vfio/common: Move vfio_host_win_add/del into spapr.c

Only spapr supports a customed host window list, other vfio driver
assume 64bit host window. So remove the check in listener callback
and move vfio_host_win_add/del into spapr.c and make it static.

With the check removed, we still need to do the same check for
VFIO_SPAPR_TCE_IOMMU which allows a single host window range
[dma32_window_start, dma32_window_size). Move vfio_find_hostwin
into spapr.c and do same check in vfio_container_add_section_window
instead.

When mapping a ram device section, if it's unaligned with
hostwin->iova_pgsizes, this mapping is bypassed. With hostwin
moved into spapr, we changed to check container->pgsizes.

Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
master
Zhenzhong Duan 2023-11-02 15:12:26 +08:00 committed by Cédric Le Goater
parent a17879f0e2
commit a2347c60a8
4 changed files with 85 additions and 89 deletions

View File

@ -245,44 +245,6 @@ bool vfio_devices_all_running_and_mig_active(VFIOContainer *container)
return true;
}
void vfio_host_win_add(VFIOContainer *container, hwaddr min_iova,
hwaddr max_iova, uint64_t iova_pgsizes)
{
VFIOHostDMAWindow *hostwin;
QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) {
if (ranges_overlap(hostwin->min_iova,
hostwin->max_iova - hostwin->min_iova + 1,
min_iova,
max_iova - min_iova + 1)) {
hw_error("%s: Overlapped IOMMU are not enabled", __func__);
}
}
hostwin = g_malloc0(sizeof(*hostwin));
hostwin->min_iova = min_iova;
hostwin->max_iova = max_iova;
hostwin->iova_pgsizes = iova_pgsizes;
QLIST_INSERT_HEAD(&container->hostwin_list, hostwin, hostwin_next);
}
int vfio_host_win_del(VFIOContainer *container,
hwaddr min_iova, hwaddr max_iova)
{
VFIOHostDMAWindow *hostwin;
QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) {
if (hostwin->min_iova == min_iova && hostwin->max_iova == max_iova) {
QLIST_REMOVE(hostwin, hostwin_next);
g_free(hostwin);
return 0;
}
}
return -1;
}
static bool vfio_listener_skipped_section(MemoryRegionSection *section)
{
return (!memory_region_is_ram(section->mr) &&
@ -531,22 +493,6 @@ static void vfio_unregister_ram_discard_listener(VFIOContainer *container,
g_free(vrdl);
}
static VFIOHostDMAWindow *vfio_find_hostwin(VFIOContainer *container,
hwaddr iova, hwaddr end)
{
VFIOHostDMAWindow *hostwin;
bool hostwin_found = false;
QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) {
if (hostwin->min_iova <= iova && end <= hostwin->max_iova) {
hostwin_found = true;
break;
}
}
return hostwin_found ? hostwin : NULL;
}
static bool vfio_known_safe_misalignment(MemoryRegionSection *section)
{
MemoryRegion *mr = section->mr;
@ -625,7 +571,6 @@ static void vfio_listener_region_add(MemoryListener *listener,
Int128 llend, llsize;
void *vaddr;
int ret;
VFIOHostDMAWindow *hostwin;
Error *err = NULL;
if (!vfio_listener_valid_section(section, "region_add")) {
@ -647,13 +592,6 @@ static void vfio_listener_region_add(MemoryListener *listener,
goto fail;
}
hostwin = vfio_find_hostwin(container, iova, end);
if (!hostwin) {
error_setg(&err, "Container %p can't map guest IOVA region"
" 0x%"HWADDR_PRIx"..0x%"HWADDR_PRIx, container, iova, end);
goto fail;
}
memory_region_ref(section->mr);
if (memory_region_is_iommu(section->mr)) {
@ -734,7 +672,7 @@ static void vfio_listener_region_add(MemoryListener *listener,
llsize = int128_sub(llend, int128_make64(iova));
if (memory_region_is_ram_device(section->mr)) {
hwaddr pgmask = (1ULL << ctz64(hostwin->iova_pgsizes)) - 1;
hwaddr pgmask = (1ULL << ctz64(container->pgsizes)) - 1;
if ((iova & pgmask) || (int128_get64(llsize) & pgmask)) {
trace_vfio_listener_region_add_no_dma_map(
@ -833,12 +771,8 @@ static void vfio_listener_region_del(MemoryListener *listener,
if (memory_region_is_ram_device(section->mr)) {
hwaddr pgmask;
VFIOHostDMAWindow *hostwin;
hostwin = vfio_find_hostwin(container, iova, end);
assert(hostwin); /* or region_add() would have failed */
pgmask = (1ULL << ctz64(hostwin->iova_pgsizes)) - 1;
pgmask = (1ULL << ctz64(container->pgsizes)) - 1;
try_unmap = !((iova & pgmask) || (int128_get64(llsize) & pgmask));
} else if (memory_region_has_ram_discard_manager(section->mr)) {
vfio_unregister_ram_discard_listener(container, section);

View File

@ -551,7 +551,6 @@ static int vfio_connect_container(VFIOGroup *group, AddressSpace *as,
container->dma_max_mappings = 0;
container->iova_ranges = NULL;
QLIST_INIT(&container->giommu_list);
QLIST_INIT(&container->hostwin_list);
QLIST_INIT(&container->vrdl_list);
ret = vfio_init_container(container, group->fd, errp);
@ -591,14 +590,6 @@ static int vfio_connect_container(VFIOGroup *group, AddressSpace *as,
vfio_get_iommu_info_migration(container, info);
g_free(info);
/*
* FIXME: We should parse VFIO_IOMMU_TYPE1_INFO_CAP_IOVA_RANGE
* information to get the actual window extent rather than assume
* a 64-bit IOVA address space.
*/
vfio_host_win_add(container, 0, (hwaddr)-1, container->pgsizes);
break;
}
case VFIO_SPAPR_TCE_v2_IOMMU:
@ -687,7 +678,6 @@ static void vfio_disconnect_container(VFIOGroup *group)
if (QLIST_EMPTY(&container->group_list)) {
VFIOAddressSpace *space = container->space;
VFIOGuestIOMMU *giommu, *tmp;
VFIOHostDMAWindow *hostwin, *next;
QLIST_REMOVE(container, next);
@ -698,12 +688,6 @@ static void vfio_disconnect_container(VFIOGroup *group)
g_free(giommu);
}
QLIST_FOREACH_SAFE(hostwin, &container->hostwin_list, hostwin_next,
next) {
QLIST_REMOVE(hostwin, hostwin_next);
g_free(hostwin);
}
trace_vfio_disconnect_container(container->fd);
close(container->fd);
vfio_free_container(container);

View File

@ -146,6 +146,60 @@ static const MemoryListener vfio_prereg_listener = {
.region_del = vfio_prereg_listener_region_del,
};
static void vfio_host_win_add(VFIOContainer *container, hwaddr min_iova,
hwaddr max_iova, uint64_t iova_pgsizes)
{
VFIOHostDMAWindow *hostwin;
QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) {
if (ranges_overlap(hostwin->min_iova,
hostwin->max_iova - hostwin->min_iova + 1,
min_iova,
max_iova - min_iova + 1)) {
hw_error("%s: Overlapped IOMMU are not enabled", __func__);
}
}
hostwin = g_malloc0(sizeof(*hostwin));
hostwin->min_iova = min_iova;
hostwin->max_iova = max_iova;
hostwin->iova_pgsizes = iova_pgsizes;
QLIST_INSERT_HEAD(&container->hostwin_list, hostwin, hostwin_next);
}
static int vfio_host_win_del(VFIOContainer *container,
hwaddr min_iova, hwaddr max_iova)
{
VFIOHostDMAWindow *hostwin;
QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) {
if (hostwin->min_iova == min_iova && hostwin->max_iova == max_iova) {
QLIST_REMOVE(hostwin, hostwin_next);
g_free(hostwin);
return 0;
}
}
return -1;
}
static VFIOHostDMAWindow *vfio_find_hostwin(VFIOContainer *container,
hwaddr iova, hwaddr end)
{
VFIOHostDMAWindow *hostwin;
bool hostwin_found = false;
QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) {
if (hostwin->min_iova <= iova && end <= hostwin->max_iova) {
hostwin_found = true;
break;
}
}
return hostwin_found ? hostwin : NULL;
}
static int vfio_spapr_remove_window(VFIOContainer *container,
hwaddr offset_within_address_space)
{
@ -267,6 +321,26 @@ int vfio_container_add_section_window(VFIOContainer *container,
hwaddr pgsize = 0;
int ret;
/*
* VFIO_SPAPR_TCE_IOMMU supports a single host window between
* [dma32_window_start, dma32_window_size), we need to ensure
* the section fall in this range.
*/
if (container->iommu_type == VFIO_SPAPR_TCE_IOMMU) {
hwaddr iova, end;
iova = section->offset_within_address_space;
end = iova + int128_get64(section->size) - 1;
if (!vfio_find_hostwin(container, iova, end)) {
error_setg(errp, "Container %p can't map guest IOVA region"
" 0x%"HWADDR_PRIx"..0x%"HWADDR_PRIx, container,
iova, end);
return -EINVAL;
}
return 0;
}
if (container->iommu_type != VFIO_SPAPR_TCE_v2_IOMMU) {
return 0;
}
@ -351,6 +425,8 @@ int vfio_spapr_container_init(VFIOContainer *container, Error **errp)
bool v2 = container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU;
int ret, fd = container->fd;
QLIST_INIT(&container->hostwin_list);
/*
* The host kernel code implementing VFIO_IOMMU_DISABLE is called
* when container fd is closed so we do not call it explicitly
@ -418,7 +494,14 @@ listener_unregister_exit:
void vfio_spapr_container_deinit(VFIOContainer *container)
{
VFIOHostDMAWindow *hostwin, *next;
if (container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU) {
memory_listener_unregister(&container->prereg_listener);
}
QLIST_FOREACH_SAFE(hostwin, &container->hostwin_list, hostwin_next,
next) {
QLIST_REMOVE(hostwin, hostwin_next);
g_free(hostwin);
}
}

View File

@ -207,11 +207,6 @@ typedef struct {
hwaddr pages;
} VFIOBitmap;
void vfio_host_win_add(VFIOContainer *container,
hwaddr min_iova, hwaddr max_iova,
uint64_t iova_pgsizes);
int vfio_host_win_del(VFIOContainer *container, hwaddr min_iova,
hwaddr max_iova);
VFIOAddressSpace *vfio_get_address_space(AddressSpace *as);
void vfio_put_address_space(VFIOAddressSpace *space);
bool vfio_devices_all_running_and_saving(VFIOContainer *container);