OpenCloudOS-Kernel/drivers/gpu/drm/armada/armada_gem.c

564 lines
13 KiB
C
Raw Normal View History

/*
* Copyright (C) 2012 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.
*/
#include <linux/dma-buf.h>
#include <linux/dma-mapping.h>
#include <linux/shmem_fs.h>
#include "armada_drm.h"
#include "armada_gem.h"
#include <drm/armada_drm.h>
#include "armada_ioctlP.h"
static vm_fault_t armada_gem_vm_fault(struct vm_fault *vmf)
{
struct drm_gem_object *gobj = vmf->vma->vm_private_data;
struct armada_gem_object *obj = drm_to_armada_gem(gobj);
unsigned long pfn = obj->phys_addr >> PAGE_SHIFT;
pfn += (vmf->address - vmf->vma->vm_start) >> PAGE_SHIFT;
return vmf_insert_pfn(vmf->vma, vmf->address, pfn);
}
const struct vm_operations_struct armada_gem_vm_ops = {
.fault = armada_gem_vm_fault,
.open = drm_gem_vm_open,
.close = drm_gem_vm_close,
};
static size_t roundup_gem_size(size_t size)
{
return roundup(size, PAGE_SIZE);
}
void armada_gem_free_object(struct drm_gem_object *obj)
{
struct armada_gem_object *dobj = drm_to_armada_gem(obj);
struct armada_private *priv = obj->dev->dev_private;
DRM_DEBUG_DRIVER("release obj %p\n", dobj);
drm_gem_free_mmap_offset(&dobj->obj);
might_lock(&priv->linear_lock);
if (dobj->page) {
/* page backed memory */
unsigned int order = get_order(dobj->obj.size);
__free_pages(dobj->page, order);
} else if (dobj->linear) {
/* linear backed memory */
mutex_lock(&priv->linear_lock);
drm_mm_remove_node(dobj->linear);
mutex_unlock(&priv->linear_lock);
kfree(dobj->linear);
if (dobj->addr)
iounmap(dobj->addr);
}
if (dobj->obj.import_attach) {
/* We only ever display imported data */
if (dobj->sgt)
dma_buf_unmap_attachment(dobj->obj.import_attach,
dobj->sgt, DMA_TO_DEVICE);
drm_prime_gem_destroy(&dobj->obj, NULL);
}
drm_gem_object_release(&dobj->obj);
kfree(dobj);
}
int
armada_gem_linear_back(struct drm_device *dev, struct armada_gem_object *obj)
{
struct armada_private *priv = dev->dev_private;
size_t size = obj->obj.size;
if (obj->page || obj->linear)
return 0;
/*
* If it is a small allocation (typically cursor, which will
* be 32x64 or 64x32 ARGB pixels) try to get it from the system.
* Framebuffers will never be this small (our minimum size for
* framebuffers is larger than this anyway.) Such objects are
* only accessed by the CPU so we don't need any special handing
* here.
*/
if (size <= 8192) {
unsigned int order = get_order(size);
struct page *p = alloc_pages(GFP_KERNEL, order);
if (p) {
obj->addr = page_address(p);
obj->phys_addr = page_to_phys(p);
obj->page = p;
memset(obj->addr, 0, PAGE_ALIGN(size));
}
}
/*
* We could grab something from CMA if it's enabled, but that
* involves building in a problem:
*
* CMA's interface uses dma_alloc_coherent(), which provides us
* with an CPU virtual address and a device address.
*
* The CPU virtual address may be either an address in the kernel
* direct mapped region (for example, as it would be on x86) or
* it may be remapped into another part of kernel memory space
* (eg, as it would be on ARM.) This means virt_to_phys() on the
* returned virtual address is invalid depending on the architecture
* implementation.
*
* The device address may also not be a physical address; it may
* be that there is some kind of remapping between the device and
* system RAM, which makes the use of the device address also
* unsafe to re-use as a physical address.
*
* This makes DRM usage of dma_alloc_coherent() in a generic way
* at best very questionable and unsafe.
*/
/* Otherwise, grab it from our linear allocation */
if (!obj->page) {
struct drm_mm_node *node;
unsigned align = min_t(unsigned, size, SZ_2M);
void __iomem *ptr;
int ret;
node = kzalloc(sizeof(*node), GFP_KERNEL);
if (!node)
return -ENOSPC;
mutex_lock(&priv->linear_lock);
drm: Improve drm_mm search (and fix topdown allocation) with rbtrees The drm_mm range manager claimed to support top-down insertion, but it was neither searching for the top-most hole that could fit the allocation request nor fitting the request to the hole correctly. In order to search the range efficiently, we create a secondary index for the holes using either their size or their address. This index allows us to find the smallest hole or the hole at the bottom or top of the range efficiently, whilst keeping the hole stack to rapidly service evictions. v2: Search for holes both high and low. Rename flags to mode. v3: Discover rb_entry_safe() and use it! v4: Kerneldoc for enum drm_mm_insert_mode. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: "Christian König" <christian.koenig@amd.com> Cc: David Airlie <airlied@linux.ie> Cc: Russell King <rmk+kernel@armlinux.org.uk> Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Sean Paul <seanpaul@chromium.org> Cc: Lucas Stach <l.stach@pengutronix.de> Cc: Christian Gmeiner <christian.gmeiner@gmail.com> Cc: Rob Clark <robdclark@gmail.com> Cc: Thierry Reding <thierry.reding@gmail.com> Cc: Stephen Warren <swarren@wwwdotorg.org> Cc: Alexandre Courbot <gnurou@gmail.com> Cc: Eric Anholt <eric@anholt.net> Cc: Sinclair Yeh <syeh@vmware.com> Cc: Thomas Hellstrom <thellstrom@vmware.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Sinclair Yeh <syeh@vmware.com> # vmwgfx Reviewed-by: Lucas Stach <l.stach@pengutronix.de> #etnaviv Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20170202210438.28702-1-chris@chris-wilson.co.uk
2017-02-03 05:04:38 +08:00
ret = drm_mm_insert_node_generic(&priv->linear, node,
size, align, 0, 0);
mutex_unlock(&priv->linear_lock);
if (ret) {
kfree(node);
return ret;
}
obj->linear = node;
/* Ensure that the memory we're returning is cleared. */
ptr = ioremap_wc(obj->linear->start, size);
if (!ptr) {
mutex_lock(&priv->linear_lock);
drm_mm_remove_node(obj->linear);
mutex_unlock(&priv->linear_lock);
kfree(obj->linear);
obj->linear = NULL;
return -ENOMEM;
}
memset_io(ptr, 0, size);
iounmap(ptr);
obj->phys_addr = obj->linear->start;
obj->dev_addr = obj->linear->start;
obj->mapped = true;
}
DRM_DEBUG_DRIVER("obj %p phys %#llx dev %#llx\n", obj,
(unsigned long long)obj->phys_addr,
(unsigned long long)obj->dev_addr);
return 0;
}
void *
armada_gem_map_object(struct drm_device *dev, struct armada_gem_object *dobj)
{
/* only linear objects need to be ioremap'd */
if (!dobj->addr && dobj->linear)
dobj->addr = ioremap_wc(dobj->phys_addr, dobj->obj.size);
return dobj->addr;
}
struct armada_gem_object *
armada_gem_alloc_private_object(struct drm_device *dev, size_t size)
{
struct armada_gem_object *obj;
size = roundup_gem_size(size);
obj = kzalloc(sizeof(*obj), GFP_KERNEL);
if (!obj)
return NULL;
drm_gem_private_object_init(dev, &obj->obj, size);
DRM_DEBUG_DRIVER("alloc private obj %p size %zu\n", obj, size);
return obj;
}
static struct armada_gem_object *armada_gem_alloc_object(struct drm_device *dev,
size_t size)
{
struct armada_gem_object *obj;
struct address_space *mapping;
size = roundup_gem_size(size);
obj = kzalloc(sizeof(*obj), GFP_KERNEL);
if (!obj)
return NULL;
if (drm_gem_object_init(dev, &obj->obj, size)) {
kfree(obj);
return NULL;
}
mapping = obj->obj.filp->f_mapping;
mapping_set_gfp_mask(mapping, GFP_HIGHUSER | __GFP_RECLAIMABLE);
DRM_DEBUG_DRIVER("alloc obj %p size %zu\n", obj, size);
return obj;
}
/* Dumb alloc support */
int armada_gem_dumb_create(struct drm_file *file, struct drm_device *dev,
struct drm_mode_create_dumb *args)
{
struct armada_gem_object *dobj;
u32 handle;
size_t size;
int ret;
args->pitch = armada_pitch(args->width, args->bpp);
args->size = size = args->pitch * args->height;
dobj = armada_gem_alloc_private_object(dev, size);
if (dobj == NULL)
return -ENOMEM;
ret = armada_gem_linear_back(dev, dobj);
if (ret)
goto err;
ret = drm_gem_handle_create(file, &dobj->obj, &handle);
if (ret)
goto err;
args->handle = handle;
/* drop reference from allocate - handle holds it now */
DRM_DEBUG_DRIVER("obj %p size %zu handle %#x\n", dobj, size, handle);
err:
drm_gem_object_put_unlocked(&dobj->obj);
return ret;
}
/* Private driver gem ioctls */
int armada_gem_create_ioctl(struct drm_device *dev, void *data,
struct drm_file *file)
{
struct drm_armada_gem_create *args = data;
struct armada_gem_object *dobj;
size_t size;
u32 handle;
int ret;
if (args->size == 0)
return -ENOMEM;
size = args->size;
dobj = armada_gem_alloc_object(dev, size);
if (dobj == NULL)
return -ENOMEM;
ret = drm_gem_handle_create(file, &dobj->obj, &handle);
if (ret)
goto err;
args->handle = handle;
/* drop reference from allocate - handle holds it now */
DRM_DEBUG_DRIVER("obj %p size %zu handle %#x\n", dobj, size, handle);
err:
drm_gem_object_put_unlocked(&dobj->obj);
return ret;
}
/* Map a shmem-backed object into process memory space */
int armada_gem_mmap_ioctl(struct drm_device *dev, void *data,
struct drm_file *file)
{
struct drm_armada_gem_mmap *args = data;
struct armada_gem_object *dobj;
unsigned long addr;
dobj = armada_gem_object_lookup(file, args->handle);
if (dobj == NULL)
return -ENOENT;
if (!dobj->obj.filp) {
drm_gem_object_put_unlocked(&dobj->obj);
return -EINVAL;
}
addr = vm_mmap(dobj->obj.filp, 0, args->size, PROT_READ | PROT_WRITE,
MAP_SHARED, args->offset);
drm_gem_object_put_unlocked(&dobj->obj);
if (IS_ERR_VALUE(addr))
return addr;
args->addr = addr;
return 0;
}
int armada_gem_pwrite_ioctl(struct drm_device *dev, void *data,
struct drm_file *file)
{
struct drm_armada_gem_pwrite *args = data;
struct armada_gem_object *dobj;
char __user *ptr;
int ret;
DRM_DEBUG_DRIVER("handle %u off %u size %u ptr 0x%llx\n",
args->handle, args->offset, args->size, args->ptr);
if (args->size == 0)
return 0;
ptr = (char __user *)(uintptr_t)args->ptr;
Remove 'type' argument from access_ok() function Nobody has actually used the type (VERIFY_READ vs VERIFY_WRITE) argument of the user address range verification function since we got rid of the old racy i386-only code to walk page tables by hand. It existed because the original 80386 would not honor the write protect bit when in kernel mode, so you had to do COW by hand before doing any user access. But we haven't supported that in a long time, and these days the 'type' argument is a purely historical artifact. A discussion about extending 'user_access_begin()' to do the range checking resulted this patch, because there is no way we're going to move the old VERIFY_xyz interface to that model. And it's best done at the end of the merge window when I've done most of my merges, so let's just get this done once and for all. This patch was mostly done with a sed-script, with manual fix-ups for the cases that weren't of the trivial 'access_ok(VERIFY_xyz' form. There were a couple of notable cases: - csky still had the old "verify_area()" name as an alias. - the iter_iov code had magical hardcoded knowledge of the actual values of VERIFY_{READ,WRITE} (not that they mattered, since nothing really used it) - microblaze used the type argument for a debug printout but other than those oddities this should be a total no-op patch. I tried to fix up all architectures, did fairly extensive grepping for access_ok() uses, and the changes are trivial, but I may have missed something. Any missed conversion should be trivially fixable, though. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-04 10:57:57 +08:00
if (!access_ok(ptr, args->size))
return -EFAULT;
ret = fault_in_pages_readable(ptr, args->size);
if (ret)
return ret;
dobj = armada_gem_object_lookup(file, args->handle);
if (dobj == NULL)
return -ENOENT;
/* Must be a kernel-mapped object */
if (!dobj->addr)
return -EINVAL;
if (args->offset > dobj->obj.size ||
args->size > dobj->obj.size - args->offset) {
DRM_ERROR("invalid size: object size %u\n", dobj->obj.size);
ret = -EINVAL;
goto unref;
}
if (copy_from_user(dobj->addr + args->offset, ptr, args->size)) {
ret = -EFAULT;
} else if (dobj->update) {
dobj->update(dobj->update_data);
ret = 0;
}
unref:
drm_gem_object_put_unlocked(&dobj->obj);
return ret;
}
/* Prime support */
static struct sg_table *
armada_gem_prime_map_dma_buf(struct dma_buf_attachment *attach,
enum dma_data_direction dir)
{
struct drm_gem_object *obj = attach->dmabuf->priv;
struct armada_gem_object *dobj = drm_to_armada_gem(obj);
struct scatterlist *sg;
struct sg_table *sgt;
int i, num;
sgt = kmalloc(sizeof(*sgt), GFP_KERNEL);
if (!sgt)
return NULL;
if (dobj->obj.filp) {
struct address_space *mapping;
int count;
count = dobj->obj.size / PAGE_SIZE;
if (sg_alloc_table(sgt, count, GFP_KERNEL))
goto free_sgt;
mapping = dobj->obj.filp->f_mapping;
for_each_sg(sgt->sgl, sg, count, i) {
struct page *page;
page = shmem_read_mapping_page(mapping, i);
if (IS_ERR(page)) {
num = i;
goto release;
}
sg_set_page(sg, page, PAGE_SIZE, 0);
}
if (dma_map_sg(attach->dev, sgt->sgl, sgt->nents, dir) == 0) {
num = sgt->nents;
goto release;
}
} else if (dobj->page) {
/* Single contiguous page */
if (sg_alloc_table(sgt, 1, GFP_KERNEL))
goto free_sgt;
sg_set_page(sgt->sgl, dobj->page, dobj->obj.size, 0);
if (dma_map_sg(attach->dev, sgt->sgl, sgt->nents, dir) == 0)
goto free_table;
} else if (dobj->linear) {
/* Single contiguous physical region - no struct page */
if (sg_alloc_table(sgt, 1, GFP_KERNEL))
goto free_sgt;
sg_dma_address(sgt->sgl) = dobj->dev_addr;
sg_dma_len(sgt->sgl) = dobj->obj.size;
} else {
goto free_sgt;
}
return sgt;
release:
for_each_sg(sgt->sgl, sg, num, i)
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 20:29:47 +08:00
put_page(sg_page(sg));
free_table:
sg_free_table(sgt);
free_sgt:
kfree(sgt);
return NULL;
}
static void armada_gem_prime_unmap_dma_buf(struct dma_buf_attachment *attach,
struct sg_table *sgt, enum dma_data_direction dir)
{
struct drm_gem_object *obj = attach->dmabuf->priv;
struct armada_gem_object *dobj = drm_to_armada_gem(obj);
int i;
if (!dobj->linear)
dma_unmap_sg(attach->dev, sgt->sgl, sgt->nents, dir);
if (dobj->obj.filp) {
struct scatterlist *sg;
for_each_sg(sgt->sgl, sg, sgt->nents, i)
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 20:29:47 +08:00
put_page(sg_page(sg));
}
sg_free_table(sgt);
kfree(sgt);
}
static void *armada_gem_dmabuf_no_kmap(struct dma_buf *buf, unsigned long n)
{
return NULL;
}
static void
armada_gem_dmabuf_no_kunmap(struct dma_buf *buf, unsigned long n, void *addr)
{
}
static int
armada_gem_dmabuf_mmap(struct dma_buf *buf, struct vm_area_struct *vma)
{
return -EINVAL;
}
static const struct dma_buf_ops armada_gem_prime_dmabuf_ops = {
.map_dma_buf = armada_gem_prime_map_dma_buf,
.unmap_dma_buf = armada_gem_prime_unmap_dma_buf,
.release = drm_gem_dmabuf_release,
.map = armada_gem_dmabuf_no_kmap,
.unmap = armada_gem_dmabuf_no_kunmap,
.mmap = armada_gem_dmabuf_mmap,
};
struct dma_buf *
armada_gem_prime_export(struct drm_device *dev, struct drm_gem_object *obj,
int flags)
{
DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
exp_info.ops = &armada_gem_prime_dmabuf_ops;
exp_info.size = obj->size;
exp_info.flags = O_RDWR;
exp_info.priv = obj;
return drm_gem_dmabuf_export(dev, &exp_info);
}
struct drm_gem_object *
armada_gem_prime_import(struct drm_device *dev, struct dma_buf *buf)
{
struct dma_buf_attachment *attach;
struct armada_gem_object *dobj;
if (buf->ops == &armada_gem_prime_dmabuf_ops) {
struct drm_gem_object *obj = buf->priv;
if (obj->dev == dev) {
/*
* Importing our own dmabuf(s) increases the
* refcount on the gem object itself.
*/
drm_gem_object_get(obj);
return obj;
}
}
attach = dma_buf_attach(buf, dev->dev);
if (IS_ERR(attach))
return ERR_CAST(attach);
dobj = armada_gem_alloc_private_object(dev, buf->size);
if (!dobj) {
dma_buf_detach(buf, attach);
return ERR_PTR(-ENOMEM);
}
dobj->obj.import_attach = attach;
get_dma_buf(buf);
/*
* Don't call dma_buf_map_attachment() here - it maps the
* scatterlist immediately for DMA, and this is not always
* an appropriate thing to do.
*/
return &dobj->obj;
}
int armada_gem_map_import(struct armada_gem_object *dobj)
{
int ret;
dobj->sgt = dma_buf_map_attachment(dobj->obj.import_attach,
DMA_TO_DEVICE);
if (IS_ERR(dobj->sgt)) {
ret = PTR_ERR(dobj->sgt);
dobj->sgt = NULL;
DRM_ERROR("dma_buf_map_attachment() error: %d\n", ret);
return ret;
}
if (dobj->sgt->nents > 1) {
DRM_ERROR("dma_buf_map_attachment() returned an (unsupported) scattered list\n");
return -EINVAL;
}
if (sg_dma_len(dobj->sgt->sgl) < dobj->obj.size) {
DRM_ERROR("dma_buf_map_attachment() returned a small buffer\n");
return -EINVAL;
}
dobj->dev_addr = sg_dma_address(dobj->sgt->sgl);
dobj->mapped = true;
return 0;
}