wifi: ath11k: fix IOMMU errors on buffer rings

virt_to_phys() doesn't work on systems with IOMMU enabled, which have
non-identity physical-to-IOVA mappings.  It leads to IO_PAGE_FAULTs like this:

[IO_PAGE_FAULT domain=0x0023 address=0x1cce00000 flags=0x0020]

And no association to the AP can be established.

This patch changes that to dma_map_single(), which works correctly. Even
virt_to_phys() documentation says device drivers should not use it:

    This function does not give bus mappings for DMA transfers. In
    almost all conceivable cases a device driver should not be using
    this function

Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1

Signed-off-by: Zhenghao Gu <imguzh@gmail.com>
Acked-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://msgid.link/20231212031914.47339-1-imguzh@gmail.com

1 files changed, 17 insertions, 2 deletions

diff --git a/drivers/net/wireless/ath/ath11k/hal.c b/drivers/net/wireless/ath/ath11k/hal.c
index c060c4b5c0cc..f3d04568c221 100644
--- a/drivers/net/wireless/ath/ath11k/hal.c
+++ b/drivers/net/wireless/ath/ath11k/hal.c
@@ -626,15 +626,30 @@ u32 *ath11k_hal_srng_dst_peek(struct ath11k_base *ab, struct hal_srng *srng)
 	return NULL;
 }
 
+static u32 *ath11k_hal_srng_dst_peek_with_dma(struct ath11k_base *ab,
+					      struct hal_srng *srng, dma_addr_t *paddr)
+{
+	lockdep_assert_held(&srng->lock);
+
+	if (srng->u.dst_ring.tp != srng->u.dst_ring.cached_hp) {
+		*paddr = srng->ring_base_paddr +
+			  sizeof(*srng->ring_base_vaddr) * srng->u.dst_ring.tp;
+		return srng->ring_base_vaddr + srng->u.dst_ring.tp;
+	}
+
+	return NULL;
+}
+
 static void ath11k_hal_srng_prefetch_desc(struct ath11k_base *ab,
 					  struct hal_srng *srng)
 {
+	dma_addr_t desc_paddr;
 	u32 *desc;
 
 	/* prefetch only if desc is available */
-	desc = ath11k_hal_srng_dst_peek(ab, srng);
+	desc = ath11k_hal_srng_dst_peek_with_dma(ab, srng, &desc_paddr);
 	if (likely(desc)) {
-		dma_sync_single_for_cpu(ab->dev, virt_to_phys(desc),
+		dma_sync_single_for_cpu(ab->dev, desc_paddr,
 					(srng->entry_size * sizeof(u32)),
 					DMA_FROM_DEVICE);
 		prefetch(desc);