convert read_kcore(), vread() to use iterators

Now we have eliminated spinlocks from the vread() case, convert read_kcore() to read_kcore_iter().

For the time being we still use a bounce buffer for vread(), however in the next patch we will convert this to interact directly with the iterator and eliminate the bounce buffer altogether.

Signed-off-by Lorenzo Stoakes <[email protected]>

---
 fs/proc/kcore.c | 58 ++++++++++++++++++++++++-------------------------
 1 file changed, 29 insertions(+), 29 deletions(-)

fs/proc/kcore.c +29 -29

index 556f310d6aa4..25e0eeb8d498 100644
@@ -24,7 +24,7 @@
 #include <linux/memblock.h>
 #include <linux/init.h>
 #include <linux/slab.h>
-#include <linux/uaccess.h>
+#include <linux/uio.h>
 #include <asm/io.h>
 #include <linux/list.h>
 #include <linux/ioport.h>

@@ -308,9 +308,12 @@ static void append_kcore_note(char *notes, size_t *i, const char *name,
 }
 
 static ssize_t
-read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
+read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter)
 {
+	struct file *file = iocb->ki_filp;
 	char *buf = file->private_data;
+	loff_t *ppos = &iocb->ki_pos;
+
 	size_t phdrs_offset, notes_offset, data_offset;
 	size_t page_offline_frozen = 1;
 	size_t phdrs_len, notes_len;

@@ -318,6 +321,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 	size_t tsz;
 	int nphdr;
 	unsigned long start;
+	size_t buflen = iov_iter_count(iter);
 	size_t orig_buflen = buflen;
 	int ret = 0;
 

@@ -333,7 +337,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 	notes_offset = phdrs_offset + phdrs_len;
 
 	/* ELF file header. */
-	if (buflen && *fpos < sizeof(struct elfhdr)) {
+	if (buflen && *ppos < sizeof(struct elfhdr)) {
 		struct elfhdr ehdr = {
 			.e_ident = {
 				[EI_MAG0] = ELFMAG0,

@@ -355,19 +359,18 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 			.e_phnum = nphdr,
 		};
 
-		tsz = min_t(size_t, buflen, sizeof(struct elfhdr) - *fpos);
-		if (copy_to_user(buffer, (char *)&ehdr + *fpos, tsz)) {
+		tsz = min_t(size_t, buflen, sizeof(struct elfhdr) - *ppos);
+		if (copy_to_iter((char *)&ehdr + *ppos, tsz, iter) != tsz) {
 			ret = -EFAULT;
 			goto out;
 		}
 
-		buffer += tsz;
 		buflen -= tsz;
-		*fpos += tsz;
+		*ppos += tsz;
 	}
 
 	/* ELF program headers. */
-	if (buflen && *fpos < phdrs_offset + phdrs_len) {
+	if (buflen && *ppos < phdrs_offset + phdrs_len) {
 		struct elf_phdr *phdrs, *phdr;
 
 		phdrs = kzalloc(phdrs_len, GFP_KERNEL);

@@ -397,22 +400,21 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 			phdr++;
 		}
 
-		tsz = min_t(size_t, buflen, phdrs_offset + phdrs_len - *fpos);
-		if (copy_to_user(buffer, (char *)phdrs + *fpos - phdrs_offset,
-				 tsz)) {
+		tsz = min_t(size_t, buflen, phdrs_offset + phdrs_len - *ppos);
+		if (copy_to_iter((char *)phdrs + *ppos - phdrs_offset, tsz,
+				 iter) != tsz) {
 			kfree(phdrs);
 			ret = -EFAULT;
 			goto out;
 		}
 		kfree(phdrs);
 
-		buffer += tsz;
 		buflen -= tsz;
-		*fpos += tsz;
+		*ppos += tsz;
 	}
 
 	/* ELF note segment. */
-	if (buflen && *fpos < notes_offset + notes_len) {
+	if (buflen && *ppos < notes_offset + notes_len) {
 		struct elf_prstatus prstatus = {};
 		struct elf_prpsinfo prpsinfo = {
 			.pr_sname = 'R',

@@ -447,24 +449,23 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 				  vmcoreinfo_data,
 				  min(vmcoreinfo_size, notes_len - i));
 
-		tsz = min_t(size_t, buflen, notes_offset + notes_len - *fpos);
-		if (copy_to_user(buffer, notes + *fpos - notes_offset, tsz)) {
+		tsz = min_t(size_t, buflen, notes_offset + notes_len - *ppos);
+		if (copy_to_iter(notes + *ppos - notes_offset, tsz, iter) != tsz) {
 			kfree(notes);
 			ret = -EFAULT;
 			goto out;
 		}
 		kfree(notes);
 
-		buffer += tsz;
 		buflen -= tsz;
-		*fpos += tsz;
+		*ppos += tsz;
 	}
 
 	/*
 	 * Check to see if our file offset matches with any of
 	 * the addresses in the elf_phdr on our list.
 	 */
-	start = kc_offset_to_vaddr(*fpos - data_offset);
+	start = kc_offset_to_vaddr(*ppos - data_offset);
 	if ((tsz = (PAGE_SIZE - (start & ~PAGE_MASK))) > buflen)
 		tsz = buflen;
 

@@ -497,7 +498,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 		}
 
 		if (!m) {
-			if (clear_user(buffer, tsz)) {
+			if (iov_iter_zero(tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}

@@ -508,14 +509,14 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 		case KCORE_VMALLOC:
 			vread(buf, (char *)start, tsz);
 			/* we have to zero-fill user buffer even if no read */
-			if (copy_to_user(buffer, buf, tsz)) {
+			if (copy_to_iter(buf, tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}
 			break;
 		case KCORE_USER:
 			/* User page is handled prior to normal kernel page: */
-			if (copy_to_user(buffer, (char *)start, tsz)) {
+			if (copy_to_iter((char *)start, tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}

@@ -531,7 +532,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 			 */
 			if (!page || PageOffline(page) ||
 			    is_page_hwpoison(page) || !pfn_is_ram(pfn)) {
-				if (clear_user(buffer, tsz)) {
+				if (iov_iter_zero(tsz, iter) != tsz) {
 					ret = -EFAULT;
 					goto out;
 				}

@@ -541,25 +542,24 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 		case KCORE_VMEMMAP:
 		case KCORE_TEXT:
 			/*
-			 * We use _copy_to_user() to bypass usermode hardening
+			 * We use _copy_to_iter() to bypass usermode hardening
 			 * which would otherwise prevent this operation.
 			 */
-			if (_copy_to_user(buffer, (char *)start, tsz)) {
+			if (_copy_to_iter((char *)start, tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}
 			break;
 		default:
 			pr_warn_once("Unhandled KCORE type: %d\n", m->type);
-			if (clear_user(buffer, tsz)) {
+			if (iov_iter_zero(tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}
 		}
 skip:
 		buflen -= tsz;
-		*fpos += tsz;
-		buffer += tsz;
+		*ppos += tsz;
 		start += tsz;
 		tsz = (buflen > PAGE_SIZE ? PAGE_SIZE : buflen);
 	}

@@ -603,7 +603,7 @@ static int release_kcore(struct inode *inode, struct file *file)
 }
 
 static const struct proc_ops kcore_proc_ops = {
-	.proc_read	= read_kcore,
+	.proc_read_iter	= read_kcore_iter,
 	.proc_open	= open_kcore,
 	.proc_release	= release_kcore,
 	.proc_lseek	= default_llseek,

Signature · 1 line

2.39.2

Hi Lorenzo,

Quoted 2 lines · Lorenzo Stoakes

Now we have eliminated spinlocks from the vread() case, convert read_kcore() to read_kcore_iter().

Sorry for late comment.

Here I could miss something important, I don't get where we have eliminated spinlocks from the vread() case. Do I misunderstand this sentence?

Quoted 212 lines

For the time being we still use a bounce buffer for vread(), however in the next patch we will convert this to interact directly with the iterator and eliminate the bounce buffer altogether.

Signed-off-by Lorenzo Stoakes <[email protected]>

---
 fs/proc/kcore.c | 58 ++++++++++++++++++++++++-------------------------
 1 file changed, 29 insertions(+), 29 deletions(-)

fs/proc/kcore.c +29 -30

index 556f310d6aa4..25e0eeb8d498 100644
@@ -24,7 +24,7 @@
 #include <linux/memblock.h>
 #include <linux/init.h>
 #include <linux/slab.h>
-#include <linux/uaccess.h>
+#include <linux/uio.h>
 #include <asm/io.h>
 #include <linux/list.h>
 #include <linux/ioport.h>

@@ -308,9 +308,12 @@ static void append_kcore_note(char *notes, size_t *i, const char *name,
 }
 
 static ssize_t
-read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
+read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter)
 {
+	struct file *file = iocb->ki_filp;
 	char *buf = file->private_data;
+	loff_t *ppos = &iocb->ki_pos;
+
 	size_t phdrs_offset, notes_offset, data_offset;
 	size_t page_offline_frozen = 1;
 	size_t phdrs_len, notes_len;

@@ -318,6 +321,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 	size_t tsz;
 	int nphdr;
 	unsigned long start;
+	size_t buflen = iov_iter_count(iter);
 	size_t orig_buflen = buflen;
 	int ret = 0;
 

@@ -333,7 +337,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 	notes_offset = phdrs_offset + phdrs_len;
 
 	/* ELF file header. */
-	if (buflen && *fpos < sizeof(struct elfhdr)) {
+	if (buflen && *ppos < sizeof(struct elfhdr)) {
 		struct elfhdr ehdr = {
 			.e_ident = {
 				[EI_MAG0] = ELFMAG0,

@@ -355,19 +359,18 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 			.e_phnum = nphdr,
 		};
 
-		tsz = min_t(size_t, buflen, sizeof(struct elfhdr) - *fpos);
-		if (copy_to_user(buffer, (char *)&ehdr + *fpos, tsz)) {
+		tsz = min_t(size_t, buflen, sizeof(struct elfhdr) - *ppos);
+		if (copy_to_iter((char *)&ehdr + *ppos, tsz, iter) != tsz) {
 			ret = -EFAULT;
 			goto out;
 		}
 
-		buffer += tsz;
 		buflen -= tsz;
-		*fpos += tsz;
+		*ppos += tsz;
 	}
 
 	/* ELF program headers. */
-	if (buflen && *fpos < phdrs_offset + phdrs_len) {
+	if (buflen && *ppos < phdrs_offset + phdrs_len) {
 		struct elf_phdr *phdrs, *phdr;
 
 		phdrs = kzalloc(phdrs_len, GFP_KERNEL);

@@ -397,22 +400,21 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 			phdr++;
 		}
 
-		tsz = min_t(size_t, buflen, phdrs_offset + phdrs_len - *fpos);
-		if (copy_to_user(buffer, (char *)phdrs + *fpos - phdrs_offset,
-				 tsz)) {
+		tsz = min_t(size_t, buflen, phdrs_offset + phdrs_len - *ppos);
+		if (copy_to_iter((char *)phdrs + *ppos - phdrs_offset, tsz,
+				 iter) != tsz) {
 			kfree(phdrs);
 			ret = -EFAULT;
 			goto out;
 		}
 		kfree(phdrs);
 
-		buffer += tsz;
 		buflen -= tsz;
-		*fpos += tsz;
+		*ppos += tsz;
 	}
 
 	/* ELF note segment. */
-	if (buflen && *fpos < notes_offset + notes_len) {
+	if (buflen && *ppos < notes_offset + notes_len) {
 		struct elf_prstatus prstatus = {};
 		struct elf_prpsinfo prpsinfo = {
 			.pr_sname = 'R',

@@ -447,24 +449,23 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 				  vmcoreinfo_data,
 				  min(vmcoreinfo_size, notes_len - i));
 
-		tsz = min_t(size_t, buflen, notes_offset + notes_len - *fpos);
-		if (copy_to_user(buffer, notes + *fpos - notes_offset, tsz)) {
+		tsz = min_t(size_t, buflen, notes_offset + notes_len - *ppos);
+		if (copy_to_iter(notes + *ppos - notes_offset, tsz, iter) != tsz) {
 			kfree(notes);
 			ret = -EFAULT;
 			goto out;
 		}
 		kfree(notes);
 
-		buffer += tsz;
 		buflen -= tsz;
-		*fpos += tsz;
+		*ppos += tsz;
 	}
 
 	/*
 	 * Check to see if our file offset matches with any of
 	 * the addresses in the elf_phdr on our list.
 	 */
-	start = kc_offset_to_vaddr(*fpos - data_offset);
+	start = kc_offset_to_vaddr(*ppos - data_offset);
 	if ((tsz = (PAGE_SIZE - (start & ~PAGE_MASK))) > buflen)
 		tsz = buflen;
 

@@ -497,7 +498,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 		}
 
 		if (!m) {
-			if (clear_user(buffer, tsz)) {
+			if (iov_iter_zero(tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}

@@ -508,14 +509,14 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 		case KCORE_VMALLOC:
 			vread(buf, (char *)start, tsz);
 			/* we have to zero-fill user buffer even if no read */
-			if (copy_to_user(buffer, buf, tsz)) {
+			if (copy_to_iter(buf, tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}
 			break;
 		case KCORE_USER:
 			/* User page is handled prior to normal kernel page: */
-			if (copy_to_user(buffer, (char *)start, tsz)) {
+			if (copy_to_iter((char *)start, tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}

@@ -531,7 +532,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 			 */
 			if (!page || PageOffline(page) ||
 			    is_page_hwpoison(page) || !pfn_is_ram(pfn)) {
-				if (clear_user(buffer, tsz)) {
+				if (iov_iter_zero(tsz, iter) != tsz) {
 					ret = -EFAULT;
 					goto out;
 				}

@@ -541,25 +542,24 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 		case KCORE_VMEMMAP:
 		case KCORE_TEXT:
 			/*
-			 * We use _copy_to_user() to bypass usermode hardening
+			 * We use _copy_to_iter() to bypass usermode hardening
 			 * which would otherwise prevent this operation.
 			 */
-			if (_copy_to_user(buffer, (char *)start, tsz)) {
+			if (_copy_to_iter((char *)start, tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}
 			break;
 		default:
 			pr_warn_once("Unhandled KCORE type: %d\n", m->type);
-			if (clear_user(buffer, tsz)) {
+			if (iov_iter_zero(tsz, iter) != tsz) {
 				ret = -EFAULT;
 				goto out;
 			}
 		}
 skip:
 		buflen -= tsz;
-		*fpos += tsz;
-		buffer += tsz;
+		*ppos += tsz;
 		start += tsz;
 		tsz = (buflen > PAGE_SIZE ? PAGE_SIZE : buflen);
 	}

@@ -603,7 +603,7 @@ static int release_kcore(struct inode *inode, struct file *file)
 }
 
 static const struct proc_ops kcore_proc_ops = {
-	.proc_read	= read_kcore,
+	.proc_read_iter	= read_kcore_iter,
 	.proc_open	= open_kcore,
 	.proc_release	= release_kcore,
 	.proc_lseek	= default_llseek,
--

2.39.2

Now we have eliminated spinlocks from the vread() case, convert read_kcore() to read_kcore_iter().

For the time being we still use a bounce buffer for vread(), however in the next patch we will convert this to interact directly with the iterator and eliminate the bounce buffer altogether.

Signed-off-by Lorenzo Stoakes <[email protected]>

---
  fs/proc/kcore.c | 58 ++++++++++++++++++++++++-------------------------
  1 file changed, 29 insertions(+), 29 deletions(-)

fs/proc/kcore.c +1 -1

index 556f310d6aa4..25e0eeb8d498 100644
@@ -24,7 +24,7 @@
  #include <linux/memblock.h>
  #include <linux/init.h>
  #include <linux/slab.h>
-#include <linux/uaccess.h>
+#include <linux/uio.h>
  #include <asm/io.h>
  #include <linux/list.h>
  #include <linux/ioport.h>

Provide an atomic context equivalent for copy_page_to_iter(). This eschews the might_fault() check copies memory in the same way that copy_page_from_iter_atomic() does.

This functions assumes a non-compound page, however this mimics the existing behaviour of copy_page_from_iter_atomic(). I am keeping the behaviour consistent between the two, deferring any such change to an explicit folio-fication effort.

This is being added in order that an iteratable form of vread() can be implemented with known prefaulted pages to avoid the need for mutex locking.

Signed-off-by Lorenzo Stoakes <[email protected]>

---
 include/linux/uio.h |  2 ++
 lib/iov_iter.c      | 28 ++++++++++++++++++++++++++++
 2 files changed, 30 insertions(+)

include/linux/uio.h +2 -0

index 27e3fd942960..fab07103090f 100644
@@ -154,6 +154,8 @@ static inline struct iovec iov_iter_iovec(const struct iov_iter *iter)
 
 size_t copy_page_from_iter_atomic(struct page *page, unsigned offset,
 				  size_t bytes, struct iov_iter *i);
+size_t copy_page_to_iter_atomic(struct page *page, unsigned offset,
+				size_t bytes, struct iov_iter *i);
 void iov_iter_advance(struct iov_iter *i, size_t bytes);
 void iov_iter_revert(struct iov_iter *i, size_t bytes);
 size_t fault_in_iov_iter_readable(const struct iov_iter *i, size_t bytes);

lib/iov_iter.c +0 -0

index 274014e4eafe..48ca1c5dfc04 100644

lib/iov_iter.c +24 -0

+	char *p = kaddr + offset;
+	size_t copied = 0;
+
+	if (!page_copy_sane(page, offset, bytes) ||
+	    WARN_ON_ONCE(i->data_source))
+		goto out;
+
+	if (unlikely(iov_iter_is_pipe(i))) {
+		copied = copy_page_to_iter_pipe(page, offset, bytes, i);
+		goto out;
+	}
+
+	iterate_and_advance(i, bytes, base, len, off,
+		copyout(base, p + off, len),
+		memcpy(base, p + off, len)
+	)
+	copied = bytes;
+
+out:
+	kunmap_local(kaddr);
+	return copied;
+}
+EXPORT_SYMBOL(copy_page_to_iter_atomic);
+
 static void pipe_advance(struct iov_iter *i, size_t size)
 {
 	struct pipe_inode_info *pipe = i->pipe;

Signature · 1 line

2.39.2

Quoted 40 lines · Lorenzo Stoakes

Provide an atomic context equivalent for copy_page_to_iter(). This eschews the might_fault() check copies memory in the same way that copy_page_from_iter_atomic() does.

This functions assumes a non-compound page, however this mimics the existing behaviour of copy_page_from_iter_atomic(). I am keeping the behaviour consistent between the two, deferring any such change to an explicit folio-fication effort.

This is being added in order that an iteratable form of vread() can be implemented with known prefaulted pages to avoid the need for mutex locking.

Signed-off-by Lorenzo Stoakes <[email protected]>

---
 include/linux/uio.h |  2 ++
 lib/iov_iter.c      | 28 ++++++++++++++++++++++++++++
 2 files changed, 30 insertions(+)

include/linux/uio.h +2 -0

index 27e3fd942960..fab07103090f 100644
@@ -154,6 +154,8 @@ static inline struct iovec iov_iter_iovec(const struct iov_iter *iter)
 
 size_t copy_page_from_iter_atomic(struct page *page, unsigned offset,
 				  size_t bytes, struct iov_iter *i);
+size_t copy_page_to_iter_atomic(struct page *page, unsigned offset,
+				size_t bytes, struct iov_iter *i);
 void iov_iter_advance(struct iov_iter *i, size_t bytes);
 void iov_iter_revert(struct iov_iter *i, size_t bytes);
 size_t fault_in_iov_iter_readable(const struct iov_iter *i, size_t bytes);

lib/iov_iter.c +0 -0

index 274014e4eafe..48ca1c5dfc04 100644
@@ -821,6 +821,34 @@ size_t copy_page_from_iter_atomic(struct page *page, unsigned offset, size_t byt
 }
 EXPORT_SYMBOL(copy_page_from_iter_atomic);
 

lib/iov_iter.c size_t copy_page_from_iter_atomic(struct page *page, unsigned offset, size_t byt · line 827

Quoted 4 lines

lib/iov_iter.c +4 -0

+size_t copy_page_to_iter_atomic(struct page *page, unsigned offset, size_t bytes,
+				struct iov_iter *i)
+{
+	char *kaddr = kmap_local_page(page);

I am a little confused about the name of this new function. In its conterpart, copy_page_from_iter_atomic(), kmap_atomic()/kunmpa_atomic() are used. With them, if CONFIG_HIGHMEM=n, it's like below:

static inline void *kmap_atomic(struct page *page)

{
        if (IS_ENABLED(CONFIG_PREEMPT_RT))
                migrate_disable();
        else
                preempt_disable();
        pagefault_disable();
        return page_address(page);
}

But kmap_local_page() is only having page_address(), the code block between kmap_local_page() and kunmap_local() is also atomic, it's a little messy in my mind.

static inline void *kmap_local_page(struct page *page)

{
        return page_address(page);
}

Quoted 30 lines

lib/iov_iter.c +24 -1

+	char *p = kaddr + offset;
+	size_t copied = 0;
+
+	if (!page_copy_sane(page, offset, bytes) ||
+	    WARN_ON_ONCE(i->data_source))
+		goto out;
+
+	if (unlikely(iov_iter_is_pipe(i))) {
+		copied = copy_page_to_iter_pipe(page, offset, bytes, i);
+		goto out;
+	}
+
+	iterate_and_advance(i, bytes, base, len, off,
+		copyout(base, p + off, len),
+		memcpy(base, p + off, len)
+	)
+	copied = bytes;
+
+out:
+	kunmap_local(kaddr);
+	return copied;
+}
+EXPORT_SYMBOL(copy_page_to_iter_atomic);
+
 static void pipe_advance(struct iov_iter *i, size_t size)
 {
 	struct pipe_inode_info *pipe = i->pipe;
--

2.39.2

Quoted 3 lines

I am a little confused about the name of this new function. In its conterpart, copy_page_from_iter_atomic(), kmap_atomic()/kunmpa_atomic() are used. With them, if CONFIG_HIGHMEM=n, it's like below:

The reason for this is that:-

1. kmap_atomic() explicitly states that it is now deprecated and must no longer be used, and kmap_local_page() should be used instead:-

* kmap_atomic - Atomically map a page for temporary usage - Deprecated!

* Do not use in new code. Use kmap_local_page() instead.

2. kmap_local_page() explicitly states that it can be used in any context:-

* Can be invoked from any context, including interrupts.

I wanted follow this advice as strictly as I could, hence the change. However, we do need preemption/pagefaults explicitly disabled in this context (we are happy to fail if the faulted in pages are unmapped in meantime), and I didn't check the internals to make sure.

So I think for safety it is better to use k[un]map_atomic() here, I'll respin and put that back in, good catch!

Quoted 51 lines

static inline void *kmap_atomic(struct page *page)

{
        if (IS_ENABLED(CONFIG_PREEMPT_RT))
                migrate_disable();
        else
                preempt_disable();
        pagefault_disable();
        return page_address(page);
}

But kmap_local_page() is only having page_address(), the code block between kmap_local_page() and kunmap_local() is also atomic, it's a little messy in my mind.

static inline void *kmap_local_page(struct page *page)

{
        return page_address(page);
}

Quoted 30 lines

lib/iov_iter.c +24 -1

+	char *p = kaddr + offset;
+	size_t copied = 0;
+
+	if (!page_copy_sane(page, offset, bytes) ||
+	    WARN_ON_ONCE(i->data_source))
+		goto out;
+
+	if (unlikely(iov_iter_is_pipe(i))) {
+		copied = copy_page_to_iter_pipe(page, offset, bytes, i);
+		goto out;
+	}
+
+	iterate_and_advance(i, bytes, base, len, off,
+		copyout(base, p + off, len),
+		memcpy(base, p + off, len)
+	)
+	copied = bytes;
+
+out:
+	kunmap_local(kaddr);
+	return copied;
+}
+EXPORT_SYMBOL(copy_page_to_iter_atomic);
+
 static void pipe_advance(struct iov_iter *i, size_t size)
 {
 	struct pipe_inode_info *pipe = i->pipe;
--

2.39.2

Quoted 17 lines · Lorenzo Stoakes

Quoted 3 lines

I am a little confused about the name of this new function. In its conterpart, copy_page_from_iter_atomic(), kmap_atomic()/kunmpa_atomic() are used. With them, if CONFIG_HIGHMEM=n, it's like below:

The reason for this is that:-

1. kmap_atomic() explicitly states that it is now deprecated and must no longer be used, and kmap_local_page() should be used instead:-

* kmap_atomic - Atomically map a page for temporary usage - Deprecated!

* Do not use in new code. Use kmap_local_page() instead.

2. kmap_local_page() explicitly states that it can be used in any context:-

* Can be invoked from any context, including interrupts.

Quoted 8 lines

I wanted follow this advice as strictly as I could, hence the change. However, we do need preemption/pagefaults explicitly disabled in this context (we are happy to fail if the faulted in pages are unmapped in meantime), and I didn't check the internals to make sure.

So I think for safety it is better to use k[un]map_atomic() here, I'll respin and put that back in, good catch!

Actually, given we have preemption disabled due to the held spinlock, I think it'd be better to add a copy_page_to_iter_nofault() that uses copy_to_user_nofault() which will disable pagefaults thus have exactly the equivalent behaviour, more explicitly and without the use of a deprecated function.

Quoted 52 lines

Quoted 52 lines

static inline void *kmap_atomic(struct page *page)

{
        if (IS_ENABLED(CONFIG_PREEMPT_RT))
                migrate_disable();
        else
                preempt_disable();
        pagefault_disable();
        return page_address(page);
}

But kmap_local_page() is only having page_address(), the code block between kmap_local_page() and kunmap_local() is also atomic, it's a little messy in my mind.

static inline void *kmap_local_page(struct page *page)

{
        return page_address(page);
}

Quoted 30 lines

lib/iov_iter.c +24 -1

+	char *p = kaddr + offset;
+	size_t copied = 0;
+
+	if (!page_copy_sane(page, offset, bytes) ||
+	    WARN_ON_ONCE(i->data_source))
+		goto out;
+
+	if (unlikely(iov_iter_is_pipe(i))) {
+		copied = copy_page_to_iter_pipe(page, offset, bytes, i);
+		goto out;
+	}
+
+	iterate_and_advance(i, bytes, base, len, off,
+		copyout(base, p + off, len),
+		memcpy(base, p + off, len)
+	)
+	copied = bytes;
+
+out:
+	kunmap_local(kaddr);
+	return copied;
+}
+EXPORT_SYMBOL(copy_page_to_iter_atomic);
+
 static void pipe_advance(struct iov_iter *i, size_t size)
 {
 	struct pipe_inode_info *pipe = i->pipe;
--

2.39.2

Quoted 41 lines · Baoquan He

Quoted 40 lines · Lorenzo Stoakes

Provide an atomic context equivalent for copy_page_to_iter(). This eschews the might_fault() check copies memory in the same way that copy_page_from_iter_atomic() does.

This functions assumes a non-compound page, however this mimics the existing behaviour of copy_page_from_iter_atomic(). I am keeping the behaviour consistent between the two, deferring any such change to an explicit folio-fication effort.

This is being added in order that an iteratable form of vread() can be implemented with known prefaulted pages to avoid the need for mutex locking.

Signed-off-by Lorenzo Stoakes <[email protected]>

---
 include/linux/uio.h |  2 ++
 lib/iov_iter.c      | 28 ++++++++++++++++++++++++++++
 2 files changed, 30 insertions(+)

include/linux/uio.h +2 -0

index 27e3fd942960..fab07103090f 100644
@@ -154,6 +154,8 @@ static inline struct iovec iov_iter_iovec(const struct iov_iter *iter)
 
 size_t copy_page_from_iter_atomic(struct page *page, unsigned offset,
 				  size_t bytes, struct iov_iter *i);
+size_t copy_page_to_iter_atomic(struct page *page, unsigned offset,
+				size_t bytes, struct iov_iter *i);
 void iov_iter_advance(struct iov_iter *i, size_t bytes);
 void iov_iter_revert(struct iov_iter *i, size_t bytes);
 size_t fault_in_iov_iter_readable(const struct iov_iter *i, size_t bytes);

lib/iov_iter.c +0 -0

index 274014e4eafe..48ca1c5dfc04 100644
@@ -821,6 +821,34 @@ size_t copy_page_from_iter_atomic(struct page *page, unsigned offset, size_t byt
 }
 EXPORT_SYMBOL(copy_page_from_iter_atomic);
 

Quoted 8 lines

lib/iov_iter.c size_t copy_page_from_iter_atomic(struct page *page, unsigned offset, size_t byt · line 827

Quoted 4 lines

lib/iov_iter.c +4 -0

+size_t copy_page_to_iter_atomic(struct page *page, unsigned offset, size_t bytes,
+				struct iov_iter *i)
+{
+	char *kaddr = kmap_local_page(page);

I am a little confused about the name of this new function. In its conterpart, copy_page_from_iter_atomic(), kmap_atomic()/kunmpa_atomic() are used. With them, if CONFIG_HIGHMEM=n, it's like below:

The reason for this is that:-

1. kmap_atomic() explicitly states that it is now deprecated and must no longer be used, and kmap_local_page() should be used instead:-

* kmap_atomic - Atomically map a page for temporary usage - Deprecated!

* Do not use in new code. Use kmap_local_page() instead.

2. kmap_local_page() explicitly states that it can be used in any context:-

* Can be invoked from any context, including interrupts.

I wanted follow this advice as strictly as I could, hence the change. However, we do need preemption/pagefaults explicitly disabled in this context (we are happy to fail if the faulted in pages are unmapped in meantime), and I didn't check the internals to make sure.

So I think for safety it is better to use k[un]map_atomic() here, I'll respin and put that back in, good catch!

Quoted 51 lines

static inline void *kmap_atomic(struct page *page)

{
        if (IS_ENABLED(CONFIG_PREEMPT_RT))
                migrate_disable();
        else
                preempt_disable();
        pagefault_disable();
        return page_address(page);
}

But kmap_local_page() is only having page_address(), the code block between kmap_local_page() and kunmap_local() is also atomic, it's a little messy in my mind.

static inline void *kmap_local_page(struct page *page)

{
        return page_address(page);
}

Quoted 30 lines

lib/iov_iter.c +24 -1

+	char *p = kaddr + offset;
+	size_t copied = 0;
+
+	if (!page_copy_sane(page, offset, bytes) ||
+	    WARN_ON_ONCE(i->data_source))
+		goto out;
+
+	if (unlikely(iov_iter_is_pipe(i))) {
+		copied = copy_page_to_iter_pipe(page, offset, bytes, i);
+		goto out;
+	}
+
+	iterate_and_advance(i, bytes, base, len, off,
+		copyout(base, p + off, len),
+		memcpy(base, p + off, len)
+	)
+	copied = bytes;
+
+out:
+	kunmap_local(kaddr);
+	return copied;
+}
+EXPORT_SYMBOL(copy_page_to_iter_atomic);
+
 static void pipe_advance(struct iov_iter *i, size_t size)
 {
 	struct pipe_inode_info *pipe = i->pipe;
--

2.39.2

Having previously laid the foundation for converting vread() to an iterator function, pull the trigger and do so.

This patch attempts to provide minimal refactoring and to reflect the existing logic as best we can, for example we continue to zero portions of memory not read, as before.

Overall, there should be no functional difference other than a performance improvement in /proc/kcore access to vmalloc regions.

Now we have eliminated the need for a bounce buffer in read_kcore_iter(), we dispense with it. We need to ensure userland pages are faulted in before proceeding, as we take spin locks.

Additionally, we must account for the fact that at any point a copy may fail if this happens, we exit indicating fewer bytes retrieved than expected.

Signed-off-by Lorenzo Stoakes <[email protected]>

---
  fs/proc/kcore.c         |  26 ++---
  include/linux/vmalloc.h |   3 +-
  mm/nommu.c              |  10 +-
  mm/vmalloc.c            | 234 +++++++++++++++++++++++++---------------
  4 files changed, 160 insertions(+), 113 deletions(-)

fs/proc/kcore.c +1 -7

index 25e0eeb8d498..221e16f75ba5 100644
@@ -307,13 +307,9 @@ static void append_kcore_note(char *notes, size_t *i, const char *name,
  	*i = ALIGN(*i + descsz, 4);
  }
 
-static ssize_t
-read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter)
+static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter)
  {
-	struct file *file = iocb->ki_filp;
-	char *buf = file->private_data;
  	loff_t *ppos = &iocb->ki_pos;
-
  	size_t phdrs_offset, notes_offset, data_offset;
  	size_t page_offline_frozen = 1;
  	size_t phdrs_len, notes_len;

@@ -507,9 +503,12 @@ read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter)
 
  		switch (m->type) {
  		case KCORE_VMALLOC:
-			vread(buf, (char *)start, tsz);
-			/* we have to zero-fill user buffer even if no read */

Quoted 49 lines · Lorenzo Stoakes

Having previously laid the foundation for converting vread() to an iterator function, pull the trigger and do so.

This patch attempts to provide minimal refactoring and to reflect the existing logic as best we can, for example we continue to zero portions of memory not read, as before.

Overall, there should be no functional difference other than a performance improvement in /proc/kcore access to vmalloc regions.

Now we have eliminated the need for a bounce buffer in read_kcore_iter(), we dispense with it. We need to ensure userland pages are faulted in before proceeding, as we take spin locks.

Additionally, we must account for the fact that at any point a copy may fail if this happens, we exit indicating fewer bytes retrieved than expected.

Signed-off-by Lorenzo Stoakes <[email protected]>

---
  fs/proc/kcore.c         |  26 ++---
  include/linux/vmalloc.h |   3 +-
  mm/nommu.c              |  10 +-
  mm/vmalloc.c            | 234 +++++++++++++++++++++++++---------------
  4 files changed, 160 insertions(+), 113 deletions(-)

fs/proc/kcore.c +1 -7

index 25e0eeb8d498..221e16f75ba5 100644
@@ -307,13 +307,9 @@ static void append_kcore_note(char *notes, size_t *i, const char *name,
  	*i = ALIGN(*i + descsz, 4);
  }
-static ssize_t
-read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter)
+static ssize_t read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter)
  {
-	struct file *file = iocb->ki_filp;
-	char *buf = file->private_data;
  	loff_t *ppos = &iocb->ki_pos;
-
  	size_t phdrs_offset, notes_offset, data_offset;
  	size_t page_offline_frozen = 1;
  	size_t phdrs_len, notes_len;

@@ -507,9 +503,12 @@ read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter)
  		switch (m->type) {
  		case KCORE_VMALLOC:
-			vread(buf, (char *)start, tsz);
-			/* we have to zero-fill user buffer even if no read */

fs/proc/kcore.c read_kcore_iter(struct kiocb *iocb, struct iov_iter *iter) · line 513

Quoted 10 lines

fs/proc/kcore.c +6 -1

-			if (copy_to_iter(buf, tsz, iter) != tsz) {
+			/*
+			 * Make sure user pages are faulted in as we acquire
+			 * spinlocks in vread_iter().
+			 */
+			if (fault_in_iov_iter_writeable(iter, tsz) ||
+			    vread_iter(iter, (char *)start, tsz) != tsz) {
  				ret = -EFAULT;
  				goto out;
  			}

What if we race with swapout after faulting the pages in? Or some other mechanism to write-protect the user space pages?

Also, "This is primarily useful when we already know that some or all of the pages in @i aren't in memory". This order of events might slow down things quite a bit if I am not wrong.

Wouldn't you want to have something like:

while (vread_iter(iter, (char *)start, tsz) != tsz) {
	if (fault_in_iov_iter_writeable(iter, tsz)) {
		ret = -EFAULT;
		goto out;
	}
}

Or am I missing something?

--
Thanks,

David / dhildenb

Additionally, we must account for the fact that at any point a copy may fail if this happens, we exit indicating fewer bytes retrieved than expected.

mm/vmalloc.c +21 -13

-static int aligned_vread(char *buf, char *addr, unsigned long count)
+/*
+ * small helper routine, copy contents to iter from addr.
+ * If the page is not present, fill zero.
+ *
+ * Returns the number of copied bytes.
+ */
+static size_t aligned_vread_iter(struct iov_iter *iter,
+				 const char *addr, size_t count)
 {
-	struct page *p;
-	int copied = 0;
+	size_t remains = count;
+	struct page *page;
 
-	while (count) {
+	while (remains > 0) {
 		unsigned long offset, length;
+		size_t copied = 0;
 
 		offset = offset_in_page(addr);
 		length = PAGE_SIZE - offset;
-		if (length > count)
-			length = count;
-		p = vmalloc_to_page(addr);
+		if (length > remains)
+			length = remains;
+		page = vmalloc_to_page(addr);
 		/*
-		 * To do safe access to this _mapped_ area, we need
-		 * lock. But adding lock here means that we need to add
-		 * overhead of vmalloc()/vfree() calls for this _debug_
-		 * interface, rarely used. Instead of that, we'll use
-		 * kmap() and get small overhead in this access function.
+		 * To do safe access to this _mapped_ area, we need lock. But
+		 * adding lock here means that we need to add overhead of
+		 * vmalloc()/vfree() calls for this _debug_ interface, rarely
+		 * used. Instead of that, we'll use an local mapping via
+		 * copy_page_to_iter_atomic() and accept a small overhead in
+		 * this access function.
 		 */
-		if (p) {

mm/vmalloc.c +49 -28

+		/* Zero anything we were unable to copy. */
+		copied += zero_iter(iter, length - copied);
+
+		addr += copied;
+		remains -= copied;
 
-		addr += length;
-		buf += length;
-		copied += length;
-		count -= length;
+		if (copied != length)
+			break;
 	}
-	return copied;
+
+	return count - remains;
 }
 
-static void vmap_ram_vread(char *buf, char *addr, int count, unsigned long flags)
+/*
+ * Read from a vm_map_ram region of memory.
+ *
+ * Returns the number of copied bytes.
+ */
+static size_t vmap_ram_vread_iter(struct iov_iter *iter, const char *addr,
+				  size_t count, unsigned long flags)
 {
 	char *start;
 	struct vmap_block *vb;
 	unsigned long offset;
-	unsigned int rs, re, n;
+	unsigned int rs, re;
+	size_t remains, n;
 
 	/*
 	 * If it's area created by vm_map_ram() interface directly, but
 	 * not further subdividing and delegating management to vmap_block,
 	 * handle it here.
 	 */
-	if (!(flags & VMAP_BLOCK)) {
-		aligned_vread(buf, addr, count);
-		return;
-	}
+	if (!(flags & VMAP_BLOCK))
+		return aligned_vread_iter(iter, addr, count);
 
 	/*
 	 * Area is split into regions and tracked with vmap_block, read out

@@ -3505,50 +3537,65 @@ static void vmap_ram_vread(char *buf, char *addr, int count, unsigned long flags
 	 */
 	vb = xa_load(&vmap_blocks, addr_to_vb_idx((unsigned long)addr));
 	if (!vb)
-		goto finished;
+		goto finished_zero;
 
 	spin_lock(&vb->lock);
 	if (bitmap_empty(vb->used_map, VMAP_BBMAP_BITS)) {
 		spin_unlock(&vb->lock);
-		goto finished;
+		goto finished_zero;
 	}
+
+	remains = count;
 	for_each_set_bitrange(rs, re, vb->used_map, VMAP_BBMAP_BITS) {
-		if (!count)
-			break;
+		size_t copied;
+
+		if (remains == 0)
+			goto finished;
+
 		start = vmap_block_vaddr(vb->va->va_start, rs);
-		while (addr < start) {
-			if (count == 0)
-				goto unlock;
-			*buf = '\0';
-			buf++;
-			addr++;
-			count--;
+
+		if (addr < start) {
+			size_t to_zero = min_t(size_t, start - addr, remains);
+			size_t zeroed = zero_iter(iter, to_zero);
+
+			addr += zeroed;
+			remains -= zeroed;
+
+			if (remains == 0 || zeroed != to_zero)
+				goto finished;
 		}
+
 		/*it could start reading from the middle of used region*/
 		offset = offset_in_page(addr);
 		n = ((re - rs + 1) << PAGE_SHIFT) - offset;
-		if (n > count)
-			n = count;
-		aligned_vread(buf, start+offset, n);
+		if (n > remains)
+			n = remains;
+
+		copied = aligned_vread_iter(iter, start + offset, n);
 
-		buf += n;
-		addr += n;
-		count -= n;
+		addr += copied;
+		remains -= copied;
+
+		if (copied != n)
+			goto finished;
 	}

mm/vmalloc.c +42 -29

+finished:
+	/* We couldn't copy/zero everything */
+	spin_unlock(&vb->lock);
+	return count - remains;
 }
 
 /**
- * vread() - read vmalloc area in a safe way.
- * @buf:     buffer for reading data
- * @addr:    vm address.
- * @count:   number of bytes to be read.
+ * vread_iter() - read vmalloc area in a safe way to an iterator.
+ * @iter:         the iterator to which data should be written.
+ * @addr:         vm address.
+ * @count:        number of bytes to be read.
  *
  * This function checks that addr is a valid vmalloc'ed area, and
  * copy data from that area to a given buffer. If the given memory range

@@ -3568,13 +3615,12 @@ static void vmap_ram_vread(char *buf, char *addr, int count, unsigned long flags
  * (same number as @count) or %0 if [addr...addr+count) doesn't
  * include any intersection with valid vmalloc area
  */
-long vread(char *buf, char *addr, unsigned long count)
+long vread_iter(struct iov_iter *iter, const char *addr, size_t count)
 {
 	struct vmap_area *va;
 	struct vm_struct *vm;
-	char *vaddr, *buf_start = buf;
-	unsigned long buflen = count;
-	unsigned long n, size, flags;
+	char *vaddr;
+	size_t n, size, flags, remains;
 
 	addr = kasan_reset_tag(addr);
 

@@ -3582,18 +3628,22 @@ long vread(char *buf, char *addr, unsigned long count)
 	if ((unsigned long) addr + count < count)
 		count = -(unsigned long) addr;
 
+	remains = count;
+
 	spin_lock(&vmap_area_lock);
 	va = find_vmap_area_exceed_addr((unsigned long)addr);
 	if (!va)
-		goto finished;
+		goto finished_zero;
 
 	/* no intersects with alive vmap_area */
-	if ((unsigned long)addr + count <= va->va_start)
-		goto finished;
+	if ((unsigned long)addr + remains <= va->va_start)
+		goto finished_zero;
 
 	list_for_each_entry_from(va, &vmap_area_list, list) {
-		if (!count)
-			break;
+		size_t copied;
+
+		if (remains == 0)
+			goto finished;
 
 		vm = va->vm;
 		flags = va->flags & VMAP_FLAGS_MASK;

@@ -3608,6 +3658,7 @@ long vread(char *buf, char *addr, unsigned long count)
 
 		if (vm && (vm->flags & VM_UNINITIALIZED))
 			continue;
+
 		/* Pair with smp_wmb() in clear_vm_uninitialized_flag() */
 		smp_rmb();
 

@@ -3616,38 +3667,45 @@ long vread(char *buf, char *addr, unsigned long count)
 
 		if (addr >= vaddr + size)
 			continue;
-		while (addr < vaddr) {
-			if (count == 0)
+
+		if (addr < vaddr) {
+			size_t to_zero = min_t(size_t, vaddr - addr, remains);
+			size_t zeroed = zero_iter(iter, to_zero);
+
+			addr += zeroed;
+			remains -= zeroed;
+
+			if (remains == 0 || zeroed != to_zero)
 				goto finished;
-			*buf = '\0';
-			buf++;
-			addr++;
-			count--;
 		}
+
 		n = vaddr + size - addr;
-		if (n > count)
-			n = count;
+		if (n > remains)
+			n = remains;
 
 		if (flags & VMAP_RAM)
-			vmap_ram_vread(buf, addr, n, flags);
+			copied = vmap_ram_vread_iter(iter, addr, n, flags);
 		else if (!(vm->flags & VM_IOREMAP))
-			aligned_vread(buf, addr, n);
+			copied = aligned_vread_iter(iter, addr, n);
 		else /* IOREMAP area is treated as memory hole */
-			memset(buf, 0, n);
-		buf += n;
-		addr += n;
-		count -= n;
+			copied = zero_iter(iter, n);
+
+		addr += copied;
+		remains -= copied;
+
+		if (copied != n)
+			goto finished;
 	}
-finished:
-	spin_unlock(&vmap_area_lock);
 

mm/vmalloc.c +2 -1

+	spin_unlock(&vmap_area_lock);
 
-	return buflen;
+	return count - remains;
 }
 
 /**

Signature · 1 line

2.39.2

On 03/21/23 at 08:54pm, Lorenzo Stoakes wrote:
......

Quoted 3 lines

Additionally, we must account for the fact that at any point a copy may fail if this happens, we exit indicating fewer bytes retrieved than expected.

......

Quoted 42 lines

mm/vmalloc.c +21 -13

-static int aligned_vread(char *buf, char *addr, unsigned long count)
+/*
+ * small helper routine, copy contents to iter from addr.
+ * If the page is not present, fill zero.
+ *
+ * Returns the number of copied bytes.
+ */
+static size_t aligned_vread_iter(struct iov_iter *iter,
+				 const char *addr, size_t count)
 {
-	struct page *p;
-	int copied = 0;
+	size_t remains = count;
+	struct page *page;
 
-	while (count) {
+	while (remains > 0) {
 		unsigned long offset, length;
+		size_t copied = 0;
 
 		offset = offset_in_page(addr);
 		length = PAGE_SIZE - offset;
-		if (length > count)
-			length = count;
-		p = vmalloc_to_page(addr);
+		if (length > remains)
+			length = remains;
+		page = vmalloc_to_page(addr);
 		/*
-		 * To do safe access to this _mapped_ area, we need
-		 * lock. But adding lock here means that we need to add
-		 * overhead of vmalloc()/vfree() calls for this _debug_
-		 * interface, rarely used. Instead of that, we'll use
-		 * kmap() and get small overhead in this access function.
+		 * To do safe access to this _mapped_ area, we need lock. But
+		 * adding lock here means that we need to add overhead of
+		 * vmalloc()/vfree() calls for this _debug_ interface, rarely
+		 * used. Instead of that, we'll use an local mapping via
+		 * copy_page_to_iter_atomic() and accept a small overhead in
+		 * this access function.
 		 */
-		if (p) {

mm/vmalloc.c void *vmalloc_32_user(unsigned long size) · line 3498

Quoted 10 lines

mm/vmalloc.c +4 -6

-			/* We can expect USER0 is not used -- see vread() */
-			void *map = kmap_atomic(p);
-			memcpy(buf, map + offset, length);
-			kunmap_atomic(map);
-		} else
-			memset(buf, 0, length);
+		if (page)
+			copied = copy_page_to_iter_atomic(page, offset, length,
+							  iter);
+

If we decided to quit at any failing copy point, indicating fewer bytes retrieved than expected, wondering why we don't quit here if copy_page_to_iter_atomic() failed?

The original zero filling is for unmapped vmalloc area, but not for reading area if failed. Not sure if I got your point.

Quoted 115 lines

mm/vmalloc.c +50 -29

+		/* Zero anything we were unable to copy. */
+		copied += zero_iter(iter, length - copied);
+
+		addr += copied;
+		remains -= copied;
 
-		addr += length;
-		buf += length;
-		copied += length;
-		count -= length;
+		if (copied != length)
+			break;
 	}
-	return copied;
+
+	return count - remains;
 }
 
-static void vmap_ram_vread(char *buf, char *addr, int count, unsigned long flags)
+/*
+ * Read from a vm_map_ram region of memory.
+ *
+ * Returns the number of copied bytes.
+ */
+static size_t vmap_ram_vread_iter(struct iov_iter *iter, const char *addr,
+				  size_t count, unsigned long flags)
 {
 	char *start;
 	struct vmap_block *vb;
 	unsigned long offset;
-	unsigned int rs, re, n;
+	unsigned int rs, re;
+	size_t remains, n;
 
 	/*
 	 * If it's area created by vm_map_ram() interface directly, but
 	 * not further subdividing and delegating management to vmap_block,
 	 * handle it here.
 	 */
-	if (!(flags & VMAP_BLOCK)) {
-		aligned_vread(buf, addr, count);
-		return;
-	}
+	if (!(flags & VMAP_BLOCK))
+		return aligned_vread_iter(iter, addr, count);
 
 	/*
 	 * Area is split into regions and tracked with vmap_block, read out

@@ -3505,50 +3537,65 @@ static void vmap_ram_vread(char *buf, char *addr, int count, unsigned long flags
 	 */
 	vb = xa_load(&vmap_blocks, addr_to_vb_idx((unsigned long)addr));
 	if (!vb)
-		goto finished;
+		goto finished_zero;
 
 	spin_lock(&vb->lock);
 	if (bitmap_empty(vb->used_map, VMAP_BBMAP_BITS)) {
 		spin_unlock(&vb->lock);
-		goto finished;
+		goto finished_zero;
 	}
+
+	remains = count;
 	for_each_set_bitrange(rs, re, vb->used_map, VMAP_BBMAP_BITS) {
-		if (!count)
-			break;
+		size_t copied;
+
+		if (remains == 0)
+			goto finished;
+
 		start = vmap_block_vaddr(vb->va->va_start, rs);
-		while (addr < start) {
-			if (count == 0)
-				goto unlock;
-			*buf = '\0';
-			buf++;
-			addr++;
-			count--;
+
+		if (addr < start) {
+			size_t to_zero = min_t(size_t, start - addr, remains);
+			size_t zeroed = zero_iter(iter, to_zero);
+
+			addr += zeroed;
+			remains -= zeroed;
+
+			if (remains == 0 || zeroed != to_zero)
+				goto finished;
 		}
+
 		/*it could start reading from the middle of used region*/
 		offset = offset_in_page(addr);
 		n = ((re - rs + 1) << PAGE_SHIFT) - offset;
-		if (n > count)
-			n = count;
-		aligned_vread(buf, start+offset, n);
+		if (n > remains)
+			n = remains;
+
+		copied = aligned_vread_iter(iter, start + offset, n);
 
-		buf += n;
-		addr += n;
-		count -= n;
+		addr += copied;
+		remains -= copied;
+
+		if (copied != n)
+			goto finished;
 	}
-unlock:
+
 	spin_unlock(&vb->lock);
 

mm/vmalloc.c static void vmap_ram_vread(char *buf, char *addr, int count, unsigned long flags · line 3587

Quoted 6 lines

mm/vmalloc.c +2 -3

-finished:
+finished_zero:
 	/* zero-fill the left dirty or free regions */
-	if (count)
-		memset(buf, 0, count);
+	return count - remains + zero_iter(iter, remains);

When it jumps to finished_zero, local varialble 'remains' is still 0, we do nothing here but return count. It doesn't look so right. You may need to move up the 'remains' assignment as 'count' line.

Quoted 121 lines

mm/vmalloc.c +42 -29

+finished:
+	/* We couldn't copy/zero everything */
+	spin_unlock(&vb->lock);
+	return count - remains;
 }
 
 /**
- * vread() - read vmalloc area in a safe way.
- * @buf:     buffer for reading data
- * @addr:    vm address.
- * @count:   number of bytes to be read.
+ * vread_iter() - read vmalloc area in a safe way to an iterator.
+ * @iter:         the iterator to which data should be written.
+ * @addr:         vm address.
+ * @count:        number of bytes to be read.
  *
  * This function checks that addr is a valid vmalloc'ed area, and
  * copy data from that area to a given buffer. If the given memory range

@@ -3568,13 +3615,12 @@ static void vmap_ram_vread(char *buf, char *addr, int count, unsigned long flags
  * (same number as @count) or %0 if [addr...addr+count) doesn't
  * include any intersection with valid vmalloc area
  */
-long vread(char *buf, char *addr, unsigned long count)
+long vread_iter(struct iov_iter *iter, const char *addr, size_t count)
 {
 	struct vmap_area *va;
 	struct vm_struct *vm;
-	char *vaddr, *buf_start = buf;
-	unsigned long buflen = count;
-	unsigned long n, size, flags;
+	char *vaddr;
+	size_t n, size, flags, remains;
 
 	addr = kasan_reset_tag(addr);
 

@@ -3582,18 +3628,22 @@ long vread(char *buf, char *addr, unsigned long count)
 	if ((unsigned long) addr + count < count)
 		count = -(unsigned long) addr;
 
+	remains = count;
+
 	spin_lock(&vmap_area_lock);
 	va = find_vmap_area_exceed_addr((unsigned long)addr);
 	if (!va)
-		goto finished;
+		goto finished_zero;
 
 	/* no intersects with alive vmap_area */
-	if ((unsigned long)addr + count <= va->va_start)
-		goto finished;
+	if ((unsigned long)addr + remains <= va->va_start)
+		goto finished_zero;
 
 	list_for_each_entry_from(va, &vmap_area_list, list) {
-		if (!count)
-			break;
+		size_t copied;
+
+		if (remains == 0)
+			goto finished;
 
 		vm = va->vm;
 		flags = va->flags & VMAP_FLAGS_MASK;

@@ -3608,6 +3658,7 @@ long vread(char *buf, char *addr, unsigned long count)
 
 		if (vm && (vm->flags & VM_UNINITIALIZED))
 			continue;
+
 		/* Pair with smp_wmb() in clear_vm_uninitialized_flag() */
 		smp_rmb();
 

@@ -3616,38 +3667,45 @@ long vread(char *buf, char *addr, unsigned long count)
 
 		if (addr >= vaddr + size)
 			continue;
-		while (addr < vaddr) {
-			if (count == 0)
+
+		if (addr < vaddr) {
+			size_t to_zero = min_t(size_t, vaddr - addr, remains);
+			size_t zeroed = zero_iter(iter, to_zero);
+
+			addr += zeroed;
+			remains -= zeroed;
+
+			if (remains == 0 || zeroed != to_zero)
 				goto finished;
-			*buf = '\0';
-			buf++;
-			addr++;
-			count--;
 		}
+
 		n = vaddr + size - addr;
-		if (n > count)
-			n = count;
+		if (n > remains)
+			n = remains;
 
 		if (flags & VMAP_RAM)
-			vmap_ram_vread(buf, addr, n, flags);
+			copied = vmap_ram_vread_iter(iter, addr, n, flags);
 		else if (!(vm->flags & VM_IOREMAP))
-			aligned_vread(buf, addr, n);
+			copied = aligned_vread_iter(iter, addr, n);
 		else /* IOREMAP area is treated as memory hole */
-			memset(buf, 0, n);
-		buf += n;
-		addr += n;
-		count -= n;
+			copied = zero_iter(iter, n);
+
+		addr += copied;
+		remains -= copied;
+
+		if (copied != n)
+			goto finished;
 	}
-finished:
-	spin_unlock(&vmap_area_lock);
 

mm/vmalloc.c long vread(char *buf, char *addr, unsigned long count) · line 3705

Quoted 10 lines

mm/vmalloc.c +5 -4

-	if (buf == buf_start)
-		return 0;
+finished_zero:
+	spin_unlock(&vmap_area_lock);
 	/* zero-fill memory holes */
-	if (buf != buf_start + buflen)
-		memset(buf, 0, buflen - (buf - buf_start));
+	return count - remains + zero_iter(iter, remains);
+finished:
+	/* Nothing remains, or We couldn't copy/zero everything. */

                              ~~~ typo, s/We/we/

Quoted 10 lines

mm/vmalloc.c +2 -2

+	spin_unlock(&vmap_area_lock);
 
-	return buflen;
+	return count - remains;
 }
 
 /**
--

2.39.2