To be used in order to create foreign mappings. This is based on the
ZONE_DEVICE facility which is used by persistent memory devices in
order to create struct pages and kernel virtual mappings for the IOMEM
areas of such devices. Note that on kernels without support for
ZONE_DEVICE Xen will fallback to use ballooned pages in order to
create foreign mappings.
The newly added helpers use the same parameters as the existing
{alloc/free}_xenballooned_pages functions, which allows for in-place
replacement of the callers. Once a memory region has been added to be
used as scratch mapping space it will no longer be released, and pages
returned are kept in a linked list. This allows to have a buffer of
pages and prevents resorting to frequent additions and removals of
regions.
If enabled (because ZONE_DEVICE is supported) the usage of the new
functionality untangles Xen balloon and RAM hotplug from the usage of
unpopulated physical memory ranges to map foreign pages, which is the
correct thing to do in order to avoid mappings of foreign pages depend
on memory hotplug.
Signed-off-by: Roger Pau Monné <[email protected]>
---
I've not added a new memory_type type and just used
MEMORY_DEVICE_DEVDAX which seems to be what we want for such memory
regions. I'm unsure whether abusing this type is fine, or if I should
instead add a specific type, maybe MEMORY_DEVICE_GENERIC? I don't
think we should be using a specific Xen type at all.
---
Cc: Oleksandr Andrushchenko <[email protected]>
Cc: David Airlie <[email protected]>
Cc: Daniel Vetter <[email protected]>
Cc: Boris Ostrovsky <[email protected]>
Cc: Juergen Gross <[email protected]>
Cc: Stefano Stabellini <[email protected]>
Cc: Dan Carpenter <[email protected]>
Cc: Roger Pau Monne <[email protected]>
Cc: Wei Liu <[email protected]>
Cc: Yan Yankovskyi <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: David Hildenbrand <[email protected]>
Cc: Michal Hocko <[email protected]>
---
drivers/gpu/drm/xen/xen_drm_front_gem.c | 8 +-
drivers/xen/Makefile | 1 +
drivers/xen/balloon.c | 4 +-
drivers/xen/grant-table.c | 4 +-
drivers/xen/privcmd.c | 4 +-
drivers/xen/unpopulated-alloc.c | 222 ++++++++++++++++++++++++
drivers/xen/xenbus/xenbus_client.c | 6 +-
drivers/xen/xlate_mmu.c | 4 +-
include/xen/xen.h | 8 +
9 files changed, 246 insertions(+), 15 deletions(-)
create mode 100644 drivers/xen/unpopulated-alloc.c
diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.c b/drivers/gpu/drm/xen/xen_drm_front_gem.c
index f0b85e094111..9dd06eae767a 100644
--- a/drivers/gpu/drm/xen/xen_drm_front_gem.c
+++ b/drivers/gpu/drm/xen/xen_drm_front_gem.c
@@ -99,8 +99,8 @@ static struct xen_gem_object *gem_create(struct drm_device *dev, size_t size)
* allocate ballooned pages which will be used to map
* grant references provided by the backend
*/
- ret = alloc_xenballooned_pages(xen_obj->num_pages,
- xen_obj->pages);
+ ret = xen_alloc_unpopulated_pages(xen_obj->num_pages,
+ xen_obj->pages);
if (ret < 0) {
DRM_ERROR("Cannot allocate %zu ballooned pages: %d\n",
xen_obj->num_pages, ret);
@@ -152,8 +152,8 @@ void xen_drm_front_gem_free_object_unlocked(struct drm_gem_object *gem_obj)
} else {
if (xen_obj->pages) {
if (xen_obj->be_alloc) {
- free_xenballooned_pages(xen_obj->num_pages,
- xen_obj->pages);
+ xen_free_unpopulated_pages(xen_obj->num_pages,
+ xen_obj->pages);
gem_free_pages_array(xen_obj);
} else {
drm_gem_put_pages(&xen_obj->base,
diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index 0d322f3d90cd..788a5d9c8ef0 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -42,3 +42,4 @@ xen-gntdev-$(CONFIG_XEN_GNTDEV_DMABUF) += gntdev-dmabuf.o
xen-gntalloc-y := gntalloc.o
xen-privcmd-y := privcmd.o privcmd-buf.o
obj-$(CONFIG_XEN_FRONT_PGDIR_SHBUF) += xen-front-pgdir-shbuf.o
+obj-$(CONFIG_ZONE_DEVICE) += unpopulated-alloc.o
diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
index b1d8b028bf80..815ef10eb2ff 100644
--- a/drivers/xen/balloon.c
+++ b/drivers/xen/balloon.c
@@ -654,7 +654,7 @@ void free_xenballooned_pages(int nr_pages, struct page **pages)
}
EXPORT_SYMBOL(free_xenballooned_pages);
-#ifdef CONFIG_XEN_PV
+#if defined(CONFIG_XEN_PV) && !defined(CONFIG_ZONE_DEVICE)
static void __init balloon_add_region(unsigned long start_pfn,
unsigned long pages)
{
@@ -708,7 +708,7 @@ static int __init balloon_init(void)
register_sysctl_table(xen_root);
#endif
-#ifdef CONFIG_XEN_PV
+#if defined(CONFIG_XEN_PV) && !defined(CONFIG_ZONE_DEVICE)
{
int i;
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index 8d06bf1cc347..523dcdf39cc9 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -801,7 +801,7 @@ int gnttab_alloc_pages(int nr_pages, struct page **pages)
{
int ret;
- ret = alloc_xenballooned_pages(nr_pages, pages);
+ ret = xen_alloc_unpopulated_pages(nr_pages, pages);
if (ret < 0)
return ret;
@@ -836,7 +836,7 @@ EXPORT_SYMBOL_GPL(gnttab_pages_clear_private);
void gnttab_free_pages(int nr_pages, struct page **pages)
{
gnttab_pages_clear_private(nr_pages, pages);
- free_xenballooned_pages(nr_pages, pages);
+ xen_free_unpopulated_pages(nr_pages, pages);
}
EXPORT_SYMBOL_GPL(gnttab_free_pages);
diff --git a/drivers/xen/privcmd.c b/drivers/xen/privcmd.c
index a250d118144a..56000ab70974 100644
--- a/drivers/xen/privcmd.c
+++ b/drivers/xen/privcmd.c
@@ -425,7 +425,7 @@ static int alloc_empty_pages(struct vm_area_struct *vma, int numpgs)
if (pages == NULL)
return -ENOMEM;
- rc = alloc_xenballooned_pages(numpgs, pages);
+ rc = xen_alloc_unpopulated_pages(numpgs, pages);
if (rc != 0) {
pr_warn("%s Could not alloc %d pfns rc:%d\n", __func__,
numpgs, rc);
@@ -900,7 +900,7 @@ static void privcmd_close(struct vm_area_struct *vma)
rc = xen_unmap_domain_gfn_range(vma, numgfns, pages);
if (rc == 0)
- free_xenballooned_pages(numpgs, pages);
+ xen_free_unpopulated_pages(numpgs, pages);
else
pr_crit("unable to unmap MFN range: leaking %d pages. rc=%d\n",
numpgs, rc);
diff --git a/drivers/xen/unpopulated-alloc.c b/drivers/xen/unpopulated-alloc.c
new file mode 100644
index 000000000000..aaa91cefbbf9
--- /dev/null
+++ b/drivers/xen/unpopulated-alloc.c
@@ -0,0 +1,222 @@
+/*
+ * Helpers to allocate unpopulated memory for foreign mappings
+ *
+ * Copyright (c) 2020, Citrix Systems R&D
+ *
+ * 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; or, when distributed
+ * separately from the Linux kernel or incorporated into other
+ * software packages, subject to the following license:
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this source file (the "Software"), to deal in the Software without
+ * restriction, including without limitation the rights to use, copy, modify,
+ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
+ * and to permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
+ * IN THE SOFTWARE.
+ */
+
+#include <linux/errno.h>
+#include <linux/gfp.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/memremap.h>
+#include <linux/slab.h>
+
+#include <asm/page.h>
+
+#include <xen/page.h>
+#include <xen/xen.h>
+
+static DEFINE_MUTEX(lock);
+static LIST_HEAD(list);
+static unsigned int count;
+
+static int fill(unsigned int nr_pages)
+{
+ struct dev_pagemap *pgmap;
+ void *vaddr;
+ unsigned int i, alloc_pages = round_up(nr_pages, PAGES_PER_SECTION);
+ int nid, ret;
+
+ pgmap = kzalloc(sizeof(*pgmap), GFP_KERNEL);
+ if (!pgmap)
+ return -ENOMEM;
+
+ pgmap->type = MEMORY_DEVICE_DEVDAX;
+ pgmap->res.name = "XEN SCRATCH";
+ pgmap->res.flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+
+ ret = allocate_resource(&iomem_resource, &pgmap->res,
+ alloc_pages * PAGE_SIZE, 0, -1,
+ PAGES_PER_SECTION * PAGE_SIZE, NULL, NULL);
+ if (ret < 0) {
+ pr_err("Cannot allocate new IOMEM resource\n");
+ kfree(pgmap);
+ return ret;
+ }
+
+ nid = memory_add_physaddr_to_nid(pgmap->res.start);
+
+#ifdef CONFIG_XEN_HAVE_PVMMU
+ /*
+ * We don't support PV MMU when Linux and Xen is using
+ * different page granularity.
+ */
+ BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
+
+ /*
+ * memremap will build page tables for the new memory so
+ * the p2m must contain invalid entries so the correct
+ * non-present PTEs will be written.
+ *
+ * If a failure occurs, the original (identity) p2m entries
+ * are not restored since this region is now known not to
+ * conflict with any devices.
+ */
+ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+ xen_pfn_t pfn = PFN_DOWN(pgmap->res.start);
+
+ for (i = 0; i < alloc_pages; i++) {
+ if (!set_phys_to_machine(pfn + i, INVALID_P2M_ENTRY)) {
+ pr_warn("set_phys_to_machine() failed, no memory added\n");
+ release_resource(&pgmap->res);
+ kfree(pgmap);
+ return -ENOMEM;
+ }
+ }
+ }
+#endif
+
+ vaddr = memremap_pages(pgmap, nid);
+ if (IS_ERR(vaddr)) {
+ pr_err("Cannot remap memory range\n");
+ release_resource(&pgmap->res);
+ kfree(pgmap);
+ return PTR_ERR(vaddr);
+ }
+
+ for (i = 0; i < alloc_pages; i++) {
+ struct page *pg = virt_to_page(vaddr + PAGE_SIZE * i);
+
+ BUG_ON(!virt_addr_valid(vaddr + PAGE_SIZE * i));
+ list_add(&pg->lru, &list);
+ count++;
+ }
+
+ return 0;
+}
+
+/**
+ * xen_alloc_unpopulated_pages - alloc unpopulated pages
+ * @nr_pages: Number of pages
+ * @pages: pages returned
+ * @return 0 on success, error otherwise
+ */
+int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages)
+{
+ unsigned int i;
+ int ret = 0;
+
+ mutex_lock(&lock);
+ if (count < nr_pages) {
+ ret = fill(nr_pages);
+ if (ret)
+ goto out;
+ }
+
+ for (i = 0; i < nr_pages; i++) {
+ struct page *pg = list_first_entry_or_null(&list, struct page,
+ lru);
+
+ BUG_ON(!pg);
+ list_del(&pg->lru);
+ count--;
+ pages[i] = pg;
+
+#ifdef CONFIG_XEN_HAVE_PVMMU
+ /*
+ * We don't support PV MMU when Linux and Xen is using
+ * different page granularity.
+ */
+ BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
+
+ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+ ret = xen_alloc_p2m_entry(page_to_pfn(pg));
+ if (ret < 0) {
+ unsigned int j;
+
+ for (j = 0; j <= i; j++) {
+ list_add(&pages[j]->lru, &list);
+ count++;
+ }
+ goto out;
+ }
+ }
+#endif
+ }
+
+out:
+ mutex_unlock(&lock);
+ return ret;
+}
+EXPORT_SYMBOL(xen_alloc_unpopulated_pages);
+
+/**
+ * xen_free_unpopulated_pages - return unpopulated pages
+ * @nr_pages: Number of pages
+ * @pages: pages to return
+ */
+void xen_free_unpopulated_pages(unsigned int nr_pages, struct page **pages)
+{
+ unsigned int i;
+
+ mutex_lock(&lock);
+ for (i = 0; i < nr_pages; i++) {
+ list_add(&pages[i]->lru, &list);
+ count++;
+ }
+ mutex_unlock(&lock);
+}
+EXPORT_SYMBOL(xen_free_unpopulated_pages);
+
+#ifdef CONFIG_XEN_PV
+static int __init init(void)
+{
+ unsigned int i;
+
+ if (!xen_domain())
+ return -ENODEV;
+
+ /*
+ * Initialize with pages from the extra memory regions (see
+ * arch/x86/xen/setup.c).
+ */
+ for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) {
+ unsigned int j;
+
+ for (j = 0; j < xen_extra_mem[i].n_pfns; j++) {
+ struct page *pg =
+ pfn_to_page(xen_extra_mem[i].start_pfn + j);
+
+ list_add(&pg->lru, &list);
+ count++;
+ }
+ }
+
+ return 0;
+}
+subsys_initcall(init);
+#endif
diff --git a/drivers/xen/xenbus/xenbus_client.c b/drivers/xen/xenbus/xenbus_client.c
index 786fbb7d8be0..70b6c4780fbd 100644
--- a/drivers/xen/xenbus/xenbus_client.c
+++ b/drivers/xen/xenbus/xenbus_client.c
@@ -615,7 +615,7 @@ static int xenbus_map_ring_hvm(struct xenbus_device *dev,
bool leaked = false;
unsigned int nr_pages = XENBUS_PAGES(nr_grefs);
- err = alloc_xenballooned_pages(nr_pages, node->hvm.pages);
+ err = xen_alloc_unpopulated_pages(nr_pages, node->hvm.pages);
if (err)
goto out_err;
@@ -656,7 +656,7 @@ static int xenbus_map_ring_hvm(struct xenbus_device *dev,
addr, nr_pages);
out_free_ballooned_pages:
if (!leaked)
- free_xenballooned_pages(nr_pages, node->hvm.pages);
+ xen_free_unpopulated_pages(nr_pages, node->hvm.pages);
out_err:
return err;
}
@@ -852,7 +852,7 @@ static int xenbus_unmap_ring_hvm(struct xenbus_device *dev, void *vaddr)
info.addrs);
if (!rv) {
vunmap(vaddr);
- free_xenballooned_pages(nr_pages, node->hvm.pages);
+ xen_free_unpopulated_pages(nr_pages, node->hvm.pages);
}
else
WARN(1, "Leaking %p, size %u page(s)\n", vaddr, nr_pages);
diff --git a/drivers/xen/xlate_mmu.c b/drivers/xen/xlate_mmu.c
index 7b1077f0abcb..34742c6e189e 100644
--- a/drivers/xen/xlate_mmu.c
+++ b/drivers/xen/xlate_mmu.c
@@ -232,7 +232,7 @@ int __init xen_xlate_map_ballooned_pages(xen_pfn_t **gfns, void **virt,
kfree(pages);
return -ENOMEM;
}
- rc = alloc_xenballooned_pages(nr_pages, pages);
+ rc = xen_alloc_unpopulated_pages(nr_pages, pages);
if (rc) {
pr_warn("%s Couldn't balloon alloc %ld pages rc:%d\n", __func__,
nr_pages, rc);
@@ -249,7 +249,7 @@ int __init xen_xlate_map_ballooned_pages(xen_pfn_t **gfns, void **virt,
if (!vaddr) {
pr_warn("%s Couldn't map %ld pages rc:%d\n", __func__,
nr_pages, rc);
- free_xenballooned_pages(nr_pages, pages);
+ xen_free_unpopulated_pages(nr_pages, pages);
kfree(pages);
kfree(pfns);
return -ENOMEM;
diff --git a/include/xen/xen.h b/include/xen/xen.h
index 19a72f591e2b..aa33bc0d933c 100644
--- a/include/xen/xen.h
+++ b/include/xen/xen.h
@@ -52,4 +52,12 @@ bool xen_biovec_phys_mergeable(const struct bio_vec *vec1,
extern u64 xen_saved_max_mem_size;
#endif
+#ifdef CONFIG_ZONE_DEVICE
+int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages);
+void xen_free_unpopulated_pages(unsigned int nr_pages, struct page **pages);
+#else
+#define xen_alloc_unpopulated_pages alloc_xenballooned_pages
+#define xen_free_unpopulated_pages free_xenballooned_pages
+#endif
+
#endif /* _XEN_XEN_H */
--
2.27.0
On 24.07.20 14:42, Roger Pau Monne wrote:
> To be used in order to create foreign mappings. This is based on the
> ZONE_DEVICE facility which is used by persistent memory devices in
> order to create struct pages and kernel virtual mappings for the IOMEM
> areas of such devices. Note that on kernels without support for
> ZONE_DEVICE Xen will fallback to use ballooned pages in order to
> create foreign mappings.
>
> The newly added helpers use the same parameters as the existing
> {alloc/free}_xenballooned_pages functions, which allows for in-place
> replacement of the callers. Once a memory region has been added to be
> used as scratch mapping space it will no longer be released, and pages
> returned are kept in a linked list. This allows to have a buffer of
> pages and prevents resorting to frequent additions and removals of
> regions.
>
> If enabled (because ZONE_DEVICE is supported) the usage of the new
> functionality untangles Xen balloon and RAM hotplug from the usage of
> unpopulated physical memory ranges to map foreign pages, which is the
> correct thing to do in order to avoid mappings of foreign pages depend
> on memory hotplug.
>
> Signed-off-by: Roger Pau Monné <[email protected]>
> ---
> I've not added a new memory_type type and just used
> MEMORY_DEVICE_DEVDAX which seems to be what we want for such memory
> regions. I'm unsure whether abusing this type is fine, or if I should
> instead add a specific type, maybe MEMORY_DEVICE_GENERIC? I don't
> think we should be using a specific Xen type at all.
> ---
> Cc: Oleksandr Andrushchenko <[email protected]>
> Cc: David Airlie <[email protected]>
> Cc: Daniel Vetter <[email protected]>
> Cc: Boris Ostrovsky <[email protected]>
> Cc: Juergen Gross <[email protected]>
> Cc: Stefano Stabellini <[email protected]>
> Cc: Dan Carpenter <[email protected]>
> Cc: Roger Pau Monne <[email protected]>
> Cc: Wei Liu <[email protected]>
> Cc: Yan Yankovskyi <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: David Hildenbrand <[email protected]>
> Cc: Michal Hocko <[email protected]>
> ---
> drivers/gpu/drm/xen/xen_drm_front_gem.c | 8 +-
> drivers/xen/Makefile | 1 +
> drivers/xen/balloon.c | 4 +-
> drivers/xen/grant-table.c | 4 +-
> drivers/xen/privcmd.c | 4 +-
> drivers/xen/unpopulated-alloc.c | 222 ++++++++++++++++++++++++
> drivers/xen/xenbus/xenbus_client.c | 6 +-
> drivers/xen/xlate_mmu.c | 4 +-
> include/xen/xen.h | 8 +
> 9 files changed, 246 insertions(+), 15 deletions(-)
> create mode 100644 drivers/xen/unpopulated-alloc.c
>
> diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.c b/drivers/gpu/drm/xen/xen_drm_front_gem.c
> index f0b85e094111..9dd06eae767a 100644
> --- a/drivers/gpu/drm/xen/xen_drm_front_gem.c
> +++ b/drivers/gpu/drm/xen/xen_drm_front_gem.c
> @@ -99,8 +99,8 @@ static struct xen_gem_object *gem_create(struct drm_device *dev, size_t size)
> * allocate ballooned pages which will be used to map
> * grant references provided by the backend
> */
> - ret = alloc_xenballooned_pages(xen_obj->num_pages,
> - xen_obj->pages);
> + ret = xen_alloc_unpopulated_pages(xen_obj->num_pages,
> + xen_obj->pages);
> if (ret < 0) {
> DRM_ERROR("Cannot allocate %zu ballooned pages: %d\n",
> xen_obj->num_pages, ret);
> @@ -152,8 +152,8 @@ void xen_drm_front_gem_free_object_unlocked(struct drm_gem_object *gem_obj)
> } else {
> if (xen_obj->pages) {
> if (xen_obj->be_alloc) {
> - free_xenballooned_pages(xen_obj->num_pages,
> - xen_obj->pages);
> + xen_free_unpopulated_pages(xen_obj->num_pages,
> + xen_obj->pages);
> gem_free_pages_array(xen_obj);
> } else {
> drm_gem_put_pages(&xen_obj->base,
> diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
> index 0d322f3d90cd..788a5d9c8ef0 100644
> --- a/drivers/xen/Makefile
> +++ b/drivers/xen/Makefile
> @@ -42,3 +42,4 @@ xen-gntdev-$(CONFIG_XEN_GNTDEV_DMABUF) += gntdev-dmabuf.o
> xen-gntalloc-y := gntalloc.o
> xen-privcmd-y := privcmd.o privcmd-buf.o
> obj-$(CONFIG_XEN_FRONT_PGDIR_SHBUF) += xen-front-pgdir-shbuf.o
> +obj-$(CONFIG_ZONE_DEVICE) += unpopulated-alloc.o
> diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
> index b1d8b028bf80..815ef10eb2ff 100644
> --- a/drivers/xen/balloon.c
> +++ b/drivers/xen/balloon.c
> @@ -654,7 +654,7 @@ void free_xenballooned_pages(int nr_pages, struct page **pages)
> }
> EXPORT_SYMBOL(free_xenballooned_pages);
>
> -#ifdef CONFIG_XEN_PV
> +#if defined(CONFIG_XEN_PV) && !defined(CONFIG_ZONE_DEVICE)
> static void __init balloon_add_region(unsigned long start_pfn,
> unsigned long pages)
> {
> @@ -708,7 +708,7 @@ static int __init balloon_init(void)
> register_sysctl_table(xen_root);
> #endif
>
> -#ifdef CONFIG_XEN_PV
> +#if defined(CONFIG_XEN_PV) && !defined(CONFIG_ZONE_DEVICE)
> {
> int i;
>
> diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
> index 8d06bf1cc347..523dcdf39cc9 100644
> --- a/drivers/xen/grant-table.c
> +++ b/drivers/xen/grant-table.c
> @@ -801,7 +801,7 @@ int gnttab_alloc_pages(int nr_pages, struct page **pages)
> {
> int ret;
>
> - ret = alloc_xenballooned_pages(nr_pages, pages);
> + ret = xen_alloc_unpopulated_pages(nr_pages, pages);
> if (ret < 0)
> return ret;
>
> @@ -836,7 +836,7 @@ EXPORT_SYMBOL_GPL(gnttab_pages_clear_private);
> void gnttab_free_pages(int nr_pages, struct page **pages)
> {
> gnttab_pages_clear_private(nr_pages, pages);
> - free_xenballooned_pages(nr_pages, pages);
> + xen_free_unpopulated_pages(nr_pages, pages);
> }
> EXPORT_SYMBOL_GPL(gnttab_free_pages);
>
> diff --git a/drivers/xen/privcmd.c b/drivers/xen/privcmd.c
> index a250d118144a..56000ab70974 100644
> --- a/drivers/xen/privcmd.c
> +++ b/drivers/xen/privcmd.c
> @@ -425,7 +425,7 @@ static int alloc_empty_pages(struct vm_area_struct *vma, int numpgs)
> if (pages == NULL)
> return -ENOMEM;
>
> - rc = alloc_xenballooned_pages(numpgs, pages);
> + rc = xen_alloc_unpopulated_pages(numpgs, pages);
> if (rc != 0) {
> pr_warn("%s Could not alloc %d pfns rc:%d\n", __func__,
> numpgs, rc);
> @@ -900,7 +900,7 @@ static void privcmd_close(struct vm_area_struct *vma)
>
> rc = xen_unmap_domain_gfn_range(vma, numgfns, pages);
> if (rc == 0)
> - free_xenballooned_pages(numpgs, pages);
> + xen_free_unpopulated_pages(numpgs, pages);
> else
> pr_crit("unable to unmap MFN range: leaking %d pages. rc=%d\n",
> numpgs, rc);
> diff --git a/drivers/xen/unpopulated-alloc.c b/drivers/xen/unpopulated-alloc.c
> new file mode 100644
> index 000000000000..aaa91cefbbf9
> --- /dev/null
> +++ b/drivers/xen/unpopulated-alloc.c
> @@ -0,0 +1,222 @@
> +/*
> + * Helpers to allocate unpopulated memory for foreign mappings
> + *
> + * Copyright (c) 2020, Citrix Systems R&D
> + *
> + * 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; or, when distributed
> + * separately from the Linux kernel or incorporated into other
> + * software packages, subject to the following license:
> + *
> + * Permission is hereby granted, free of charge, to any person obtaining a copy
> + * of this source file (the "Software"), to deal in the Software without
> + * restriction, including without limitation the rights to use, copy, modify,
> + * merge, publish, distribute, sublicense, and/or sell copies of the Software,
> + * and to permit persons to whom the Software is furnished to do so, subject to
> + * the following conditions:
> + *
> + * The above copyright notice and this permission notice shall be included in
> + * all copies or substantial portions of the Software.
> + *
> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
> + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
> + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
> + * IN THE SOFTWARE.
> + */
Please use:
// SPDX-License-Identifier: GPL-2.0-only
instead of the long GPL sermon.
> +
> +#include <linux/errno.h>
> +#include <linux/gfp.h>
> +#include <linux/kernel.h>
> +#include <linux/mm.h>
> +#include <linux/memremap.h>
> +#include <linux/slab.h>
> +
> +#include <asm/page.h>
> +
> +#include <xen/page.h>
> +#include <xen/xen.h>
> +
> +static DEFINE_MUTEX(lock);
> +static LIST_HEAD(list);
> +static unsigned int count;
> +
> +static int fill(unsigned int nr_pages)
> +{
> + struct dev_pagemap *pgmap;
> + void *vaddr;
> + unsigned int i, alloc_pages = round_up(nr_pages, PAGES_PER_SECTION);
> + int nid, ret;
> +
> + pgmap = kzalloc(sizeof(*pgmap), GFP_KERNEL);
> + if (!pgmap)
> + return -ENOMEM;
> +
> + pgmap->type = MEMORY_DEVICE_DEVDAX;
> + pgmap->res.name = "XEN SCRATCH";
> + pgmap->res.flags = IORESOURCE_MEM | IORESOURCE_BUSY;
> +
> + ret = allocate_resource(&iomem_resource, &pgmap->res,
> + alloc_pages * PAGE_SIZE, 0, -1,
> + PAGES_PER_SECTION * PAGE_SIZE, NULL, NULL);
> + if (ret < 0) {
> + pr_err("Cannot allocate new IOMEM resource\n");
> + kfree(pgmap);
> + return ret;
> + }
> +
> + nid = memory_add_physaddr_to_nid(pgmap->res.start);
> +
> +#ifdef CONFIG_XEN_HAVE_PVMMU
> + /*
> + * We don't support PV MMU when Linux and Xen is using
> + * different page granularity.
> + */
> + BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
Drop that, please. PV MMU is x86 only and we surely don't want to add
it to another architecture. On x86 this will never trigger.
> +
> + /*
> + * memremap will build page tables for the new memory so
> + * the p2m must contain invalid entries so the correct
> + * non-present PTEs will be written.
> + *
> + * If a failure occurs, the original (identity) p2m entries
> + * are not restored since this region is now known not to
> + * conflict with any devices.
> + */
> + if (!xen_feature(XENFEAT_auto_translated_physmap)) {
> + xen_pfn_t pfn = PFN_DOWN(pgmap->res.start);
> +
> + for (i = 0; i < alloc_pages; i++) {
> + if (!set_phys_to_machine(pfn + i, INVALID_P2M_ENTRY)) {
> + pr_warn("set_phys_to_machine() failed, no memory added\n");
> + release_resource(&pgmap->res);
> + kfree(pgmap);
> + return -ENOMEM;
> + }
> + }
> + }
> +#endif
> +
> + vaddr = memremap_pages(pgmap, nid);
> + if (IS_ERR(vaddr)) {
> + pr_err("Cannot remap memory range\n");
> + release_resource(&pgmap->res);
> + kfree(pgmap);
> + return PTR_ERR(vaddr);
> + }
> +
> + for (i = 0; i < alloc_pages; i++) {
> + struct page *pg = virt_to_page(vaddr + PAGE_SIZE * i);
> +
> + BUG_ON(!virt_addr_valid(vaddr + PAGE_SIZE * i));
> + list_add(&pg->lru, &list);
> + count++;
> + }
> +
> + return 0;
> +}
> +
> +/**
> + * xen_alloc_unpopulated_pages - alloc unpopulated pages
> + * @nr_pages: Number of pages
> + * @pages: pages returned
> + * @return 0 on success, error otherwise
> + */
> +int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages)
> +{
> + unsigned int i;
> + int ret = 0;
> +
> + mutex_lock(&lock);
> + if (count < nr_pages) {
> + ret = fill(nr_pages);
I'd rather use: ret = fill(nr_pages - count);
> + if (ret)
> + goto out;
> + }
> +
> + for (i = 0; i < nr_pages; i++) {
> + struct page *pg = list_first_entry_or_null(&list, struct page,
> + lru);
> +
> + BUG_ON(!pg);
> + list_del(&pg->lru);
> + count--;
> + pages[i] = pg;
> +
> +#ifdef CONFIG_XEN_HAVE_PVMMU
> + /*
> + * We don't support PV MMU when Linux and Xen is using
> + * different page granularity.
> + */
> + BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
Having two identical BUILD_BUG_ON() in the same source is really not
wanted.
> +
> + if (!xen_feature(XENFEAT_auto_translated_physmap)) {
> + ret = xen_alloc_p2m_entry(page_to_pfn(pg));
> + if (ret < 0) {
> + unsigned int j;
> +
> + for (j = 0; j <= i; j++) {
> + list_add(&pages[j]->lru, &list);
> + count++;
> + }
> + goto out;
> + }
> + }
> +#endif
> + }
> +
> +out:
> + mutex_unlock(&lock);
> + return ret;
> +}
> +EXPORT_SYMBOL(xen_alloc_unpopulated_pages);
> +
> +/**
> + * xen_free_unpopulated_pages - return unpopulated pages
> + * @nr_pages: Number of pages
> + * @pages: pages to return
> + */
> +void xen_free_unpopulated_pages(unsigned int nr_pages, struct page **pages)
> +{
> + unsigned int i;
> +
> + mutex_lock(&lock);
> + for (i = 0; i < nr_pages; i++) {
> + list_add(&pages[i]->lru, &list);
> + count++;
> + }
> + mutex_unlock(&lock);
> +}
> +EXPORT_SYMBOL(xen_free_unpopulated_pages);
> +
> +#ifdef CONFIG_XEN_PV
> +static int __init init(void)
> +{
> + unsigned int i;
> +
> + if (!xen_domain())
> + return -ENODEV;
> +
> + /*
> + * Initialize with pages from the extra memory regions (see
> + * arch/x86/xen/setup.c).
> + */
> + for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) {
> + unsigned int j;
> +
> + for (j = 0; j < xen_extra_mem[i].n_pfns; j++) {
> + struct page *pg =
> + pfn_to_page(xen_extra_mem[i].start_pfn + j);
> +
> + list_add(&pg->lru, &list);
> + count++;
> + }
> + }
> +
> + return 0;
> +}
> +subsys_initcall(init);
> +#endif
> diff --git a/drivers/xen/xenbus/xenbus_client.c b/drivers/xen/xenbus/xenbus_client.c
> index 786fbb7d8be0..70b6c4780fbd 100644
> --- a/drivers/xen/xenbus/xenbus_client.c
> +++ b/drivers/xen/xenbus/xenbus_client.c
> @@ -615,7 +615,7 @@ static int xenbus_map_ring_hvm(struct xenbus_device *dev,
> bool leaked = false;
> unsigned int nr_pages = XENBUS_PAGES(nr_grefs);
>
> - err = alloc_xenballooned_pages(nr_pages, node->hvm.pages);
> + err = xen_alloc_unpopulated_pages(nr_pages, node->hvm.pages);
> if (err)
> goto out_err;
>
> @@ -656,7 +656,7 @@ static int xenbus_map_ring_hvm(struct xenbus_device *dev,
> addr, nr_pages);
> out_free_ballooned_pages:
> if (!leaked)
> - free_xenballooned_pages(nr_pages, node->hvm.pages);
> + xen_free_unpopulated_pages(nr_pages, node->hvm.pages);
> out_err:
> return err;
> }
> @@ -852,7 +852,7 @@ static int xenbus_unmap_ring_hvm(struct xenbus_device *dev, void *vaddr)
> info.addrs);
> if (!rv) {
> vunmap(vaddr);
> - free_xenballooned_pages(nr_pages, node->hvm.pages);
> + xen_free_unpopulated_pages(nr_pages, node->hvm.pages);
> }
> else
> WARN(1, "Leaking %p, size %u page(s)\n", vaddr, nr_pages);
> diff --git a/drivers/xen/xlate_mmu.c b/drivers/xen/xlate_mmu.c
> index 7b1077f0abcb..34742c6e189e 100644
> --- a/drivers/xen/xlate_mmu.c
> +++ b/drivers/xen/xlate_mmu.c
> @@ -232,7 +232,7 @@ int __init xen_xlate_map_ballooned_pages(xen_pfn_t **gfns, void **virt,
> kfree(pages);
> return -ENOMEM;
> }
> - rc = alloc_xenballooned_pages(nr_pages, pages);
> + rc = xen_alloc_unpopulated_pages(nr_pages, pages);
> if (rc) {
> pr_warn("%s Couldn't balloon alloc %ld pages rc:%d\n", __func__,
> nr_pages, rc);
> @@ -249,7 +249,7 @@ int __init xen_xlate_map_ballooned_pages(xen_pfn_t **gfns, void **virt,
> if (!vaddr) {
> pr_warn("%s Couldn't map %ld pages rc:%d\n", __func__,
> nr_pages, rc);
> - free_xenballooned_pages(nr_pages, pages);
> + xen_free_unpopulated_pages(nr_pages, pages);
> kfree(pages);
> kfree(pfns);
> return -ENOMEM;
> diff --git a/include/xen/xen.h b/include/xen/xen.h
> index 19a72f591e2b..aa33bc0d933c 100644
> --- a/include/xen/xen.h
> +++ b/include/xen/xen.h
> @@ -52,4 +52,12 @@ bool xen_biovec_phys_mergeable(const struct bio_vec *vec1,
> extern u64 xen_saved_max_mem_size;
> #endif
>
> +#ifdef CONFIG_ZONE_DEVICE
> +int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages);
> +void xen_free_unpopulated_pages(unsigned int nr_pages, struct page **pages);
> +#else
> +#define xen_alloc_unpopulated_pages alloc_xenballooned_pages
> +#define xen_free_unpopulated_pages free_xenballooned_pages
> +#endif
> +
> #endif /* _XEN_XEN_H */
>
Juergen
CCing Dan
On 24.07.20 14:42, Roger Pau Monne wrote:
> To be used in order to create foreign mappings. This is based on the
> ZONE_DEVICE facility which is used by persistent memory devices in
> order to create struct pages and kernel virtual mappings for the IOMEM
> areas of such devices. Note that on kernels without support for
> ZONE_DEVICE Xen will fallback to use ballooned pages in order to
> create foreign mappings.
>
> The newly added helpers use the same parameters as the existing
> {alloc/free}_xenballooned_pages functions, which allows for in-place
> replacement of the callers. Once a memory region has been added to be
> used as scratch mapping space it will no longer be released, and pages
> returned are kept in a linked list. This allows to have a buffer of
> pages and prevents resorting to frequent additions and removals of
> regions.
>
> If enabled (because ZONE_DEVICE is supported) the usage of the new
> functionality untangles Xen balloon and RAM hotplug from the usage of
> unpopulated physical memory ranges to map foreign pages, which is the
> correct thing to do in order to avoid mappings of foreign pages depend
> on memory hotplug.
>
> Signed-off-by: Roger Pau Monné <[email protected]>
> ---
> I've not added a new memory_type type and just used
> MEMORY_DEVICE_DEVDAX which seems to be what we want for such memory
> regions. I'm unsure whether abusing this type is fine, or if I should
> instead add a specific type, maybe MEMORY_DEVICE_GENERIC? I don't
> think we should be using a specific Xen type at all.
> ---
> Cc: Oleksandr Andrushchenko <[email protected]>
> Cc: David Airlie <[email protected]>
> Cc: Daniel Vetter <[email protected]>
> Cc: Boris Ostrovsky <[email protected]>
> Cc: Juergen Gross <[email protected]>
> Cc: Stefano Stabellini <[email protected]>
> Cc: Dan Carpenter <[email protected]>
> Cc: Roger Pau Monne <[email protected]>
> Cc: Wei Liu <[email protected]>
> Cc: Yan Yankovskyi <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: David Hildenbrand <[email protected]>
> Cc: Michal Hocko <[email protected]>
> ---
> drivers/gpu/drm/xen/xen_drm_front_gem.c | 8 +-
> drivers/xen/Makefile | 1 +
> drivers/xen/balloon.c | 4 +-
> drivers/xen/grant-table.c | 4 +-
> drivers/xen/privcmd.c | 4 +-
> drivers/xen/unpopulated-alloc.c | 222 ++++++++++++++++++++++++
> drivers/xen/xenbus/xenbus_client.c | 6 +-
> drivers/xen/xlate_mmu.c | 4 +-
> include/xen/xen.h | 8 +
> 9 files changed, 246 insertions(+), 15 deletions(-)
> create mode 100644 drivers/xen/unpopulated-alloc.c
>
> diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.c b/drivers/gpu/drm/xen/xen_drm_front_gem.c
> index f0b85e094111..9dd06eae767a 100644
> --- a/drivers/gpu/drm/xen/xen_drm_front_gem.c
> +++ b/drivers/gpu/drm/xen/xen_drm_front_gem.c
> @@ -99,8 +99,8 @@ static struct xen_gem_object *gem_create(struct drm_device *dev, size_t size)
> * allocate ballooned pages which will be used to map
> * grant references provided by the backend
> */
> - ret = alloc_xenballooned_pages(xen_obj->num_pages,
> - xen_obj->pages);
> + ret = xen_alloc_unpopulated_pages(xen_obj->num_pages,
> + xen_obj->pages);
> if (ret < 0) {
> DRM_ERROR("Cannot allocate %zu ballooned pages: %d\n",
> xen_obj->num_pages, ret);
> @@ -152,8 +152,8 @@ void xen_drm_front_gem_free_object_unlocked(struct drm_gem_object *gem_obj)
> } else {
> if (xen_obj->pages) {
> if (xen_obj->be_alloc) {
> - free_xenballooned_pages(xen_obj->num_pages,
> - xen_obj->pages);
> + xen_free_unpopulated_pages(xen_obj->num_pages,
> + xen_obj->pages);
> gem_free_pages_array(xen_obj);
> } else {
> drm_gem_put_pages(&xen_obj->base,
> diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
> index 0d322f3d90cd..788a5d9c8ef0 100644
> --- a/drivers/xen/Makefile
> +++ b/drivers/xen/Makefile
> @@ -42,3 +42,4 @@ xen-gntdev-$(CONFIG_XEN_GNTDEV_DMABUF) += gntdev-dmabuf.o
> xen-gntalloc-y := gntalloc.o
> xen-privcmd-y := privcmd.o privcmd-buf.o
> obj-$(CONFIG_XEN_FRONT_PGDIR_SHBUF) += xen-front-pgdir-shbuf.o
> +obj-$(CONFIG_ZONE_DEVICE) += unpopulated-alloc.o
> diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c
> index b1d8b028bf80..815ef10eb2ff 100644
> --- a/drivers/xen/balloon.c
> +++ b/drivers/xen/balloon.c
> @@ -654,7 +654,7 @@ void free_xenballooned_pages(int nr_pages, struct page **pages)
> }
> EXPORT_SYMBOL(free_xenballooned_pages);
>
> -#ifdef CONFIG_XEN_PV
> +#if defined(CONFIG_XEN_PV) && !defined(CONFIG_ZONE_DEVICE)
> static void __init balloon_add_region(unsigned long start_pfn,
> unsigned long pages)
> {
> @@ -708,7 +708,7 @@ static int __init balloon_init(void)
> register_sysctl_table(xen_root);
> #endif
>
> -#ifdef CONFIG_XEN_PV
> +#if defined(CONFIG_XEN_PV) && !defined(CONFIG_ZONE_DEVICE)
> {
> int i;
>
> diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
> index 8d06bf1cc347..523dcdf39cc9 100644
> --- a/drivers/xen/grant-table.c
> +++ b/drivers/xen/grant-table.c
> @@ -801,7 +801,7 @@ int gnttab_alloc_pages(int nr_pages, struct page **pages)
> {
> int ret;
>
> - ret = alloc_xenballooned_pages(nr_pages, pages);
> + ret = xen_alloc_unpopulated_pages(nr_pages, pages);
> if (ret < 0)
> return ret;
>
> @@ -836,7 +836,7 @@ EXPORT_SYMBOL_GPL(gnttab_pages_clear_private);
> void gnttab_free_pages(int nr_pages, struct page **pages)
> {
> gnttab_pages_clear_private(nr_pages, pages);
> - free_xenballooned_pages(nr_pages, pages);
> + xen_free_unpopulated_pages(nr_pages, pages);
> }
> EXPORT_SYMBOL_GPL(gnttab_free_pages);
>
> diff --git a/drivers/xen/privcmd.c b/drivers/xen/privcmd.c
> index a250d118144a..56000ab70974 100644
> --- a/drivers/xen/privcmd.c
> +++ b/drivers/xen/privcmd.c
> @@ -425,7 +425,7 @@ static int alloc_empty_pages(struct vm_area_struct *vma, int numpgs)
> if (pages == NULL)
> return -ENOMEM;
>
> - rc = alloc_xenballooned_pages(numpgs, pages);
> + rc = xen_alloc_unpopulated_pages(numpgs, pages);
> if (rc != 0) {
> pr_warn("%s Could not alloc %d pfns rc:%d\n", __func__,
> numpgs, rc);
> @@ -900,7 +900,7 @@ static void privcmd_close(struct vm_area_struct *vma)
>
> rc = xen_unmap_domain_gfn_range(vma, numgfns, pages);
> if (rc == 0)
> - free_xenballooned_pages(numpgs, pages);
> + xen_free_unpopulated_pages(numpgs, pages);
> else
> pr_crit("unable to unmap MFN range: leaking %d pages. rc=%d\n",
> numpgs, rc);
> diff --git a/drivers/xen/unpopulated-alloc.c b/drivers/xen/unpopulated-alloc.c
> new file mode 100644
> index 000000000000..aaa91cefbbf9
> --- /dev/null
> +++ b/drivers/xen/unpopulated-alloc.c
> @@ -0,0 +1,222 @@
> +/*
> + * Helpers to allocate unpopulated memory for foreign mappings
> + *
> + * Copyright (c) 2020, Citrix Systems R&D
> + *
> + * 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; or, when distributed
> + * separately from the Linux kernel or incorporated into other
> + * software packages, subject to the following license:
> + *
> + * Permission is hereby granted, free of charge, to any person obtaining a copy
> + * of this source file (the "Software"), to deal in the Software without
> + * restriction, including without limitation the rights to use, copy, modify,
> + * merge, publish, distribute, sublicense, and/or sell copies of the Software,
> + * and to permit persons to whom the Software is furnished to do so, subject to
> + * the following conditions:
> + *
> + * The above copyright notice and this permission notice shall be included in
> + * all copies or substantial portions of the Software.
> + *
> + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
> + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
> + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
> + * IN THE SOFTWARE.
> + */
> +
> +#include <linux/errno.h>
> +#include <linux/gfp.h>
> +#include <linux/kernel.h>
> +#include <linux/mm.h>
> +#include <linux/memremap.h>
> +#include <linux/slab.h>
> +
> +#include <asm/page.h>
> +
> +#include <xen/page.h>
> +#include <xen/xen.h>
> +
> +static DEFINE_MUTEX(lock);
> +static LIST_HEAD(list);
> +static unsigned int count;
> +
> +static int fill(unsigned int nr_pages)
> +{
> + struct dev_pagemap *pgmap;
> + void *vaddr;
> + unsigned int i, alloc_pages = round_up(nr_pages, PAGES_PER_SECTION);
> + int nid, ret;
> +
> + pgmap = kzalloc(sizeof(*pgmap), GFP_KERNEL);
> + if (!pgmap)
> + return -ENOMEM;
> +
> + pgmap->type = MEMORY_DEVICE_DEVDAX;
> + pgmap->res.name = "XEN SCRATCH";
> + pgmap->res.flags = IORESOURCE_MEM | IORESOURCE_BUSY;
> +
> + ret = allocate_resource(&iomem_resource, &pgmap->res,
> + alloc_pages * PAGE_SIZE, 0, -1,
> + PAGES_PER_SECTION * PAGE_SIZE, NULL, NULL);
> + if (ret < 0) {
> + pr_err("Cannot allocate new IOMEM resource\n");
> + kfree(pgmap);
> + return ret;
> + }
> +
> + nid = memory_add_physaddr_to_nid(pgmap->res.start);
> +
> +#ifdef CONFIG_XEN_HAVE_PVMMU
> + /*
> + * We don't support PV MMU when Linux and Xen is using
> + * different page granularity.
> + */
> + BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
> +
> + /*
> + * memremap will build page tables for the new memory so
> + * the p2m must contain invalid entries so the correct
> + * non-present PTEs will be written.
> + *
> + * If a failure occurs, the original (identity) p2m entries
> + * are not restored since this region is now known not to
> + * conflict with any devices.
> + */
> + if (!xen_feature(XENFEAT_auto_translated_physmap)) {
> + xen_pfn_t pfn = PFN_DOWN(pgmap->res.start);
> +
> + for (i = 0; i < alloc_pages; i++) {
> + if (!set_phys_to_machine(pfn + i, INVALID_P2M_ENTRY)) {
> + pr_warn("set_phys_to_machine() failed, no memory added\n");
> + release_resource(&pgmap->res);
> + kfree(pgmap);
> + return -ENOMEM;
> + }
> + }
> + }
> +#endif
> +
> + vaddr = memremap_pages(pgmap, nid);
> + if (IS_ERR(vaddr)) {
> + pr_err("Cannot remap memory range\n");
> + release_resource(&pgmap->res);
> + kfree(pgmap);
> + return PTR_ERR(vaddr);
> + }
> +
> + for (i = 0; i < alloc_pages; i++) {
> + struct page *pg = virt_to_page(vaddr + PAGE_SIZE * i);
> +
> + BUG_ON(!virt_addr_valid(vaddr + PAGE_SIZE * i));
> + list_add(&pg->lru, &list);
> + count++;
> + }
> +
> + return 0;
> +}
> +
> +/**
> + * xen_alloc_unpopulated_pages - alloc unpopulated pages
> + * @nr_pages: Number of pages
> + * @pages: pages returned
> + * @return 0 on success, error otherwise
> + */
> +int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages)
> +{
> + unsigned int i;
> + int ret = 0;
> +
> + mutex_lock(&lock);
> + if (count < nr_pages) {
> + ret = fill(nr_pages);
> + if (ret)
> + goto out;
> + }
> +
> + for (i = 0; i < nr_pages; i++) {
> + struct page *pg = list_first_entry_or_null(&list, struct page,
> + lru);
> +
> + BUG_ON(!pg);
> + list_del(&pg->lru);
> + count--;
> + pages[i] = pg;
> +
> +#ifdef CONFIG_XEN_HAVE_PVMMU
> + /*
> + * We don't support PV MMU when Linux and Xen is using
> + * different page granularity.
> + */
> + BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
> +
> + if (!xen_feature(XENFEAT_auto_translated_physmap)) {
> + ret = xen_alloc_p2m_entry(page_to_pfn(pg));
> + if (ret < 0) {
> + unsigned int j;
> +
> + for (j = 0; j <= i; j++) {
> + list_add(&pages[j]->lru, &list);
> + count++;
> + }
> + goto out;
> + }
> + }
> +#endif
> + }
> +
> +out:
> + mutex_unlock(&lock);
> + return ret;
> +}
> +EXPORT_SYMBOL(xen_alloc_unpopulated_pages);
> +
> +/**
> + * xen_free_unpopulated_pages - return unpopulated pages
> + * @nr_pages: Number of pages
> + * @pages: pages to return
> + */
> +void xen_free_unpopulated_pages(unsigned int nr_pages, struct page **pages)
> +{
> + unsigned int i;
> +
> + mutex_lock(&lock);
> + for (i = 0; i < nr_pages; i++) {
> + list_add(&pages[i]->lru, &list);
> + count++;
> + }
> + mutex_unlock(&lock);
> +}
> +EXPORT_SYMBOL(xen_free_unpopulated_pages);
> +
> +#ifdef CONFIG_XEN_PV
> +static int __init init(void)
> +{
> + unsigned int i;
> +
> + if (!xen_domain())
> + return -ENODEV;
> +
> + /*
> + * Initialize with pages from the extra memory regions (see
> + * arch/x86/xen/setup.c).
> + */
> + for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) {
> + unsigned int j;
> +
> + for (j = 0; j < xen_extra_mem[i].n_pfns; j++) {
> + struct page *pg =
> + pfn_to_page(xen_extra_mem[i].start_pfn + j);
> +
> + list_add(&pg->lru, &list);
> + count++;
> + }
> + }
> +
> + return 0;
> +}
> +subsys_initcall(init);
> +#endif
> diff --git a/drivers/xen/xenbus/xenbus_client.c b/drivers/xen/xenbus/xenbus_client.c
> index 786fbb7d8be0..70b6c4780fbd 100644
> --- a/drivers/xen/xenbus/xenbus_client.c
> +++ b/drivers/xen/xenbus/xenbus_client.c
> @@ -615,7 +615,7 @@ static int xenbus_map_ring_hvm(struct xenbus_device *dev,
> bool leaked = false;
> unsigned int nr_pages = XENBUS_PAGES(nr_grefs);
>
> - err = alloc_xenballooned_pages(nr_pages, node->hvm.pages);
> + err = xen_alloc_unpopulated_pages(nr_pages, node->hvm.pages);
> if (err)
> goto out_err;
>
> @@ -656,7 +656,7 @@ static int xenbus_map_ring_hvm(struct xenbus_device *dev,
> addr, nr_pages);
> out_free_ballooned_pages:
> if (!leaked)
> - free_xenballooned_pages(nr_pages, node->hvm.pages);
> + xen_free_unpopulated_pages(nr_pages, node->hvm.pages);
> out_err:
> return err;
> }
> @@ -852,7 +852,7 @@ static int xenbus_unmap_ring_hvm(struct xenbus_device *dev, void *vaddr)
> info.addrs);
> if (!rv) {
> vunmap(vaddr);
> - free_xenballooned_pages(nr_pages, node->hvm.pages);
> + xen_free_unpopulated_pages(nr_pages, node->hvm.pages);
> }
> else
> WARN(1, "Leaking %p, size %u page(s)\n", vaddr, nr_pages);
> diff --git a/drivers/xen/xlate_mmu.c b/drivers/xen/xlate_mmu.c
> index 7b1077f0abcb..34742c6e189e 100644
> --- a/drivers/xen/xlate_mmu.c
> +++ b/drivers/xen/xlate_mmu.c
> @@ -232,7 +232,7 @@ int __init xen_xlate_map_ballooned_pages(xen_pfn_t **gfns, void **virt,
> kfree(pages);
> return -ENOMEM;
> }
> - rc = alloc_xenballooned_pages(nr_pages, pages);
> + rc = xen_alloc_unpopulated_pages(nr_pages, pages);
> if (rc) {
> pr_warn("%s Couldn't balloon alloc %ld pages rc:%d\n", __func__,
> nr_pages, rc);
> @@ -249,7 +249,7 @@ int __init xen_xlate_map_ballooned_pages(xen_pfn_t **gfns, void **virt,
> if (!vaddr) {
> pr_warn("%s Couldn't map %ld pages rc:%d\n", __func__,
> nr_pages, rc);
> - free_xenballooned_pages(nr_pages, pages);
> + xen_free_unpopulated_pages(nr_pages, pages);
> kfree(pages);
> kfree(pfns);
> return -ENOMEM;
> diff --git a/include/xen/xen.h b/include/xen/xen.h
> index 19a72f591e2b..aa33bc0d933c 100644
> --- a/include/xen/xen.h
> +++ b/include/xen/xen.h
> @@ -52,4 +52,12 @@ bool xen_biovec_phys_mergeable(const struct bio_vec *vec1,
> extern u64 xen_saved_max_mem_size;
> #endif
>
> +#ifdef CONFIG_ZONE_DEVICE
> +int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages);
> +void xen_free_unpopulated_pages(unsigned int nr_pages, struct page **pages);
> +#else
> +#define xen_alloc_unpopulated_pages alloc_xenballooned_pages
> +#define xen_free_unpopulated_pages free_xenballooned_pages
> +#endif
> +
> #endif /* _XEN_XEN_H */
>
--
Thanks,
David / dhildenb
Hi Roger,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on xen-tip/linux-next]
[also build test ERROR on linus/master v5.8-rc6 next-20200724]
[cannot apply to linux/master]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]
url: https://github.com/0day-ci/linux/commits/Roger-Pau-Monne/xen-balloon-fixes-for-memory-hotplug/20200724-204452
base: https://git.kernel.org/pub/scm/linux/kernel/git/xen/tip.git linux-next
config: i386-debian-10.3 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-14) 9.3.0
reproduce (this is a W=1 build):
# save the attached .config to linux build tree
make W=1 ARCH=i386
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <[email protected]>
All errors (new ones prefixed by >>):
drivers/gpu/drm/xen/xen_drm_front_gem.c: In function 'gem_create':
>> drivers/gpu/drm/xen/xen_drm_front_gem.c:102:9: error: implicit declaration of function 'xen_alloc_unpopulated_pages' [-Werror=implicit-function-declaration]
102 | ret = xen_alloc_unpopulated_pages(xen_obj->num_pages,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/xen/xen_drm_front_gem.c: In function 'xen_drm_front_gem_free_object_unlocked':
>> drivers/gpu/drm/xen/xen_drm_front_gem.c:155:5: error: implicit declaration of function 'xen_free_unpopulated_pages' [-Werror=implicit-function-declaration]
155 | xen_free_unpopulated_pages(xen_obj->num_pages,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: some warnings being treated as errors
vim +/xen_alloc_unpopulated_pages +102 drivers/gpu/drm/xen/xen_drm_front_gem.c
77
78 static struct xen_gem_object *gem_create(struct drm_device *dev, size_t size)
79 {
80 struct xen_drm_front_drm_info *drm_info = dev->dev_private;
81 struct xen_gem_object *xen_obj;
82 int ret;
83
84 size = round_up(size, PAGE_SIZE);
85 xen_obj = gem_create_obj(dev, size);
86 if (IS_ERR_OR_NULL(xen_obj))
87 return xen_obj;
88
89 if (drm_info->front_info->cfg.be_alloc) {
90 /*
91 * backend will allocate space for this buffer, so
92 * only allocate array of pointers to pages
93 */
94 ret = gem_alloc_pages_array(xen_obj, size);
95 if (ret < 0)
96 goto fail;
97
98 /*
99 * allocate ballooned pages which will be used to map
100 * grant references provided by the backend
101 */
> 102 ret = xen_alloc_unpopulated_pages(xen_obj->num_pages,
103 xen_obj->pages);
104 if (ret < 0) {
105 DRM_ERROR("Cannot allocate %zu ballooned pages: %d\n",
106 xen_obj->num_pages, ret);
107 gem_free_pages_array(xen_obj);
108 goto fail;
109 }
110
111 xen_obj->be_alloc = true;
112 return xen_obj;
113 }
114 /*
115 * need to allocate backing pages now, so we can share those
116 * with the backend
117 */
118 xen_obj->num_pages = DIV_ROUND_UP(size, PAGE_SIZE);
119 xen_obj->pages = drm_gem_get_pages(&xen_obj->base);
120 if (IS_ERR_OR_NULL(xen_obj->pages)) {
121 ret = PTR_ERR(xen_obj->pages);
122 xen_obj->pages = NULL;
123 goto fail;
124 }
125
126 return xen_obj;
127
128 fail:
129 DRM_ERROR("Failed to allocate buffer with size %zu\n", size);
130 return ERR_PTR(ret);
131 }
132
133 struct drm_gem_object *xen_drm_front_gem_create(struct drm_device *dev,
134 size_t size)
135 {
136 struct xen_gem_object *xen_obj;
137
138 xen_obj = gem_create(dev, size);
139 if (IS_ERR_OR_NULL(xen_obj))
140 return ERR_CAST(xen_obj);
141
142 return &xen_obj->base;
143 }
144
145 void xen_drm_front_gem_free_object_unlocked(struct drm_gem_object *gem_obj)
146 {
147 struct xen_gem_object *xen_obj = to_xen_gem_obj(gem_obj);
148
149 if (xen_obj->base.import_attach) {
150 drm_prime_gem_destroy(&xen_obj->base, xen_obj->sgt_imported);
151 gem_free_pages_array(xen_obj);
152 } else {
153 if (xen_obj->pages) {
154 if (xen_obj->be_alloc) {
> 155 xen_free_unpopulated_pages(xen_obj->num_pages,
156 xen_obj->pages);
157 gem_free_pages_array(xen_obj);
158 } else {
159 drm_gem_put_pages(&xen_obj->base,
160 xen_obj->pages, true, false);
161 }
162 }
163 }
164 drm_gem_object_release(gem_obj);
165 kfree(xen_obj);
166 }
167
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/[email protected]
On 7/24/20 10:34 AM, David Hildenbrand wrote:
> CCing Dan
>
> On 24.07.20 14:42, Roger Pau Monne wrote:
>> diff --git a/drivers/xen/unpopulated-alloc.c b/drivers/xen/unpopulated-alloc.c
>> new file mode 100644
>> index 000000000000..aaa91cefbbf9
>> --- /dev/null
>> +++ b/drivers/xen/unpopulated-alloc.c
>> @@ -0,0 +1,222 @@
>> + */
>> +
>> +#include <linux/errno.h>
>> +#include <linux/gfp.h>
>> +#include <linux/kernel.h>
>> +#include <linux/mm.h>
>> +#include <linux/memremap.h>
>> +#include <linux/slab.h>
>> +
>> +#include <asm/page.h>
>> +
>> +#include <xen/page.h>
>> +#include <xen/xen.h>
>> +
>> +static DEFINE_MUTEX(lock);
>> +static LIST_HEAD(list);
>> +static unsigned int count;
>> +
>> +static int fill(unsigned int nr_pages)
Less generic names? How about list_lock, pg_list, pg_count,
fill_pglist()? (But these are bad too, so maybe you can come up with
something better)
>> +{
>> + struct dev_pagemap *pgmap;
>> + void *vaddr;
>> + unsigned int i, alloc_pages = round_up(nr_pages, PAGES_PER_SECTION);
>> + int nid, ret;
>> +
>> + pgmap = kzalloc(sizeof(*pgmap), GFP_KERNEL);
>> + if (!pgmap)
>> + return -ENOMEM;
>> +
>> + pgmap->type = MEMORY_DEVICE_DEVDAX;
>> + pgmap->res.name = "XEN SCRATCH";
Typically iomem resources only capitalize first letters.
>> + pgmap->res.flags = IORESOURCE_MEM | IORESOURCE_BUSY;
>> +
>> + ret = allocate_resource(&iomem_resource, &pgmap->res,
>> + alloc_pages * PAGE_SIZE, 0, -1,
>> + PAGES_PER_SECTION * PAGE_SIZE, NULL, NULL);
Are we not going to end up with a whole bunch of "Xen scratch" resource
ranges for each miss in the page list? Or do we expect them to get merged?
>> + if (ret < 0) {
>> + pr_err("Cannot allocate new IOMEM resource\n");
>> + kfree(pgmap);
>> + return ret;
>> + }
>> +
>> + nid = memory_add_physaddr_to_nid(pgmap->res.start);
Should we consider page range crossing node boundaries?
>> +
>> +#ifdef CONFIG_XEN_HAVE_PVMMU
>> + /*
>> + * We don't support PV MMU when Linux and Xen is using
>> + * different page granularity.
>> + */
>> + BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
>> +
>> + /*
>> + * memremap will build page tables for the new memory so
>> + * the p2m must contain invalid entries so the correct
>> + * non-present PTEs will be written.
>> + *
>> + * If a failure occurs, the original (identity) p2m entries
>> + * are not restored since this region is now known not to
>> + * conflict with any devices.
>> + */
>> + if (!xen_feature(XENFEAT_auto_translated_physmap)) {
>> + xen_pfn_t pfn = PFN_DOWN(pgmap->res.start);
>> +
>> + for (i = 0; i < alloc_pages; i++) {
>> + if (!set_phys_to_machine(pfn + i, INVALID_P2M_ENTRY)) {
>> + pr_warn("set_phys_to_machine() failed, no memory added\n");
>> + release_resource(&pgmap->res);
>> + kfree(pgmap);
>> + return -ENOMEM;
>> + }
>> + }
>> + }
>> +#endif
>> +
>> + vaddr = memremap_pages(pgmap, nid);
>> + if (IS_ERR(vaddr)) {
>> + pr_err("Cannot remap memory range\n");
>> + release_resource(&pgmap->res);
>> + kfree(pgmap);
>> + return PTR_ERR(vaddr);
>> + }
>> +
>> + for (i = 0; i < alloc_pages; i++) {
>> + struct page *pg = virt_to_page(vaddr + PAGE_SIZE * i);
>> +
>> + BUG_ON(!virt_addr_valid(vaddr + PAGE_SIZE * i));
>> + list_add(&pg->lru, &list);
>> + count++;
>> + }
>> +
>> + return 0;
>> +}
>> +
>> +/**
>> + * xen_alloc_unpopulated_pages - alloc unpopulated pages
>> + * @nr_pages: Number of pages
>> + * @pages: pages returned
>> + * @return 0 on success, error otherwise
>> + */
>> +int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages)
>> +{
>> + unsigned int i;
>> + int ret = 0;
>> +
>> + mutex_lock(&lock);
>> + if (count < nr_pages) {
>> + ret = fill(nr_pages);
(nr_pages - count) ?
>> + if (ret)
>> + goto out;
>> + }
>> +
>> + for (i = 0; i < nr_pages; i++) {
>> + struct page *pg = list_first_entry_or_null(&list, struct page,
>> + lru);
>> +
>> + BUG_ON(!pg);
>> + list_del(&pg->lru);
>> + count--;
>> + pages[i] = pg;
>> +
>> +#ifdef CONFIG_XEN_HAVE_PVMMU
>> + /*
>> + * We don't support PV MMU when Linux and Xen is using
>> + * different page granularity.
>> + */
>> + BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
>> +
>> + if (!xen_feature(XENFEAT_auto_translated_physmap)) {
>> + ret = xen_alloc_p2m_entry(page_to_pfn(pg));
>> + if (ret < 0) {
>> + unsigned int j;
>> +
>> + for (j = 0; j <= i; j++) {
>> + list_add(&pages[j]->lru, &list);
>> + count++;
>> + }
>> + goto out;
>> + }
>> + }
>> +#endif
>> + }
>> +
>> +out:
>> + mutex_unlock(&lock);
>> + return ret;
>> +}
>> +EXPORT_SYMBOL(xen_alloc_unpopulated_pages);
>> +
>> +
>> +#ifdef CONFIG_XEN_PV
>> +static int __init init(void)
>> +{
>> + unsigned int i;
>> +
>> + if (!xen_domain())
>> + return -ENODEV;
>> +
>> + /*
>> + * Initialize with pages from the extra memory regions (see
>> + * arch/x86/xen/setup.c).
>> + */
This loop will be executing only for PV guests so we can just bail out
for non-PV guests here.
-boris
>> + for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) {
>> + unsigned int j;
>> +
>> + for (j = 0; j < xen_extra_mem[i].n_pfns; j++) {
>> + struct page *pg =
>> + pfn_to_page(xen_extra_mem[i].start_pfn + j);
>> +
>> + list_add(&pg->lru, &list);
>> + count++;
>> + }
>> + }
>> +
>> + return 0;
>> +}
>> +subsys_initcall(init);
Hi Roger,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on xen-tip/linux-next]
[also build test ERROR on linus/master v5.8-rc6 next-20200724]
[cannot apply to linux/master]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]
url: https://github.com/0day-ci/linux/commits/Roger-Pau-Monne/xen-balloon-fixes-for-memory-hotplug/20200724-204452
base: https://git.kernel.org/pub/scm/linux/kernel/git/xen/tip.git linux-next
config: x86_64-allyesconfig (attached as .config)
compiler: clang version 12.0.0 (https://github.com/llvm/llvm-project 1d09ecf36175f7910ffedd6d497c07b5c74c22fb)
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# install x86_64 cross compiling tool for clang build
# apt-get install binutils-x86-64-linux-gnu
# save the attached .config to linux build tree
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross ARCH=x86_64
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <[email protected]>
All errors (new ones prefixed by >>):
>> drivers/gpu/drm/xen/xen_drm_front_gem.c:102:9: error: implicit declaration of function 'xen_alloc_unpopulated_pages' [-Werror,-Wimplicit-function-declaration]
ret = xen_alloc_unpopulated_pages(xen_obj->num_pages,
^
>> drivers/gpu/drm/xen/xen_drm_front_gem.c:155:5: error: implicit declaration of function 'xen_free_unpopulated_pages' [-Werror,-Wimplicit-function-declaration]
xen_free_unpopulated_pages(xen_obj->num_pages,
^
2 errors generated.
vim +/xen_alloc_unpopulated_pages +102 drivers/gpu/drm/xen/xen_drm_front_gem.c
77
78 static struct xen_gem_object *gem_create(struct drm_device *dev, size_t size)
79 {
80 struct xen_drm_front_drm_info *drm_info = dev->dev_private;
81 struct xen_gem_object *xen_obj;
82 int ret;
83
84 size = round_up(size, PAGE_SIZE);
85 xen_obj = gem_create_obj(dev, size);
86 if (IS_ERR_OR_NULL(xen_obj))
87 return xen_obj;
88
89 if (drm_info->front_info->cfg.be_alloc) {
90 /*
91 * backend will allocate space for this buffer, so
92 * only allocate array of pointers to pages
93 */
94 ret = gem_alloc_pages_array(xen_obj, size);
95 if (ret < 0)
96 goto fail;
97
98 /*
99 * allocate ballooned pages which will be used to map
100 * grant references provided by the backend
101 */
> 102 ret = xen_alloc_unpopulated_pages(xen_obj->num_pages,
103 xen_obj->pages);
104 if (ret < 0) {
105 DRM_ERROR("Cannot allocate %zu ballooned pages: %d\n",
106 xen_obj->num_pages, ret);
107 gem_free_pages_array(xen_obj);
108 goto fail;
109 }
110
111 xen_obj->be_alloc = true;
112 return xen_obj;
113 }
114 /*
115 * need to allocate backing pages now, so we can share those
116 * with the backend
117 */
118 xen_obj->num_pages = DIV_ROUND_UP(size, PAGE_SIZE);
119 xen_obj->pages = drm_gem_get_pages(&xen_obj->base);
120 if (IS_ERR_OR_NULL(xen_obj->pages)) {
121 ret = PTR_ERR(xen_obj->pages);
122 xen_obj->pages = NULL;
123 goto fail;
124 }
125
126 return xen_obj;
127
128 fail:
129 DRM_ERROR("Failed to allocate buffer with size %zu\n", size);
130 return ERR_PTR(ret);
131 }
132
133 struct drm_gem_object *xen_drm_front_gem_create(struct drm_device *dev,
134 size_t size)
135 {
136 struct xen_gem_object *xen_obj;
137
138 xen_obj = gem_create(dev, size);
139 if (IS_ERR_OR_NULL(xen_obj))
140 return ERR_CAST(xen_obj);
141
142 return &xen_obj->base;
143 }
144
145 void xen_drm_front_gem_free_object_unlocked(struct drm_gem_object *gem_obj)
146 {
147 struct xen_gem_object *xen_obj = to_xen_gem_obj(gem_obj);
148
149 if (xen_obj->base.import_attach) {
150 drm_prime_gem_destroy(&xen_obj->base, xen_obj->sgt_imported);
151 gem_free_pages_array(xen_obj);
152 } else {
153 if (xen_obj->pages) {
154 if (xen_obj->be_alloc) {
> 155 xen_free_unpopulated_pages(xen_obj->num_pages,
156 xen_obj->pages);
157 gem_free_pages_array(xen_obj);
158 } else {
159 drm_gem_put_pages(&xen_obj->base,
160 xen_obj->pages, true, false);
161 }
162 }
163 }
164 drm_gem_object_release(gem_obj);
165 kfree(xen_obj);
166 }
167
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/[email protected]
On 24.07.20 18:36, Boris Ostrovsky wrote:
> On 7/24/20 10:34 AM, David Hildenbrand wrote:
>> CCing Dan
>>
>> On 24.07.20 14:42, Roger Pau Monne wrote:
>>> diff --git a/drivers/xen/unpopulated-alloc.c b/drivers/xen/unpopulated-alloc.c
>>> new file mode 100644
>>> index 000000000000..aaa91cefbbf9
>>> --- /dev/null
>>> +++ b/drivers/xen/unpopulated-alloc.c
>>> @@ -0,0 +1,222 @@
>
>
>
>>> + */
>>> +
>>> +#include <linux/errno.h>
>>> +#include <linux/gfp.h>
>>> +#include <linux/kernel.h>
>>> +#include <linux/mm.h>
>>> +#include <linux/memremap.h>
>>> +#include <linux/slab.h>
>>> +
>>> +#include <asm/page.h>
>>> +
>>> +#include <xen/page.h>
>>> +#include <xen/xen.h>
>>> +
>>> +static DEFINE_MUTEX(lock);
>>> +static LIST_HEAD(list);
>>> +static unsigned int count;
>>> +
>>> +static int fill(unsigned int nr_pages)
>
>
> Less generic names? How about list_lock, pg_list, pg_count,
> fill_pglist()? (But these are bad too, so maybe you can come up with
> something better)
>
>
>>> +{
>>> + struct dev_pagemap *pgmap;
>>> + void *vaddr;
>>> + unsigned int i, alloc_pages = round_up(nr_pages, PAGES_PER_SECTION);
>>> + int nid, ret;
>>> +
>>> + pgmap = kzalloc(sizeof(*pgmap), GFP_KERNEL);
>>> + if (!pgmap)
>>> + return -ENOMEM;
>>> +
>>> + pgmap->type = MEMORY_DEVICE_DEVDAX;
>>> + pgmap->res.name = "XEN SCRATCH";
>
>
> Typically iomem resources only capitalize first letters.
>
>
>>> + pgmap->res.flags = IORESOURCE_MEM | IORESOURCE_BUSY;
>>> +
>>> + ret = allocate_resource(&iomem_resource, &pgmap->res,
>>> + alloc_pages * PAGE_SIZE, 0, -1,
>>> + PAGES_PER_SECTION * PAGE_SIZE, NULL, NULL);
>
>
> Are we not going to end up with a whole bunch of "Xen scratch" resource
> ranges for each miss in the page list? Or do we expect them to get merged?
>
AFAIK, no resources will get merged (and it's in the general case not
safe to do). The old approach (add_memory_resource()) will end up with
the same situation ("Xen Scratch" vs. "System RAM") one new resource per
added memory block/section.
FWIW, I am looking into merging selected resources in the context of
virtio-mem _after_ adding succeeded (not directly when adding the
resource to the tree). Interface might look something like
void merge_child_mem_resources(struct resource *parent, const char *name);
So I can, for example, trigger merging of all "System RAM (virtio_mem)"
resources, that are located under a device node (e.g., "virtio0").
I also thought about tagging each mergeable resource via something like
"IORESOURCE_MERGEABLE" - whereby the user agrees that it does not hold
any pointers to such a resource. But I don't see yet a copelling reason
to sacrifice space for a new flag.
So with this in place, this code could call once adding succeeded
merge_child_mem_resources(&iomem_resource, "Xen Scratch");
--
Thanks,
David / dhildenb
On Fri, Jul 24, 2020 at 12:36:33PM -0400, Boris Ostrovsky wrote:
> On 7/24/20 10:34 AM, David Hildenbrand wrote:
> > CCing Dan
> >
> > On 24.07.20 14:42, Roger Pau Monne wrote:
> >> +
> >> +#include <linux/errno.h>
> >> +#include <linux/gfp.h>
> >> +#include <linux/kernel.h>
> >> +#include <linux/mm.h>
> >> +#include <linux/memremap.h>
> >> +#include <linux/slab.h>
> >> +
> >> +#include <asm/page.h>
> >> +
> >> +#include <xen/page.h>
> >> +#include <xen/xen.h>
> >> +
> >> +static DEFINE_MUTEX(lock);
> >> +static LIST_HEAD(list);
> >> +static unsigned int count;
> >> +
> >> +static int fill(unsigned int nr_pages)
>
>
> Less generic names? How about list_lock, pg_list, pg_count,
> fill_pglist()? (But these are bad too, so maybe you can come up with
> something better)
OK, I have to admit I like using such short names when the code allows
to, for example this code is so simple that it didn't seem to warrant
using longer names. Will rename on next version.
> >> +{
> >> + struct dev_pagemap *pgmap;
> >> + void *vaddr;
> >> + unsigned int i, alloc_pages = round_up(nr_pages, PAGES_PER_SECTION);
> >> + int nid, ret;
> >> +
> >> + pgmap = kzalloc(sizeof(*pgmap), GFP_KERNEL);
> >> + if (!pgmap)
> >> + return -ENOMEM;
> >> +
> >> + pgmap->type = MEMORY_DEVICE_DEVDAX;
> >> + pgmap->res.name = "XEN SCRATCH";
>
>
> Typically iomem resources only capitalize first letters.
>
>
> >> + pgmap->res.flags = IORESOURCE_MEM | IORESOURCE_BUSY;
> >> +
> >> + ret = allocate_resource(&iomem_resource, &pgmap->res,
> >> + alloc_pages * PAGE_SIZE, 0, -1,
> >> + PAGES_PER_SECTION * PAGE_SIZE, NULL, NULL);
>
>
> Are we not going to end up with a whole bunch of "Xen scratch" resource
> ranges for each miss in the page list? Or do we expect them to get merged?
PAGES_PER_SECTION is IMO big enough to prevent ending up with a lot of
separated ranges. I think the value is 32 or 64MiB on x86, so while we
are likely to end up with more than one resource added, I don't think
it's going to be massive.
>
> >> + if (ret < 0) {
> >> + pr_err("Cannot allocate new IOMEM resource\n");
> >> + kfree(pgmap);
> >> + return ret;
> >> + }
> >> +
> >> + nid = memory_add_physaddr_to_nid(pgmap->res.start);
>
>
> Should we consider page range crossing node boundaries?
I'm not sure whether this is possible (I would think allocate_resource
should return a range from a single node), but then it would greatly
complicate the code to perform the memremap_pages, as we would have to
split the region into multiple dev_pagemap structs.
FWIW the current code in the balloon driver does exactly the same
(which doesn't mean it's correct, but that's where I got the logic
from).
> >> +
> >> +#ifdef CONFIG_XEN_HAVE_PVMMU
> >> + /*
> >> + * We don't support PV MMU when Linux and Xen is using
> >> + * different page granularity.
> >> + */
> >> + BUILD_BUG_ON(XEN_PAGE_SIZE != PAGE_SIZE);
> >> +
> >> + /*
> >> + * memremap will build page tables for the new memory so
> >> + * the p2m must contain invalid entries so the correct
> >> + * non-present PTEs will be written.
> >> + *
> >> + * If a failure occurs, the original (identity) p2m entries
> >> + * are not restored since this region is now known not to
> >> + * conflict with any devices.
> >> + */
> >> + if (!xen_feature(XENFEAT_auto_translated_physmap)) {
> >> + xen_pfn_t pfn = PFN_DOWN(pgmap->res.start);
> >> +
> >> + for (i = 0; i < alloc_pages; i++) {
> >> + if (!set_phys_to_machine(pfn + i, INVALID_P2M_ENTRY)) {
> >> + pr_warn("set_phys_to_machine() failed, no memory added\n");
> >> + release_resource(&pgmap->res);
> >> + kfree(pgmap);
> >> + return -ENOMEM;
> >> + }
> >> + }
> >> + }
> >> +#endif
> >> +
> >> + vaddr = memremap_pages(pgmap, nid);
> >> + if (IS_ERR(vaddr)) {
> >> + pr_err("Cannot remap memory range\n");
> >> + release_resource(&pgmap->res);
> >> + kfree(pgmap);
> >> + return PTR_ERR(vaddr);
> >> + }
> >> +
> >> + for (i = 0; i < alloc_pages; i++) {
> >> + struct page *pg = virt_to_page(vaddr + PAGE_SIZE * i);
> >> +
> >> + BUG_ON(!virt_addr_valid(vaddr + PAGE_SIZE * i));
> >> + list_add(&pg->lru, &list);
> >> + count++;
> >> + }
> >> +
> >> + return 0;
> >> +}
> >> +
> >> +/**
> >> + * xen_alloc_unpopulated_pages - alloc unpopulated pages
> >> + * @nr_pages: Number of pages
> >> + * @pages: pages returned
> >> + * @return 0 on success, error otherwise
> >> + */
> >> +int xen_alloc_unpopulated_pages(unsigned int nr_pages, struct page **pages)
> >> +{
> >> + unsigned int i;
> >> + int ret = 0;
> >> +
> >> + mutex_lock(&lock);
> >> + if (count < nr_pages) {
> >> + ret = fill(nr_pages);
>
>
> (nr_pages - count) ?
Yup, already fixed as Juergen also pointed it out.
> >> +
> >> +#ifdef CONFIG_XEN_PV
> >> +static int __init init(void)
> >> +{
> >> + unsigned int i;
> >> +
> >> + if (!xen_domain())
> >> + return -ENODEV;
> >> +
> >> + /*
> >> + * Initialize with pages from the extra memory regions (see
> >> + * arch/x86/xen/setup.c).
> >> + */
>
>
> This loop will be executing only for PV guests so we can just bail out
> for non-PV guests here.
Sure.
Thanks, Roger.