hw/nvme: fix endianness issue for shadow doorbells

In commit 2fda072 ("hw/nvme: fix missing endian conversions for
doorbell buffers"), we fixed shadow doorbells for big-endian guests
running on little endian hosts. But I did not fix little-endian guests
on big-endian hosts. Fix this.

Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1765
Fixes: 3f7fe8d ("hw/nvme: Implement shadow doorbell buffer support")
Cc: qemu-stable@nongnu.org
Reported-by: Thomas Huth <thuth@redhat.com>
Tested-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Keith Busch <kbusch@kernel.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Klaus Jensen <k.jensen@samsung.com>
(cherry picked from commit ea3c76f)
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>

hw/nvme/ctrl.c

@@ -6767,6 +6767,7 @@ static uint16_t nvme_dbbuf_config(NvmeCtrl *n, const NvmeRequest *req)
     PCIDevice *pci = PCI_DEVICE(n);
     uint64_t dbs_addr = le64_to_cpu(req->cmd.dptr.prp1);
     uint64_t eis_addr = le64_to_cpu(req->cmd.dptr.prp2);
+    uint32_t v;
     int i;
 
     /* Address should be page aligned */
@@ -6784,14 +6785,16 @@ static uint16_t nvme_dbbuf_config(NvmeCtrl *n, const NvmeRequest *req)
         NvmeCQueue *cq = n->cq[i];
 
         if (sq) {
+            v = cpu_to_le32(sq->tail);
+
             /*
              * CAP.DSTRD is 0, so offset of ith sq db_addr is (i<<3)
              * nvme_process_db() uses this hard-coded way to calculate
              * doorbell offsets. Be consistent with that here.
              */
             sq->db_addr = dbs_addr + (i << 3);
             sq->ei_addr = eis_addr + (i << 3);
-            pci_dma_write(pci, sq->db_addr, &sq->tail, sizeof(sq->tail));
+            pci_dma_write(pci, sq->db_addr, &v, sizeof(sq->tail));
 
             if (n->params.ioeventfd && sq->sqid != 0) {
                 if (!nvme_init_sq_ioeventfd(sq)) {
@@ -6801,10 +6804,12 @@ static uint16_t nvme_dbbuf_config(NvmeCtrl *n, const NvmeRequest *req)
         }
 
         if (cq) {
+            v = cpu_to_le32(cq->head);
+
             /* CAP.DSTRD is 0, so offset of ith cq db_addr is (i<<3)+(1<<2) */
             cq->db_addr = dbs_addr + (i << 3) + (1 << 2);
             cq->ei_addr = eis_addr + (i << 3) + (1 << 2);
-            pci_dma_write(pci, cq->db_addr, &cq->head, sizeof(cq->head));
+            pci_dma_write(pci, cq->db_addr, &v, sizeof(cq->head));
 
             if (n->params.ioeventfd && cq->cqid != 0) {
                 if (!nvme_init_cq_ioeventfd(cq)) {
@@ -7555,7 +7560,7 @@ static uint64_t nvme_mmio_read(void *opaque, hwaddr addr, unsigned size)
 static void nvme_process_db(NvmeCtrl *n, hwaddr addr, int val)
 {
     PCIDevice *pci = PCI_DEVICE(n);
-    uint32_t qid;
+    uint32_t qid, v;
 
     if (unlikely(addr & ((1 << 2) - 1))) {
         NVME_GUEST_ERR(pci_nvme_ub_db_wr_misaligned,
@@ -7622,7 +7627,8 @@ static void nvme_process_db(NvmeCtrl *n, hwaddr addr, int val)
         start_sqs = nvme_cq_full(cq) ? 1 : 0;
         cq->head = new_head;
         if (!qid && n->dbbuf_enabled) {
-            pci_dma_write(pci, cq->db_addr, &cq->head, sizeof(cq->head));
+            v = cpu_to_le32(cq->head);
+            pci_dma_write(pci, cq->db_addr, &v, sizeof(cq->head));
         }
         if (start_sqs) {
             NvmeSQueue *sq;
@@ -7682,6 +7688,8 @@ static void nvme_process_db(NvmeCtrl *n, hwaddr addr, int val)
 
         sq->tail = new_tail;
         if (!qid && n->dbbuf_enabled) {
+            v = cpu_to_le32(sq->tail);
+
             /*
              * The spec states "the host shall also update the controller's
              * corresponding doorbell property to match the value of that entry
@@ -7695,7 +7703,7 @@ static void nvme_process_db(NvmeCtrl *n, hwaddr addr, int val)
              * including ones that run on Linux, are not updating Admin Queues,
              * so we can't trust reading it for an appropriate sq tail.
              */
-            pci_dma_write(pci, sq->db_addr, &sq->tail, sizeof(sq->tail));
+            pci_dma_write(pci, sq->db_addr, &v, sizeof(sq->tail));
         }
 
         qemu_bh_schedule(sq->bh);