dma/idxd: fix truncated error code in status check

When checking if the DMA device is active, the result of the operand will
always be zero since the err_code is truncated to 8 bits which makes
checking the 31st bit impossible.

This is fixed by changing the type of err_code to uint32_t so that it is
not truncated.

Coverity issue: 373657
Fixes: 9449330a8458 ("dma/idxd: create dmadev instances on PCI probe")

Signed-off-by: Kevin Laatz <kevin.laatz@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Reviewed-by: Conor Walsh <conor.walsh@intel.com>

diff --git a/drivers/dma/idxd/idxd_pci.c b/drivers/dma/idxd/idxd_pci.c
index 1aabace..9ca1ec6 100644 (file)
--- a/drivers/dma/idxd/idxd_pci.c
+++ b/drivers/dma/idxd/idxd_pci.c
@@ -22,7 +22,7 @@ const struct rte_pci_id pci_id_idxd_map[] = {
 static inline int
 idxd_pci_dev_command(struct idxd_dmadev *idxd, enum rte_idxd_cmds command)
 {
-       uint8_t err_code;
+       uint32_t err_code;
        uint16_t qid = idxd->qid;
        int i = 0;
 
@@ -37,7 +37,8 @@ idxd_pci_dev_command(struct idxd_dmadev *idxd, enum rte_idxd_cmds command)
                if (++i >= 1000) {
                        IDXD_PMD_ERR("Timeout waiting for command response from HW");
                        rte_spinlock_unlock(&idxd->u.pci->lk);
-                       return err_code;
+                       err_code &= CMDSTATUS_ERR_MASK;
+                       return -err_code;
                }
        } while (err_code & CMDSTATUS_ACTIVE_MASK);
        rte_spinlock_unlock(&idxd->u.pci->lk);