2005-04-17 00:20:36 +02:00
|
|
|
/*
|
2008-09-25 16:59:19 +02:00
|
|
|
* linux/arch/arm/mm/dma-mapping.c
|
2005-04-17 00:20:36 +02:00
|
|
|
*
|
|
|
|
* Copyright (C) 2000-2004 Russell King
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License version 2 as
|
|
|
|
* published by the Free Software Foundation.
|
|
|
|
*
|
|
|
|
* DMA uncached mapping support.
|
|
|
|
*/
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/mm.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 09:04:11 +01:00
|
|
|
#include <linux/gfp.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/list.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/device.h>
|
|
|
|
#include <linux/dma-mapping.h>
|
2010-12-15 21:14:45 +01:00
|
|
|
#include <linux/highmem.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2006-04-02 01:07:39 +02:00
|
|
|
#include <asm/memory.h>
|
2009-03-13 03:52:09 +01:00
|
|
|
#include <asm/highmem.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <asm/cacheflush.h>
|
|
|
|
#include <asm/tlbflush.h>
|
2006-01-12 17:12:21 +01:00
|
|
|
#include <asm/sizes.h>
|
|
|
|
|
2009-07-24 13:35:02 +02:00
|
|
|
static u64 get_coherent_dma_mask(struct device *dev)
|
|
|
|
{
|
|
|
|
u64 mask = ISA_DMA_THRESHOLD;
|
|
|
|
|
|
|
|
if (dev) {
|
|
|
|
mask = dev->coherent_dma_mask;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Sanity check the DMA mask - it must be non-zero, and
|
|
|
|
* must be able to be satisfied by a DMA allocation.
|
|
|
|
*/
|
|
|
|
if (mask == 0) {
|
|
|
|
dev_warn(dev, "coherent DMA mask is unset\n");
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((~mask) & ISA_DMA_THRESHOLD) {
|
|
|
|
dev_warn(dev, "coherent DMA mask %#llx is smaller "
|
|
|
|
"than system GFP_DMA mask %#llx\n",
|
|
|
|
mask, (unsigned long long)ISA_DMA_THRESHOLD);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2009-07-24 13:35:02 +02:00
|
|
|
return mask;
|
|
|
|
}
|
|
|
|
|
2009-11-19 16:31:07 +01:00
|
|
|
/*
|
|
|
|
* Allocate a DMA buffer for 'dev' of size 'size' using the
|
|
|
|
* specified gfp mask. Note that 'size' must be page aligned.
|
|
|
|
*/
|
|
|
|
static struct page *__dma_alloc_buffer(struct device *dev, size_t size, gfp_t gfp)
|
|
|
|
{
|
|
|
|
unsigned long order = get_order(size);
|
|
|
|
struct page *page, *p, *e;
|
|
|
|
void *ptr;
|
|
|
|
u64 mask = get_coherent_dma_mask(dev);
|
|
|
|
|
|
|
|
#ifdef CONFIG_DMA_API_DEBUG
|
|
|
|
u64 limit = (mask + 1) & ~mask;
|
|
|
|
if (limit && size >= limit) {
|
|
|
|
dev_warn(dev, "coherent allocation too big (requested %#x mask %#llx)\n",
|
|
|
|
size, mask);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
if (!mask)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
if (mask < 0xffffffffULL)
|
|
|
|
gfp |= GFP_DMA;
|
|
|
|
|
|
|
|
page = alloc_pages(gfp, order);
|
|
|
|
if (!page)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Now split the huge page and free the excess pages
|
|
|
|
*/
|
|
|
|
split_page(page, order);
|
|
|
|
for (p = page + (size >> PAGE_SHIFT), e = page + (1 << order); p < e; p++)
|
|
|
|
__free_page(p);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Ensure that the allocated pages are zeroed, and that any data
|
|
|
|
* lurking in the kernel direct-mapped region is invalidated.
|
|
|
|
*/
|
|
|
|
ptr = page_address(page);
|
|
|
|
memset(ptr, 0, size);
|
|
|
|
dmac_flush_range(ptr, ptr + size);
|
|
|
|
outer_flush_range(__pa(ptr), __pa(ptr) + size);
|
|
|
|
|
|
|
|
return page;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Free a DMA buffer. 'size' must be page aligned.
|
|
|
|
*/
|
|
|
|
static void __dma_free_buffer(struct page *page, size_t size)
|
|
|
|
{
|
|
|
|
struct page *e = page + (size >> PAGE_SHIFT);
|
|
|
|
|
|
|
|
while (page < e) {
|
|
|
|
__free_page(page);
|
|
|
|
page++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-07-24 13:35:02 +02:00
|
|
|
#ifdef CONFIG_MMU
|
2010-06-21 16:09:06 +02:00
|
|
|
/* Sanity check size */
|
|
|
|
#if (CONSISTENT_DMA_SIZE % SZ_2M)
|
|
|
|
#error "CONSISTENT_DMA_SIZE must be multiple of 2MiB"
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#define CONSISTENT_OFFSET(x) (((unsigned long)(x) - CONSISTENT_BASE) >> PAGE_SHIFT)
|
|
|
|
#define CONSISTENT_PTE_INDEX(x) (((unsigned long)(x) - CONSISTENT_BASE) >> PGDIR_SHIFT)
|
|
|
|
#define NUM_CONSISTENT_PTES (CONSISTENT_DMA_SIZE >> PGDIR_SHIFT)
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
/*
|
2006-01-12 17:12:21 +01:00
|
|
|
* These are the page tables (2MB each) covering uncached, DMA consistent allocations
|
2005-04-17 00:20:36 +02:00
|
|
|
*/
|
2006-01-12 17:12:21 +01:00
|
|
|
static pte_t *consistent_pte[NUM_CONSISTENT_PTES];
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2009-11-19 16:07:04 +01:00
|
|
|
#include "vmregion.h"
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2009-11-19 16:07:04 +01:00
|
|
|
static struct arm_vmregion_head consistent_head = {
|
|
|
|
.vm_lock = __SPIN_LOCK_UNLOCKED(&consistent_head.vm_lock),
|
2005-04-17 00:20:36 +02:00
|
|
|
.vm_list = LIST_HEAD_INIT(consistent_head.vm_list),
|
|
|
|
.vm_start = CONSISTENT_BASE,
|
|
|
|
.vm_end = CONSISTENT_END,
|
|
|
|
};
|
|
|
|
|
|
|
|
#ifdef CONFIG_HUGETLB_PAGE
|
|
|
|
#error ARM Coherent DMA allocator does not (yet) support huge TLB
|
|
|
|
#endif
|
|
|
|
|
2009-11-19 17:46:02 +01:00
|
|
|
/*
|
|
|
|
* Initialise the consistent memory allocation.
|
|
|
|
*/
|
|
|
|
static int __init consistent_init(void)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
pgd_t *pgd;
|
|
|
|
pmd_t *pmd;
|
|
|
|
pte_t *pte;
|
|
|
|
int i = 0;
|
|
|
|
u32 base = CONSISTENT_BASE;
|
|
|
|
|
|
|
|
do {
|
|
|
|
pgd = pgd_offset(&init_mm, base);
|
|
|
|
pmd = pmd_alloc(&init_mm, pgd, base);
|
|
|
|
if (!pmd) {
|
|
|
|
printk(KERN_ERR "%s: no pmd tables\n", __func__);
|
|
|
|
ret = -ENOMEM;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
WARN_ON(!pmd_none(*pmd));
|
|
|
|
|
|
|
|
pte = pte_alloc_kernel(pmd, base);
|
|
|
|
if (!pte) {
|
|
|
|
printk(KERN_ERR "%s: no pte tables\n", __func__);
|
|
|
|
ret = -ENOMEM;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
consistent_pte[i++] = pte;
|
|
|
|
base += (1 << PGDIR_SHIFT);
|
|
|
|
} while (base < CONSISTENT_END);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
core_initcall(consistent_init);
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
static void *
|
2009-11-19 22:12:17 +01:00
|
|
|
__dma_alloc_remap(struct page *page, size_t size, gfp_t gfp, pgprot_t prot)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2009-11-19 16:07:04 +01:00
|
|
|
struct arm_vmregion *c;
|
2010-07-25 09:57:02 +02:00
|
|
|
size_t align;
|
|
|
|
int bit;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2009-11-19 21:58:31 +01:00
|
|
|
if (!consistent_pte[0]) {
|
|
|
|
printk(KERN_ERR "%s: not initialised\n", __func__);
|
|
|
|
dump_stack();
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2010-07-25 09:57:02 +02:00
|
|
|
/*
|
|
|
|
* Align the virtual region allocation - maximum alignment is
|
|
|
|
* a section size, minimum is a page size. This helps reduce
|
|
|
|
* fragmentation of the DMA space, and also prevents allocations
|
|
|
|
* smaller than a section from crossing a section boundary.
|
|
|
|
*/
|
2010-11-03 17:00:15 +01:00
|
|
|
bit = fls(size - 1);
|
2010-07-25 09:57:02 +02:00
|
|
|
if (bit > SECTION_SHIFT)
|
|
|
|
bit = SECTION_SHIFT;
|
|
|
|
align = 1 << bit;
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* Allocate a virtual address in the consistent mapping region.
|
|
|
|
*/
|
2010-07-25 09:57:02 +02:00
|
|
|
c = arm_vmregion_alloc(&consistent_head, align, size,
|
2005-04-17 00:20:36 +02:00
|
|
|
gfp & ~(__GFP_DMA | __GFP_HIGHMEM));
|
|
|
|
if (c) {
|
2006-01-12 17:12:21 +01:00
|
|
|
pte_t *pte;
|
|
|
|
int idx = CONSISTENT_PTE_INDEX(c->vm_start);
|
|
|
|
u32 off = CONSISTENT_OFFSET(c->vm_start) & (PTRS_PER_PTE-1);
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2006-01-12 17:12:21 +01:00
|
|
|
pte = consistent_pte[idx] + off;
|
2005-04-17 00:20:36 +02:00
|
|
|
c->vm_pages = page;
|
|
|
|
|
|
|
|
do {
|
|
|
|
BUG_ON(!pte_none(*pte));
|
|
|
|
|
2006-12-13 15:34:43 +01:00
|
|
|
set_pte_ext(pte, mk_pte(page, prot), 0);
|
2005-04-17 00:20:36 +02:00
|
|
|
page++;
|
|
|
|
pte++;
|
2006-01-12 17:12:21 +01:00
|
|
|
off++;
|
|
|
|
if (off >= PTRS_PER_PTE) {
|
|
|
|
off = 0;
|
|
|
|
pte = consistent_pte[++idx];
|
|
|
|
}
|
2005-04-17 00:20:36 +02:00
|
|
|
} while (size -= PAGE_SIZE);
|
|
|
|
|
ARM: Ensure PTE modifications via dma_alloc_coherent are visible
Dave Hylands reports:
| We've observed a problem with dma_alloc_writecombine when the system
| is under heavy load (heavy bus traffic). We've managed to reduce the
| problem to the following snippet, which is run from a kthread in a
| continuous loop:
|
| void *virtAddr;
| dma_addr_t physAddr;
| unsigned int numBytes = 256;
|
| for (;;) {
| virtAddr = dma_alloc_writecombine(NULL,
| numBytes, &physAddr, GFP_KERNEL);
| if (virtAddr == NULL) {
| printk(KERN_ERR "Running out of memory\n");
| break;
| }
|
| /* access DMA memory allocated */
| tmp = virtAddr;
| *tmp = 0x77;
|
| /* free DMA memory */
| dma_free_writecombine(NULL,
| numBytes, virtAddr, physAddr);
|
| ...sleep here...
| }
|
| By itself, the code will run forever with no issues. However, as we
| increase our bus traffic (typically using DMA) then the *tmp = 0x77
| line will eventually cause a page fault. If we add a small delay (a
| few microseconds) before the *tmp = 0x77, then we don't see a page
| fault, even under heavy load.
A dsb() is required after modifying the PTE entries to ensure that they
will always be visible. Add this dsb().
Reported-by: Dave Hylands <dhylands@gmail.com>
Tested-by: Dave Hylands <dhylands@gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2010-09-08 17:27:56 +02:00
|
|
|
dsb();
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
return (void *)c->vm_start;
|
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
2009-11-19 17:31:39 +01:00
|
|
|
|
|
|
|
static void __dma_free_remap(void *cpu_addr, size_t size)
|
|
|
|
{
|
|
|
|
struct arm_vmregion *c;
|
|
|
|
unsigned long addr;
|
|
|
|
pte_t *ptep;
|
|
|
|
int idx;
|
|
|
|
u32 off;
|
|
|
|
|
|
|
|
c = arm_vmregion_find_remove(&consistent_head, (unsigned long)cpu_addr);
|
|
|
|
if (!c) {
|
|
|
|
printk(KERN_ERR "%s: trying to free invalid coherent area: %p\n",
|
|
|
|
__func__, cpu_addr);
|
|
|
|
dump_stack();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((c->vm_end - c->vm_start) != size) {
|
|
|
|
printk(KERN_ERR "%s: freeing wrong coherent size (%ld != %d)\n",
|
|
|
|
__func__, c->vm_end - c->vm_start, size);
|
|
|
|
dump_stack();
|
|
|
|
size = c->vm_end - c->vm_start;
|
|
|
|
}
|
|
|
|
|
|
|
|
idx = CONSISTENT_PTE_INDEX(c->vm_start);
|
|
|
|
off = CONSISTENT_OFFSET(c->vm_start) & (PTRS_PER_PTE-1);
|
|
|
|
ptep = consistent_pte[idx] + off;
|
|
|
|
addr = c->vm_start;
|
|
|
|
do {
|
|
|
|
pte_t pte = ptep_get_and_clear(&init_mm, addr, ptep);
|
|
|
|
|
|
|
|
ptep++;
|
|
|
|
addr += PAGE_SIZE;
|
|
|
|
off++;
|
|
|
|
if (off >= PTRS_PER_PTE) {
|
|
|
|
off = 0;
|
|
|
|
ptep = consistent_pte[++idx];
|
|
|
|
}
|
|
|
|
|
2009-11-20 19:19:52 +01:00
|
|
|
if (pte_none(pte) || !pte_present(pte))
|
|
|
|
printk(KERN_CRIT "%s: bad page in kernel page table\n",
|
|
|
|
__func__);
|
2009-11-19 17:31:39 +01:00
|
|
|
} while (size -= PAGE_SIZE);
|
|
|
|
|
|
|
|
flush_tlb_kernel_range(c->vm_start, c->vm_end);
|
|
|
|
|
|
|
|
arm_vmregion_free(&consistent_head, c);
|
|
|
|
}
|
|
|
|
|
2009-07-24 13:35:02 +02:00
|
|
|
#else /* !CONFIG_MMU */
|
2009-11-19 17:31:39 +01:00
|
|
|
|
2009-11-19 22:12:17 +01:00
|
|
|
#define __dma_alloc_remap(page, size, gfp, prot) page_address(page)
|
|
|
|
#define __dma_free_remap(addr, size) do { } while (0)
|
|
|
|
|
|
|
|
#endif /* CONFIG_MMU */
|
|
|
|
|
2009-07-24 13:35:02 +02:00
|
|
|
static void *
|
|
|
|
__dma_alloc(struct device *dev, size_t size, dma_addr_t *handle, gfp_t gfp,
|
|
|
|
pgprot_t prot)
|
|
|
|
{
|
2009-11-19 16:54:45 +01:00
|
|
|
struct page *page;
|
2009-11-19 22:12:17 +01:00
|
|
|
void *addr;
|
2009-07-24 13:35:02 +02:00
|
|
|
|
2009-11-19 16:54:45 +01:00
|
|
|
*handle = ~0;
|
|
|
|
size = PAGE_ALIGN(size);
|
2009-07-24 13:35:02 +02:00
|
|
|
|
2009-11-19 16:54:45 +01:00
|
|
|
page = __dma_alloc_buffer(dev, size, gfp);
|
|
|
|
if (!page)
|
|
|
|
return NULL;
|
2009-07-24 13:35:02 +02:00
|
|
|
|
2009-11-19 22:12:17 +01:00
|
|
|
if (!arch_is_coherent())
|
|
|
|
addr = __dma_alloc_remap(page, size, gfp, prot);
|
|
|
|
else
|
|
|
|
addr = page_address(page);
|
2009-11-19 17:31:39 +01:00
|
|
|
|
2009-11-19 22:12:17 +01:00
|
|
|
if (addr)
|
2011-01-03 01:00:17 +01:00
|
|
|
*handle = pfn_to_dma(dev, page_to_pfn(page));
|
2009-11-19 17:31:39 +01:00
|
|
|
|
2009-11-19 22:12:17 +01:00
|
|
|
return addr;
|
|
|
|
}
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Allocate DMA-coherent memory space and return both the kernel remapped
|
|
|
|
* virtual and bus address for that space.
|
|
|
|
*/
|
|
|
|
void *
|
2005-10-21 09:20:58 +02:00
|
|
|
dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *handle, gfp_t gfp)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2008-07-18 11:30:14 +02:00
|
|
|
void *memory;
|
|
|
|
|
|
|
|
if (dma_alloc_from_coherent(dev, size, handle, &memory))
|
|
|
|
return memory;
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
return __dma_alloc(dev, size, handle, gfp,
|
2009-11-20 22:06:43 +01:00
|
|
|
pgprot_dmacoherent(pgprot_kernel));
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_alloc_coherent);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Allocate a writecombining region, in much the same way as
|
|
|
|
* dma_alloc_coherent above.
|
|
|
|
*/
|
|
|
|
void *
|
2005-10-21 09:20:58 +02:00
|
|
|
dma_alloc_writecombine(struct device *dev, size_t size, dma_addr_t *handle, gfp_t gfp)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
return __dma_alloc(dev, size, handle, gfp,
|
|
|
|
pgprot_writecombine(pgprot_kernel));
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_alloc_writecombine);
|
|
|
|
|
|
|
|
static int dma_mmap(struct device *dev, struct vm_area_struct *vma,
|
|
|
|
void *cpu_addr, dma_addr_t dma_addr, size_t size)
|
|
|
|
{
|
2009-07-24 13:35:02 +02:00
|
|
|
int ret = -ENXIO;
|
|
|
|
#ifdef CONFIG_MMU
|
2009-11-19 16:07:04 +01:00
|
|
|
unsigned long user_size, kern_size;
|
|
|
|
struct arm_vmregion *c;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
user_size = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
|
|
|
|
|
2009-11-19 16:07:04 +01:00
|
|
|
c = arm_vmregion_find(&consistent_head, (unsigned long)cpu_addr);
|
2005-04-17 00:20:36 +02:00
|
|
|
if (c) {
|
|
|
|
unsigned long off = vma->vm_pgoff;
|
|
|
|
|
|
|
|
kern_size = (c->vm_end - c->vm_start) >> PAGE_SHIFT;
|
|
|
|
|
|
|
|
if (off < kern_size &&
|
|
|
|
user_size <= (kern_size - off)) {
|
|
|
|
ret = remap_pfn_range(vma, vma->vm_start,
|
|
|
|
page_to_pfn(c->vm_pages) + off,
|
|
|
|
user_size << PAGE_SHIFT,
|
|
|
|
vma->vm_page_prot);
|
|
|
|
}
|
|
|
|
}
|
2009-07-24 13:35:02 +02:00
|
|
|
#endif /* CONFIG_MMU */
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
int dma_mmap_coherent(struct device *dev, struct vm_area_struct *vma,
|
|
|
|
void *cpu_addr, dma_addr_t dma_addr, size_t size)
|
|
|
|
{
|
2009-11-20 22:06:43 +01:00
|
|
|
vma->vm_page_prot = pgprot_dmacoherent(vma->vm_page_prot);
|
2005-04-17 00:20:36 +02:00
|
|
|
return dma_mmap(dev, vma, cpu_addr, dma_addr, size);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_mmap_coherent);
|
|
|
|
|
|
|
|
int dma_mmap_writecombine(struct device *dev, struct vm_area_struct *vma,
|
|
|
|
void *cpu_addr, dma_addr_t dma_addr, size_t size)
|
|
|
|
{
|
|
|
|
vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot);
|
|
|
|
return dma_mmap(dev, vma, cpu_addr, dma_addr, size);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_mmap_writecombine);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* free a page as defined by the above mapping.
|
2005-11-25 16:52:51 +01:00
|
|
|
* Must not be called with IRQs disabled.
|
2005-04-17 00:20:36 +02:00
|
|
|
*/
|
|
|
|
void dma_free_coherent(struct device *dev, size_t size, void *cpu_addr, dma_addr_t handle)
|
|
|
|
{
|
2005-11-25 16:52:51 +01:00
|
|
|
WARN_ON(irqs_disabled());
|
|
|
|
|
2008-07-18 11:30:14 +02:00
|
|
|
if (dma_release_from_coherent(dev, get_order(size), cpu_addr))
|
|
|
|
return;
|
|
|
|
|
2009-11-19 16:38:12 +01:00
|
|
|
size = PAGE_ALIGN(size);
|
|
|
|
|
2009-11-19 17:31:39 +01:00
|
|
|
if (!arch_is_coherent())
|
|
|
|
__dma_free_remap(cpu_addr, size);
|
2009-11-19 16:31:07 +01:00
|
|
|
|
2011-01-03 01:00:17 +01:00
|
|
|
__dma_free_buffer(pfn_to_page(dma_to_pfn(dev, handle)), size);
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_free_coherent);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Make an area consistent for devices.
|
2006-11-21 22:57:23 +01:00
|
|
|
* Note: Drivers should NOT use this function directly, as it will break
|
|
|
|
* platforms with CONFIG_DMABOUNCE.
|
|
|
|
* Use the driver DMA support - see dma-mapping.h (dma_sync_*)
|
2005-04-17 00:20:36 +02:00
|
|
|
*/
|
2009-11-24 17:27:17 +01:00
|
|
|
void ___dma_single_cpu_to_dev(const void *kaddr, size_t size,
|
|
|
|
enum dma_data_direction dir)
|
|
|
|
{
|
2009-10-31 17:52:16 +01:00
|
|
|
unsigned long paddr;
|
|
|
|
|
2009-11-26 17:19:58 +01:00
|
|
|
BUG_ON(!virt_addr_valid(kaddr) || !virt_addr_valid(kaddr + size - 1));
|
|
|
|
|
|
|
|
dmac_map_area(kaddr, size, dir);
|
2009-10-31 17:52:16 +01:00
|
|
|
|
|
|
|
paddr = __pa(kaddr);
|
|
|
|
if (dir == DMA_FROM_DEVICE) {
|
|
|
|
outer_inv_range(paddr, paddr + size);
|
|
|
|
} else {
|
|
|
|
outer_clean_range(paddr, paddr + size);
|
|
|
|
}
|
|
|
|
/* FIXME: non-speculating: flush on bidirectional mappings? */
|
2009-11-24 17:27:17 +01:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(___dma_single_cpu_to_dev);
|
|
|
|
|
|
|
|
void ___dma_single_dev_to_cpu(const void *kaddr, size_t size,
|
|
|
|
enum dma_data_direction dir)
|
|
|
|
{
|
2009-11-26 17:19:58 +01:00
|
|
|
BUG_ON(!virt_addr_valid(kaddr) || !virt_addr_valid(kaddr + size - 1));
|
|
|
|
|
2009-10-31 17:52:16 +01:00
|
|
|
/* FIXME: non-speculating: not required */
|
|
|
|
/* don't bother invalidating if DMA to device */
|
|
|
|
if (dir != DMA_TO_DEVICE) {
|
|
|
|
unsigned long paddr = __pa(kaddr);
|
|
|
|
outer_inv_range(paddr, paddr + size);
|
|
|
|
}
|
|
|
|
|
2009-11-26 17:19:58 +01:00
|
|
|
dmac_unmap_area(kaddr, size, dir);
|
2009-11-24 17:27:17 +01:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(___dma_single_dev_to_cpu);
|
2008-09-25 17:30:57 +02:00
|
|
|
|
2009-11-24 17:27:17 +01:00
|
|
|
static void dma_cache_maint_page(struct page *page, unsigned long offset,
|
2009-11-26 17:19:58 +01:00
|
|
|
size_t size, enum dma_data_direction dir,
|
|
|
|
void (*op)(const void *, size_t, int))
|
2009-03-13 03:52:09 +01:00
|
|
|
{
|
|
|
|
/*
|
|
|
|
* A single sg entry may refer to multiple physically contiguous
|
|
|
|
* pages. But we still need to process highmem pages individually.
|
|
|
|
* If highmem is not configured then the bulk of this loop gets
|
|
|
|
* optimized out.
|
|
|
|
*/
|
|
|
|
size_t left = size;
|
|
|
|
do {
|
|
|
|
size_t len = left;
|
2009-11-24 15:41:01 +01:00
|
|
|
void *vaddr;
|
|
|
|
|
|
|
|
if (PageHighMem(page)) {
|
|
|
|
if (len + offset > PAGE_SIZE) {
|
|
|
|
if (offset >= PAGE_SIZE) {
|
|
|
|
page += offset / PAGE_SIZE;
|
|
|
|
offset %= PAGE_SIZE;
|
|
|
|
}
|
|
|
|
len = PAGE_SIZE - offset;
|
|
|
|
}
|
|
|
|
vaddr = kmap_high_get(page);
|
|
|
|
if (vaddr) {
|
|
|
|
vaddr += offset;
|
2009-11-26 17:19:58 +01:00
|
|
|
op(vaddr, len, dir);
|
2009-11-24 15:41:01 +01:00
|
|
|
kunmap_high(page);
|
2010-03-29 22:46:02 +02:00
|
|
|
} else if (cache_is_vipt()) {
|
2010-12-15 21:14:45 +01:00
|
|
|
/* unmapped pages might still be cached */
|
|
|
|
vaddr = kmap_atomic(page);
|
2010-03-29 22:46:02 +02:00
|
|
|
op(vaddr + offset, len, dir);
|
2010-12-15 21:14:45 +01:00
|
|
|
kunmap_atomic(vaddr);
|
2009-03-13 03:52:09 +01:00
|
|
|
}
|
2009-11-24 15:41:01 +01:00
|
|
|
} else {
|
|
|
|
vaddr = page_address(page) + offset;
|
2009-11-26 17:19:58 +01:00
|
|
|
op(vaddr, len, dir);
|
2009-03-13 03:52:09 +01:00
|
|
|
}
|
|
|
|
offset = 0;
|
|
|
|
page++;
|
|
|
|
left -= len;
|
|
|
|
} while (left);
|
|
|
|
}
|
2009-11-24 17:27:17 +01:00
|
|
|
|
|
|
|
void ___dma_page_cpu_to_dev(struct page *page, unsigned long off,
|
|
|
|
size_t size, enum dma_data_direction dir)
|
|
|
|
{
|
2009-11-24 18:53:33 +01:00
|
|
|
unsigned long paddr;
|
|
|
|
|
2009-11-26 17:19:58 +01:00
|
|
|
dma_cache_maint_page(page, off, size, dir, dmac_map_area);
|
2009-11-24 18:53:33 +01:00
|
|
|
|
|
|
|
paddr = page_to_phys(page) + off;
|
2009-10-31 17:52:16 +01:00
|
|
|
if (dir == DMA_FROM_DEVICE) {
|
|
|
|
outer_inv_range(paddr, paddr + size);
|
|
|
|
} else {
|
|
|
|
outer_clean_range(paddr, paddr + size);
|
|
|
|
}
|
|
|
|
/* FIXME: non-speculating: flush on bidirectional mappings? */
|
2009-11-24 17:27:17 +01:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(___dma_page_cpu_to_dev);
|
|
|
|
|
|
|
|
void ___dma_page_dev_to_cpu(struct page *page, unsigned long off,
|
|
|
|
size_t size, enum dma_data_direction dir)
|
|
|
|
{
|
2009-10-31 17:52:16 +01:00
|
|
|
unsigned long paddr = page_to_phys(page) + off;
|
|
|
|
|
|
|
|
/* FIXME: non-speculating: not required */
|
|
|
|
/* don't bother invalidating if DMA to device */
|
|
|
|
if (dir != DMA_TO_DEVICE)
|
|
|
|
outer_inv_range(paddr, paddr + size);
|
|
|
|
|
2009-11-26 17:19:58 +01:00
|
|
|
dma_cache_maint_page(page, off, size, dir, dmac_unmap_area);
|
2010-09-13 16:57:36 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Mark the D-cache clean for this page to avoid extra flushing.
|
|
|
|
*/
|
|
|
|
if (dir != DMA_TO_DEVICE && off == 0 && size >= PAGE_SIZE)
|
|
|
|
set_bit(PG_dcache_clean, &page->flags);
|
2009-11-24 17:27:17 +01:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(___dma_page_dev_to_cpu);
|
2009-03-13 03:52:09 +01:00
|
|
|
|
2008-09-25 17:30:57 +02:00
|
|
|
/**
|
|
|
|
* dma_map_sg - map a set of SG buffers for streaming mode DMA
|
|
|
|
* @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
|
|
|
|
* @sg: list of buffers
|
|
|
|
* @nents: number of buffers to map
|
|
|
|
* @dir: DMA transfer direction
|
|
|
|
*
|
|
|
|
* Map a set of buffers described by scatterlist in streaming mode for DMA.
|
|
|
|
* This is the scatter-gather version of the dma_map_single interface.
|
|
|
|
* Here the scatter gather list elements are each tagged with the
|
|
|
|
* appropriate dma address and length. They are obtained via
|
|
|
|
* sg_dma_{address,length}.
|
|
|
|
*
|
|
|
|
* Device ownership issues as mentioned for dma_map_single are the same
|
|
|
|
* here.
|
|
|
|
*/
|
|
|
|
int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
|
|
|
|
enum dma_data_direction dir)
|
|
|
|
{
|
|
|
|
struct scatterlist *s;
|
2008-09-25 22:05:02 +02:00
|
|
|
int i, j;
|
2008-09-25 17:30:57 +02:00
|
|
|
|
2011-01-03 12:29:28 +01:00
|
|
|
BUG_ON(!valid_dma_direction(dir));
|
|
|
|
|
2008-09-25 17:30:57 +02:00
|
|
|
for_each_sg(sg, s, nents, i) {
|
2011-01-03 12:29:28 +01:00
|
|
|
s->dma_address = __dma_map_page(dev, sg_page(s), s->offset,
|
2008-09-25 22:05:02 +02:00
|
|
|
s->length, dir);
|
|
|
|
if (dma_mapping_error(dev, s->dma_address))
|
|
|
|
goto bad_mapping;
|
2008-09-25 17:30:57 +02:00
|
|
|
}
|
2011-01-03 12:29:28 +01:00
|
|
|
debug_dma_map_sg(dev, sg, nents, nents, dir);
|
2008-09-25 17:30:57 +02:00
|
|
|
return nents;
|
2008-09-25 22:05:02 +02:00
|
|
|
|
|
|
|
bad_mapping:
|
|
|
|
for_each_sg(sg, s, i, j)
|
2011-01-03 12:29:28 +01:00
|
|
|
__dma_unmap_page(dev, sg_dma_address(s), sg_dma_len(s), dir);
|
2008-09-25 22:05:02 +02:00
|
|
|
return 0;
|
2008-09-25 17:30:57 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_map_sg);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* dma_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg
|
|
|
|
* @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
|
|
|
|
* @sg: list of buffers
|
2011-01-12 18:50:37 +01:00
|
|
|
* @nents: number of buffers to unmap (same as was passed to dma_map_sg)
|
2008-09-25 17:30:57 +02:00
|
|
|
* @dir: DMA transfer direction (same as was passed to dma_map_sg)
|
|
|
|
*
|
|
|
|
* Unmap a set of streaming mode DMA translations. Again, CPU access
|
|
|
|
* rules concerning calls here are the same as for dma_unmap_single().
|
|
|
|
*/
|
|
|
|
void dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nents,
|
|
|
|
enum dma_data_direction dir)
|
|
|
|
{
|
2008-09-25 22:05:02 +02:00
|
|
|
struct scatterlist *s;
|
|
|
|
int i;
|
|
|
|
|
2011-01-03 12:29:28 +01:00
|
|
|
debug_dma_unmap_sg(dev, sg, nents, dir);
|
|
|
|
|
2008-09-25 22:05:02 +02:00
|
|
|
for_each_sg(sg, s, nents, i)
|
2011-01-03 12:29:28 +01:00
|
|
|
__dma_unmap_page(dev, sg_dma_address(s), sg_dma_len(s), dir);
|
2008-09-25 17:30:57 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_unmap_sg);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* dma_sync_sg_for_cpu
|
|
|
|
* @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
|
|
|
|
* @sg: list of buffers
|
|
|
|
* @nents: number of buffers to map (returned from dma_map_sg)
|
|
|
|
* @dir: DMA transfer direction (same as was passed to dma_map_sg)
|
|
|
|
*/
|
|
|
|
void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
|
|
|
|
int nents, enum dma_data_direction dir)
|
|
|
|
{
|
|
|
|
struct scatterlist *s;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for_each_sg(sg, s, nents, i) {
|
2009-10-31 17:52:16 +01:00
|
|
|
if (!dmabounce_sync_for_cpu(dev, sg_dma_address(s), 0,
|
|
|
|
sg_dma_len(s), dir))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
__dma_page_dev_to_cpu(sg_page(s), s->offset,
|
|
|
|
s->length, dir);
|
2008-09-25 17:30:57 +02:00
|
|
|
}
|
2011-01-03 12:29:28 +01:00
|
|
|
|
|
|
|
debug_dma_sync_sg_for_cpu(dev, sg, nents, dir);
|
2008-09-25 17:30:57 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_sync_sg_for_cpu);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* dma_sync_sg_for_device
|
|
|
|
* @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
|
|
|
|
* @sg: list of buffers
|
|
|
|
* @nents: number of buffers to map (returned from dma_map_sg)
|
|
|
|
* @dir: DMA transfer direction (same as was passed to dma_map_sg)
|
|
|
|
*/
|
|
|
|
void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
|
|
|
|
int nents, enum dma_data_direction dir)
|
|
|
|
{
|
|
|
|
struct scatterlist *s;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for_each_sg(sg, s, nents, i) {
|
2008-09-25 22:38:41 +02:00
|
|
|
if (!dmabounce_sync_for_device(dev, sg_dma_address(s), 0,
|
|
|
|
sg_dma_len(s), dir))
|
|
|
|
continue;
|
|
|
|
|
2009-10-31 17:52:16 +01:00
|
|
|
__dma_page_cpu_to_dev(sg_page(s), s->offset,
|
|
|
|
s->length, dir);
|
2008-09-25 17:30:57 +02:00
|
|
|
}
|
2011-01-03 12:29:28 +01:00
|
|
|
|
|
|
|
debug_dma_sync_sg_for_device(dev, sg, nents, dir);
|
2008-09-25 17:30:57 +02:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(dma_sync_sg_for_device);
|
2011-01-03 12:29:28 +01:00
|
|
|
|
|
|
|
#define PREALLOC_DMA_DEBUG_ENTRIES 4096
|
|
|
|
|
|
|
|
static int __init dma_debug_do_init(void)
|
|
|
|
{
|
|
|
|
dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
fs_initcall(dma_debug_do_init);
|