linux/drivers/xen/swiotlb-xen.c
Michael Kelley 7296f2301a swiotlb: reduce swiotlb pool lookups
With CONFIG_SWIOTLB_DYNAMIC enabled, each round-trip map/unmap pair
in the swiotlb results in 6 calls to swiotlb_find_pool(). In multiple
places, the pool is found and used in one function, and then must
be found again in the next function that is called because only the
tlb_addr is passed as an argument. These are the six call sites:

dma_direct_map_page:
 1. swiotlb_map -> swiotlb_tbl_map_single -> swiotlb_bounce

dma_direct_unmap_page:
 2. dma_direct_sync_single_for_cpu -> is_swiotlb_buffer
 3. dma_direct_sync_single_for_cpu -> swiotlb_sync_single_for_cpu ->
	swiotlb_bounce
 4. is_swiotlb_buffer
 5. swiotlb_tbl_unmap_single -> swiotlb_del_transient
 6. swiotlb_tbl_unmap_single -> swiotlb_release_slots

Reduce the number of calls by finding the pool at a higher level, and
passing it as an argument instead of searching again. A key change is
for is_swiotlb_buffer() to return a pool pointer instead of a boolean,
and then pass this pool pointer to subsequent swiotlb functions.

There are 9 occurrences of is_swiotlb_buffer() used to test if a buffer
is a swiotlb buffer before calling a swiotlb function. To reduce code
duplication in getting the pool pointer and passing it as an argument,
introduce inline wrappers for this pattern. The generated code is
essentially unchanged.

Since is_swiotlb_buffer() no longer returns a boolean, rename some
functions to reflect the change:

 * swiotlb_find_pool() becomes __swiotlb_find_pool()
 * is_swiotlb_buffer() becomes swiotlb_find_pool()
 * is_xen_swiotlb_buffer() becomes xen_swiotlb_find_pool()

With these changes, a round-trip map/unmap pair requires only 2 pool
lookups (listed using the new names and wrappers):

dma_direct_unmap_page:
 1. dma_direct_sync_single_for_cpu -> swiotlb_find_pool
 2. swiotlb_tbl_unmap_single -> swiotlb_find_pool

These changes come from noticing the inefficiencies in a code review,
not from performance measurements. With CONFIG_SWIOTLB_DYNAMIC,
__swiotlb_find_pool() is not trivial, and it uses an RCU read lock,
so avoiding the redundant calls helps performance in a hot path.
When CONFIG_SWIOTLB_DYNAMIC is *not* set, the code size reduction
is minimal and the perf benefits are likely negligible, but no
harm is done.

No functional change is intended.

Signed-off-by: Michael Kelley <mhklinux@outlook.com>
Reviewed-by: Petr Tesarik <petr@tesarici.cz>
Signed-off-by: Christoph Hellwig <hch@lst.de>
2024-07-10 07:59:03 +02:00

419 lines
12 KiB
C

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright 2010
* by Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
*
* This code provides a IOMMU for Xen PV guests with PCI passthrough.
*
* PV guests under Xen are running in an non-contiguous memory architecture.
*
* When PCI pass-through is utilized, this necessitates an IOMMU for
* translating bus (DMA) to virtual and vice-versa and also providing a
* mechanism to have contiguous pages for device drivers operations (say DMA
* operations).
*
* Specifically, under Xen the Linux idea of pages is an illusion. It
* assumes that pages start at zero and go up to the available memory. To
* help with that, the Linux Xen MMU provides a lookup mechanism to
* translate the page frame numbers (PFN) to machine frame numbers (MFN)
* and vice-versa. The MFN are the "real" frame numbers. Furthermore
* memory is not contiguous. Xen hypervisor stitches memory for guests
* from different pools, which means there is no guarantee that PFN==MFN
* and PFN+1==MFN+1. Lastly with Xen 4.0, pages (in debug mode) are
* allocated in descending order (high to low), meaning the guest might
* never get any MFN's under the 4GB mark.
*/
#define pr_fmt(fmt) "xen:" KBUILD_MODNAME ": " fmt
#include <linux/memblock.h>
#include <linux/dma-direct.h>
#include <linux/dma-map-ops.h>
#include <linux/export.h>
#include <xen/swiotlb-xen.h>
#include <xen/page.h>
#include <xen/xen-ops.h>
#include <xen/hvc-console.h>
#include <asm/dma-mapping.h>
#include <trace/events/swiotlb.h>
#define MAX_DMA_BITS 32
/*
* Quick lookup value of the bus address of the IOTLB.
*/
static inline phys_addr_t xen_phys_to_bus(struct device *dev, phys_addr_t paddr)
{
unsigned long bfn = pfn_to_bfn(XEN_PFN_DOWN(paddr));
phys_addr_t baddr = (phys_addr_t)bfn << XEN_PAGE_SHIFT;
baddr |= paddr & ~XEN_PAGE_MASK;
return baddr;
}
static inline dma_addr_t xen_phys_to_dma(struct device *dev, phys_addr_t paddr)
{
return phys_to_dma(dev, xen_phys_to_bus(dev, paddr));
}
static inline phys_addr_t xen_bus_to_phys(struct device *dev,
phys_addr_t baddr)
{
unsigned long xen_pfn = bfn_to_pfn(XEN_PFN_DOWN(baddr));
phys_addr_t paddr = (xen_pfn << XEN_PAGE_SHIFT) |
(baddr & ~XEN_PAGE_MASK);
return paddr;
}
static inline phys_addr_t xen_dma_to_phys(struct device *dev,
dma_addr_t dma_addr)
{
return xen_bus_to_phys(dev, dma_to_phys(dev, dma_addr));
}
static inline int range_straddles_page_boundary(phys_addr_t p, size_t size)
{
unsigned long next_bfn, xen_pfn = XEN_PFN_DOWN(p);
unsigned int i, nr_pages = XEN_PFN_UP(xen_offset_in_page(p) + size);
next_bfn = pfn_to_bfn(xen_pfn);
for (i = 1; i < nr_pages; i++)
if (pfn_to_bfn(++xen_pfn) != ++next_bfn)
return 1;
return 0;
}
static struct io_tlb_pool *xen_swiotlb_find_pool(struct device *dev,
dma_addr_t dma_addr)
{
unsigned long bfn = XEN_PFN_DOWN(dma_to_phys(dev, dma_addr));
unsigned long xen_pfn = bfn_to_local_pfn(bfn);
phys_addr_t paddr = (phys_addr_t)xen_pfn << XEN_PAGE_SHIFT;
/* If the address is outside our domain, it CAN
* have the same virtual address as another address
* in our domain. Therefore _only_ check address within our domain.
*/
if (pfn_valid(PFN_DOWN(paddr)))
return swiotlb_find_pool(dev, paddr);
return NULL;
}
#ifdef CONFIG_X86
int xen_swiotlb_fixup(void *buf, unsigned long nslabs)
{
int rc;
unsigned int order = get_order(IO_TLB_SEGSIZE << IO_TLB_SHIFT);
unsigned int i, dma_bits = order + PAGE_SHIFT;
dma_addr_t dma_handle;
phys_addr_t p = virt_to_phys(buf);
BUILD_BUG_ON(IO_TLB_SEGSIZE & (IO_TLB_SEGSIZE - 1));
BUG_ON(nslabs % IO_TLB_SEGSIZE);
i = 0;
do {
do {
rc = xen_create_contiguous_region(
p + (i << IO_TLB_SHIFT), order,
dma_bits, &dma_handle);
} while (rc && dma_bits++ < MAX_DMA_BITS);
if (rc)
return rc;
i += IO_TLB_SEGSIZE;
} while (i < nslabs);
return 0;
}
static void *
xen_swiotlb_alloc_coherent(struct device *dev, size_t size,
dma_addr_t *dma_handle, gfp_t flags, unsigned long attrs)
{
u64 dma_mask = dev->coherent_dma_mask;
int order = get_order(size);
phys_addr_t phys;
void *ret;
/* Align the allocation to the Xen page size */
size = 1UL << (order + XEN_PAGE_SHIFT);
ret = (void *)__get_free_pages(flags, get_order(size));
if (!ret)
return ret;
phys = virt_to_phys(ret);
*dma_handle = xen_phys_to_dma(dev, phys);
if (*dma_handle + size - 1 > dma_mask ||
range_straddles_page_boundary(phys, size)) {
if (xen_create_contiguous_region(phys, order, fls64(dma_mask),
dma_handle) != 0)
goto out_free_pages;
SetPageXenRemapped(virt_to_page(ret));
}
memset(ret, 0, size);
return ret;
out_free_pages:
free_pages((unsigned long)ret, get_order(size));
return NULL;
}
static void
xen_swiotlb_free_coherent(struct device *dev, size_t size, void *vaddr,
dma_addr_t dma_handle, unsigned long attrs)
{
phys_addr_t phys = virt_to_phys(vaddr);
int order = get_order(size);
/* Convert the size to actually allocated. */
size = 1UL << (order + XEN_PAGE_SHIFT);
if (WARN_ON_ONCE(dma_handle + size - 1 > dev->coherent_dma_mask) ||
WARN_ON_ONCE(range_straddles_page_boundary(phys, size)))
return;
if (TestClearPageXenRemapped(virt_to_page(vaddr)))
xen_destroy_contiguous_region(phys, order);
free_pages((unsigned long)vaddr, get_order(size));
}
#endif /* CONFIG_X86 */
/*
* Map a single buffer of the indicated size for DMA in streaming mode. The
* physical address to use is returned.
*
* Once the device is given the dma address, the device owns this memory until
* either xen_swiotlb_unmap_page or xen_swiotlb_dma_sync_single is performed.
*/
static dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
unsigned long offset, size_t size,
enum dma_data_direction dir,
unsigned long attrs)
{
phys_addr_t map, phys = page_to_phys(page) + offset;
dma_addr_t dev_addr = xen_phys_to_dma(dev, phys);
BUG_ON(dir == DMA_NONE);
/*
* If the address happens to be in the device's DMA window,
* we can safely return the device addr and not worry about bounce
* buffering it.
*/
if (dma_capable(dev, dev_addr, size, true) &&
!range_straddles_page_boundary(phys, size) &&
!xen_arch_need_swiotlb(dev, phys, dev_addr) &&
!is_swiotlb_force_bounce(dev))
goto done;
/*
* Oh well, have to allocate and map a bounce buffer.
*/
trace_swiotlb_bounced(dev, dev_addr, size);
map = swiotlb_tbl_map_single(dev, phys, size, 0, dir, attrs);
if (map == (phys_addr_t)DMA_MAPPING_ERROR)
return DMA_MAPPING_ERROR;
phys = map;
dev_addr = xen_phys_to_dma(dev, map);
/*
* Ensure that the address returned is DMA'ble
*/
if (unlikely(!dma_capable(dev, dev_addr, size, true))) {
__swiotlb_tbl_unmap_single(dev, map, size, dir,
attrs | DMA_ATTR_SKIP_CPU_SYNC,
swiotlb_find_pool(dev, map));
return DMA_MAPPING_ERROR;
}
done:
if (!dev_is_dma_coherent(dev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC)) {
if (pfn_valid(PFN_DOWN(dma_to_phys(dev, dev_addr))))
arch_sync_dma_for_device(phys, size, dir);
else
xen_dma_sync_for_device(dev, dev_addr, size, dir);
}
return dev_addr;
}
/*
* Unmap a single streaming mode DMA translation. The dma_addr and size must
* match what was provided for in a previous xen_swiotlb_map_page call. All
* other usages are undefined.
*
* After this call, reads by the cpu to the buffer are guaranteed to see
* whatever the device wrote there.
*/
static void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
size_t size, enum dma_data_direction dir, unsigned long attrs)
{
phys_addr_t paddr = xen_dma_to_phys(hwdev, dev_addr);
struct io_tlb_pool *pool;
BUG_ON(dir == DMA_NONE);
if (!dev_is_dma_coherent(hwdev) && !(attrs & DMA_ATTR_SKIP_CPU_SYNC)) {
if (pfn_valid(PFN_DOWN(dma_to_phys(hwdev, dev_addr))))
arch_sync_dma_for_cpu(paddr, size, dir);
else
xen_dma_sync_for_cpu(hwdev, dev_addr, size, dir);
}
/* NOTE: We use dev_addr here, not paddr! */
pool = xen_swiotlb_find_pool(hwdev, dev_addr);
if (pool)
__swiotlb_tbl_unmap_single(hwdev, paddr, size, dir,
attrs, pool);
}
static void
xen_swiotlb_sync_single_for_cpu(struct device *dev, dma_addr_t dma_addr,
size_t size, enum dma_data_direction dir)
{
phys_addr_t paddr = xen_dma_to_phys(dev, dma_addr);
struct io_tlb_pool *pool;
if (!dev_is_dma_coherent(dev)) {
if (pfn_valid(PFN_DOWN(dma_to_phys(dev, dma_addr))))
arch_sync_dma_for_cpu(paddr, size, dir);
else
xen_dma_sync_for_cpu(dev, dma_addr, size, dir);
}
pool = xen_swiotlb_find_pool(dev, dma_addr);
if (pool)
__swiotlb_sync_single_for_cpu(dev, paddr, size, dir, pool);
}
static void
xen_swiotlb_sync_single_for_device(struct device *dev, dma_addr_t dma_addr,
size_t size, enum dma_data_direction dir)
{
phys_addr_t paddr = xen_dma_to_phys(dev, dma_addr);
struct io_tlb_pool *pool;
pool = xen_swiotlb_find_pool(dev, dma_addr);
if (pool)
__swiotlb_sync_single_for_device(dev, paddr, size, dir, pool);
if (!dev_is_dma_coherent(dev)) {
if (pfn_valid(PFN_DOWN(dma_to_phys(dev, dma_addr))))
arch_sync_dma_for_device(paddr, size, dir);
else
xen_dma_sync_for_device(dev, dma_addr, size, dir);
}
}
/*
* Unmap a set of streaming mode DMA translations. Again, cpu read rules
* concerning calls here are the same as for swiotlb_unmap_page() above.
*/
static void
xen_swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sgl, int nelems,
enum dma_data_direction dir, unsigned long attrs)
{
struct scatterlist *sg;
int i;
BUG_ON(dir == DMA_NONE);
for_each_sg(sgl, sg, nelems, i)
xen_swiotlb_unmap_page(hwdev, sg->dma_address, sg_dma_len(sg),
dir, attrs);
}
static int
xen_swiotlb_map_sg(struct device *dev, struct scatterlist *sgl, int nelems,
enum dma_data_direction dir, unsigned long attrs)
{
struct scatterlist *sg;
int i;
BUG_ON(dir == DMA_NONE);
for_each_sg(sgl, sg, nelems, i) {
sg->dma_address = xen_swiotlb_map_page(dev, sg_page(sg),
sg->offset, sg->length, dir, attrs);
if (sg->dma_address == DMA_MAPPING_ERROR)
goto out_unmap;
sg_dma_len(sg) = sg->length;
}
return nelems;
out_unmap:
xen_swiotlb_unmap_sg(dev, sgl, i, dir, attrs | DMA_ATTR_SKIP_CPU_SYNC);
sg_dma_len(sgl) = 0;
return -EIO;
}
static void
xen_swiotlb_sync_sg_for_cpu(struct device *dev, struct scatterlist *sgl,
int nelems, enum dma_data_direction dir)
{
struct scatterlist *sg;
int i;
for_each_sg(sgl, sg, nelems, i) {
xen_swiotlb_sync_single_for_cpu(dev, sg->dma_address,
sg->length, dir);
}
}
static void
xen_swiotlb_sync_sg_for_device(struct device *dev, struct scatterlist *sgl,
int nelems, enum dma_data_direction dir)
{
struct scatterlist *sg;
int i;
for_each_sg(sgl, sg, nelems, i) {
xen_swiotlb_sync_single_for_device(dev, sg->dma_address,
sg->length, dir);
}
}
/*
* Return whether the given device DMA address mask can be supported
* properly. For example, if your device can only drive the low 24-bits
* during bus mastering, then you would pass 0x00ffffff as the mask to
* this function.
*/
static int
xen_swiotlb_dma_supported(struct device *hwdev, u64 mask)
{
return xen_phys_to_dma(hwdev, default_swiotlb_limit()) <= mask;
}
const struct dma_map_ops xen_swiotlb_dma_ops = {
#ifdef CONFIG_X86
.alloc = xen_swiotlb_alloc_coherent,
.free = xen_swiotlb_free_coherent,
#else
.alloc = dma_direct_alloc,
.free = dma_direct_free,
#endif
.sync_single_for_cpu = xen_swiotlb_sync_single_for_cpu,
.sync_single_for_device = xen_swiotlb_sync_single_for_device,
.sync_sg_for_cpu = xen_swiotlb_sync_sg_for_cpu,
.sync_sg_for_device = xen_swiotlb_sync_sg_for_device,
.map_sg = xen_swiotlb_map_sg,
.unmap_sg = xen_swiotlb_unmap_sg,
.map_page = xen_swiotlb_map_page,
.unmap_page = xen_swiotlb_unmap_page,
.dma_supported = xen_swiotlb_dma_supported,
.mmap = dma_common_mmap,
.get_sgtable = dma_common_get_sgtable,
.alloc_pages_op = dma_common_alloc_pages,
.free_pages = dma_common_free_pages,
.max_mapping_size = swiotlb_max_mapping_size,
};