[PATCH v3 9/9] hw/block/nvme: allow open to close zone transitions by controller

[Klaus Jensen <its@irrelevant.dk>]

From: Klaus Jensen <k.jensen@samsung.com>

Allow the controller to release open resources by transitioning
implicitly and explicitly opened zones to closed. This is done using a
naive "least recently opened" strategy.

Signed-off-by: Klaus Jensen <k.jensen@samsung.com>
---
 hw/block/nvme-ns.h    |  5 ++++
 hw/block/nvme-ns.c    |  5 ++++
 hw/block/nvme.c       | 57 ++++++++++++++++++++++++++++++++++++++++---
 hw/block/trace-events |  1 +
 4 files changed, 65 insertions(+), 3 deletions(-)

diff --git a/hw/block/nvme-ns.h b/hw/block/nvme-ns.h
index 3d0269eef6f0..5d8523c047d8 100644
--- a/hw/block/nvme-ns.h
+++ b/hw/block/nvme-ns.h
@@ -38,6 +38,8 @@ typedef struct NvmeZone {
     uint8_t            *zde;
 
     uint64_t wp_staging;
+
+    QTAILQ_ENTRY(NvmeZone) lru_entry;
 } NvmeZone;
 
 typedef struct NvmeNamespace {
@@ -64,6 +66,9 @@ typedef struct NvmeNamespace {
         struct {
             uint32_t open;
             uint32_t active;
+
+            QTAILQ_HEAD(, NvmeZone) lru_open;
+            QTAILQ_HEAD(, NvmeZone) lru_active;
         } resources;
     } zns;
 } NvmeNamespace;
diff --git a/hw/block/nvme-ns.c b/hw/block/nvme-ns.c
index a01cc5eeb445..cb8b44a78450 100644
--- a/hw/block/nvme-ns.c
+++ b/hw/block/nvme-ns.c
@@ -135,6 +135,9 @@ void nvme_ns_zns_init_zone_state(NvmeNamespace *ns)
     ns->zns.resources.open = ns->params.zns.mor != 0xffffffff ?
         ns->params.zns.mor + 1 : ns->zns.num_zones;
 
+    QTAILQ_INIT(&ns->zns.resources.lru_open);
+    QTAILQ_INIT(&ns->zns.resources.lru_active);
+
     for (int i = 0; i < ns->zns.num_zones; i++) {
         NvmeZone *zone = &ns->zns.zones[i];
         zone->zd = &ns->zns.zd[i];
@@ -158,6 +161,8 @@ void nvme_ns_zns_init_zone_state(NvmeNamespace *ns)
 
             if (ns->zns.resources.active) {
                 ns->zns.resources.active--;
+                QTAILQ_INSERT_TAIL(&ns->zns.resources.lru_active, zone,
+                                   lru_entry);
                 break;
             }
 
diff --git a/hw/block/nvme.c b/hw/block/nvme.c
index cc637b3a68e9..1fab9d69261c 100644
--- a/hw/block/nvme.c
+++ b/hw/block/nvme.c
@@ -1105,11 +1105,47 @@ static inline void nvme_zone_reset_wp(NvmeZone *zone)
     zone->wp_staging = nvme_zslba(zone);
 }
 
+static uint16_t nvme_zrm_transition(NvmeNamespace *ns, NvmeZone *zone,
+                                    NvmeZoneState to);
+
+static uint16_t nvme_zrm_release_open(NvmeNamespace *ns)
+{
+    NvmeZone *candidate;
+    NvmeZoneState zs;
+    uint16_t status;
+
+    trace_pci_nvme_zrm_release_open(ns->params.nsid);
+
+    QTAILQ_FOREACH(candidate, &ns->zns.resources.lru_open, lru_entry) {
+        zs = nvme_zs(candidate);
+
+        /* skip explicitly opened zones */
+        if (zs == NVME_ZS_ZSEO) {
+            continue;
+        }
+
+        /* skip zones that have in-flight writes */
+        if (candidate->wp_staging != nvme_wp(candidate)) {
+            continue;
+        }
+
+        status = nvme_zrm_transition(ns, candidate, NVME_ZS_ZSC);
+        if (status) {
+            return status;
+        }
+
+        return NVME_SUCCESS;
+    }
+
+    return NVME_TOO_MANY_OPEN_ZONES;
+}
+
 static uint16_t nvme_zrm_transition(NvmeNamespace *ns, NvmeZone *zone,
                                     NvmeZoneState to)
 {
     NvmeZoneState from = nvme_zs(zone);
     NvmeZoneDescriptor *zd = zone->zd;
+    uint16_t status;
 
     trace_pci_nvme_zrm_transition(ns->params.nsid, nvme_zslba(zone), from, to);
 
@@ -1131,6 +1167,7 @@ static uint16_t nvme_zrm_transition(NvmeNamespace *ns, NvmeZone *zone,
             }
 
             ns->zns.resources.active--;
+            QTAILQ_INSERT_TAIL(&ns->zns.resources.lru_active, zone, lru_entry);
 
             break;
 
@@ -1141,11 +1178,15 @@ static uint16_t nvme_zrm_transition(NvmeNamespace *ns, NvmeZone *zone,
             }
 
             if (!ns->zns.resources.open) {
-                return NVME_TOO_MANY_OPEN_ZONES;
+                status = nvme_zrm_release_open(ns);
+                if (status) {
+                    return status;
+                }
             }
 
             ns->zns.resources.active--;
             ns->zns.resources.open--;
+            QTAILQ_INSERT_TAIL(&ns->zns.resources.lru_open, zone, lru_entry);
 
             break;
 
@@ -1172,11 +1213,15 @@ static uint16_t nvme_zrm_transition(NvmeNamespace *ns, NvmeZone *zone,
         case NVME_ZS_ZSF:
         case NVME_ZS_ZSRO:
             ns->zns.resources.active++;
+            ns->zns.resources.open++;
+            QTAILQ_REMOVE(&ns->zns.resources.lru_open, zone, lru_entry);
 
-            /* fallthrough */
+            break;
 
         case NVME_ZS_ZSC:
             ns->zns.resources.open++;
+            QTAILQ_REMOVE(&ns->zns.resources.lru_open, zone, lru_entry);
+            QTAILQ_INSERT_TAIL(&ns->zns.resources.lru_active, zone, lru_entry);
 
             break;
 
@@ -1201,16 +1246,22 @@ static uint16_t nvme_zrm_transition(NvmeNamespace *ns, NvmeZone *zone,
         case NVME_ZS_ZSF:
         case NVME_ZS_ZSRO:
             ns->zns.resources.active++;
+            QTAILQ_REMOVE(&ns->zns.resources.lru_active, zone, lru_entry);
 
             break;
 
         case NVME_ZS_ZSIO:
         case NVME_ZS_ZSEO:
             if (!ns->zns.resources.open) {
-                return NVME_TOO_MANY_OPEN_ZONES;
+                status = nvme_zrm_release_open(ns);
+                if (status) {
+                    return status;
+                }
             }
 
             ns->zns.resources.open--;
+            QTAILQ_REMOVE(&ns->zns.resources.lru_active, zone, lru_entry);
+            QTAILQ_INSERT_TAIL(&ns->zns.resources.lru_open, zone, lru_entry);
 
             break;
 
diff --git a/hw/block/trace-events b/hw/block/trace-events
index 2363412a9375..0064fedf31ae 100644
--- a/hw/block/trace-events
+++ b/hw/block/trace-events
@@ -83,6 +83,7 @@ pci_nvme_enqueue_event_masked(uint8_t typ) "type 0x%"PRIx8""
 pci_nvme_no_outstanding_aers(void) "ignoring event; no outstanding AERs"
 pci_nvme_enqueue_req_completion(uint16_t cid, uint16_t cqid, uint16_t status) "cid %"PRIu16" cqid %"PRIu16" status 0x%"PRIx16""
 pci_nvme_zrm_transition(uint32_t nsid, uint64_t zslba, uint8_t from, uint8_t to) "nsid %"PRIu32" zslba 0x%"PRIx64" from 0x%"PRIx8" to 0x%"PRIx8""
+pci_nvme_zrm_release_open(uint32_t nsid) "nsid %"PRIu32""
 pci_nvme_mmio_read(uint64_t addr) "addr 0x%"PRIx64""
 pci_nvme_mmio_write(uint64_t addr, uint64_t data) "addr 0x%"PRIx64" data 0x%"PRIx64""
 pci_nvme_mmio_doorbell_cq(uint16_t cqid, uint16_t new_head) "cqid %"PRIu16" new_head %"PRIu16""
-- 
2.28.0