[PATCH 8/8] hw/block/nvme: add PMR RDS/WDS support

[Klaus Jensen]

From: Naveen Nagar <naveen.n1@samsung.com>

Add support for the PMRMSCL and PMRMSCU MMIO registers. This allows
adding RDS/WDS support for PMR as well.

Signed-off-by: Naveen Nagar <naveen.n1@samsung.com>
Signed-off-by: Klaus Jensen <k.jensen@samsung.com>
---
 hw/block/nvme.h      |  1 +
 include/block/nvme.h |  1 +
 hw/block/nvme.c      | 88 ++++++++++++++++++++++++++++++++++++++++++--
 3 files changed, 87 insertions(+), 3 deletions(-)

diff --git a/hw/block/nvme.h b/hw/block/nvme.h
index b1b273ce535e..0dc119dedc00 100644
--- a/hw/block/nvme.h
+++ b/hw/block/nvme.h
@@ -142,6 +142,7 @@ typedef struct NvmeCtrl {
     uint64_t    timestamp_set_qemu_clock_ms;    /* QEMU clock time */
     uint64_t    starttime_ms;
     uint16_t    temperature;
+    bool        pmr_cmse;
 
     HostMemoryBackend *pmrdev;
 
diff --git a/include/block/nvme.h b/include/block/nvme.h
index 686e2541a587..ed645b6cb4f2 100644
--- a/include/block/nvme.h
+++ b/include/block/nvme.h
@@ -62,6 +62,7 @@ enum NvmeCapMask {
 #define NVME_CAP_CSS(cap)   (((cap) >> CAP_CSS_SHIFT)    & CAP_CSS_MASK)
 #define NVME_CAP_MPSMIN(cap)(((cap) >> CAP_MPSMIN_SHIFT) & CAP_MPSMIN_MASK)
 #define NVME_CAP_MPSMAX(cap)(((cap) >> CAP_MPSMAX_SHIFT) & CAP_MPSMAX_MASK)
+#define NVME_CAP_PMRS(cap)  (((cap) >> CAP_PMRS_SHIFT)   & CAP_PMRS_MASK)
 
 #define NVME_CAP_SET_MQES(cap, val)   (cap |= (uint64_t)(val & CAP_MQES_MASK)  
\
                                                            << CAP_MQES_SHIFT)
diff --git a/hw/block/nvme.c b/hw/block/nvme.c
index c04d6e69e4df..055b9c69bafb 100644
--- a/hw/block/nvme.c
+++ b/hw/block/nvme.c
@@ -142,6 +142,31 @@ static inline void *nvme_addr_to_cmb(NvmeCtrl *n, hwaddr 
addr)
     return &n->cmbuf[addr - n->ctrl_mem.addr];
 }
 
+static inline bool nvme_addr_is_pmr(NvmeCtrl *n, hwaddr addr)
+{
+    hwaddr hi, low;
+
+    if (!n->pmrdev || !n->pmrdev->mr.enabled) {
+        return false;
+    }
+
+    low = NVME_PMRMSC_CBA(n->bar.pmrmsc) << PMRMSC_CBA_SHIFT;
+    hi  = low + int128_get64(n->pmrdev->mr.size);
+
+    return addr >= low && addr < hi;
+}
+
+static inline void *nvme_addr_to_pmr(NvmeCtrl *n, hwaddr addr)
+{
+    hwaddr cba;
+
+    assert(nvme_addr_is_pmr(n, addr));
+
+    cba = NVME_PMRMSC_CBA(n->bar.pmrmsc) << PMRMSC_CBA_SHIFT;
+
+    return memory_region_get_ram_ptr(&n->pmrdev->mr) + (addr - cba);
+}
+
 static int nvme_addr_read(NvmeCtrl *n, hwaddr addr, void *buf, int size)
 {
     hwaddr hi = addr + size - 1;
@@ -154,6 +179,11 @@ static int nvme_addr_read(NvmeCtrl *n, hwaddr addr, void 
*buf, int size)
         return 0;
     }
 
+    if (nvme_addr_is_pmr(n, addr) && nvme_addr_is_pmr(n, hi)) {
+        memcpy(buf, nvme_addr_to_pmr(n, addr), size);
+        return 0;
+    }
+
     return pci_dma_read(&n->parent_obj, addr, buf, size);
 }
 
@@ -275,6 +305,22 @@ static uint16_t nvme_map_addr_cmb(NvmeCtrl *n, 
QEMUIOVector *iov, hwaddr addr,
     return NVME_SUCCESS;
 }
 
+static uint16_t nvme_map_addr_pmr(NvmeCtrl *n, QEMUIOVector *iov, hwaddr addr,
+    size_t len)
+{
+    if (!len) {
+        return NVME_SUCCESS;
+    }
+
+    if (!nvme_addr_is_pmr(n, addr) || !nvme_addr_is_pmr(n, addr + len - 1)) {
+        return NVME_DATA_TRAS_ERROR;
+    }
+
+    qemu_iovec_add(iov, nvme_addr_to_pmr(n, addr), len);
+
+    return NVME_SUCCESS;
+}
+
 static uint16_t nvme_map_addr(NvmeCtrl *n, QEMUSGList *qsg, QEMUIOVector *iov,
                               hwaddr addr, size_t len)
 {
@@ -297,6 +343,19 @@ static uint16_t nvme_map_addr(NvmeCtrl *n, QEMUSGList 
*qsg, QEMUIOVector *iov,
 
         return nvme_map_addr_cmb(n, iov, addr, len);
     }
+    if (nvme_addr_is_pmr(n, addr)) {
+        if (qsg && qsg->sg) {
+            return NVME_INVALID_FIELD | NVME_DNR;
+        }
+
+        assert(iov);
+
+        if (!iov->iov) {
+            qemu_iovec_init(iov, 1);
+        }
+
+        return nvme_map_addr_pmr(n, iov, addr, len);
+    }
 
     if (iov && iov->iov) {
         return NVME_INVALID_USE_OF_CMB | NVME_DNR;
@@ -328,7 +387,7 @@ static uint16_t nvme_map_prp(NvmeCtrl *n, uint64_t prp1, 
uint64_t prp2,
 
     trace_pci_nvme_map_prp(trans_len, len, prp1, prp2, num_prps);
 
-    if (nvme_addr_is_cmb(n, prp1)) {
+    if (nvme_addr_is_cmb(n, prp1) || (nvme_addr_is_pmr(n, prp1))) {
         qemu_iovec_init(iov, num_prps);
     } else {
         pci_dma_sglist_init(qsg, &n->parent_obj, num_prps);
@@ -2571,8 +2630,28 @@ static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, 
uint64_t data,
         NVME_GUEST_ERR(pci_nvme_ub_mmiowr_pmrswtp_readonly,
                        "invalid write to PMRSWTP register, ignored");
         return;
-    case 0xE14: /* TODO PMRMSC */
-        break;
+    case 0xE14: /* PMRMSCL */
+        if (!NVME_CAP_PMRS(n->bar.cap)) {
+            return;
+        }
+
+        n->bar.pmrmsc |= data & 0xffffffff;
+
+        if (NVME_PMRMSC_CMSE(n->bar.pmrmsc)) {
+            hwaddr cba = NVME_PMRMSC_CBA(n->bar.pmrmsc) << PMRMSC_CBA_SHIFT;
+            if (cba + int128_get64(n->pmrdev->mr.size) < cba) {
+                NVME_PMRSTS_SET_CBAI(n->bar.pmrsts, 1);
+            }
+        }
+
+        return;
+    case 0xE18: /* PMRMSCU */
+        if (!NVME_CAP_PMRS(n->bar.cap)) {
+            return;
+        }
+
+        n->bar.pmrmsc |= data << 32;
+        return;
     default:
         NVME_GUEST_ERR(pci_nvme_ub_mmiowr_invalid,
                        "invalid MMIO write,"
@@ -2919,8 +2998,11 @@ 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);
     NVME_PMRCAP_SET_PMRWBM(n->bar.pmrcap, 0x02);
+    NVME_PMRCAP_SET_CMSS(n->bar.pmrcap, 1);
 
     pci_register_bar(pci_dev, NVME_PMRCAP_BIR(n->bar.pmrcap),
                      PCI_BASE_ADDRESS_SPACE_MEMORY |
-- 
2.29.2