This is a note to let you know that I've just added the patch titled
swiotlb: factor out a nr_slots helper
to the 5.10-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
swiotlb-factor-out-a-nr_slots-helper.patch
and it can be found in the queue-5.10 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From foo@baz Wed May 5 09:27:45 AM CEST 2021
From: Jianxiong Gao <jxgao@xxxxxxxxxx>
Date: Thu, 29 Apr 2021 17:33:10 +0000
Subject: swiotlb: factor out a nr_slots helper
To: stable@xxxxxxxxxxxxxxx, hch@xxxxxx, marcorr@xxxxxxxxxx, sashal@xxxxxxxxxx
Cc: Jianxiong Gao <jxgao@xxxxxxxxxx>, Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
Message-ID: <20210429173315.1252465-5-jxgao@xxxxxxxxxx>
From: Jianxiong Gao <jxgao@xxxxxxxxxx>
commit: c32a77fd18780a5192dfb6eec69f239faebf28fd
Factor out a helper to find the number of slots for a given size.
Signed-off-by: Christoph Hellwig <hch@xxxxxx>
Acked-by: Jianxiong Gao <jxgao@xxxxxxxxxx>
Tested-by: Jianxiong Gao <jxgao@xxxxxxxxxx>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
Signed-off-by: Jianxiong Gao <jxgao@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
kernel/dma/swiotlb.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
--- a/kernel/dma/swiotlb.c
+++ b/kernel/dma/swiotlb.c
@@ -178,6 +178,11 @@ static inline unsigned long io_tlb_offse
return val & (IO_TLB_SEGSIZE - 1);
}
+static inline unsigned long nr_slots(u64 val)
+{
+ return DIV_ROUND_UP(val, IO_TLB_SIZE);
+}
+
/*
* Early SWIOTLB allocation may be too early to allow an architecture to
* perform the desired operations. This function allows the architecture to
@@ -477,20 +482,20 @@ phys_addr_t swiotlb_tbl_map_single(struc
tbl_dma_addr &= mask;
- offset_slots = ALIGN(tbl_dma_addr, IO_TLB_SIZE) >> IO_TLB_SHIFT;
+ offset_slots = nr_slots(tbl_dma_addr);
/*
* Carefully handle integer overflow which can occur when mask == ~0UL.
*/
max_slots = mask + 1
- ? ALIGN(mask + 1, IO_TLB_SIZE) >> IO_TLB_SHIFT
+ ? nr_slots(mask + 1)
: 1UL << (BITS_PER_LONG - IO_TLB_SHIFT);
/*
* For mappings greater than or equal to a page, we limit the stride
* (and hence alignment) to a page size.
*/
- nslots = ALIGN(alloc_size, IO_TLB_SIZE) >> IO_TLB_SHIFT;
+ nslots = nr_slots(alloc_size);
if (alloc_size >= PAGE_SIZE)
stride = (1 << (PAGE_SHIFT - IO_TLB_SHIFT));
else
@@ -586,7 +591,7 @@ void swiotlb_tbl_unmap_single(struct dev
enum dma_data_direction dir, unsigned long attrs)
{
unsigned long flags;
- int i, count, nslots = ALIGN(alloc_size, IO_TLB_SIZE) >> IO_TLB_SHIFT;
+ int i, count, nslots = nr_slots(alloc_size);
int index = (tlb_addr - io_tlb_start) >> IO_TLB_SHIFT;
phys_addr_t orig_addr = io_tlb_orig_addr[index];
Patches currently in stable-queue which might be from jxgao@xxxxxxxxxx are
queue-5.10/swiotlb-respect-min_align_mask.patch
queue-5.10/swiotlb-factor-out-a-nr_slots-helper.patch
queue-5.10/swiotlb-add-a-io_tlb_size-define.patch
queue-5.10/swiotlb-clean-up-swiotlb_tbl_unmap_single.patch
queue-5.10/swiotlb-factor-out-an-io_tlb_offset-helper.patch
queue-5.10/swiotlb-don-t-modify-orig_addr-in-swiotlb_tbl_sync_single.patch
queue-5.10/swiotlb-refactor-swiotlb_tbl_map_single.patch
queue-5.10/driver-core-add-a-min_align_mask-field-to-struct-device_dma_parameters.patch
queue-5.10/nvme-pci-set-min_align_mask.patch
