Merge remote-tracking branch 'remotes/nvme/tags/nvme-fixes-20210412-pull-request' into staging

emulated nvme docs and fixes for -rc3

- documentation
- fixes

# gpg: Signature made Mon 12 Apr 2021 07:56:09 BST
# gpg:                using RSA key 522833AA75E2DCE6A24766C04DE1AF316D4F0DE9
# gpg: Good signature from "Klaus Jensen <its@irrelevant.dk>" [unknown]
# gpg:                 aka "Klaus Jensen <k.jensen@samsung.com>" [unknown]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: DDCA 4D9C 9EF9 31CC 3468  4272 63D5 6FC5 E55D A838
#      Subkey fingerprint: 5228 33AA 75E2 DCE6 A247  66C0 4DE1 AF31 6D4F 0DE9

* remotes/nvme/tags/nvme-fixes-20210412-pull-request:
  hw/block/nvme: drain namespaces on sq deletion
  hw/block/nvme: store aiocb in compare
  hw/block/nvme: map prp fix if prp2 contains non-zero offset
  docs: add nvme emulation documentation

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

1 files changed, 32 insertions, 6 deletions

diff --git a/hw/block/nvme.c b/hw/block/nvme.c
index 6b1f056a0e..624a1431d0 100644
--- a/hw/block/nvme.c
+++ b/hw/block/nvme.c
@@ -470,6 +470,7 @@ static void nvme_req_clear(NvmeRequest *req)
 {
     req->ns = NULL;
     req->opaque = NULL;
+    req->aiocb = NULL;
     memset(&req->cqe, 0x0, sizeof(req->cqe));
     req->status = NVME_SUCCESS;
 }
@@ -655,7 +656,12 @@ static uint16_t nvme_map_prp(NvmeCtrl *n, NvmeSg *sg, uint64_t prp1,
             uint32_t nents, prp_trans;
             int i = 0;
 
-            nents = (len + n->page_size - 1) >> n->page_bits;
+            /*
+             * The first PRP list entry, pointed to by PRP2 may contain offset.
+             * Hence, we need to calculate the number of entries in based on
+             * that offset.
+             */
+            nents = (n->page_size - (prp2 & (n->page_size - 1))) >> 3;
             prp_trans = MIN(n->max_prp_ents, nents) * sizeof(uint64_t);
             ret = nvme_addr_read(n, prp2, (void *)prp_list, prp_trans);
             if (ret) {
@@ -666,7 +672,7 @@ static uint16_t nvme_map_prp(NvmeCtrl *n, NvmeSg *sg, uint64_t prp1,
             while (len != 0) {
                 uint64_t prp_ent = le64_to_cpu(prp_list[i]);
 
-                if (i == n->max_prp_ents - 1 && len > n->page_size) {
+                if (i == nents - 1 && len > n->page_size) {
                     if (unlikely(prp_ent & (n->page_size - 1))) {
                         trace_pci_nvme_err_invalid_prplist_ent(prp_ent);
                         status = NVME_INVALID_PRP_OFFSET | NVME_DNR;
@@ -675,7 +681,8 @@ static uint16_t nvme_map_prp(NvmeCtrl *n, NvmeSg *sg, uint64_t prp1,
 
                     i = 0;
                     nents = (len + n->page_size - 1) >> n->page_bits;
-                    prp_trans = MIN(n->max_prp_ents, nents) * sizeof(uint64_t);
+                    nents = MIN(nents, n->max_prp_ents);
+                    prp_trans = nents * sizeof(uint64_t);
                     ret = nvme_addr_read(n, prp_ent, (void *)prp_list,
                                          prp_trans);
                     if (ret) {
@@ -2837,7 +2844,8 @@ static uint16_t nvme_compare(NvmeCtrl *n, NvmeRequest *req)
 
     block_acct_start(blk_get_stats(blk), &req->acct, data_len,
                      BLOCK_ACCT_READ);
-    blk_aio_preadv(blk, offset, &ctx->data.iov, 0, nvme_compare_data_cb, req);
+    req->aiocb = blk_aio_preadv(blk, offset, &ctx->data.iov, 0,
+                                nvme_compare_data_cb, req);
 
     return NVME_NO_COMPLETE;
 }
@@ -3680,6 +3688,7 @@ static uint16_t nvme_del_sq(NvmeCtrl *n, NvmeRequest *req)
     NvmeSQueue *sq;
     NvmeCQueue *cq;
     uint16_t qid = le16_to_cpu(c->qid);
+    uint32_t nsid;
 
     if (unlikely(!qid || nvme_check_sqid(n, qid))) {
         trace_pci_nvme_err_invalid_del_sq(qid);
@@ -3691,9 +3700,26 @@ static uint16_t nvme_del_sq(NvmeCtrl *n, NvmeRequest *req)
     sq = n->sq[qid];
     while (!QTAILQ_EMPTY(&sq->out_req_list)) {
         r = QTAILQ_FIRST(&sq->out_req_list);
-        assert(r->aiocb);
-        blk_aio_cancel(r->aiocb);
+        if (r->aiocb) {
+            blk_aio_cancel(r->aiocb);
+        }
+    }
+
+    /*
+     * Drain all namespaces if there are still outstanding requests that we
+     * could not cancel explicitly.
+     */
+    if (!QTAILQ_EMPTY(&sq->out_req_list)) {
+        for (nsid = 1; nsid <= NVME_MAX_NAMESPACES; nsid++) {
+            NvmeNamespace *ns = nvme_ns(n, nsid);
+            if (ns) {
+                nvme_ns_drain(ns);
+            }
+        }
     }
+
+    assert(QTAILQ_EMPTY(&sq->out_req_list));
+
     if (!nvme_check_cqid(n, sq->cqid)) {
         cq = n->cq[sq->cqid];
         QTAILQ_REMOVE(&cq->sq_list, sq, entry);