raw/ioat: fix work-queue config size

According to latest DSA spec[1], the work-queue config register size
should be based off a value read from the WQ capabilities register.
Update driver to read this value and base the start of each WQ config
off that value.

[1] https://software.intel.com/content/www/us/en/develop/download/intel-data-streaming-accelerator-preliminary-architecture-specification.html

Fixes: ff06fa2cf3ba ("raw/ioat: probe idxd PCI")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Tested-by: Kevin Laatz <kevin.laatz@intel.com>

diff --git a/drivers/raw/ioat/idxd_pci.c b/drivers/raw/ioat/idxd_pci.c
index d4d87b1..99ecfbb 100644 (file)
--- a/drivers/raw/ioat/idxd_pci.c
+++ b/drivers/raw/ioat/idxd_pci.c
@@ -44,10 +44,16 @@ idxd_pci_dev_command(struct idxd_rawdev *idxd, enum rte_idxd_cmds command)
        return err_code & CMDSTATUS_ERR_MASK;
 }
 
+static uint32_t *
+idxd_get_wq_cfg(struct idxd_pci_common *pci, uint8_t wq_idx)
+{
+       return RTE_PTR_ADD(pci->wq_regs_base, wq_idx << (5 + pci->wq_cfg_sz));
+}
+
 static int
 idxd_is_wq_enabled(struct idxd_rawdev *idxd)
 {
-       uint32_t state = idxd->u.pci->wq_regs[idxd->qid].wqcfg[WQ_STATE_IDX];
+       uint32_t state = idxd_get_wq_cfg(idxd->u.pci, idxd->qid)[WQ_STATE_IDX];
        return ((state >> WQ_STATE_SHIFT) & WQ_STATE_MASK) == 0x1;
 }
 
@@ -137,8 +143,9 @@ init_pci_device(struct rte_pci_device *dev, struct idxd_rawdev *idxd)
        grp_offset = (uint16_t)pci->regs->offsets[0];
        pci->grp_regs = RTE_PTR_ADD(pci->regs, grp_offset * 0x100);
        wq_offset = (uint16_t)(pci->regs->offsets[0] >> 16);
-       pci->wq_regs = RTE_PTR_ADD(pci->regs, wq_offset * 0x100);
+       pci->wq_regs_base = RTE_PTR_ADD(pci->regs, wq_offset * 0x100);
        pci->portals = dev->mem_resource[2].addr;
+       pci->wq_cfg_sz = (pci->regs->wqcap >> 24) & 0x0F;
 
        /* sanity check device status */
        if (pci->regs->gensts & GENSTS_DEV_STATE_MASK) {
@@ -169,7 +176,7 @@ init_pci_device(struct rte_pci_device *dev, struct idxd_rawdev *idxd)
                pci->grp_regs[i].grpwqcfg[0] = 0;
        }
        for (i = 0; i < nb_wqs; i++)
-               pci->wq_regs[i].wqcfg[0] = 0;
+               idxd_get_wq_cfg(pci, i)[0] = 0;
 
        /* put each engine into a separate group to avoid reordering */
        if (nb_groups > nb_engines)
@@ -194,10 +201,10 @@ init_pci_device(struct rte_pci_device *dev, struct idxd_rawdev *idxd)
                                i, i % nb_groups);
                pci->grp_regs[i % nb_groups].grpwqcfg[0] |= (1ULL << i);
                /* now configure it, in terms of size, max batch, mode */
-               pci->wq_regs[i].wqcfg[WQ_SIZE_IDX] = wq_size;
-               pci->wq_regs[i].wqcfg[WQ_MODE_IDX] = (1 << WQ_PRIORITY_SHIFT) |
+               idxd_get_wq_cfg(pci, i)[WQ_SIZE_IDX] = wq_size;
+               idxd_get_wq_cfg(pci, i)[WQ_MODE_IDX] = (1 << WQ_PRIORITY_SHIFT) |
                                WQ_MODE_DEDICATED;
-               pci->wq_regs[i].wqcfg[WQ_SIZES_IDX] = lg2_max_copy_size |
+               idxd_get_wq_cfg(pci, i)[WQ_SIZES_IDX] = lg2_max_copy_size |
                                (lg2_max_batch << WQ_BATCH_SZ_SHIFT);
        }
 

diff --git a/drivers/raw/ioat/ioat_private.h b/drivers/raw/ioat/ioat_private.h
index a74bc04..6c42381 100644 (file)
--- a/drivers/raw/ioat/ioat_private.h
+++ b/drivers/raw/ioat/ioat_private.h
@@ -30,8 +30,10 @@ extern int ioat_pmd_logtype;
 
 struct idxd_pci_common {
        rte_spinlock_t lk;
+
+       uint8_t wq_cfg_sz;
        volatile struct rte_idxd_bar0 *regs;
-       volatile struct rte_idxd_wqcfg *wq_regs;
+       volatile uint32_t *wq_regs_base;
        volatile struct rte_idxd_grpcfg *grp_regs;
        volatile void *portals;
 };

diff --git a/drivers/raw/ioat/ioat_spec.h b/drivers/raw/ioat/ioat_spec.h
index 1aa768b..6aa467e 100644 (file)
--- a/drivers/raw/ioat/ioat_spec.h
+++ b/drivers/raw/ioat/ioat_spec.h
@@ -302,10 +302,7 @@ struct rte_idxd_bar0 {
        uint64_t __rte_aligned(0x20) swerror[4]; /* offset 0xC0 */
 };
 
-struct rte_idxd_wqcfg {
-       uint32_t wqcfg[8] __rte_aligned(32); /* 32-byte register */
-};
-
+/* workqueue config is provided by array of uint32_t. */
 #define WQ_SIZE_IDX      0 /* size is in first 32-bit value */
 #define WQ_THRESHOLD_IDX 1 /* WQ threshold second 32-bits */
 #define WQ_MODE_IDX      2 /* WQ mode and other flags */