forge/linux-uconsole
1
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

Revert "BACKPORT: FROMGIT: userfaultfd: support minor fault handling for shmem"

Browse source 
This reverts commit d672123ec4 as an
updated version of the patch-set will be merged later.

Signed-off-by: Lokesh Gidra <lokeshgidra@google.com>
Bug: 187930641
Change-Id: I765fe86a2dc0305482a0590c14143dee27840b8a
This commit is contained in:
Lokesh Gidra 2021-05-13 05:11:10 -07:00 • committed by Suren Baghdasaryan
commit 84330a5f50
6 changed files with 84 additions and 81 deletions
Download patch file Download diff file Expand all files Collapse all files

6
fs/userfaultfd.c
View file

@ -1273,7 +1273,8 @@ static inline bool vma_can_userfault(struct vm_area_struct *vma,
}
if (vm_flags & VM_UFFD_MINOR) {
if (!(is_vm_hugetlb_page(vma) || vma_is_shmem(vma)))
/* FIXME: Add minor fault interception for shmem. */
if (!is_vm_hugetlb_page(vma))
return false;
}
@ -1952,8 +1953,7 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
/* report all available features and ioctls to userland */
uffdio_api.features = UFFD_API_FEATURES;
#ifndef CONFIG_HAVE_ARCH_USERFAULTFD_MINOR
uffdio_api.features &=
~(UFFD_FEATURE_MINOR_HUGETLBFS | UFFD_FEATURE_MINOR_SHMEM);
uffdio_api.features &= ~UFFD_FEATURE_MINOR_HUGETLBFS;
#endif
uffdio_api.ioctls = UFFD_API_IOCTLS;
ret = -EFAULT;

26
include/linux/shmem_fs.h
View file

@ -9,7 +9,6 @@
#include <linux/percpu_counter.h>
#include <linux/xattr.h>
#include <linux/fs_parser.h>
#include <linux/userfaultfd_k.h>
/* inode in-kernel data */
@ -119,16 +118,21 @@ static inline bool shmem_file(struct file *file)
extern bool shmem_charge(struct inode *inode, long pages);
extern void shmem_uncharge(struct inode *inode, long pages);
#ifdef CONFIG_USERFAULTFD
#ifdef CONFIG_SHMEM
int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd,
struct vm_area_struct *dst_vma,
unsigned long dst_addr, unsigned long src_addr,
enum mcopy_atomic_mode mode, struct page **pagep);
#else /* !CONFIG_SHMEM */
#define shmem_mcopy_atomic_pte(dst_mm, dst_pmd, dst_vma, dst_addr, \
src_addr, mode, pagep) ({ BUG(); 0; })
#endif /* CONFIG_SHMEM */
#endif /* CONFIG_USERFAULTFD */
extern int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd,
struct vm_area_struct *dst_vma,
unsigned long dst_addr,
unsigned long src_addr,
struct page **pagep);
extern int shmem_mfill_zeropage_pte(struct mm_struct *dst_mm,
pmd_t *dst_pmd,
struct vm_area_struct *dst_vma,
unsigned long dst_addr);
#else
#define shmem_mcopy_atomic_pte(dst_mm, dst_pte, dst_vma, dst_addr, \
src_addr, pagep) ({ BUG(); 0; })
#define shmem_mfill_zeropage_pte(dst_mm, dst_pmd, dst_vma, \
dst_addr) ({ BUG(); 0; })
#endif
#endif

4
include/uapi/linux/userfaultfd.h
View file

@ -31,8 +31,7 @@
UFFD_FEATURE_MISSING_SHMEM | \
UFFD_FEATURE_SIGBUS | \
UFFD_FEATURE_THREAD_ID | \
UFFD_FEATURE_MINOR_HUGETLBFS | \
UFFD_FEATURE_MINOR_SHMEM)
UFFD_FEATURE_MINOR_HUGETLBFS)
#define UFFD_API_IOCTLS \
((__u64)1 << _UFFDIO_REGISTER | \
(__u64)1 << _UFFDIO_UNREGISTER | \
@ -197,7 +196,6 @@ struct uffdio_api {
#define UFFD_FEATURE_SIGBUS (1<<7)
#define UFFD_FEATURE_THREAD_ID (1<<8)
#define UFFD_FEATURE_MINOR_HUGETLBFS (1<<9)
#define UFFD_FEATURE_MINOR_SHMEM (1<<10)
__u64 features;
__u64 ioctls;

8
mm/memory.c
View file

@ -4244,11 +4244,9 @@ static vm_fault_t do_read_fault(struct vm_fault *vmf)
* something).
*/
if (vma->vm_ops->map_pages && fault_around_bytes >> PAGE_SHIFT > 1) {
if (likely(!userfaultfd_minor(vmf->vma))) {
ret = do_fault_around(vmf);
if (ret)
return ret;
}
ret = do_fault_around(vmf);
if (ret)
return ret;
}
ret = __do_fault(vmf);

94
mm/shmem.c
View file

@ -77,6 +77,7 @@ static struct vfsmount *shm_mnt;
#include <linux/syscalls.h>
#include <linux/fcntl.h>
#include <uapi/linux/memfd.h>
#include <linux/userfaultfd_k.h>
#include <linux/rmap.h>
#include <linux/uuid.h>
@ -1786,8 +1787,8 @@ unlock:
* vm. If we swap it in we mark it dirty since we also free the swap
* entry since a page cannot live in both the swap and page cache.
*
* vma, vmf, and fault_type are only supplied by shmem_fault: otherwise they
* are NULL.
* vmf and fault_type are only supplied by shmem_fault:
* otherwise they are NULL.
*/
static int shmem_getpage_gfp(struct inode *inode, pgoff_t index,
struct page **pagep, enum sgp_type sgp, gfp_t gfp,
@ -1829,12 +1830,6 @@ repeat:
return error;
}
if (page && vma && userfaultfd_minor(vma)) {
unlock_page(page);
*fault_type = handle_userfault(vmf, VM_UFFD_MINOR);
return 0;
}
if (page)
hindex = page->index;
if (page && sgp == SGP_WRITE)
@ -2348,13 +2343,14 @@ bool shmem_mapping(struct address_space *mapping)
return mapping->a_ops == &shmem_aops;
}
#ifdef CONFIG_USERFAULTFD
int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd,
struct vm_area_struct *dst_vma,
unsigned long dst_addr, unsigned long src_addr,
enum mcopy_atomic_mode mode, struct page **pagep)
static int shmem_mfill_atomic_pte(struct mm_struct *dst_mm,
pmd_t *dst_pmd,
struct vm_area_struct *dst_vma,
unsigned long dst_addr,
unsigned long src_addr,
bool zeropage,
struct page **pagep)
{
bool is_continue = (mode == MCOPY_ATOMIC_CONTINUE);
struct inode *inode = file_inode(dst_vma->vm_file);
struct shmem_inode_info *info = SHMEM_I(inode);
struct address_space *mapping = inode->i_mapping;
@ -2381,17 +2377,12 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd,
goto out;
}
if (is_continue) {
ret = -EFAULT;
page = find_lock_page(mapping, pgoff);
if (!page)
goto out_unacct_blocks;
} else if (!*pagep) {
if (!*pagep) {
page = shmem_alloc_page(gfp, info, pgoff);
if (!page)
goto out_unacct_blocks;
if (mode == MCOPY_ATOMIC_NORMAL) { /* mcopy_atomic */
if (!zeropage) { /* mcopy_atomic */
page_kaddr = kmap_atomic(page);
ret = copy_from_user(page_kaddr,
(const void __user *)src_addr,
@ -2405,7 +2396,7 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd,
/* don't free the page */
return -ENOENT;
}
} else { /* zeropage */
} else { /* mfill_zeropage_atomic */
clear_highpage(page);
}
} else {
@ -2413,13 +2404,10 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd,
*pagep = NULL;
}
if (!is_continue) {
VM_BUG_ON(PageSwapBacked(page));
VM_BUG_ON(PageLocked(page));
__SetPageLocked(page);
__SetPageSwapBacked(page);
__SetPageUptodate(page);
}
VM_BUG_ON(PageLocked(page) || PageSwapBacked(page));
__SetPageLocked(page);
__SetPageSwapBacked(page);
__SetPageUptodate(page);
ret = -EFAULT;
offset = linear_page_index(dst_vma, dst_addr);
@ -2427,13 +2415,10 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd,
if (unlikely(offset >= max_off))
goto out_release;
/* If page wasn't already in the page cache, add it. */
if (!is_continue) {
ret = shmem_add_to_page_cache(page, mapping, pgoff, NULL,
gfp & GFP_RECLAIM_MASK, dst_mm);
if (ret)
goto out_release;
}
ret = shmem_add_to_page_cache(page, mapping, pgoff, NULL,
gfp & GFP_RECLAIM_MASK, dst_mm);
if (ret)
goto out_release;
_dst_pte = mk_pte(page, dst_vma->vm_page_prot);
if (dst_vma->vm_flags & VM_WRITE)
@ -2460,15 +2445,13 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd,
if (!pte_none(*dst_pte))
goto out_release_unlock;
if (!is_continue) {
lru_cache_add(page);
lru_cache_add(page);
spin_lock_irq(&info->lock);
info->alloced++;
inode->i_blocks += BLOCKS_PER_PAGE;
shmem_recalc_inode(inode);
spin_unlock_irq(&info->lock);
}
spin_lock_irq(&info->lock);
info->alloced++;
inode->i_blocks += BLOCKS_PER_PAGE;
shmem_recalc_inode(inode);
spin_unlock_irq(&info->lock);
inc_mm_counter(dst_mm, mm_counter_file(page));
page_add_file_rmap(page, false);
@ -2492,7 +2475,28 @@ out_unacct_blocks:
shmem_inode_unacct_blocks(inode, 1);
goto out;
}
#endif /* CONFIG_USERFAULTFD */
int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm,
pmd_t *dst_pmd,
struct vm_area_struct *dst_vma,
unsigned long dst_addr,
unsigned long src_addr,
struct page **pagep)
{
return shmem_mfill_atomic_pte(dst_mm, dst_pmd, dst_vma,
dst_addr, src_addr, false, pagep);
}
int shmem_mfill_zeropage_pte(struct mm_struct *dst_mm,
pmd_t *dst_pmd,
struct vm_area_struct *dst_vma,
unsigned long dst_addr)
{
struct page *page = NULL;
return shmem_mfill_atomic_pte(dst_mm, dst_pmd, dst_vma,
dst_addr, 0, true, &page);
}
#ifdef CONFIG_TMPFS
static const struct inode_operations shmem_symlink_inode_operations;

27
mm/userfaultfd.c
View file

@ -415,7 +415,7 @@ static __always_inline ssize_t mfill_atomic_pte(struct mm_struct *dst_mm,
unsigned long dst_addr,
unsigned long src_addr,
struct page **page,
enum mcopy_atomic_mode mode,
bool zeropage,
bool wp_copy)
{
ssize_t err;
@ -431,24 +431,22 @@ static __always_inline ssize_t mfill_atomic_pte(struct mm_struct *dst_mm,
* and not in the radix tree.
*/
if (!(dst_vma->vm_flags & VM_SHARED)) {
switch (mode) {
case MCOPY_ATOMIC_NORMAL:
if (!zeropage)
err = mcopy_atomic_pte(dst_mm, dst_pmd, dst_vma,
dst_addr, src_addr, page,
wp_copy);
break;
case MCOPY_ATOMIC_ZEROPAGE:
else
err = mfill_zeropage_pte(dst_mm, dst_pmd,
dst_vma, dst_addr);
break;
case MCOPY_ATOMIC_CONTINUE:
err = -EINVAL;
break;
}
} else {
VM_WARN_ON_ONCE(wp_copy);
err = shmem_mcopy_atomic_pte(dst_mm, dst_pmd, dst_vma, dst_addr,
src_addr, mode, page);
if (!zeropage)
err = shmem_mcopy_atomic_pte(dst_mm, dst_pmd,
dst_vma, dst_addr,
src_addr, page);
else
err = shmem_mfill_zeropage_pte(dst_mm, dst_pmd,
dst_vma, dst_addr);
}
return err;
@ -469,6 +467,7 @@ static __always_inline ssize_t __mcopy_atomic(struct mm_struct *dst_mm,
long copied;
struct page *page;
bool wp_copy;
bool zeropage = (mcopy_mode == MCOPY_ATOMIC_ZEROPAGE);
/*
* Sanitize the command parameters:
@ -531,7 +530,7 @@ retry:
if (!vma_is_anonymous(dst_vma) && !vma_is_shmem(dst_vma))
goto out_unlock;
if (!vma_is_shmem(dst_vma) && mcopy_mode == MCOPY_ATOMIC_CONTINUE)
if (mcopy_mode == MCOPY_ATOMIC_CONTINUE)
goto out_unlock;
/*
@ -579,7 +578,7 @@ retry:
BUG_ON(pmd_trans_huge(*dst_pmd));
err = mfill_atomic_pte(dst_mm, dst_pmd, dst_vma, dst_addr,
src_addr, &page, mcopy_mode, wp_copy);
src_addr, &page, zeropage, wp_copy);
cond_resched();
if (unlikely(err == -ENOENT)) {