| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <20210724193449.361667-2-agruenba@redhat.com>
Date: Sat, 24 Jul 2021 21:34:42 +0200
From: Andreas Gruenbacher <agruenba@...hat.com>
To: Linus Torvalds <torvalds@...ux-foundation.org>,
Alexander Viro <viro@...iv.linux.org.uk>,
Christoph Hellwig <hch@...radead.org>,
"Darrick J. Wong" <djwong@...nel.org>
Cc: Jan Kara <jack@...e.cz>, Matthew Wilcox <willy@...radead.org>,
cluster-devel@...hat.com, linux-fsdevel@...r.kernel.org,
linux-kernel@...r.kernel.org, ocfs2-devel@....oracle.com,
Andreas Gruenbacher <agruenba@...hat.com>
Subject: [PATCH v4 1/8] iov_iter: Introduce iov_iter_fault_in_writeable helper
Introduce a new iov_iter_fault_in_writeable helper for faulting in an iterator
for writing. The pages are faulted in manually, without writing to them, so
this function is non-destructive.
We'll use iov_iter_fault_in_writeable in gfs2 once we've determined that part
of the iterator isn't in memory.
Signed-off-by: Andreas Gruenbacher <agruenba@...hat.com>
---
include/linux/mm.h | 3 +++
include/linux/uio.h | 1 +
lib/iov_iter.c | 40 +++++++++++++++++++++++++++++++
mm/gup.c | 57 +++++++++++++++++++++++++++++++++++++++++++++
4 files changed, 101 insertions(+)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7ca22e6e694a..14b1353995e2 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1840,6 +1840,9 @@ int get_user_pages_fast(unsigned long start, int nr_pages,
int pin_user_pages_fast(unsigned long start, int nr_pages,
unsigned int gup_flags, struct page **pages);
+unsigned long fault_in_user_pages(unsigned long start, unsigned long len,
+ bool write);
+
int account_locked_vm(struct mm_struct *mm, unsigned long pages, bool inc);
int __account_locked_vm(struct mm_struct *mm, unsigned long pages, bool inc,
struct task_struct *task, bool bypass_rlim);
diff --git a/include/linux/uio.h b/include/linux/uio.h
index 82c3c3e819e0..8e469b8b862f 100644
--- a/include/linux/uio.h
+++ b/include/linux/uio.h
@@ -120,6 +120,7 @@ size_t copy_page_from_iter_atomic(struct page *page, unsigned offset,
void iov_iter_advance(struct iov_iter *i, size_t bytes);
void iov_iter_revert(struct iov_iter *i, size_t bytes);
int iov_iter_fault_in_readable(const struct iov_iter *i, size_t bytes);
+int iov_iter_fault_in_writeable(const struct iov_iter *i, size_t bytes);
size_t iov_iter_single_seg_count(const struct iov_iter *i);
size_t copy_page_to_iter(struct page *page, size_t offset, size_t bytes,
struct iov_iter *i);
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index 20dc3d800573..ccf1ee8d4edf 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -460,6 +460,46 @@ int iov_iter_fault_in_readable(const struct iov_iter *i, size_t bytes)
}
EXPORT_SYMBOL(iov_iter_fault_in_readable);
+/**
+ * iov_iter_fault_in_writeable - fault in an iov iterator for writing
+ * @i: iterator
+ * @bytes: maximum length
+ *
+ * Faults in part or all of the iterator. This is primarily useful when we
+ * already know that some or all of the pages in @i aren't in memory.
+ *
+ * This function uses fault_in_user_pages() to fault in the pages, which
+ * internally uses get_user_pages(), so it doesn't trigger hardware page
+ * faults. Unlike fault_in_pages_writeable() which writes to the memory to
+ * fault it in, this function is non-destructive.
+ *
+ * Returns 0 on success, and a non-zero error code if the memory could not be
+ * accessed (i.e. because it is an invalid address).
+ */
+int iov_iter_fault_in_writeable(const struct iov_iter *i, size_t bytes)
+{
+ if (iter_is_iovec(i)) {
+ const struct iovec *p;
+ size_t skip;
+
+ if (bytes > i->count)
+ bytes = i->count;
+ for (p = i->iov, skip = i->iov_offset; bytes; p++, skip = 0) {
+ unsigned long len = min(bytes, p->iov_len - skip);
+ unsigned long start;
+
+ if (unlikely(!len))
+ continue;
+ start = (unsigned long)p->iov_base + skip;
+ if (fault_in_user_pages(start, len, true) != len)
+ return -EFAULT;
+ bytes -= len;
+ }
+ }
+ return 0;
+}
+EXPORT_SYMBOL(iov_iter_fault_in_writeable);
+
void iov_iter_init(struct iov_iter *i, unsigned int direction,
const struct iovec *iov, unsigned long nr_segs,
size_t count)
diff --git a/mm/gup.c b/mm/gup.c
index 42b8b1fa6521..784809c232f1 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -1669,6 +1669,63 @@ static long __get_user_pages_locked(struct mm_struct *mm, unsigned long start,
}
#endif /* !CONFIG_MMU */
+/**
+ * fault_in_user_pages - fault in an address range for reading / writing
+ * @start: start of address range
+ * @len: length of address range
+ * @write: fault in for writing
+ *
+ * Note that we don't pin or otherwise hold the pages referenced that we fault
+ * in. There's no guarantee that they'll stay in memory for any duration of
+ * time.
+ *
+ * Returns the number of bytes faulted in from @start.
+ */
+unsigned long fault_in_user_pages(unsigned long start, unsigned long len,
+ bool write)
+{
+ struct mm_struct *mm = current->mm;
+ struct vm_area_struct *vma = NULL;
+ unsigned long end, nstart, nend;
+ int locked = 0;
+ int gup_flags;
+
+ gup_flags = FOLL_TOUCH | FOLL_POPULATE;
+ if (write)
+ gup_flags |= FOLL_WRITE;
+
+ end = PAGE_ALIGN(start + len);
+ for (nstart = start & PAGE_MASK; nstart < end; nstart = nend) {
+ unsigned long nr_pages;
+ long ret;
+
+ if (!locked) {
+ locked = 1;
+ mmap_read_lock(mm);
+ vma = find_vma(mm, nstart);
+ } else if (nstart >= vma->vm_end)
+ vma = vma->vm_next;
+ if (!vma || vma->vm_start >= end)
+ break;
+ nend = min(end, vma->vm_end);
+ if (vma->vm_flags & (VM_IO | VM_PFNMAP))
+ continue;
+ if (nstart < vma->vm_start)
+ nstart = vma->vm_start;
+ nr_pages = (nend - nstart) / PAGE_SIZE;
+ ret = __get_user_pages_locked(mm, nstart, nr_pages,
+ NULL, NULL, &locked, gup_flags);
+ if (ret <= 0)
+ break;
+ nend = nstart + ret * PAGE_SIZE;
+ }
+ if (locked)
+ mmap_read_unlock(mm);
+ if (nstart > start)
+ return min(nstart - start, len);
+ return 0;
+}
+
/**
* get_dump_page() - pin user page in memory while writing it to core dump
* @addr: user address
--
2.26.3
Powered by blists - more mailing lists