swiotlb: manipulate orig_addr when tlb_addr has offset

in case of driver wants to sync part of ranges with offset,
swiotlb_tbl_sync_single() copies from orig_addr base to tlb_addr with
offset and ends up with data mismatch.

It was removed from
"swiotlb: don't modify orig_addr in swiotlb_tbl_sync_single",
but said logic has to be added back in.

From Linus's email:
"That commit which the removed the offset calculation entirely, because the old

        (unsigned long)tlb_addr & (IO_TLB_SIZE - 1)

was wrong, but instead of removing it, I think it should have just
fixed it to be

        (tlb_addr - mem->start) & (IO_TLB_SIZE - 1);

instead. That way the slot offset always matches the slot index calculation."

(Unfortunatly that broke NVMe).

The use-case that drivers are hitting is as follow:

1. Get dma_addr_t from dma_map_single()

dma_addr_t tlb_addr = dma_map_single(dev, vaddr, vsize, DMA_TO_DEVICE);

    |<---------------vsize------------->|
    +-----------------------------------+
    |                                   | original buffer
    +-----------------------------------+
  vaddr

 swiotlb_align_offset
     |<----->|<---------------vsize------------->|
     +-------+-----------------------------------+
     |       |                                   | swiotlb buffer
     +-------+-----------------------------------+
          tlb_addr

2. Do something
3. Sync dma_addr_t through dma_sync_single_for_device(..)

dma_sync_single_for_device(dev, tlb_addr + offset, size, DMA_TO_DEVICE);

  Error case.
    Copy data to original buffer but it is from base addr (instead of
  base addr + offset) in original buffer:

 swiotlb_align_offset
     |<----->|<- offset ->|<- size ->|
     +-------+-----------------------------------+
     |       |            |##########|           | swiotlb buffer
     +-------+-----------------------------------+
          tlb_addr

    |<- size ->|
    +-----------------------------------+
    |##########|                        | original buffer
    +-----------------------------------+
  vaddr

The fix is to copy the data to the original buffer and take into
account the offset, like so:

 swiotlb_align_offset
     |<----->|<- offset ->|<- size ->|
     +-------+-----------------------------------+
     |       |            |##########|           | swiotlb buffer
     +-------+-----------------------------------+
          tlb_addr

    |<- offset ->|<- size ->|
    +-----------------------------------+
    |            |##########|           | original buffer
    +-----------------------------------+
  vaddr

[One fix which was Linus's that made more sense to as it created a
symmetry would break NVMe. The reason for that is the:
 unsigned int offset = (tlb_addr - mem->start) & (IO_TLB_SIZE - 1);

would come up with the proper offset, but it would lose the
alignment (which this patch contains).]

Fixes: 16fc3ce ("swiotlb: don't modify orig_addr in swiotlb_tbl_sync_single")
Signed-off-by: Bumyong Lee <[email protected]>
Signed-off-by: Chanho Park <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reported-by: Dominique MARTINET <[email protected]>
Reported-by: Horia Geantă <[email protected]>
Tested-by: Horia Geantă <[email protected]>
CC: [email protected]
Signed-off-by: Konrad Rzeszutek Wilk <[email protected]>

Author: bumyonglee

kernel/dma/swiotlb.c

@@ -334,6 +334,14 @@ void __init swiotlb_exit(void)
 	io_tlb_default_mem = NULL;
 }
 
+/*
+ * Return the offset into a iotlb slot required to keep the device happy.
+ */
+static unsigned int swiotlb_align_offset(struct device *dev, u64 addr)
+{
+	return addr & dma_get_min_align_mask(dev) & (IO_TLB_SIZE - 1);
+}
+
 /*
  * Bounce: copy the swiotlb buffer from or back to the original dma location
  */
@@ -346,10 +354,17 @@ static void swiotlb_bounce(struct device *dev, phys_addr_t tlb_addr, size_t size
 	size_t alloc_size = mem->slots[index].alloc_size;
 	unsigned long pfn = PFN_DOWN(orig_addr);
 	unsigned char *vaddr = phys_to_virt(tlb_addr);
+	unsigned int tlb_offset;
 
 	if (orig_addr == INVALID_PHYS_ADDR)
 		return;
 
+	tlb_offset = (tlb_addr & (IO_TLB_SIZE - 1)) -
+		     swiotlb_align_offset(dev, orig_addr);
+
+	orig_addr += tlb_offset;
+	alloc_size -= tlb_offset;
+
 	if (size > alloc_size) {
 		dev_WARN_ONCE(dev, 1,
 			"Buffer overflow detected. Allocation size: %zu. Mapping size: %zu.\n",
@@ -390,14 +405,6 @@ static void swiotlb_bounce(struct device *dev, phys_addr_t tlb_addr, size_t size
 
 #define slot_addr(start, idx)	((start) + ((idx) << IO_TLB_SHIFT))
 
-/*
- * Return the offset into a iotlb slot required to keep the device happy.
- */
-static unsigned int swiotlb_align_offset(struct device *dev, u64 addr)
-{
-	return addr & dma_get_min_align_mask(dev) & (IO_TLB_SIZE - 1);
-}
-
 /*
  * Carefully handle integer overflow which can occur when boundary_mask == ~0UL.
  */