2019-06-04 16:11:33 +08:00
|
|
|
// SPDX-License-Identifier: GPL-2.0-only
|
2013-05-21 11:33:10 +08:00
|
|
|
/*
|
|
|
|
* VFIO: IOMMU DMA mapping support for TCE on POWER
|
|
|
|
*
|
|
|
|
* Copyright (C) 2013 IBM Corp. All rights reserved.
|
|
|
|
* Author: Alexey Kardashevskiy <aik@ozlabs.ru>
|
2022-12-05 23:29:18 +08:00
|
|
|
* Copyright Gavin Shan, IBM Corporation 2014.
|
2013-05-21 11:33:10 +08:00
|
|
|
*
|
|
|
|
* Derived from original vfio_iommu_type1.c:
|
|
|
|
* Copyright (C) 2012 Red Hat, Inc. All rights reserved.
|
|
|
|
* Author: Alex Williamson <alex.williamson@redhat.com>
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/pci.h>
|
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/uaccess.h>
|
|
|
|
#include <linux/err.h>
|
|
|
|
#include <linux/vfio.h>
|
2015-06-05 14:35:25 +08:00
|
|
|
#include <linux/vmalloc.h>
|
2017-02-09 01:51:29 +08:00
|
|
|
#include <linux/sched/mm.h>
|
2017-02-09 01:51:30 +08:00
|
|
|
#include <linux/sched/signal.h>
|
2019-07-17 07:30:54 +08:00
|
|
|
#include <linux/mm.h>
|
2021-09-24 23:56:59 +08:00
|
|
|
#include "vfio.h"
|
2017-02-09 01:51:29 +08:00
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
#include <asm/iommu.h>
|
|
|
|
#include <asm/tce.h>
|
2015-06-05 14:35:25 +08:00
|
|
|
#include <asm/mmu_context.h>
|
2013-05-21 11:33:10 +08:00
|
|
|
|
|
|
|
#define DRIVER_VERSION "0.1"
|
|
|
|
#define DRIVER_AUTHOR "aik@ozlabs.ru"
|
|
|
|
#define DRIVER_DESC "VFIO IOMMU SPAPR TCE"
|
|
|
|
|
|
|
|
static void tce_iommu_detach_group(void *iommu_data,
|
|
|
|
struct iommu_group *iommu_group);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* VFIO IOMMU fd for SPAPR_TCE IOMMU implementation
|
|
|
|
*
|
|
|
|
* This code handles mapping and unmapping of user data buffers
|
|
|
|
* into DMA'ble space using the IOMMU
|
|
|
|
*/
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
struct tce_iommu_group {
|
|
|
|
struct list_head next;
|
|
|
|
struct iommu_group *grp;
|
|
|
|
};
|
|
|
|
|
2016-11-30 14:52:05 +08:00
|
|
|
/*
|
|
|
|
* A container needs to remember which preregistered region it has
|
|
|
|
* referenced to do proper cleanup at the userspace process exit.
|
|
|
|
*/
|
|
|
|
struct tce_iommu_prereg {
|
|
|
|
struct list_head next;
|
|
|
|
struct mm_iommu_table_group_mem_t *mem;
|
|
|
|
};
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
/*
|
|
|
|
* The container descriptor supports only a single group per container.
|
|
|
|
* Required by the API as the container is not supplied with the IOMMU group
|
|
|
|
* at the moment of initialization.
|
|
|
|
*/
|
|
|
|
struct tce_container {
|
|
|
|
struct mutex lock;
|
|
|
|
bool enabled;
|
2015-06-05 14:35:25 +08:00
|
|
|
bool v2;
|
2016-11-30 14:52:03 +08:00
|
|
|
bool def_window_pending;
|
2015-06-05 14:35:01 +08:00
|
|
|
unsigned long locked_pages;
|
2016-11-30 14:52:04 +08:00
|
|
|
struct mm_struct *mm;
|
2015-06-05 14:35:25 +08:00
|
|
|
struct iommu_table *tables[IOMMU_TABLE_GROUP_MAX_TABLES];
|
|
|
|
struct list_head group_list;
|
2016-11-30 14:52:05 +08:00
|
|
|
struct list_head prereg_list;
|
2013-05-21 11:33:10 +08:00
|
|
|
};
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
static long tce_iommu_mm_set(struct tce_container *container)
|
|
|
|
{
|
|
|
|
if (container->mm) {
|
|
|
|
if (container->mm == current->mm)
|
|
|
|
return 0;
|
|
|
|
return -EPERM;
|
|
|
|
}
|
|
|
|
BUG_ON(!current->mm);
|
|
|
|
container->mm = current->mm;
|
2019-12-29 23:42:57 +08:00
|
|
|
mmgrab(container->mm);
|
2016-11-30 14:52:04 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-11-30 14:52:05 +08:00
|
|
|
static long tce_iommu_prereg_free(struct tce_container *container,
|
|
|
|
struct tce_iommu_prereg *tcemem)
|
|
|
|
{
|
|
|
|
long ret;
|
|
|
|
|
|
|
|
ret = mm_iommu_put(container->mm, tcemem->mem);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
list_del(&tcemem->next);
|
|
|
|
kfree(tcemem);
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
static long tce_iommu_unregister_pages(struct tce_container *container,
|
|
|
|
__u64 vaddr, __u64 size)
|
|
|
|
{
|
|
|
|
struct mm_iommu_table_group_mem_t *mem;
|
2016-11-30 14:52:05 +08:00
|
|
|
struct tce_iommu_prereg *tcemem;
|
|
|
|
bool found = false;
|
2018-12-19 16:52:14 +08:00
|
|
|
long ret;
|
2015-06-05 14:35:25 +08:00
|
|
|
|
|
|
|
if ((vaddr & ~PAGE_MASK) || (size & ~PAGE_MASK))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2018-12-19 16:52:14 +08:00
|
|
|
mem = mm_iommu_get(container->mm, vaddr, size >> PAGE_SHIFT);
|
2015-06-05 14:35:25 +08:00
|
|
|
if (!mem)
|
|
|
|
return -ENOENT;
|
|
|
|
|
2016-11-30 14:52:05 +08:00
|
|
|
list_for_each_entry(tcemem, &container->prereg_list, next) {
|
|
|
|
if (tcemem->mem == mem) {
|
|
|
|
found = true;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!found)
|
2018-12-19 16:52:14 +08:00
|
|
|
ret = -ENOENT;
|
|
|
|
else
|
|
|
|
ret = tce_iommu_prereg_free(container, tcemem);
|
2016-11-30 14:52:05 +08:00
|
|
|
|
2018-12-19 16:52:14 +08:00
|
|
|
mm_iommu_put(container->mm, mem);
|
|
|
|
|
|
|
|
return ret;
|
2015-06-05 14:35:25 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static long tce_iommu_register_pages(struct tce_container *container,
|
|
|
|
__u64 vaddr, __u64 size)
|
|
|
|
{
|
|
|
|
long ret = 0;
|
|
|
|
struct mm_iommu_table_group_mem_t *mem = NULL;
|
2016-11-30 14:52:05 +08:00
|
|
|
struct tce_iommu_prereg *tcemem;
|
2015-06-05 14:35:25 +08:00
|
|
|
unsigned long entries = size >> PAGE_SHIFT;
|
|
|
|
|
|
|
|
if ((vaddr & ~PAGE_MASK) || (size & ~PAGE_MASK) ||
|
|
|
|
((vaddr + size) < vaddr))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2018-12-19 16:52:14 +08:00
|
|
|
mem = mm_iommu_get(container->mm, vaddr, entries);
|
2016-11-30 14:52:05 +08:00
|
|
|
if (mem) {
|
|
|
|
list_for_each_entry(tcemem, &container->prereg_list, next) {
|
2018-12-19 16:52:14 +08:00
|
|
|
if (tcemem->mem == mem) {
|
|
|
|
ret = -EBUSY;
|
|
|
|
goto put_exit;
|
|
|
|
}
|
2016-11-30 14:52:05 +08:00
|
|
|
}
|
2018-12-19 16:52:14 +08:00
|
|
|
} else {
|
|
|
|
ret = mm_iommu_new(container->mm, vaddr, entries, &mem);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
2016-11-30 14:52:05 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
tcemem = kzalloc(sizeof(*tcemem), GFP_KERNEL);
|
2017-03-27 11:23:40 +08:00
|
|
|
if (!tcemem) {
|
2018-12-19 16:52:14 +08:00
|
|
|
ret = -ENOMEM;
|
|
|
|
goto put_exit;
|
2017-03-27 11:23:40 +08:00
|
|
|
}
|
|
|
|
|
2016-11-30 14:52:05 +08:00
|
|
|
tcemem->mem = mem;
|
|
|
|
list_add(&tcemem->next, &container->prereg_list);
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
container->enabled = true;
|
|
|
|
|
|
|
|
return 0;
|
2018-12-19 16:52:14 +08:00
|
|
|
|
|
|
|
put_exit:
|
|
|
|
mm_iommu_put(container->mm, mem);
|
|
|
|
return ret;
|
2015-06-05 14:35:25 +08:00
|
|
|
}
|
|
|
|
|
2018-12-19 16:52:15 +08:00
|
|
|
static bool tce_page_is_contained(struct mm_struct *mm, unsigned long hpa,
|
2019-09-24 06:34:28 +08:00
|
|
|
unsigned int it_page_shift)
|
2015-06-05 14:34:59 +08:00
|
|
|
{
|
2018-12-19 16:52:15 +08:00
|
|
|
struct page *page;
|
|
|
|
unsigned long size = 0;
|
|
|
|
|
2019-09-24 06:34:28 +08:00
|
|
|
if (mm_iommu_is_devmem(mm, hpa, it_page_shift, &size))
|
|
|
|
return size == (1UL << it_page_shift);
|
2018-12-19 16:52:15 +08:00
|
|
|
|
|
|
|
page = pfn_to_page(hpa >> PAGE_SHIFT);
|
2015-06-05 14:34:59 +08:00
|
|
|
/*
|
|
|
|
* Check that the TCE table granularity is not bigger than the size of
|
|
|
|
* a page we just found. Otherwise the hardware can get access to
|
|
|
|
* a bigger memory chunk that it should.
|
|
|
|
*/
|
2019-09-24 06:34:28 +08:00
|
|
|
return page_shift(compound_head(page)) >= it_page_shift;
|
2015-06-05 14:34:59 +08:00
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
static inline bool tce_groups_attached(struct tce_container *container)
|
|
|
|
{
|
|
|
|
return !list_empty(&container->group_list);
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:09 +08:00
|
|
|
static long tce_iommu_find_table(struct tce_container *container,
|
|
|
|
phys_addr_t ioba, struct iommu_table **ptbl)
|
|
|
|
{
|
|
|
|
long i;
|
|
|
|
|
|
|
|
for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
|
2015-06-05 14:35:25 +08:00
|
|
|
struct iommu_table *tbl = container->tables[i];
|
2015-06-05 14:35:09 +08:00
|
|
|
|
|
|
|
if (tbl) {
|
|
|
|
unsigned long entry = ioba >> tbl->it_page_shift;
|
|
|
|
unsigned long start = tbl->it_offset;
|
|
|
|
unsigned long end = start + tbl->it_size;
|
|
|
|
|
|
|
|
if ((start <= entry) && (entry < end)) {
|
|
|
|
*ptbl = tbl;
|
|
|
|
return i;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:26 +08:00
|
|
|
static int tce_iommu_find_free_table(struct tce_container *container)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
|
|
|
for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
|
|
|
|
if (!container->tables[i])
|
|
|
|
return i;
|
|
|
|
}
|
|
|
|
|
|
|
|
return -ENOSPC;
|
|
|
|
}
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
static int tce_iommu_enable(struct tce_container *container)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
2015-06-05 14:35:01 +08:00
|
|
|
unsigned long locked;
|
2015-06-05 14:35:09 +08:00
|
|
|
struct iommu_table_group *table_group;
|
2015-06-05 14:35:25 +08:00
|
|
|
struct tce_iommu_group *tcegrp;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
|
|
|
if (container->enabled)
|
|
|
|
return -EBUSY;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* When userspace pages are mapped into the IOMMU, they are effectively
|
|
|
|
* locked memory, so, theoretically, we need to update the accounting
|
|
|
|
* of locked pages on each map and unmap. For powerpc, the map unmap
|
|
|
|
* paths can be very hot, though, and the accounting would kill
|
|
|
|
* performance, especially since it would be difficult to impossible
|
|
|
|
* to handle the accounting in real mode only.
|
|
|
|
*
|
|
|
|
* To address that, rather than precisely accounting every page, we
|
|
|
|
* instead account for a worst case on locked memory when the iommu is
|
|
|
|
* enabled and disabled. The worst case upper bound on locked memory
|
|
|
|
* is the size of the whole iommu window, which is usually relatively
|
|
|
|
* small (compared to total memory sizes) on POWER hardware.
|
|
|
|
*
|
|
|
|
* Also we don't have a nice way to fail on H_PUT_TCE due to ulimits,
|
|
|
|
* that would effectively kill the guest at random points, much better
|
|
|
|
* enforcing the limit based on the max that the guest can map.
|
2015-06-05 14:35:01 +08:00
|
|
|
*
|
|
|
|
* Unfortunately at the moment it counts whole tables, no matter how
|
|
|
|
* much memory the guest has. I.e. for 4GB guest and 4 IOMMU groups
|
|
|
|
* each with 2GB DMA window, 8GB will be counted here. The reason for
|
|
|
|
* this is that we cannot tell here the amount of RAM used by the guest
|
|
|
|
* as this information is only available from KVM and VFIO is
|
|
|
|
* KVM agnostic.
|
2015-06-05 14:35:20 +08:00
|
|
|
*
|
|
|
|
* So we do not allow enabling a container without a group attached
|
|
|
|
* as there is no way to know how much we should increment
|
|
|
|
* the locked_vm counter.
|
2013-05-21 11:33:10 +08:00
|
|
|
*/
|
2015-06-05 14:35:25 +08:00
|
|
|
if (!tce_groups_attached(container))
|
|
|
|
return -ENODEV;
|
|
|
|
|
|
|
|
tcegrp = list_first_entry(&container->group_list,
|
|
|
|
struct tce_iommu_group, next);
|
|
|
|
table_group = iommu_group_get_iommudata(tcegrp->grp);
|
2015-06-05 14:35:09 +08:00
|
|
|
if (!table_group)
|
|
|
|
return -ENODEV;
|
|
|
|
|
2015-06-05 14:35:20 +08:00
|
|
|
if (!table_group->tce32_size)
|
|
|
|
return -EPERM;
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
ret = tce_iommu_mm_set(container);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
2015-06-05 14:35:20 +08:00
|
|
|
locked = table_group->tce32_size >> PAGE_SHIFT;
|
2019-07-17 07:30:54 +08:00
|
|
|
ret = account_locked_vm(container->mm, locked, true);
|
2015-06-05 14:35:01 +08:00
|
|
|
if (ret)
|
|
|
|
return ret;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2015-06-05 14:35:01 +08:00
|
|
|
container->locked_pages = locked;
|
|
|
|
|
|
|
|
container->enabled = true;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void tce_iommu_disable(struct tce_container *container)
|
|
|
|
{
|
|
|
|
if (!container->enabled)
|
|
|
|
return;
|
|
|
|
|
|
|
|
container->enabled = false;
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
BUG_ON(!container->mm);
|
2019-07-17 07:30:54 +08:00
|
|
|
account_locked_vm(container->mm, container->locked_pages, false);
|
2013-05-21 11:33:10 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static void *tce_iommu_open(unsigned long arg)
|
|
|
|
{
|
|
|
|
struct tce_container *container;
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
if ((arg != VFIO_SPAPR_TCE_IOMMU) && (arg != VFIO_SPAPR_TCE_v2_IOMMU)) {
|
2013-05-21 11:33:10 +08:00
|
|
|
pr_err("tce_vfio: Wrong IOMMU type\n");
|
|
|
|
return ERR_PTR(-EINVAL);
|
|
|
|
}
|
|
|
|
|
|
|
|
container = kzalloc(sizeof(*container), GFP_KERNEL);
|
|
|
|
if (!container)
|
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
|
|
|
|
mutex_init(&container->lock);
|
2015-06-05 14:35:25 +08:00
|
|
|
INIT_LIST_HEAD_RCU(&container->group_list);
|
2016-11-30 14:52:05 +08:00
|
|
|
INIT_LIST_HEAD_RCU(&container->prereg_list);
|
2015-06-05 14:35:25 +08:00
|
|
|
|
|
|
|
container->v2 = arg == VFIO_SPAPR_TCE_v2_IOMMU;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
|
|
|
return container;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
static int tce_iommu_clear(struct tce_container *container,
|
|
|
|
struct iommu_table *tbl,
|
|
|
|
unsigned long entry, unsigned long pages);
|
2016-11-30 14:52:04 +08:00
|
|
|
static void tce_iommu_free_table(struct tce_container *container,
|
|
|
|
struct iommu_table *tbl);
|
2015-06-05 14:35:25 +08:00
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
static void tce_iommu_release(void *iommu_data)
|
|
|
|
{
|
|
|
|
struct tce_container *container = iommu_data;
|
2015-06-05 14:35:25 +08:00
|
|
|
struct tce_iommu_group *tcegrp;
|
2018-10-02 11:22:31 +08:00
|
|
|
struct tce_iommu_prereg *tcemem, *tmtmp;
|
2015-06-05 14:35:25 +08:00
|
|
|
long i;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
while (tce_groups_attached(container)) {
|
|
|
|
tcegrp = list_first_entry(&container->group_list,
|
|
|
|
struct tce_iommu_group, next);
|
|
|
|
tce_iommu_detach_group(iommu_data, tcegrp->grp);
|
|
|
|
}
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
/*
|
|
|
|
* If VFIO created a table, it was not disposed
|
|
|
|
* by tce_iommu_detach_group() so do it now.
|
|
|
|
*/
|
|
|
|
for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
|
|
|
|
struct iommu_table *tbl = container->tables[i];
|
|
|
|
|
|
|
|
if (!tbl)
|
|
|
|
continue;
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
tce_iommu_free_table(container, tbl);
|
2015-06-05 14:35:25 +08:00
|
|
|
}
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2018-10-02 11:22:31 +08:00
|
|
|
list_for_each_entry_safe(tcemem, tmtmp, &container->prereg_list, next)
|
|
|
|
WARN_ON(tce_iommu_prereg_free(container, tcemem));
|
2016-11-30 14:52:05 +08:00
|
|
|
|
2015-06-05 14:35:03 +08:00
|
|
|
tce_iommu_disable(container);
|
2016-11-30 14:52:04 +08:00
|
|
|
if (container->mm)
|
|
|
|
mmdrop(container->mm);
|
2013-05-21 11:33:10 +08:00
|
|
|
mutex_destroy(&container->lock);
|
|
|
|
|
|
|
|
kfree(container);
|
|
|
|
}
|
|
|
|
|
2022-07-02 14:46:13 +08:00
|
|
|
static void tce_iommu_unuse_page(unsigned long hpa)
|
2015-06-05 14:35:03 +08:00
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
|
2015-06-05 14:35:15 +08:00
|
|
|
page = pfn_to_page(hpa >> PAGE_SHIFT);
|
2020-07-28 03:43:38 +08:00
|
|
|
unpin_user_page(page);
|
2015-06-05 14:35:03 +08:00
|
|
|
}
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
static int tce_iommu_prereg_ua_to_hpa(struct tce_container *container,
|
2018-07-17 15:19:12 +08:00
|
|
|
unsigned long tce, unsigned long shift,
|
2015-06-05 14:35:25 +08:00
|
|
|
unsigned long *phpa, struct mm_iommu_table_group_mem_t **pmem)
|
|
|
|
{
|
|
|
|
long ret = 0;
|
|
|
|
struct mm_iommu_table_group_mem_t *mem;
|
|
|
|
|
2018-07-17 15:19:12 +08:00
|
|
|
mem = mm_iommu_lookup(container->mm, tce, 1ULL << shift);
|
2015-06-05 14:35:25 +08:00
|
|
|
if (!mem)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2018-07-17 15:19:13 +08:00
|
|
|
ret = mm_iommu_ua_to_hpa(mem, tce, shift, phpa);
|
2015-06-05 14:35:25 +08:00
|
|
|
if (ret)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
*pmem = mem;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
static void tce_iommu_unuse_page_v2(struct tce_container *container,
|
|
|
|
struct iommu_table *tbl, unsigned long entry)
|
2015-06-05 14:35:25 +08:00
|
|
|
{
|
|
|
|
struct mm_iommu_table_group_mem_t *mem = NULL;
|
|
|
|
int ret;
|
|
|
|
unsigned long hpa = 0;
|
2018-10-15 18:08:41 +08:00
|
|
|
__be64 *pua = IOMMU_TABLE_USERSPACE_ENTRY_RO(tbl, entry);
|
2015-06-05 14:35:25 +08:00
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
if (!pua)
|
2015-06-05 14:35:25 +08:00
|
|
|
return;
|
|
|
|
|
2018-07-04 14:13:46 +08:00
|
|
|
ret = tce_iommu_prereg_ua_to_hpa(container, be64_to_cpu(*pua),
|
2018-08-13 13:59:06 +08:00
|
|
|
tbl->it_page_shift, &hpa, &mem);
|
2015-06-05 14:35:25 +08:00
|
|
|
if (ret)
|
2018-07-04 14:13:46 +08:00
|
|
|
pr_debug("%s: tce %llx at #%lx was not cached, ret=%d\n",
|
|
|
|
__func__, be64_to_cpu(*pua), entry, ret);
|
2015-06-05 14:35:25 +08:00
|
|
|
if (mem)
|
|
|
|
mm_iommu_mapped_dec(mem);
|
|
|
|
|
2018-07-04 14:13:46 +08:00
|
|
|
*pua = cpu_to_be64(0);
|
2015-06-05 14:35:25 +08:00
|
|
|
}
|
|
|
|
|
2015-06-05 14:34:58 +08:00
|
|
|
static int tce_iommu_clear(struct tce_container *container,
|
|
|
|
struct iommu_table *tbl,
|
|
|
|
unsigned long entry, unsigned long pages)
|
|
|
|
{
|
2015-06-05 14:35:15 +08:00
|
|
|
unsigned long oldhpa;
|
|
|
|
long ret;
|
|
|
|
enum dma_data_direction direction;
|
2019-08-29 16:52:50 +08:00
|
|
|
unsigned long lastentry = entry + pages, firstentry = entry;
|
2018-10-15 18:08:41 +08:00
|
|
|
|
|
|
|
for ( ; entry < lastentry; ++entry) {
|
|
|
|
if (tbl->it_indirect_levels && tbl->it_userspace) {
|
|
|
|
/*
|
|
|
|
* For multilevel tables, we can take a shortcut here
|
|
|
|
* and skip some TCEs as we know that the userspace
|
|
|
|
* addresses cache is a mirror of the real TCE table
|
|
|
|
* and if it is missing some indirect levels, then
|
|
|
|
* the hardware table does not have them allocated
|
|
|
|
* either and therefore does not require updating.
|
|
|
|
*/
|
|
|
|
__be64 *pua = IOMMU_TABLE_USERSPACE_ENTRY_RO(tbl,
|
|
|
|
entry);
|
|
|
|
if (!pua) {
|
|
|
|
/* align to level_size which is power of two */
|
|
|
|
entry |= tbl->it_level_size - 1;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
}
|
2015-06-05 14:34:58 +08:00
|
|
|
|
2017-10-03 02:39:11 +08:00
|
|
|
cond_resched();
|
|
|
|
|
2015-06-05 14:35:15 +08:00
|
|
|
direction = DMA_NONE;
|
|
|
|
oldhpa = 0;
|
2019-08-29 16:52:50 +08:00
|
|
|
ret = iommu_tce_xchg_no_kill(container->mm, tbl, entry, &oldhpa,
|
2018-12-19 16:52:15 +08:00
|
|
|
&direction);
|
2015-06-05 14:35:15 +08:00
|
|
|
if (ret)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
if (direction == DMA_NONE)
|
2015-06-05 14:34:58 +08:00
|
|
|
continue;
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
if (container->v2) {
|
2016-11-30 14:52:04 +08:00
|
|
|
tce_iommu_unuse_page_v2(container, tbl, entry);
|
2015-06-05 14:35:25 +08:00
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
2022-07-02 14:46:13 +08:00
|
|
|
tce_iommu_unuse_page(oldhpa);
|
2015-06-05 14:34:58 +08:00
|
|
|
}
|
|
|
|
|
2019-08-29 16:52:50 +08:00
|
|
|
iommu_tce_kill(tbl, firstentry, pages);
|
|
|
|
|
2015-06-05 14:34:58 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:03 +08:00
|
|
|
static int tce_iommu_use_page(unsigned long tce, unsigned long *hpa)
|
|
|
|
{
|
|
|
|
struct page *page = NULL;
|
|
|
|
enum dma_data_direction direction = iommu_tce_direction(tce);
|
|
|
|
|
2020-07-28 03:43:38 +08:00
|
|
|
if (pin_user_pages_fast(tce & PAGE_MASK, 1,
|
2019-05-14 08:17:11 +08:00
|
|
|
direction != DMA_TO_DEVICE ? FOLL_WRITE : 0,
|
|
|
|
&page) != 1)
|
2015-06-05 14:35:03 +08:00
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
*hpa = __pa((unsigned long) page_address(page));
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:34:58 +08:00
|
|
|
static long tce_iommu_build(struct tce_container *container,
|
|
|
|
struct iommu_table *tbl,
|
2015-06-05 14:35:15 +08:00
|
|
|
unsigned long entry, unsigned long tce, unsigned long pages,
|
|
|
|
enum dma_data_direction direction)
|
2015-06-05 14:34:58 +08:00
|
|
|
{
|
|
|
|
long i, ret = 0;
|
2015-06-05 14:35:03 +08:00
|
|
|
unsigned long hpa;
|
2015-06-05 14:35:15 +08:00
|
|
|
enum dma_data_direction dirtmp;
|
2015-06-05 14:34:58 +08:00
|
|
|
|
|
|
|
for (i = 0; i < pages; ++i) {
|
|
|
|
unsigned long offset = tce & IOMMU_PAGE_MASK(tbl) & ~PAGE_MASK;
|
|
|
|
|
2015-06-05 14:35:03 +08:00
|
|
|
ret = tce_iommu_use_page(tce, &hpa);
|
|
|
|
if (ret)
|
2015-06-05 14:34:58 +08:00
|
|
|
break;
|
2015-06-05 14:34:59 +08:00
|
|
|
|
2018-12-19 16:52:15 +08:00
|
|
|
if (!tce_page_is_contained(container->mm, hpa,
|
|
|
|
tbl->it_page_shift)) {
|
2015-06-05 14:34:59 +08:00
|
|
|
ret = -EPERM;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:03 +08:00
|
|
|
hpa |= offset;
|
2015-06-05 14:35:15 +08:00
|
|
|
dirtmp = direction;
|
2019-08-29 16:52:50 +08:00
|
|
|
ret = iommu_tce_xchg_no_kill(container->mm, tbl, entry + i,
|
|
|
|
&hpa, &dirtmp);
|
2015-06-05 14:34:58 +08:00
|
|
|
if (ret) {
|
2022-07-02 14:46:13 +08:00
|
|
|
tce_iommu_unuse_page(hpa);
|
2015-06-05 14:34:58 +08:00
|
|
|
pr_err("iommu_tce: %s failed ioba=%lx, tce=%lx, ret=%ld\n",
|
|
|
|
__func__, entry << tbl->it_page_shift,
|
|
|
|
tce, ret);
|
|
|
|
break;
|
|
|
|
}
|
2015-06-05 14:35:15 +08:00
|
|
|
|
|
|
|
if (dirtmp != DMA_NONE)
|
2022-07-02 14:46:13 +08:00
|
|
|
tce_iommu_unuse_page(hpa);
|
2015-06-05 14:35:15 +08:00
|
|
|
|
2015-06-05 14:35:00 +08:00
|
|
|
tce += IOMMU_PAGE_SIZE(tbl);
|
2015-06-05 14:34:58 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
if (ret)
|
|
|
|
tce_iommu_clear(container, tbl, entry, i);
|
2019-08-29 16:52:50 +08:00
|
|
|
else
|
|
|
|
iommu_tce_kill(tbl, entry, pages);
|
2015-06-05 14:34:58 +08:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
static long tce_iommu_build_v2(struct tce_container *container,
|
|
|
|
struct iommu_table *tbl,
|
|
|
|
unsigned long entry, unsigned long tce, unsigned long pages,
|
|
|
|
enum dma_data_direction direction)
|
|
|
|
{
|
|
|
|
long i, ret = 0;
|
|
|
|
unsigned long hpa;
|
|
|
|
enum dma_data_direction dirtmp;
|
|
|
|
|
|
|
|
for (i = 0; i < pages; ++i) {
|
|
|
|
struct mm_iommu_table_group_mem_t *mem = NULL;
|
2018-07-04 14:13:46 +08:00
|
|
|
__be64 *pua = IOMMU_TABLE_USERSPACE_ENTRY(tbl, entry + i);
|
2015-06-05 14:35:25 +08:00
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
ret = tce_iommu_prereg_ua_to_hpa(container,
|
2018-07-17 15:19:12 +08:00
|
|
|
tce, tbl->it_page_shift, &hpa, &mem);
|
2015-06-05 14:35:25 +08:00
|
|
|
if (ret)
|
|
|
|
break;
|
|
|
|
|
2018-12-19 16:52:15 +08:00
|
|
|
if (!tce_page_is_contained(container->mm, hpa,
|
|
|
|
tbl->it_page_shift)) {
|
2015-06-05 14:35:25 +08:00
|
|
|
ret = -EPERM;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Preserve offset within IOMMU page */
|
|
|
|
hpa |= tce & IOMMU_PAGE_MASK(tbl) & ~PAGE_MASK;
|
|
|
|
dirtmp = direction;
|
|
|
|
|
|
|
|
/* The registered region is being unregistered */
|
|
|
|
if (mm_iommu_mapped_inc(mem))
|
|
|
|
break;
|
|
|
|
|
2019-08-29 16:52:50 +08:00
|
|
|
ret = iommu_tce_xchg_no_kill(container->mm, tbl, entry + i,
|
|
|
|
&hpa, &dirtmp);
|
2015-06-05 14:35:25 +08:00
|
|
|
if (ret) {
|
|
|
|
/* dirtmp cannot be DMA_NONE here */
|
2016-11-30 14:52:04 +08:00
|
|
|
tce_iommu_unuse_page_v2(container, tbl, entry + i);
|
2015-06-05 14:35:25 +08:00
|
|
|
pr_err("iommu_tce: %s failed ioba=%lx, tce=%lx, ret=%ld\n",
|
|
|
|
__func__, entry << tbl->it_page_shift,
|
|
|
|
tce, ret);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (dirtmp != DMA_NONE)
|
2016-11-30 14:52:04 +08:00
|
|
|
tce_iommu_unuse_page_v2(container, tbl, entry + i);
|
2015-06-05 14:35:25 +08:00
|
|
|
|
2018-07-04 14:13:46 +08:00
|
|
|
*pua = cpu_to_be64(tce);
|
2015-06-05 14:35:25 +08:00
|
|
|
|
|
|
|
tce += IOMMU_PAGE_SIZE(tbl);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (ret)
|
|
|
|
tce_iommu_clear(container, tbl, entry, i);
|
2019-08-29 16:52:50 +08:00
|
|
|
else
|
|
|
|
iommu_tce_kill(tbl, entry, pages);
|
2015-06-05 14:35:25 +08:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:23 +08:00
|
|
|
static long tce_iommu_create_table(struct tce_container *container,
|
|
|
|
struct iommu_table_group *table_group,
|
|
|
|
int num,
|
|
|
|
__u32 page_shift,
|
|
|
|
__u64 window_size,
|
|
|
|
__u32 levels,
|
|
|
|
struct iommu_table **ptbl)
|
|
|
|
{
|
|
|
|
long ret, table_size;
|
|
|
|
|
|
|
|
table_size = table_group->ops->get_table_size(page_shift, window_size,
|
|
|
|
levels);
|
|
|
|
if (!table_size)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2019-07-17 07:30:54 +08:00
|
|
|
ret = account_locked_vm(container->mm, table_size >> PAGE_SHIFT, true);
|
2015-06-05 14:35:23 +08:00
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
ret = table_group->ops->create_table(table_group, num,
|
|
|
|
page_shift, window_size, levels, ptbl);
|
|
|
|
|
|
|
|
WARN_ON(!ret && !(*ptbl)->it_ops->free);
|
2018-07-04 14:13:49 +08:00
|
|
|
WARN_ON(!ret && ((*ptbl)->it_allocated_size > table_size));
|
2015-06-05 14:35:23 +08:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
static void tce_iommu_free_table(struct tce_container *container,
|
|
|
|
struct iommu_table *tbl)
|
2015-06-05 14:35:23 +08:00
|
|
|
{
|
|
|
|
unsigned long pages = tbl->it_allocated_size >> PAGE_SHIFT;
|
|
|
|
|
2017-03-22 12:21:50 +08:00
|
|
|
iommu_tce_table_put(tbl);
|
2019-07-17 07:30:54 +08:00
|
|
|
account_locked_vm(container->mm, pages, false);
|
2015-06-05 14:35:23 +08:00
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:26 +08:00
|
|
|
static long tce_iommu_create_window(struct tce_container *container,
|
|
|
|
__u32 page_shift, __u64 window_size, __u32 levels,
|
|
|
|
__u64 *start_addr)
|
|
|
|
{
|
|
|
|
struct tce_iommu_group *tcegrp;
|
|
|
|
struct iommu_table_group *table_group;
|
|
|
|
struct iommu_table *tbl = NULL;
|
|
|
|
long ret, num;
|
|
|
|
|
|
|
|
num = tce_iommu_find_free_table(container);
|
|
|
|
if (num < 0)
|
|
|
|
return num;
|
|
|
|
|
|
|
|
/* Get the first group for ops::create_table */
|
|
|
|
tcegrp = list_first_entry(&container->group_list,
|
|
|
|
struct tce_iommu_group, next);
|
|
|
|
table_group = iommu_group_get_iommudata(tcegrp->grp);
|
|
|
|
if (!table_group)
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (!(table_group->pgsizes & (1ULL << page_shift)))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (!table_group->ops->set_window || !table_group->ops->unset_window ||
|
|
|
|
!table_group->ops->get_table_size ||
|
|
|
|
!table_group->ops->create_table)
|
|
|
|
return -EPERM;
|
|
|
|
|
|
|
|
/* Create TCE table */
|
|
|
|
ret = tce_iommu_create_table(container, table_group, num,
|
|
|
|
page_shift, window_size, levels, &tbl);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
BUG_ON(!tbl->it_ops->free);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Program the table to every group.
|
|
|
|
* Groups have been tested for compatibility at the attach time.
|
|
|
|
*/
|
|
|
|
list_for_each_entry(tcegrp, &container->group_list, next) {
|
|
|
|
table_group = iommu_group_get_iommudata(tcegrp->grp);
|
|
|
|
|
|
|
|
ret = table_group->ops->set_window(table_group, num, tbl);
|
|
|
|
if (ret)
|
|
|
|
goto unset_exit;
|
|
|
|
}
|
|
|
|
|
|
|
|
container->tables[num] = tbl;
|
|
|
|
|
|
|
|
/* Return start address assigned by platform in create_table() */
|
|
|
|
*start_addr = tbl->it_offset << tbl->it_page_shift;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
unset_exit:
|
|
|
|
list_for_each_entry(tcegrp, &container->group_list, next) {
|
|
|
|
table_group = iommu_group_get_iommudata(tcegrp->grp);
|
|
|
|
table_group->ops->unset_window(table_group, num);
|
|
|
|
}
|
2016-11-30 14:52:04 +08:00
|
|
|
tce_iommu_free_table(container, tbl);
|
2015-06-05 14:35:26 +08:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static long tce_iommu_remove_window(struct tce_container *container,
|
|
|
|
__u64 start_addr)
|
|
|
|
{
|
|
|
|
struct iommu_table_group *table_group = NULL;
|
|
|
|
struct iommu_table *tbl;
|
|
|
|
struct tce_iommu_group *tcegrp;
|
|
|
|
int num;
|
|
|
|
|
|
|
|
num = tce_iommu_find_table(container, start_addr, &tbl);
|
|
|
|
if (num < 0)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
BUG_ON(!tbl->it_size);
|
|
|
|
|
2024-06-24 20:38:58 +08:00
|
|
|
tce_iommu_clear(container, tbl, tbl->it_offset, tbl->it_size);
|
|
|
|
|
2015-06-05 14:35:26 +08:00
|
|
|
/* Detach groups from IOMMUs */
|
|
|
|
list_for_each_entry(tcegrp, &container->group_list, next) {
|
|
|
|
table_group = iommu_group_get_iommudata(tcegrp->grp);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* SPAPR TCE IOMMU exposes the default DMA window to
|
|
|
|
* the guest via dma32_window_start/size of
|
|
|
|
* VFIO_IOMMU_SPAPR_TCE_GET_INFO. Some platforms allow
|
|
|
|
* the userspace to remove this window, some do not so
|
|
|
|
* here we check for the platform capability.
|
|
|
|
*/
|
|
|
|
if (!table_group->ops || !table_group->ops->unset_window)
|
|
|
|
return -EPERM;
|
|
|
|
|
|
|
|
table_group->ops->unset_window(table_group, num);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Free table */
|
2016-11-30 14:52:04 +08:00
|
|
|
tce_iommu_free_table(container, tbl);
|
2015-06-05 14:35:26 +08:00
|
|
|
container->tables[num] = NULL;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-11-30 14:52:02 +08:00
|
|
|
static long tce_iommu_create_default_window(struct tce_container *container)
|
|
|
|
{
|
|
|
|
long ret;
|
|
|
|
__u64 start_addr = 0;
|
|
|
|
struct tce_iommu_group *tcegrp;
|
|
|
|
struct iommu_table_group *table_group;
|
|
|
|
|
2016-11-30 14:52:03 +08:00
|
|
|
if (!container->def_window_pending)
|
|
|
|
return 0;
|
|
|
|
|
2016-11-30 14:52:02 +08:00
|
|
|
if (!tce_groups_attached(container))
|
|
|
|
return -ENODEV;
|
|
|
|
|
|
|
|
tcegrp = list_first_entry(&container->group_list,
|
|
|
|
struct tce_iommu_group, next);
|
|
|
|
table_group = iommu_group_get_iommudata(tcegrp->grp);
|
|
|
|
if (!table_group)
|
|
|
|
return -ENODEV;
|
|
|
|
|
|
|
|
ret = tce_iommu_create_window(container, IOMMU_PAGE_SHIFT_4K,
|
|
|
|
table_group->tce32_size, 1, &start_addr);
|
|
|
|
WARN_ON_ONCE(!ret && start_addr);
|
|
|
|
|
2016-11-30 14:52:03 +08:00
|
|
|
if (!ret)
|
|
|
|
container->def_window_pending = false;
|
|
|
|
|
2016-11-30 14:52:02 +08:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2022-12-05 23:29:18 +08:00
|
|
|
static long vfio_spapr_ioctl_eeh_pe_op(struct iommu_group *group,
|
|
|
|
unsigned long arg)
|
|
|
|
{
|
|
|
|
struct eeh_pe *pe;
|
|
|
|
struct vfio_eeh_pe_op op;
|
|
|
|
unsigned long minsz;
|
|
|
|
|
|
|
|
pe = eeh_iommu_group_to_pe(group);
|
|
|
|
if (!pe)
|
|
|
|
return -ENODEV;
|
|
|
|
|
|
|
|
minsz = offsetofend(struct vfio_eeh_pe_op, op);
|
|
|
|
if (copy_from_user(&op, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
if (op.argsz < minsz || op.flags)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
switch (op.op) {
|
|
|
|
case VFIO_EEH_PE_DISABLE:
|
|
|
|
return eeh_pe_set_option(pe, EEH_OPT_DISABLE);
|
|
|
|
case VFIO_EEH_PE_ENABLE:
|
|
|
|
return eeh_pe_set_option(pe, EEH_OPT_ENABLE);
|
|
|
|
case VFIO_EEH_PE_UNFREEZE_IO:
|
|
|
|
return eeh_pe_set_option(pe, EEH_OPT_THAW_MMIO);
|
|
|
|
case VFIO_EEH_PE_UNFREEZE_DMA:
|
|
|
|
return eeh_pe_set_option(pe, EEH_OPT_THAW_DMA);
|
|
|
|
case VFIO_EEH_PE_GET_STATE:
|
|
|
|
return eeh_pe_get_state(pe);
|
|
|
|
break;
|
|
|
|
case VFIO_EEH_PE_RESET_DEACTIVATE:
|
|
|
|
return eeh_pe_reset(pe, EEH_RESET_DEACTIVATE, true);
|
|
|
|
case VFIO_EEH_PE_RESET_HOT:
|
|
|
|
return eeh_pe_reset(pe, EEH_RESET_HOT, true);
|
|
|
|
case VFIO_EEH_PE_RESET_FUNDAMENTAL:
|
|
|
|
return eeh_pe_reset(pe, EEH_RESET_FUNDAMENTAL, true);
|
|
|
|
case VFIO_EEH_PE_CONFIGURE:
|
|
|
|
return eeh_pe_configure(pe);
|
|
|
|
case VFIO_EEH_PE_INJECT_ERR:
|
|
|
|
minsz = offsetofend(struct vfio_eeh_pe_op, err.mask);
|
|
|
|
if (op.argsz < minsz)
|
|
|
|
return -EINVAL;
|
|
|
|
if (copy_from_user(&op, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
return eeh_pe_inject_err(pe, op.err.type, op.err.func,
|
|
|
|
op.err.addr, op.err.mask);
|
|
|
|
default:
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
static long tce_iommu_ioctl(void *iommu_data,
|
|
|
|
unsigned int cmd, unsigned long arg)
|
|
|
|
{
|
|
|
|
struct tce_container *container = iommu_data;
|
2015-06-05 14:35:26 +08:00
|
|
|
unsigned long minsz, ddwsz;
|
2013-05-21 11:33:10 +08:00
|
|
|
long ret;
|
|
|
|
|
|
|
|
switch (cmd) {
|
|
|
|
case VFIO_CHECK_EXTENSION:
|
2014-06-10 09:41:57 +08:00
|
|
|
switch (arg) {
|
|
|
|
case VFIO_SPAPR_TCE_IOMMU:
|
2015-06-05 14:35:25 +08:00
|
|
|
case VFIO_SPAPR_TCE_v2_IOMMU:
|
2022-12-05 23:29:17 +08:00
|
|
|
return 1;
|
|
|
|
case VFIO_EEH:
|
|
|
|
return eeh_enabled();
|
2014-06-10 09:41:57 +08:00
|
|
|
default:
|
2022-12-05 23:29:17 +08:00
|
|
|
return 0;
|
2014-06-10 09:41:57 +08:00
|
|
|
}
|
2016-11-30 14:52:04 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Sanity check to prevent one userspace from manipulating
|
|
|
|
* another userspace mm.
|
|
|
|
*/
|
|
|
|
BUG_ON(!container);
|
|
|
|
if (container->mm && container->mm != current->mm)
|
|
|
|
return -EPERM;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
switch (cmd) {
|
2013-05-21 11:33:10 +08:00
|
|
|
case VFIO_IOMMU_SPAPR_TCE_GET_INFO: {
|
|
|
|
struct vfio_iommu_spapr_tce_info info;
|
2015-06-05 14:35:25 +08:00
|
|
|
struct tce_iommu_group *tcegrp;
|
2015-06-05 14:35:09 +08:00
|
|
|
struct iommu_table_group *table_group;
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
if (!tce_groups_attached(container))
|
2015-06-05 14:35:09 +08:00
|
|
|
return -ENXIO;
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
tcegrp = list_first_entry(&container->group_list,
|
|
|
|
struct tce_iommu_group, next);
|
|
|
|
table_group = iommu_group_get_iommudata(tcegrp->grp);
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2015-06-05 14:35:20 +08:00
|
|
|
if (!table_group)
|
2013-05-21 11:33:10 +08:00
|
|
|
return -ENXIO;
|
|
|
|
|
|
|
|
minsz = offsetofend(struct vfio_iommu_spapr_tce_info,
|
|
|
|
dma32_window_size);
|
|
|
|
|
|
|
|
if (copy_from_user(&info, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (info.argsz < minsz)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2015-06-05 14:35:20 +08:00
|
|
|
info.dma32_window_start = table_group->tce32_start;
|
|
|
|
info.dma32_window_size = table_group->tce32_size;
|
2013-05-21 11:33:10 +08:00
|
|
|
info.flags = 0;
|
2015-06-05 14:35:26 +08:00
|
|
|
memset(&info.ddw, 0, sizeof(info.ddw));
|
|
|
|
|
|
|
|
if (table_group->max_dynamic_windows_supported &&
|
|
|
|
container->v2) {
|
|
|
|
info.flags |= VFIO_IOMMU_SPAPR_INFO_DDW;
|
|
|
|
info.ddw.pgsizes = table_group->pgsizes;
|
|
|
|
info.ddw.max_dynamic_windows_supported =
|
|
|
|
table_group->max_dynamic_windows_supported;
|
|
|
|
info.ddw.levels = table_group->max_levels;
|
|
|
|
}
|
|
|
|
|
|
|
|
ddwsz = offsetofend(struct vfio_iommu_spapr_tce_info, ddw);
|
|
|
|
|
|
|
|
if (info.argsz >= ddwsz)
|
|
|
|
minsz = ddwsz;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
|
|
|
if (copy_to_user((void __user *)arg, &info, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
case VFIO_IOMMU_MAP_DMA: {
|
|
|
|
struct vfio_iommu_type1_dma_map param;
|
2015-06-05 14:35:09 +08:00
|
|
|
struct iommu_table *tbl = NULL;
|
|
|
|
long num;
|
2015-06-05 14:35:15 +08:00
|
|
|
enum dma_data_direction direction;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2015-06-05 14:35:02 +08:00
|
|
|
if (!container->enabled)
|
|
|
|
return -EPERM;
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
minsz = offsetofend(struct vfio_iommu_type1_dma_map, size);
|
|
|
|
|
|
|
|
if (copy_from_user(¶m, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (param.argsz < minsz)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (param.flags & ~(VFIO_DMA_MAP_FLAG_READ |
|
|
|
|
VFIO_DMA_MAP_FLAG_WRITE))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2016-11-30 14:52:03 +08:00
|
|
|
ret = tce_iommu_create_default_window(container);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
2015-06-05 14:35:09 +08:00
|
|
|
num = tce_iommu_find_table(container, param.iova, &tbl);
|
|
|
|
if (num < 0)
|
|
|
|
return -ENXIO;
|
|
|
|
|
2015-06-05 14:35:00 +08:00
|
|
|
if ((param.size & ~IOMMU_PAGE_MASK(tbl)) ||
|
|
|
|
(param.vaddr & ~IOMMU_PAGE_MASK(tbl)))
|
2013-05-21 11:33:10 +08:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
/* iova is checked by the IOMMU API */
|
2015-06-05 14:35:15 +08:00
|
|
|
if (param.flags & VFIO_DMA_MAP_FLAG_READ) {
|
|
|
|
if (param.flags & VFIO_DMA_MAP_FLAG_WRITE)
|
|
|
|
direction = DMA_BIDIRECTIONAL;
|
|
|
|
else
|
|
|
|
direction = DMA_TO_DEVICE;
|
|
|
|
} else {
|
|
|
|
if (param.flags & VFIO_DMA_MAP_FLAG_WRITE)
|
|
|
|
direction = DMA_FROM_DEVICE;
|
|
|
|
else
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2015-06-05 14:35:15 +08:00
|
|
|
ret = iommu_tce_put_param_check(tbl, param.iova, param.vaddr);
|
2013-05-21 11:33:10 +08:00
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
if (container->v2)
|
|
|
|
ret = tce_iommu_build_v2(container, tbl,
|
|
|
|
param.iova >> tbl->it_page_shift,
|
|
|
|
param.vaddr,
|
|
|
|
param.size >> tbl->it_page_shift,
|
|
|
|
direction);
|
|
|
|
else
|
|
|
|
ret = tce_iommu_build(container, tbl,
|
|
|
|
param.iova >> tbl->it_page_shift,
|
|
|
|
param.vaddr,
|
|
|
|
param.size >> tbl->it_page_shift,
|
|
|
|
direction);
|
2013-05-21 11:33:10 +08:00
|
|
|
|
|
|
|
iommu_flush_tce(tbl);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
case VFIO_IOMMU_UNMAP_DMA: {
|
|
|
|
struct vfio_iommu_type1_dma_unmap param;
|
2015-06-05 14:35:09 +08:00
|
|
|
struct iommu_table *tbl = NULL;
|
|
|
|
long num;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2015-06-05 14:35:02 +08:00
|
|
|
if (!container->enabled)
|
|
|
|
return -EPERM;
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
minsz = offsetofend(struct vfio_iommu_type1_dma_unmap,
|
|
|
|
size);
|
|
|
|
|
|
|
|
if (copy_from_user(¶m, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (param.argsz < minsz)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
/* No flag is supported now */
|
|
|
|
if (param.flags)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2016-11-30 14:52:03 +08:00
|
|
|
ret = tce_iommu_create_default_window(container);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
2015-06-05 14:35:09 +08:00
|
|
|
num = tce_iommu_find_table(container, param.iova, &tbl);
|
|
|
|
if (num < 0)
|
|
|
|
return -ENXIO;
|
|
|
|
|
2015-06-05 14:35:00 +08:00
|
|
|
if (param.size & ~IOMMU_PAGE_MASK(tbl))
|
2013-05-21 11:33:10 +08:00
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
ret = iommu_tce_clear_param_check(tbl, param.iova, 0,
|
2015-06-05 14:35:00 +08:00
|
|
|
param.size >> tbl->it_page_shift);
|
2013-05-21 11:33:10 +08:00
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
2015-06-05 14:34:58 +08:00
|
|
|
ret = tce_iommu_clear(container, tbl,
|
2015-06-05 14:35:00 +08:00
|
|
|
param.iova >> tbl->it_page_shift,
|
|
|
|
param.size >> tbl->it_page_shift);
|
2013-05-21 11:33:10 +08:00
|
|
|
iommu_flush_tce(tbl);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
2015-06-05 14:35:25 +08:00
|
|
|
case VFIO_IOMMU_SPAPR_REGISTER_MEMORY: {
|
|
|
|
struct vfio_iommu_spapr_register_memory param;
|
|
|
|
|
|
|
|
if (!container->v2)
|
|
|
|
break;
|
|
|
|
|
|
|
|
minsz = offsetofend(struct vfio_iommu_spapr_register_memory,
|
|
|
|
size);
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
ret = tce_iommu_mm_set(container);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
if (copy_from_user(¶m, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (param.argsz < minsz)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
/* No flag is supported now */
|
|
|
|
if (param.flags)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
mutex_lock(&container->lock);
|
|
|
|
ret = tce_iommu_register_pages(container, param.vaddr,
|
|
|
|
param.size);
|
|
|
|
mutex_unlock(&container->lock);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
case VFIO_IOMMU_SPAPR_UNREGISTER_MEMORY: {
|
|
|
|
struct vfio_iommu_spapr_register_memory param;
|
|
|
|
|
|
|
|
if (!container->v2)
|
|
|
|
break;
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
if (!container->mm)
|
|
|
|
return -EPERM;
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
minsz = offsetofend(struct vfio_iommu_spapr_register_memory,
|
|
|
|
size);
|
|
|
|
|
|
|
|
if (copy_from_user(¶m, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (param.argsz < minsz)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
/* No flag is supported now */
|
|
|
|
if (param.flags)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
mutex_lock(&container->lock);
|
|
|
|
ret = tce_iommu_unregister_pages(container, param.vaddr,
|
|
|
|
param.size);
|
|
|
|
mutex_unlock(&container->lock);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
2013-05-21 11:33:10 +08:00
|
|
|
case VFIO_IOMMU_ENABLE:
|
2015-06-05 14:35:25 +08:00
|
|
|
if (container->v2)
|
|
|
|
break;
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
mutex_lock(&container->lock);
|
|
|
|
ret = tce_iommu_enable(container);
|
|
|
|
mutex_unlock(&container->lock);
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
|
|
|
|
case VFIO_IOMMU_DISABLE:
|
2015-06-05 14:35:25 +08:00
|
|
|
if (container->v2)
|
|
|
|
break;
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
mutex_lock(&container->lock);
|
|
|
|
tce_iommu_disable(container);
|
|
|
|
mutex_unlock(&container->lock);
|
|
|
|
return 0;
|
2014-06-10 09:41:57 +08:00
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
case VFIO_EEH_PE_OP: {
|
|
|
|
struct tce_iommu_group *tcegrp;
|
|
|
|
|
|
|
|
ret = 0;
|
|
|
|
list_for_each_entry(tcegrp, &container->group_list, next) {
|
2022-12-05 23:29:18 +08:00
|
|
|
ret = vfio_spapr_ioctl_eeh_pe_op(tcegrp->grp, arg);
|
2015-06-05 14:35:25 +08:00
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:26 +08:00
|
|
|
case VFIO_IOMMU_SPAPR_TCE_CREATE: {
|
|
|
|
struct vfio_iommu_spapr_tce_create create;
|
|
|
|
|
|
|
|
if (!container->v2)
|
|
|
|
break;
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
ret = tce_iommu_mm_set(container);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
2015-06-05 14:35:26 +08:00
|
|
|
if (!tce_groups_attached(container))
|
|
|
|
return -ENXIO;
|
|
|
|
|
|
|
|
minsz = offsetofend(struct vfio_iommu_spapr_tce_create,
|
|
|
|
start_addr);
|
|
|
|
|
|
|
|
if (copy_from_user(&create, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (create.argsz < minsz)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (create.flags)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
mutex_lock(&container->lock);
|
|
|
|
|
2016-11-30 14:52:03 +08:00
|
|
|
ret = tce_iommu_create_default_window(container);
|
2017-02-01 11:26:16 +08:00
|
|
|
if (!ret)
|
|
|
|
ret = tce_iommu_create_window(container,
|
|
|
|
create.page_shift,
|
|
|
|
create.window_size, create.levels,
|
|
|
|
&create.start_addr);
|
2015-06-05 14:35:26 +08:00
|
|
|
|
|
|
|
mutex_unlock(&container->lock);
|
|
|
|
|
|
|
|
if (!ret && copy_to_user((void __user *)arg, &create, minsz))
|
|
|
|
ret = -EFAULT;
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
case VFIO_IOMMU_SPAPR_TCE_REMOVE: {
|
|
|
|
struct vfio_iommu_spapr_tce_remove remove;
|
|
|
|
|
|
|
|
if (!container->v2)
|
|
|
|
break;
|
|
|
|
|
2016-11-30 14:52:04 +08:00
|
|
|
ret = tce_iommu_mm_set(container);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
2015-06-05 14:35:26 +08:00
|
|
|
if (!tce_groups_attached(container))
|
|
|
|
return -ENXIO;
|
|
|
|
|
|
|
|
minsz = offsetofend(struct vfio_iommu_spapr_tce_remove,
|
|
|
|
start_addr);
|
|
|
|
|
|
|
|
if (copy_from_user(&remove, (void __user *)arg, minsz))
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
if (remove.argsz < minsz)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (remove.flags)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2016-11-30 14:52:03 +08:00
|
|
|
if (container->def_window_pending && !remove.start_addr) {
|
|
|
|
container->def_window_pending = false;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:26 +08:00
|
|
|
mutex_lock(&container->lock);
|
|
|
|
|
|
|
|
ret = tce_iommu_remove_window(container, remove.start_addr);
|
|
|
|
|
|
|
|
mutex_unlock(&container->lock);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
2013-05-21 11:33:10 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return -ENOTTY;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:10 +08:00
|
|
|
static void tce_iommu_release_ownership(struct tce_container *container,
|
|
|
|
struct iommu_table_group *table_group)
|
|
|
|
{
|
2015-06-05 14:35:23 +08:00
|
|
|
long i;
|
|
|
|
|
|
|
|
if (!table_group->ops->unset_window) {
|
|
|
|
WARN_ON_ONCE(1);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2024-06-24 20:38:58 +08:00
|
|
|
for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
|
|
|
|
if (container->tables[i]) {
|
|
|
|
tce_iommu_clear(container, container->tables[i],
|
|
|
|
container->tables[i]->it_offset,
|
|
|
|
container->tables[i]->it_size);
|
2019-02-11 15:49:17 +08:00
|
|
|
table_group->ops->unset_window(table_group, i);
|
2024-06-24 20:38:58 +08:00
|
|
|
}
|
|
|
|
}
|
2015-06-05 14:35:10 +08:00
|
|
|
}
|
|
|
|
|
2023-03-07 01:30:20 +08:00
|
|
|
static long tce_iommu_take_ownership(struct tce_container *container,
|
2015-06-05 14:35:10 +08:00
|
|
|
struct iommu_table_group *table_group)
|
|
|
|
{
|
2017-02-07 14:26:57 +08:00
|
|
|
long i, ret = 0;
|
|
|
|
|
|
|
|
/* Set all windows to the new group */
|
|
|
|
for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
|
|
|
|
struct iommu_table *tbl = container->tables[i];
|
|
|
|
|
|
|
|
if (!tbl)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
ret = table_group->ops->set_window(table_group, i, tbl);
|
|
|
|
if (ret)
|
|
|
|
goto release_exit;
|
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
return 0;
|
2017-02-07 14:26:57 +08:00
|
|
|
|
|
|
|
release_exit:
|
|
|
|
for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i)
|
|
|
|
table_group->ops->unset_window(table_group, i);
|
|
|
|
|
|
|
|
return ret;
|
2015-06-05 14:35:10 +08:00
|
|
|
}
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
static int tce_iommu_attach_group(void *iommu_data,
|
2021-09-24 23:57:01 +08:00
|
|
|
struct iommu_group *iommu_group, enum vfio_group_type type)
|
2013-05-21 11:33:10 +08:00
|
|
|
{
|
2019-08-19 09:51:17 +08:00
|
|
|
int ret = 0;
|
2013-05-21 11:33:10 +08:00
|
|
|
struct tce_container *container = iommu_data;
|
2015-06-05 14:35:09 +08:00
|
|
|
struct iommu_table_group *table_group;
|
2015-06-05 14:35:25 +08:00
|
|
|
struct tce_iommu_group *tcegrp = NULL;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2021-09-24 23:57:02 +08:00
|
|
|
if (type == VFIO_EMULATED_IOMMU)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2013-05-21 11:33:10 +08:00
|
|
|
mutex_lock(&container->lock);
|
|
|
|
|
|
|
|
/* pr_debug("tce_vfio: Attaching group #%u to iommu %p\n",
|
|
|
|
iommu_group_id(iommu_group), iommu_group); */
|
2015-06-05 14:35:25 +08:00
|
|
|
table_group = iommu_group_get_iommudata(iommu_group);
|
2017-01-25 00:50:26 +08:00
|
|
|
if (!table_group) {
|
|
|
|
ret = -ENODEV;
|
|
|
|
goto unlock_exit;
|
|
|
|
}
|
2015-06-05 14:35:25 +08:00
|
|
|
|
2023-03-07 01:30:20 +08:00
|
|
|
/* v2 requires full support of dynamic DMA windows */
|
|
|
|
if (container->v2 && table_group->max_dynamic_windows_supported == 0) {
|
|
|
|
ret = -EINVAL;
|
|
|
|
goto unlock_exit;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* v1 reuses TCE tables and does not share them among PEs */
|
|
|
|
if (!container->v2 && tce_groups_attached(container)) {
|
2013-05-21 11:33:10 +08:00
|
|
|
ret = -EBUSY;
|
2015-06-05 14:35:04 +08:00
|
|
|
goto unlock_exit;
|
|
|
|
}
|
|
|
|
|
2022-07-14 16:09:12 +08:00
|
|
|
/*
|
|
|
|
* Check if new group has the same iommu_table_group_ops
|
|
|
|
* (i.e. compatible)
|
|
|
|
*/
|
2015-06-05 14:35:25 +08:00
|
|
|
list_for_each_entry(tcegrp, &container->group_list, next) {
|
|
|
|
struct iommu_table_group *table_group_tmp;
|
|
|
|
|
|
|
|
if (tcegrp->grp == iommu_group) {
|
|
|
|
pr_warn("tce_vfio: Group %d is already attached\n",
|
|
|
|
iommu_group_id(iommu_group));
|
|
|
|
ret = -EBUSY;
|
|
|
|
goto unlock_exit;
|
|
|
|
}
|
|
|
|
table_group_tmp = iommu_group_get_iommudata(tcegrp->grp);
|
2016-04-29 16:55:15 +08:00
|
|
|
if (table_group_tmp->ops->create_table !=
|
|
|
|
table_group->ops->create_table) {
|
2015-06-05 14:35:25 +08:00
|
|
|
pr_warn("tce_vfio: Group %d is incompatible with group %d\n",
|
|
|
|
iommu_group_id(iommu_group),
|
|
|
|
iommu_group_id(tcegrp->grp));
|
|
|
|
ret = -EPERM;
|
|
|
|
goto unlock_exit;
|
|
|
|
}
|
2013-05-21 11:33:10 +08:00
|
|
|
}
|
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
tcegrp = kzalloc(sizeof(*tcegrp), GFP_KERNEL);
|
|
|
|
if (!tcegrp) {
|
|
|
|
ret = -ENOMEM;
|
2015-06-05 14:35:09 +08:00
|
|
|
goto unlock_exit;
|
|
|
|
}
|
|
|
|
|
2023-03-07 01:30:20 +08:00
|
|
|
ret = tce_iommu_take_ownership(container, table_group);
|
|
|
|
if (!tce_groups_attached(container) && !container->tables[0])
|
|
|
|
container->def_window_pending = true;
|
2015-06-05 14:35:10 +08:00
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
if (!ret) {
|
|
|
|
tcegrp->grp = iommu_group;
|
|
|
|
list_add(&tcegrp->next, &container->group_list);
|
|
|
|
}
|
2015-06-05 14:35:04 +08:00
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
if (ret && tcegrp)
|
|
|
|
kfree(tcegrp);
|
|
|
|
|
2019-08-19 09:51:17 +08:00
|
|
|
unlock_exit:
|
2013-05-21 11:33:10 +08:00
|
|
|
mutex_unlock(&container->lock);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void tce_iommu_detach_group(void *iommu_data,
|
|
|
|
struct iommu_group *iommu_group)
|
|
|
|
{
|
|
|
|
struct tce_container *container = iommu_data;
|
2015-06-05 14:35:09 +08:00
|
|
|
struct iommu_table_group *table_group;
|
2015-06-05 14:35:25 +08:00
|
|
|
bool found = false;
|
|
|
|
struct tce_iommu_group *tcegrp;
|
2013-05-21 11:33:10 +08:00
|
|
|
|
|
|
|
mutex_lock(&container->lock);
|
2015-06-05 14:35:25 +08:00
|
|
|
|
|
|
|
list_for_each_entry(tcegrp, &container->group_list, next) {
|
|
|
|
if (tcegrp->grp == iommu_group) {
|
|
|
|
found = true;
|
|
|
|
break;
|
|
|
|
}
|
2015-06-05 14:35:04 +08:00
|
|
|
}
|
2013-05-21 11:33:10 +08:00
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
if (!found) {
|
|
|
|
pr_warn("tce_vfio: detaching unattached group #%u\n",
|
|
|
|
iommu_group_id(iommu_group));
|
|
|
|
goto unlock_exit;
|
2013-05-21 11:33:10 +08:00
|
|
|
}
|
2015-06-05 14:35:04 +08:00
|
|
|
|
2015-06-05 14:35:25 +08:00
|
|
|
list_del(&tcegrp->next);
|
|
|
|
kfree(tcegrp);
|
2015-06-05 14:35:09 +08:00
|
|
|
|
|
|
|
table_group = iommu_group_get_iommudata(iommu_group);
|
|
|
|
BUG_ON(!table_group);
|
|
|
|
|
2023-03-07 01:30:20 +08:00
|
|
|
tce_iommu_release_ownership(container, table_group);
|
2015-06-05 14:35:04 +08:00
|
|
|
|
|
|
|
unlock_exit:
|
2013-05-21 11:33:10 +08:00
|
|
|
mutex_unlock(&container->lock);
|
|
|
|
}
|
|
|
|
|
2019-04-04 02:36:21 +08:00
|
|
|
static const struct vfio_iommu_driver_ops tce_iommu_driver_ops = {
|
2013-05-21 11:33:10 +08:00
|
|
|
.name = "iommu-vfio-powerpc",
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.open = tce_iommu_open,
|
|
|
|
.release = tce_iommu_release,
|
|
|
|
.ioctl = tce_iommu_ioctl,
|
|
|
|
.attach_group = tce_iommu_attach_group,
|
|
|
|
.detach_group = tce_iommu_detach_group,
|
|
|
|
};
|
|
|
|
|
|
|
|
static int __init tce_iommu_init(void)
|
|
|
|
{
|
|
|
|
return vfio_register_iommu_driver(&tce_iommu_driver_ops);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __exit tce_iommu_cleanup(void)
|
|
|
|
{
|
|
|
|
vfio_unregister_iommu_driver(&tce_iommu_driver_ops);
|
|
|
|
}
|
|
|
|
|
|
|
|
module_init(tce_iommu_init);
|
|
|
|
module_exit(tce_iommu_cleanup);
|
|
|
|
|
|
|
|
MODULE_VERSION(DRIVER_VERSION);
|
|
|
|
MODULE_LICENSE("GPL v2");
|
|
|
|
MODULE_AUTHOR(DRIVER_AUTHOR);
|
|
|
|
MODULE_DESCRIPTION(DRIVER_DESC);
|
|
|
|
|