocs_nvm; /* fall through */ @@ -4357,7 +4372,7 @@ static uint16_t nvme_create_cq(NvmeCtrl *n, NvmeRequest *req) trace_pci_nvme_err_invalid_create_cq_cqid(cqid); return NVME_INVALID_QID | NVME_DNR; } - if (unlikely(!qsize || qsize > NVME_CAP_MQES(n->bar.cap))) { + if (unlikely(!qsize || qsize > NVME_CAP_MQES(ldq_le_p(&n->bar.cap)))) { trace_pci_nvme_err_invalid_create_cq_size(qsize); return NVME_MAX_QSIZE_EXCEEDED | NVME_DNR; } @@ -5150,17 +5165,19 @@ static void nvme_update_dmrsl(NvmeCtrl *n) static void nvme_select_iocs_ns(NvmeCtrl *n, NvmeNamespace *ns) { + uint32_t cc = ldl_le_p(&n->bar.cc); + ns->iocs = nvme_cse_iocs_none; switch (ns->csi) { case NVME_CSI_NVM: - if (NVME_CC_CSS(n->bar.cc) != NVME_CC_CSS_ADMIN_ONLY) { + if (NVME_CC_CSS(cc) != NVME_CC_CSS_ADMIN_ONLY) { ns->iocs = nvme_cse_iocs_nvm; } break; case NVME_CSI_ZONED: - if (NVME_CC_CSS(n->bar.cc) == NVME_CC_CSS_CSI) { + if (NVME_CC_CSS(cc) == NVME_CC_CSS_CSI) { ns->iocs = nvme_cse_iocs_zoned; - } else if (NVME_CC_CSS(n->bar.cc) == NVME_CC_CSS_NVM) { + } else if (NVME_CC_CSS(cc) == NVME_CC_CSS_NVM) { ns->iocs = nvme_cse_iocs_nvm; } break; @@ -5497,7 +5514,7 @@ static void nvme_process_sq(void *opaque) if (nvme_addr_read(n, addr, (void *)&cmd, sizeof(cmd))) { trace_pci_nvme_err_addr_read(addr); trace_pci_nvme_err_cfs(); - n->bar.csts = NVME_CSTS_FAILED; + stl_le_p(&n->bar.csts, NVME_CSTS_FAILED); break; } nvme_inc_sq_head(sq); @@ -5552,8 +5569,6 @@ static void nvme_ctrl_reset(NvmeCtrl *n) n->aer_queued = 0; n->outstanding_aers = 0; n->qs_created = false; - - n->bar.cc = 0; } static void nvme_ctrl_shutdown(NvmeCtrl *n) @@ -5592,7 +5607,12 @@ static void nvme_select_iocs(NvmeCtrl *n) static int nvme_start_ctrl(NvmeCtrl *n) { - uint32_t page_bits = NVME_CC_MPS(n->bar.cc) + 12; + uint64_t cap = ldq_le_p(&n->bar.cap); + uint32_t cc = ldl_le_p(&n->bar.cc); + uint32_t aqa = ldl_le_p(&n->bar.aqa); + uint64_t asq = ldq_le_p(&n->bar.asq); + uint64_t acq = ldq_le_p(&n->bar.acq); + uint32_t page_bits = NVME_CC_MPS(cc) + 12; uint32_t page_size = 1 << page_bits; if (unlikely(n->cq[0])) { @@ -5603,73 +5623,72 @@ static int nvme_start_ctrl(NvmeCtrl *n) trace_pci_nvme_err_startfail_sq(); return -1; } - if (unlikely(!n->bar.asq)) { + if (unlikely(!asq)) { trace_pci_nvme_err_startfail_nbarasq(); return -1; } - if (unlikely(!n->bar.acq)) { + if (unlikely(!acq)) { trace_pci_nvme_err_startfail_nbaracq(); return -1; } - if (unlikely(n->bar.asq & (page_size - 1))) { - trace_pci_nvme_err_startfail_asq_misaligned(n->bar.asq); + if (unlikely(asq & (page_size - 1))) { + trace_pci_nvme_err_startfail_asq_misaligned(asq); return -1; } - if (unlikely(n->bar.acq & (page_size - 1))) { - trace_pci_nvme_err_startfail_acq_misaligned(n->bar.acq); + if (unlikely(acq & (page_size - 1))) { + trace_pci_nvme_err_startfail_acq_misaligned(acq); return -1; } - if (unlikely(!(NVME_CAP_CSS(n->bar.cap) & (1 << NVME_CC_CSS(n->bar.cc))))) { - trace_pci_nvme_err_startfail_css(NVME_CC_CSS(n->bar.cc)); + if (unlikely(!(NVME_CAP_CSS(cap) & (1 << NVME_CC_CSS(cc))))) { + trace_pci_nvme_err_startfail_css(NVME_CC_CSS(cc)); return -1; } - if (unlikely(NVME_CC_MPS(n->bar.cc) < - NVME_CAP_MPSMIN(n->bar.cap))) { + if (unlikely(NVME_CC_MPS(cc) < NVME_CAP_MPSMIN(cap))) { trace_pci_nvme_err_startfail_page_too_small( - NVME_CC_MPS(n->bar.cc), - NVME_CAP_MPSMIN(n->bar.cap)); + NVME_CC_MPS(cc), + NVME_CAP_MPSMIN(cap)); return -1; } - if (unlikely(NVME_CC_MPS(n->bar.cc) > - NVME_CAP_MPSMAX(n->bar.cap))) { + if (unlikely(NVME_CC_MPS(cc) > + NVME_CAP_MPSMAX(cap))) { trace_pci_nvme_err_startfail_page_too_large( - NVME_CC_MPS(n->bar.cc), - NVME_CAP_MPSMAX(n->bar.cap)); + NVME_CC_MPS(cc), + NVME_CAP_MPSMAX(cap)); return -1; } - if (unlikely(NVME_CC_IOCQES(n->bar.cc) < + if (unlikely(NVME_CC_IOCQES(cc) < NVME_CTRL_CQES_MIN(n->id_ctrl.cqes))) { trace_pci_nvme_err_startfail_cqent_too_small( - NVME_CC_IOCQES(n->bar.cc), - NVME_CTRL_CQES_MIN(n->bar.cap)); + NVME_CC_IOCQES(cc), + NVME_CTRL_CQES_MIN(cap)); return -1; } - if (unlikely(NVME_CC_IOCQES(n->bar.cc) > + if (unlikely(NVME_CC_IOCQES(cc) > NVME_CTRL_CQES_MAX(n->id_ctrl.cqes))) { trace_pci_nvme_err_startfail_cqent_too_large( - NVME_CC_IOCQES(n->bar.cc), - NVME_CTRL_CQES_MAX(n->bar.cap)); + NVME_CC_IOCQES(cc), + NVME_CTRL_CQES_MAX(cap)); return -1; } - if (unlikely(NVME_CC_IOSQES(n->bar.cc) < + if (unlikely(NVME_CC_IOSQES(cc) < NVME_CTRL_SQES_MIN(n->id_ctrl.sqes))) { trace_pci_nvme_err_startfail_sqent_too_small( - NVME_CC_IOSQES(n->bar.cc), - NVME_CTRL_SQES_MIN(n->bar.cap)); + NVME_CC_IOSQES(cc), + NVME_CTRL_SQES_MIN(cap)); return -1; } - if (unlikely(NVME_CC_IOSQES(n->bar.cc) > + if (unlikely(NVME_CC_IOSQES(cc) > NVME_CTRL_SQES_MAX(n->id_ctrl.sqes))) { trace_pci_nvme_err_startfail_sqent_too_large( - NVME_CC_IOSQES(n->bar.cc), - NVME_CTRL_SQES_MAX(n->bar.cap)); + NVME_CC_IOSQES(cc), + NVME_CTRL_SQES_MAX(cap)); return -1; } - if (unlikely(!NVME_AQA_ASQS(n->bar.aqa))) { + if (unlikely(!NVME_AQA_ASQS(aqa))) { trace_pci_nvme_err_startfail_asqent_sz_zero(); return -1; } - if (unlikely(!NVME_AQA_ACQS(n->bar.aqa))) { + if (unlikely(!NVME_AQA_ACQS(aqa))) { trace_pci_nvme_err_startfail_acqent_sz_zero(); return -1; } @@ -5677,12 +5696,10 @@ static int nvme_start_ctrl(NvmeCtrl *n) n->page_bits = page_bits; n->page_size = page_size; n->max_prp_ents = n->page_size / sizeof(uint64_t); - n->cqe_size = 1 << NVME_CC_IOCQES(n->bar.cc); - n->sqe_size = 1 << NVME_CC_IOSQES(n->bar.cc); - nvme_init_cq(&n->admin_cq, n, n->bar.acq, 0, 0, - NVME_AQA_ACQS(n->bar.aqa) + 1, 1); - nvme_init_sq(&n->admin_sq, n, n->bar.asq, 0, 0, - NVME_AQA_ASQS(n->bar.aqa) + 1); + n->cqe_size = 1 << NVME_CC_IOCQES(cc); + n->sqe_size = 1 << NVME_CC_IOSQES(cc); + nvme_init_cq(&n->admin_cq, n, acq, 0, 0, NVME_AQA_ACQS(aqa) + 1, 1); + nvme_init_sq(&n->admin_sq, n, asq, 0, 0, NVME_AQA_ASQS(aqa) + 1); nvme_set_timestamp(n, 0ULL); @@ -5695,22 +5712,33 @@ static int nvme_start_ctrl(NvmeCtrl *n) static void nvme_cmb_enable_regs(NvmeCtrl *n) { - NVME_CMBLOC_SET_CDPCILS(n->bar.cmbloc, 1); - NVME_CMBLOC_SET_CDPMLS(n->bar.cmbloc, 1); - NVME_CMBLOC_SET_BIR(n->bar.cmbloc, NVME_CMB_BIR); + uint32_t cmbloc = ldl_le_p(&n->bar.cmbloc); + uint32_t cmbsz = ldl_le_p(&n->bar.cmbsz); + + NVME_CMBLOC_SET_CDPCILS(cmbloc, 1); + NVME_CMBLOC_SET_CDPMLS(cmbloc, 1); + NVME_CMBLOC_SET_BIR(cmbloc, NVME_CMB_BIR); + stl_le_p(&n->bar.cmbloc, cmbloc); - NVME_CMBSZ_SET_SQS(n->bar.cmbsz, 1); - NVME_CMBSZ_SET_CQS(n->bar.cmbsz, 0); - NVME_CMBSZ_SET_LISTS(n->bar.cmbsz, 1); - NVME_CMBSZ_SET_RDS(n->bar.cmbsz, 1); - NVME_CMBSZ_SET_WDS(n->bar.cmbsz, 1); - NVME_CMBSZ_SET_SZU(n->bar.cmbsz, 2); /* MBs */ - NVME_CMBSZ_SET_SZ(n->bar.cmbsz, n->params.cmb_size_mb); + NVME_CMBSZ_SET_SQS(cmbsz, 1); + NVME_CMBSZ_SET_CQS(cmbsz, 0); + NVME_CMBSZ_SET_LISTS(cmbsz, 1); + NVME_CMBSZ_SET_RDS(cmbsz, 1); + NVME_CMBSZ_SET_WDS(cmbsz, 1); + NVME_CMBSZ_SET_SZU(cmbsz, 2); /* MBs */ + NVME_CMBSZ_SET_SZ(cmbsz, n->params.cmb_size_mb); + stl_le_p(&n->bar.cmbsz, cmbsz); } static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, uint64_t data, unsigned size) { + uint64_t cap = ldq_le_p(&n->bar.cap); + uint32_t cc = ldl_le_p(&n->bar.cc); + uint32_t intms = ldl_le_p(&n->bar.intms); + uint32_t csts = ldl_le_p(&n->bar.csts); + uint32_t pmrsts = ldl_le_p(&n->bar.pmrsts); + if (unlikely(offset & (sizeof(uint32_t) - 1))) { NVME_GUEST_ERR(pci_nvme_ub_mmiowr_misaligned32, "MMIO write not 32-bit aligned," @@ -5727,65 +5755,77 @@ static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, uint64_t data, } switch (offset) { - case 0xc: /* INTMS */ + case NVME_REG_INTMS: if (unlikely(msix_enabled(&(n->parent_obj)))) { NVME_GUEST_ERR(pci_nvme_ub_mmiowr_intmask_with_msix, "undefined access to interrupt mask set" " when MSI-X is enabled"); /* should be ignored, fall through for now */ } - n->bar.intms |= data & 0xffffffff; + intms |= data; + stl_le_p(&n->bar.intms, intms); n->bar.intmc = n->bar.intms; - trace_pci_nvme_mmio_intm_set(data & 0xffffffff, n->bar.intmc); + trace_pci_nvme_mmio_intm_set(data & 0xffffffff, intms); nvme_irq_check(n); break; - case 0x10: /* INTMC */ + case NVME_REG_INTMC: if (unlikely(msix_enabled(&(n->parent_obj)))) { NVME_GUEST_ERR(pci_nvme_ub_mmiowr_intmask_with_msix, "undefined access to interrupt mask clr" " when MSI-X is enabled"); /* should be ignored, fall through for now */ } - n->bar.intms &= ~(data & 0xffffffff); + intms &= ~data; + stl_le_p(&n->bar.intms, intms); n->bar.intmc = n->bar.intms; - trace_pci_nvme_mmio_intm_clr(data & 0xffffffff, n->bar.intmc); + trace_pci_nvme_mmio_intm_clr(data & 0xffffffff, intms); nvme_irq_check(n); break; - case 0x14: /* CC */ + case NVME_REG_CC: trace_pci_nvme_mmio_cfg(data & 0xffffffff); + /* Windows first sends data, then sends enable bit */ - if (!NVME_CC_EN(data) && !NVME_CC_EN(n->bar.cc) && - !NVME_CC_SHN(data) && !NVME_CC_SHN(n->bar.cc)) + if (!NVME_CC_EN(data) && !NVME_CC_EN(cc) && + !NVME_CC_SHN(data) && !NVME_CC_SHN(cc)) { - n->bar.cc = data; + cc = data; } - if (NVME_CC_EN(data) && !NVME_CC_EN(n->bar.cc)) { - n->bar.cc = data; + if (NVME_CC_EN(data) && !NVME_CC_EN(cc)) { + cc = data; + + /* flush CC since nvme_start_ctrl() needs the value */ + stl_le_p(&n->bar.cc, cc); if (unlikely(nvme_start_ctrl(n))) { trace_pci_nvme_err_startfail(); - n->bar.csts = NVME_CSTS_FAILED; + csts = NVME_CSTS_FAILED; } else { trace_pci_nvme_mmio_start_success(); - n->bar.csts = NVME_CSTS_READY; + csts = NVME_CSTS_READY; } - } else if (!NVME_CC_EN(data) && NVME_CC_EN(n->bar.cc)) { + } else if (!NVME_CC_EN(data) && NVME_CC_EN(cc)) { trace_pci_nvme_mmio_stopped(); nvme_ctrl_reset(n); - n->bar.csts &= ~NVME_CSTS_READY; + cc = 0; + csts &= ~NVME_CSTS_READY; } - if (NVME_CC_SHN(data) && !(NVME_CC_SHN(n->bar.cc))) { + + if (NVME_CC_SHN(data) && !(NVME_CC_SHN(cc))) { trace_pci_nvme_mmio_shutdown_set(); nvme_ctrl_shutdown(n); - n->bar.cc = data; - n->bar.csts |= NVME_CSTS_SHST_COMPLETE; - } else if (!NVME_CC_SHN(data) && NVME_CC_SHN(n->bar.cc)) { + cc = data; + csts |= NVME_CSTS_SHST_COMPLETE; + } else if (!NVME_CC_SHN(data) && NVME_CC_SHN(cc)) { trace_pci_nvme_mmio_shutdown_cleared(); - n->bar.csts &= ~NVME_CSTS_SHST_COMPLETE; - n->bar.cc = data; + csts &= ~NVME_CSTS_SHST_COMPLETE; + cc = data; } + + stl_le_p(&n->bar.cc, cc); + stl_le_p(&n->bar.csts, csts); + break; - case 0x1c: /* CSTS */ + case NVME_REG_CSTS: if (data & (1 << 4)) { NVME_GUEST_ERR(pci_nvme_ub_mmiowr_ssreset_w1c_unsupported, "attempted to W1C CSTS.NSSRO" @@ -5796,7 +5836,7 @@ static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, uint64_t data, " of controller status"); } break; - case 0x20: /* NSSR */ + case NVME_REG_NSSR: if (data == 0x4e564d65) { trace_pci_nvme_ub_mmiowr_ssreset_unsupported(); } else { @@ -5804,53 +5844,53 @@ static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, uint64_t data, return; } break; - case 0x24: /* AQA */ - n->bar.aqa = data & 0xffffffff; + case NVME_REG_AQA: + stl_le_p(&n->bar.aqa, data); trace_pci_nvme_mmio_aqattr(data & 0xffffffff); break; - case 0x28: /* ASQ */ - n->bar.asq = size == 8 ? data : - (n->bar.asq & ~0xffffffffULL) | (data & 0xffffffff); + case NVME_REG_ASQ: + stn_le_p(&n->bar.asq, size, data); trace_pci_nvme_mmio_asqaddr(data); break; - case 0x2c: /* ASQ hi */ - n->bar.asq = (n->bar.asq & 0xffffffff) | (data << 32); - trace_pci_nvme_mmio_asqaddr_hi(data, n->bar.asq); + case NVME_REG_ASQ + 4: + stl_le_p((uint8_t *)&n->bar.asq + 4, data); + trace_pci_nvme_mmio_asqaddr_hi(data, ldq_le_p(&n->bar.asq)); break; - case 0x30: /* ACQ */ + case NVME_REG_ACQ: trace_pci_nvme_mmio_acqaddr(data); - n->bar.acq = size == 8 ? data : - (n->bar.acq & ~0xffffffffULL) | (data & 0xffffffff); + stn_le_p(&n->bar.acq, size, data); break; - case 0x34: /* ACQ hi */ - n->bar.acq = (n->bar.acq & 0xffffffff) | (data << 32); - trace_pci_nvme_mmio_acqaddr_hi(data, n->bar.acq); + case NVME_REG_ACQ + 4: + stl_le_p((uint8_t *)&n->bar.acq + 4, data); + trace_pci_nvme_mmio_acqaddr_hi(data, ldq_le_p(&n->bar.acq)); break; - case 0x38: /* CMBLOC */ + case NVME_REG_CMBLOC: NVME_GUEST_ERR(pci_nvme_ub_mmiowr_cmbloc_reserved, "invalid write to reserved CMBLOC" " when CMBSZ is zero, ignored"); return; - case 0x3C: /* CMBSZ */ + case NVME_REG_CMBSZ: NVME_GUEST_ERR(pci_nvme_ub_mmiowr_cmbsz_readonly, "invalid write to read only CMBSZ, ignored"); return; - case 0x50: /* CMBMSC */ - if (!NVME_CAP_CMBS(n->bar.cap)) { + case NVME_REG_CMBMSC: + if (!NVME_CAP_CMBS(cap)) { return; } - n->bar.cmbmsc = size == 8 ? data : - (n->bar.cmbmsc & ~0xffffffff) | (data & 0xffffffff); + stn_le_p(&n->bar.cmbmsc, size, data); n->cmb.cmse = false; if (NVME_CMBMSC_CRE(data)) { nvme_cmb_enable_regs(n); if (NVME_CMBMSC_CMSE(data)) { - hwaddr cba = NVME_CMBMSC_CBA(data) << CMBMSC_CBA_SHIFT; + uint64_t cmbmsc = ldq_le_p(&n->bar.cmbmsc); + hwaddr cba = NVME_CMBMSC_CBA(cmbmsc) << CMBMSC_CBA_SHIFT; if (cba + int128_get64(n->cmb.mem.size) < cba) { - NVME_CMBSTS_SET_CBAI(n->bar.cmbsts, 1); + uint32_t cmbsts = ldl_le_p(&n->bar.cmbsts); + NVME_CMBSTS_SET_CBAI(cmbsts, 1); + stl_le_p(&n->bar.cmbsts, cmbsts); return; } @@ -5863,53 +5903,57 @@ static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, uint64_t data, } return; - case 0x54: /* CMBMSC hi */ - n->bar.cmbmsc = (n->bar.cmbmsc & 0xffffffff) | (data << 32); + case NVME_REG_CMBMSC + 4: + stl_le_p((uint8_t *)&n->bar.cmbmsc + 4, data); return; - case 0xe00: /* PMRCAP */ + case NVME_REG_PMRCAP: NVME_GUEST_ERR(pci_nvme_ub_mmiowr_pmrcap_readonly, "invalid write to PMRCAP register, ignored"); return; - case 0xe04: /* PMRCTL */ - if (!NVME_CAP_PMRS(n->bar.cap)) { + case NVME_REG_PMRCTL: + if (!NVME_CAP_PMRS(cap)) { return; } - n->bar.pmrctl = data; + stl_le_p(&n->bar.pmrctl, data); if (NVME_PMRCTL_EN(data)) { memory_region_set_enabled(&n->pmr.dev->mr, true); - n->bar.pmrsts = 0; + pmrsts = 0; } else { memory_region_set_enabled(&n->pmr.dev->mr, false); - NVME_PMRSTS_SET_NRDY(n->bar.pmrsts, 1); + NVME_PMRSTS_SET_NRDY(pmrsts, 1); n->pmr.cmse = false; } + stl_le_p(&n->bar.pmrsts, pmrsts); return; - case 0xe08: /* PMRSTS */ + case NVME_REG_PMRSTS: NVME_GUEST_ERR(pci_nvme_ub_mmiowr_pmrsts_readonly, "invalid write to PMRSTS register, ignored"); return; - case 0xe0C: /* PMREBS */ + case NVME_REG_PMREBS: NVME_GUEST_ERR(pci_nvme_ub_mmiowr_pmrebs_readonly, "invalid write to PMREBS register, ignored"); return; - case 0xe10: /* PMRSWTP */ + case NVME_REG_PMRSWTP: NVME_GUEST_ERR(pci_nvme_ub_mmiowr_pmrswtp_readonly, "invalid write to PMRSWTP register, ignored"); return; - case 0xe14: /* PMRMSCL */ - if (!NVME_CAP_PMRS(n->bar.cap)) { + case NVME_REG_PMRMSCL: + if (!NVME_CAP_PMRS(cap)) { return; } - n->bar.pmrmsc = (n->bar.pmrmsc & ~0xffffffff) | (data & 0xffffffff); + stl_le_p(&n->bar.pmrmscl, data); n->pmr.cmse = false; - if (NVME_PMRMSC_CMSE(n->bar.pmrmsc)) { - hwaddr cba = NVME_PMRMSC_CBA(n->bar.pmrmsc) << PMRMSC_CBA_SHIFT; + if (NVME_PMRMSCL_CMSE(data)) { + uint64_t pmrmscu = ldl_le_p(&n->bar.pmrmscu); + hwaddr cba = pmrmscu << 32 | + (NVME_PMRMSCL_CBA(data) << PMRMSCL_CBA_SHIFT); if (cba + int128_get64(n->pmr.dev->mr.size) < cba) { - NVME_PMRSTS_SET_CBAI(n->bar.pmrsts, 1); + NVME_PMRSTS_SET_CBAI(pmrsts, 1); + stl_le_p(&n->bar.pmrsts, pmrsts); return; } @@ -5918,12 +5962,12 @@ static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, uint64_t data, } return; - case 0xe18: /* PMRMSCU */ - if (!NVME_CAP_PMRS(n->bar.cap)) { + case NVME_REG_PMRMSCU: + if (!NVME_CAP_PMRS(cap)) { return; } - n->bar.pmrmsc = (n->bar.pmrmsc & 0xffffffff) | (data << 32); + stl_le_p(&n->bar.pmrmscu, data); return; default: NVME_GUEST_ERR(pci_nvme_ub_mmiowr_invalid, @@ -5938,7 +5982,6 @@ static uint64_t nvme_mmio_read(void *opaque, hwaddr addr, unsigned size) { NvmeCtrl *n = (NvmeCtrl *)opaque; uint8_t *ptr = (uint8_t *)&n->bar; - uint64_t val = 0; trace_pci_nvme_mmio_read(addr, size); @@ -5954,24 +5997,25 @@ static uint64_t nvme_mmio_read(void *opaque, hwaddr addr, unsigned size) /* should RAZ, fall through for now */ } - if (addr < sizeof(n->bar)) { - /* - * When PMRWBM bit 1 is set then read from - * from PMRSTS should ensure prior writes - * made it to persistent media - */ - if (addr == 0xe08 && - (NVME_PMRCAP_PMRWBM(n->bar.pmrcap) & 0x02)) { - memory_region_msync(&n->pmr.dev->mr, 0, n->pmr.dev->size); - } - memcpy(&val, ptr + addr, size); - } else { + if (addr > sizeof(n->bar) - size) { NVME_GUEST_ERR(pci_nvme_ub_mmiord_invalid_ofs, "MMIO read beyond last register," " offset=0x%"PRIx64", returning 0", addr); + + return 0; + } + + /* + * When PMRWBM bit 1 is set then read from + * from PMRSTS should ensure prior writes + * made it to persistent media + */ + if (addr == NVME_REG_PMRSTS && + (NVME_PMRCAP_PMRWBM(ldl_le_p(&n->bar.pmrcap)) & 0x02)) { + memory_region_msync(&n->pmr.dev->mr, 0, n->pmr.dev->size); } - return val; + return ldn_le_p(ptr + addr, size); } static void nvme_process_db(NvmeCtrl *n, hwaddr addr, int val) @@ -6229,6 +6273,7 @@ static void nvme_init_state(NvmeCtrl *n) static void nvme_init_cmb(NvmeCtrl *n, PCIDevice *pci_dev) { uint64_t cmb_size = n->params.cmb_size_mb * MiB; + uint64_t cap = ldq_le_p(&n->bar.cap); n->cmb.buf = g_malloc0(cmb_size); memory_region_init_io(&n->cmb.mem, OBJECT(n), &nvme_cmb_ops, n, @@ -6238,7 +6283,8 @@ static void nvme_init_cmb(NvmeCtrl *n, PCIDevice *pci_dev) PCI_BASE_ADDRESS_MEM_TYPE_64 | PCI_BASE_ADDRESS_MEM_PREFETCH, &n->cmb.mem); - NVME_CAP_SET_CMBS(n->bar.cap, 1); + NVME_CAP_SET_CMBS(cap, 1); + stq_le_p(&n->bar.cap, cap); if (n->params.legacy_cmb) { nvme_cmb_enable_regs(n); @@ -6248,14 +6294,17 @@ static void nvme_init_cmb(NvmeCtrl *n, PCIDevice *pci_dev) static void nvme_init_pmr(NvmeCtrl *n, PCIDevice *pci_dev) { - NVME_PMRCAP_SET_RDS(n->bar.pmrcap, 1); - NVME_PMRCAP_SET_WDS(n->bar.pmrcap, 1); - NVME_PMRCAP_SET_BIR(n->bar.pmrcap, NVME_PMR_BIR); + uint32_t pmrcap = ldl_le_p(&n->bar.pmrcap); + + NVME_PMRCAP_SET_RDS(pmrcap, 1); + NVME_PMRCAP_SET_WDS(pmrcap, 1); + NVME_PMRCAP_SET_BIR(pmrcap, NVME_PMR_BIR); /* Turn on bit 1 support */ - NVME_PMRCAP_SET_PMRWBM(n->bar.pmrcap, 0x02); - NVME_PMRCAP_SET_CMSS(n->bar.pmrcap, 1); + NVME_PMRCAP_SET_PMRWBM(pmrcap, 0x02); + NVME_PMRCAP_SET_CMSS(pmrcap, 1); + stl_le_p(&n->bar.pmrcap, pmrcap); - pci_register_bar(pci_dev, NVME_PMRCAP_BIR(n->bar.pmrcap), + pci_register_bar(pci_dev, NVME_PMR_BIR, PCI_BASE_ADDRESS_SPACE_MEMORY | PCI_BASE_ADDRESS_MEM_TYPE_64 | PCI_BASE_ADDRESS_MEM_PREFETCH, &n->pmr.dev->mr); @@ -6345,6 +6394,7 @@ static void nvme_init_ctrl(NvmeCtrl *n, PCIDevice *pci_dev) { NvmeIdCtrl *id = &n->id_ctrl; uint8_t *pci_conf = pci_dev->config; + uint64_t cap = ldq_le_p(&n->bar.cap); id->vid = cpu_to_le16(pci_get_word(pci_conf + PCI_VENDOR_ID)); id->ssvid = cpu_to_le16(pci_get_word(pci_conf + PCI_SUBSYSTEM_VENDOR_ID)); @@ -6423,17 +6473,18 @@ static void nvme_init_ctrl(NvmeCtrl *n, PCIDevice *pci_dev) id->cmic |= NVME_CMIC_MULTI_CTRL; } - NVME_CAP_SET_MQES(n->bar.cap, 0x7ff); - NVME_CAP_SET_CQR(n->bar.cap, 1); - NVME_CAP_SET_TO(n->bar.cap, 0xf); - NVME_CAP_SET_CSS(n->bar.cap, NVME_CAP_CSS_NVM); - NVME_CAP_SET_CSS(n->bar.cap, NVME_CAP_CSS_CSI_SUPP); - NVME_CAP_SET_CSS(n->bar.cap, NVME_CAP_CSS_ADMIN_ONLY); - NVME_CAP_SET_MPSMAX(n->bar.cap, 4); - NVME_CAP_SET_CMBS(n->bar.cap, n->params.cmb_size_mb ? 1 : 0); - NVME_CAP_SET_PMRS(n->bar.cap, n->pmr.dev ? 1 : 0); + NVME_CAP_SET_MQES(cap, 0x7ff); + NVME_CAP_SET_CQR(cap, 1); + NVME_CAP_SET_TO(cap, 0xf); + NVME_CAP_SET_CSS(cap, NVME_CAP_CSS_NVM); + NVME_CAP_SET_CSS(cap, NVME_CAP_CSS_CSI_SUPP); + NVME_CAP_SET_CSS(cap, NVME_CAP_CSS_ADMIN_ONLY); + NVME_CAP_SET_MPSMAX(cap, 4); + NVME_CAP_SET_CMBS(cap, n->params.cmb_size_mb ? 1 : 0); + NVME_CAP_SET_PMRS(cap, n->pmr.dev ? 1 : 0); + stq_le_p(&n->bar.cap, cap); - n->bar.vs = NVME_SPEC_VER; + stl_le_p(&n->bar.vs, NVME_SPEC_VER); n->bar.intmc = n->bar.intms = 0; } @@ -6498,7 +6549,7 @@ static void nvme_realize(PCIDevice *pci_dev, Error **errp) ns = &n->namespace; ns->params.nsid = 1; - if (nvme_ns_setup(n, ns, errp)) { + if (nvme_ns_setup(ns, errp)) { return; } @@ -6514,13 +6565,15 @@ static void nvme_exit(PCIDevice *pci_dev) nvme_ctrl_reset(n); - for (i = 1; i <= NVME_MAX_NAMESPACES; i++) { - ns = nvme_ns(n, i); - if (!ns) { - continue; + if (n->subsys) { + for (i = 1; i <= NVME_MAX_NAMESPACES; i++) { + ns = nvme_ns(n, i); + if (ns) { + ns->attached--; + } } - nvme_ns_cleanup(ns); + nvme_subsys_unregister_ctrl(n->subsys, n); } g_free(n->cq); @@ -6582,7 +6635,7 @@ static void nvme_set_smart_warning(Object *obj, Visitor *v, const char *name, cap = NVME_SMART_SPARE | NVME_SMART_TEMPERATURE | NVME_SMART_RELIABILITY | NVME_SMART_MEDIA_READ_ONLY | NVME_SMART_FAILED_VOLATILE_MEDIA; - if (NVME_CAP_PMRS(n->bar.cap)) { + if (NVME_CAP_PMRS(ldq_le_p(&n->bar.cap))) { cap |= NVME_SMART_PMR_UNRELIABLE; } diff --git a/hw/nvme/ns.c b/hw/nvme/ns.c index 4275c3d..b7cf149 100644 --- a/hw/nvme/ns.c +++ b/hw/nvme/ns.c @@ -346,8 +346,7 @@ static void nvme_zoned_ns_shutdown(NvmeNamespace *ns) assert(ns->nr_open_zones == 0); } -static int nvme_ns_check_constraints(NvmeCtrl *n, NvmeNamespace *ns, - Error **errp) +static int nvme_ns_check_constraints(NvmeNamespace *ns, Error **errp) { if (!ns->blkconf.blk) { error_setg(errp, "block backend not configured"); @@ -366,20 +365,6 @@ static int nvme_ns_check_constraints(NvmeCtrl *n, NvmeNamespace *ns, return -1; } - if (!n->subsys) { - if (ns->params.detached) { - error_setg(errp, "detached requires that the nvme device is " - "linked to an nvme-subsys device"); - return -1; - } - - if (ns->params.shared) { - error_setg(errp, "shared requires that the nvme device is " - "linked to an nvme-subsys device"); - return -1; - } - } - if (ns->params.zoned) { if (ns->params.max_active_zones) { if (ns->params.max_open_zones > ns->params.max_active_zones) { @@ -411,9 +396,9 @@ static int nvme_ns_check_constraints(NvmeCtrl *n, NvmeNamespace *ns, return 0; } -int nvme_ns_setup(NvmeCtrl *n, NvmeNamespace *ns, Error **errp) +int nvme_ns_setup(NvmeNamespace *ns, Error **errp) { - if (nvme_ns_check_constraints(n, ns, errp)) { + if (nvme_ns_check_constraints(ns, errp)) { return -1; } @@ -456,6 +441,15 @@ void nvme_ns_cleanup(NvmeNamespace *ns) } } +static void nvme_ns_unrealize(DeviceState *dev) +{ + NvmeNamespace *ns = NVME_NS(dev); + + nvme_ns_drain(ns); + nvme_ns_shutdown(ns); + nvme_ns_cleanup(ns); +} + static void nvme_ns_realize(DeviceState *dev, Error **errp) { NvmeNamespace *ns = NVME_NS(dev); @@ -465,7 +459,29 @@ static void nvme_ns_realize(DeviceState *dev, Error **errp) uint32_t nsid = ns->params.nsid; int i; - if (nvme_ns_setup(n, ns, errp)) { + if (!n->subsys) { + if (ns->params.detached) { + error_setg(errp, "detached requires that the nvme device is " + "linked to an nvme-subsys device"); + return; + } + + if (ns->params.shared) { + error_setg(errp, "shared requires that the nvme device is " + "linked to an nvme-subsys device"); + return; + } + } else { + /* + * If this namespace belongs to a subsystem (through a link on the + * controller device), reparent the device. + */ + if (!qdev_set_parent_bus(dev, &subsys->bus.parent_bus, errp)) { + return; + } + } + + if (nvme_ns_setup(ns, errp)) { return; } @@ -553,6 +569,7 @@ static void nvme_ns_class_init(ObjectClass *oc, void *data) dc->bus_type = TYPE_NVME_BUS; dc->realize = nvme_ns_realize; + dc->unrealize = nvme_ns_unrealize; device_class_set_props(dc, nvme_ns_props); dc->desc = "Virtual NVMe namespace"; } diff --git a/hw/nvme/nvme.h b/hw/nvme/nvme.h index 56f8ece..83ffaba 100644 --- a/hw/nvme/nvme.h +++ b/hw/nvme/nvme.h @@ -33,12 +33,20 @@ QEMU_BUILD_BUG_ON(NVME_MAX_NAMESPACES > NVME_NSID_BROADCAST - 1); typedef struct NvmeCtrl NvmeCtrl; typedef struct NvmeNamespace NvmeNamespace; +#define TYPE_NVME_BUS "nvme-bus" +OBJECT_DECLARE_SIMPLE_TYPE(NvmeBus, NVME_BUS) + +typedef struct NvmeBus { + BusState parent_bus; +} NvmeBus; + #define TYPE_NVME_SUBSYS "nvme-subsys" #define NVME_SUBSYS(obj) \ OBJECT_CHECK(NvmeSubsystem, (obj), TYPE_NVME_SUBSYS) typedef struct NvmeSubsystem { DeviceState parent_obj; + NvmeBus bus; uint8_t subnqn[256]; NvmeCtrl *ctrls[NVME_MAX_CONTROLLERS]; @@ -50,6 +58,7 @@ typedef struct NvmeSubsystem { } NvmeSubsystem; int nvme_subsys_register_ctrl(NvmeCtrl *n, Error **errp); +void nvme_subsys_unregister_ctrl(NvmeSubsystem *subsys, NvmeCtrl *n); static inline NvmeCtrl *nvme_subsys_ctrl(NvmeSubsystem *subsys, uint32_t cntlid) @@ -246,7 +255,7 @@ static inline void nvme_aor_dec_active(NvmeNamespace *ns) } void nvme_ns_init_format(NvmeNamespace *ns); -int nvme_ns_setup(NvmeCtrl *n, NvmeNamespace *ns, Error **errp); +int nvme_ns_setup(NvmeNamespace *ns, Error **errp); void nvme_ns_drain(NvmeNamespace *ns); void nvme_ns_shutdown(NvmeNamespace *ns); void nvme_ns_cleanup(NvmeNamespace *ns); @@ -364,13 +373,6 @@ typedef struct NvmeCQueue { QTAILQ_HEAD(, NvmeRequest) req_list; } NvmeCQueue; -#define TYPE_NVME_BUS "nvme-bus" -#define NVME_BUS(obj) OBJECT_CHECK(NvmeBus, (obj), TYPE_NVME_BUS) - -typedef struct NvmeBus { - BusState parent_bus; -} NvmeBus; - #define TYPE_NVME "nvme" #define NVME(obj) \ OBJECT_CHECK(NvmeCtrl, (obj), TYPE_NVME) diff --git a/hw/nvme/subsys.c b/hw/nvme/subsys.c index 192223d..93c3595 100644 --- a/hw/nvme/subsys.c +++ b/hw/nvme/subsys.c @@ -32,6 +32,11 @@ int nvme_subsys_register_ctrl(NvmeCtrl *n, Error **errp) return cntlid; } +void nvme_subsys_unregister_ctrl(NvmeSubsystem *subsys, NvmeCtrl *n) +{ + subsys->ctrls[n->cntlid] = NULL; +} + static void nvme_subsys_setup(NvmeSubsystem *subsys) { const char *nqn = subsys->params.nqn ? @@ -45,6 +50,9 @@ static void nvme_subsys_realize(DeviceState *dev, Error **errp) { NvmeSubsystem *subsys = NVME_SUBSYS(dev); + qbus_create_inplace(&subsys->bus, sizeof(NvmeBus), TYPE_NVME_BUS, dev, + dev->id); + nvme_subsys_setup(subsys); } @@ -61,6 +69,7 @@ static void nvme_subsys_class_init(ObjectClass *oc, void *data) dc->realize = nvme_subsys_realize; dc->desc = "Virtual NVMe subsystem"; + dc->hotpluggable = false; device_class_set_props(dc, nvme_subsystem_props); } diff --git a/hw/nvme/trace-events b/hw/nvme/trace-events index f9a1f14..430eeb3 100644 --- a/hw/nvme/trace-events +++ b/hw/nvme/trace-events @@ -199,3 +199,4 @@ pci_nvme_ub_db_wr_invalid_cqhead(uint32_t qid, uint16_t new_head) "completion qu pci_nvme_ub_db_wr_invalid_sq(uint32_t qid) "submission queue doorbell write for nonexistent queue, sqid=%"PRIu32", ignoring" pci_nvme_ub_db_wr_invalid_sqtail(uint32_t qid, uint16_t new_tail) "submission queue doorbell write value beyond queue size, sqid=%"PRIu32", new_head=%"PRIu16", ignoring" pci_nvme_ub_unknown_css_value(void) "unknown value in cc.css field" +pci_nvme_ub_too_many_mappings(void) "too many prp/sgl mappings" |