Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1222471 > unrolled thread

[PATCH] f2fs crypto: allocate buffer for decrypting filename

Started byJaegeuk Kim <jaegeuk@kernel.org>
First post2015-09-11 03:50 +0200
Last post2015-09-17 14:30 +0200
Articles 6 — 2 participants

Back to article view | Back to linux.kernel

Contents

  [PATCH] f2fs crypto: allocate buffer for decrypting filename Jaegeuk Kim <jaegeuk@kernel.org> - 2015-09-11 03:50 +0200
    Re: [PATCH v2] f2fs crypto: allocate buffer for decrypting filename Jaegeuk Kim <jaegeuk@kernel.org> - 2015-09-12 00:50 +0200
      Re: [f2fs-dev] [PATCH v3] f2fs crypto: allocate buffer for  decrypting filename Jaegeuk Kim <jaegeuk@kernel.org> - 2015-09-15 19:00 +0200
        RE: [f2fs-dev] [PATCH v3] f2fs crypto: allocate buffer for decrypting  filename Chao Yu <chao2.yu@samsung.com> - 2015-09-16 12:30 +0200
          Re: [f2fs-dev] [PATCH v4] f2fs crypto: allocate buffer for  decrypting filename Jaegeuk Kim <jaegeuk@kernel.org> - 2015-09-16 20:00 +0200
            RE: [f2fs-dev] [PATCH v4] f2fs crypto: allocate buffer for decrypting  filename Chao Yu <chao2.yu@samsung.com> - 2015-09-17 14:30 +0200

#1222471 — [PATCH] f2fs crypto: allocate buffer for decrypting filename

FromJaegeuk Kim <jaegeuk@kernel.org>
Date2015-09-11 03:50 +0200
Subject[PATCH] f2fs crypto: allocate buffer for decrypting filename
Message-ID<q7lDs-3it-7@gated-at.bofh.it>
We got dentry pages from high_mem, and its address space directly goes into the
decryption path via f2fs_fname_disk_to_usr.
But, sg_init_one assumes the address is not from high_mem, so we can get this
panic since it doesn't call kmap_high but kunmap_high is triggered at the end.

kernel BUG at ../../../../../../kernel/mm/highmem.c:290!
Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
...
 (kunmap_high+0xb0/0xb8) from [<c0114534>] (__kunmap_atomic+0xa0/0xa4)
 (__kunmap_atomic+0xa0/0xa4) from [<c035f028>] (blkcipher_walk_done+0x128/0x1ec)
 (blkcipher_walk_done+0x128/0x1ec) from [<c0366c24>] (crypto_cbc_decrypt+0xc0/0x170)
 (crypto_cbc_decrypt+0xc0/0x170) from [<c0367148>] (crypto_cts_decrypt+0xc0/0x114)
 (crypto_cts_decrypt+0xc0/0x114) from [<c035ea98>] (async_decrypt+0x40/0x48)
 (async_decrypt+0x40/0x48) from [<c032ca34>] (f2fs_fname_disk_to_usr+0x124/0x304)
 (f2fs_fname_disk_to_usr+0x124/0x304) from [<c03056fc>] (f2fs_fill_dentries+0xac/0x188)
 (f2fs_fill_dentries+0xac/0x188) from [<c03059c8>] (f2fs_readdir+0x1f0/0x300)
 (f2fs_readdir+0x1f0/0x300) from [<c0218054>] (vfs_readdir+0x90/0xb4)
 (vfs_readdir+0x90/0xb4) from [<c0218418>] (SyS_getdents64+0x64/0xcc)
 (SyS_getdents64+0x64/0xcc) from [<c0105ba0>] (ret_fast_syscall+0x0/0x30)

Cc: <stable@vger.kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/dir.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 8f15fc1..f3fe1a1 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -788,8 +788,9 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 			d_type = DT_UNKNOWN;
 
 		/* encrypted case */
-		de_name.name = d->filename[bit_pos];
 		de_name.len = le16_to_cpu(de->name_len);
+		de_name.name = kmalloc(de_name.len, GFP_NOFS);
+		memcpy(de_name.name, d->filename[bit_pos], de_name.len);
 
 		if (f2fs_encrypted_inode(d->inode)) {
 			int save_len = fstr->len;
@@ -799,13 +800,18 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 							&de_name, fstr);
 			de_name = *fstr;
 			fstr->len = save_len;
-			if (ret < 0)
+			if (ret < 0) {
+				kfree(de_name.name);
 				return true;
+			}
 		}
 
 		if (!dir_emit(ctx, de_name.name, de_name.len,
-					le32_to_cpu(de->ino), d_type))
+					le32_to_cpu(de->ino), d_type)) {
+			kfree(de_name.name);
 			return true;
+		}
+		kfree(de_name.name);
 
 		bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len));
 		ctx->pos = start_pos + bit_pos;
-- 
2.1.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1223141 — Re: [PATCH v2] f2fs crypto: allocate buffer for decrypting filename

FromJaegeuk Kim <jaegeuk@kernel.org>
Date2015-09-12 00:50 +0200
SubjectRe: [PATCH v2] f2fs crypto: allocate buffer for decrypting filename
Message-ID<q7FiN-75P-9@gated-at.bofh.it>
In reply to#1222471
Change log from v1:
 o fix wrong pointer assignment

From a0f3577cab882adc70d71e733d62c48aab61c3a4 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Thu, 3 Sep 2015 13:38:23 -0700
Subject: [PATCH] f2fs crypto: allocate buffer for decrypting filename

We got dentry pages from high_mem, and its address space directly goes into the
decryption path via f2fs_fname_disk_to_usr.
But, sg_init_one assumes the address is not from high_mem, so we can get this
panic since it doesn't call kmap_high but kunmap_high is triggered at the end.

kernel BUG at ../../../../../../kernel/mm/highmem.c:290!
Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
...
 (kunmap_high+0xb0/0xb8) from [<c0114534>] (__kunmap_atomic+0xa0/0xa4)
 (__kunmap_atomic+0xa0/0xa4) from [<c035f028>] (blkcipher_walk_done+0x128/0x1ec)
 (blkcipher_walk_done+0x128/0x1ec) from [<c0366c24>] (crypto_cbc_decrypt+0xc0/0x170)
 (crypto_cbc_decrypt+0xc0/0x170) from [<c0367148>] (crypto_cts_decrypt+0xc0/0x114)
 (crypto_cts_decrypt+0xc0/0x114) from [<c035ea98>] (async_decrypt+0x40/0x48)
 (async_decrypt+0x40/0x48) from [<c032ca34>] (f2fs_fname_disk_to_usr+0x124/0x304)
 (f2fs_fname_disk_to_usr+0x124/0x304) from [<c03056fc>] (f2fs_fill_dentries+0xac/0x188)
 (f2fs_fill_dentries+0xac/0x188) from [<c03059c8>] (f2fs_readdir+0x1f0/0x300)
 (f2fs_readdir+0x1f0/0x300) from [<c0218054>] (vfs_readdir+0x90/0xb4)
 (vfs_readdir+0x90/0xb4) from [<c0218418>] (SyS_getdents64+0x64/0xcc)
 (SyS_getdents64+0x64/0xcc) from [<c0105ba0>] (ret_fast_syscall+0x0/0x30)

Cc: <stable@vger.kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/dir.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 8f15fc1..cce512c 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -773,6 +773,7 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 	unsigned int bit_pos;
 	struct f2fs_dir_entry *de = NULL;
 	struct f2fs_str de_name = FSTR_INIT(NULL, 0);
+	char *name = NULL;
 
 	bit_pos = ((unsigned long)ctx->pos % d->max);
 
@@ -788,8 +789,10 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 			d_type = DT_UNKNOWN;
 
 		/* encrypted case */
-		de_name.name = d->filename[bit_pos];
 		de_name.len = le16_to_cpu(de->name_len);
+		name = kmalloc(de_name.len, GFP_NOFS);
+		memcpy(name, d->filename[bit_pos], de_name.len);
+		de_name.name = name;
 
 		if (f2fs_encrypted_inode(d->inode)) {
 			int save_len = fstr->len;
@@ -799,13 +802,18 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 							&de_name, fstr);
 			de_name = *fstr;
 			fstr->len = save_len;
-			if (ret < 0)
+			if (ret < 0) {
+				kfree(name);
 				return true;
+			}
 		}
 
 		if (!dir_emit(ctx, de_name.name, de_name.len,
-					le32_to_cpu(de->ino), d_type))
+					le32_to_cpu(de->ino), d_type)) {
+			kfree(name);
 			return true;
+		}
+		kfree(name);
 
 		bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len));
 		ctx->pos = start_pos + bit_pos;
-- 
2.1.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1225442 — Re: [f2fs-dev] [PATCH v3] f2fs crypto: allocate buffer for decrypting filename

FromJaegeuk Kim <jaegeuk@kernel.org>
Date2015-09-15 19:00 +0200
SubjectRe: [f2fs-dev] [PATCH v3] f2fs crypto: allocate buffer for decrypting filename
Message-ID<q91Kk-2gb-41@gated-at.bofh.it>
In reply to#1223141
Change log from v1:
  o fix wrong pointer assignment

Chang log from v2:
  o add one more missing call path: f2fs_encrypted_follow_link

From 84574dd5c3e8ed9ca9fdfcbd251b354cdbc5ecab Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Thu, 3 Sep 2015 13:38:23 -0700
Subject: [PATCH 1/3] f2fs crypto: allocate buffer for decrypting filename

We got dentry pages from high_mem, and its address space directly goes into the
decryption path via f2fs_fname_disk_to_usr.
But, sg_init_one assumes the address is not from high_mem, so we can get this
panic since it doesn't call kmap_high but kunmap_high is triggered at the end.

kernel BUG at ../../../../../../kernel/mm/highmem.c:290!
Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
...
 (kunmap_high+0xb0/0xb8) from [<c0114534>] (__kunmap_atomic+0xa0/0xa4)
 (__kunmap_atomic+0xa0/0xa4) from [<c035f028>] (blkcipher_walk_done+0x128/0x1ec)
 (blkcipher_walk_done+0x128/0x1ec) from [<c0366c24>] (crypto_cbc_decrypt+0xc0/0x170)
 (crypto_cbc_decrypt+0xc0/0x170) from [<c0367148>] (crypto_cts_decrypt+0xc0/0x114)
 (crypto_cts_decrypt+0xc0/0x114) from [<c035ea98>] (async_decrypt+0x40/0x48)
 (async_decrypt+0x40/0x48) from [<c032ca34>] (f2fs_fname_disk_to_usr+0x124/0x304)
 (f2fs_fname_disk_to_usr+0x124/0x304) from [<c03056fc>] (f2fs_fill_dentries+0xac/0x188)
 (f2fs_fill_dentries+0xac/0x188) from [<c03059c8>] (f2fs_readdir+0x1f0/0x300)
 (f2fs_readdir+0x1f0/0x300) from [<c0218054>] (vfs_readdir+0x90/0xb4)
 (vfs_readdir+0x90/0xb4) from [<c0218418>] (SyS_getdents64+0x64/0xcc)
 (SyS_getdents64+0x64/0xcc) from [<c0105ba0>] (ret_fast_syscall+0x0/0x30)

Cc: <stable@vger.kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/dir.c   | 14 +++++++++++---
 fs/f2fs/namei.c |  8 +++++++-
 2 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 8f15fc1..cce512c 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -773,6 +773,7 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 	unsigned int bit_pos;
 	struct f2fs_dir_entry *de = NULL;
 	struct f2fs_str de_name = FSTR_INIT(NULL, 0);
+	char *name = NULL;
 
 	bit_pos = ((unsigned long)ctx->pos % d->max);
 
@@ -788,8 +789,10 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 			d_type = DT_UNKNOWN;
 
 		/* encrypted case */
-		de_name.name = d->filename[bit_pos];
 		de_name.len = le16_to_cpu(de->name_len);
+		name = kmalloc(de_name.len, GFP_NOFS);
+		memcpy(name, d->filename[bit_pos], de_name.len);
+		de_name.name = name;
 
 		if (f2fs_encrypted_inode(d->inode)) {
 			int save_len = fstr->len;
@@ -799,13 +802,18 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 							&de_name, fstr);
 			de_name = *fstr;
 			fstr->len = save_len;
-			if (ret < 0)
+			if (ret < 0) {
+				kfree(name);
 				return true;
+			}
 		}
 
 		if (!dir_emit(ctx, de_name.name, de_name.len,
-					le32_to_cpu(de->ino), d_type))
+					le32_to_cpu(de->ino), d_type)) {
+			kfree(name);
 			return true;
+		}
+		kfree(name);
 
 		bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len));
 		ctx->pos = start_pos + bit_pos;
diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index a680bf3..ebd612e 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -933,6 +933,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 	struct f2fs_encrypted_symlink_data *sd;
 	loff_t size = min_t(loff_t, i_size_read(inode), PAGE_SIZE - 1);
 	u32 max_size = inode->i_sb->s_blocksize;
+	char *name = NULL;
 	int res;
 
 	res = f2fs_get_encryption_info(inode);
@@ -947,8 +948,10 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 
 	/* Symlink is encrypted */
 	sd = (struct f2fs_encrypted_symlink_data *)caddr;
-	cstr.name = sd->encrypted_path;
 	cstr.len = le16_to_cpu(sd->len);
+	name = kmalloc(cstr.len, GFP_NOFS);
+	memcpy(name, sd->encrypted_path, cstr.len);
+	cstr.name = name;
 
 	/* this is broken symlink case */
 	if (cstr.name[0] == 0 && cstr.len == 0) {
@@ -970,6 +973,8 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 	if (res < 0)
 		goto errout;
 
+	kfree(name);
+
 	paddr = pstr.name;
 
 	/* Null-terminate the name */
@@ -979,6 +984,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 	page_cache_release(cpage);
 	return *cookie = paddr;
 errout:
+	kfree(name);
 	f2fs_fname_crypto_free_buffer(&pstr);
 	kunmap(cpage);
 	page_cache_release(cpage);
-- 
2.1.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1225961 — RE: [f2fs-dev] [PATCH v3] f2fs crypto: allocate buffer for decrypting filename

FromChao Yu <chao2.yu@samsung.com>
Date2015-09-16 12:30 +0200
SubjectRE: [f2fs-dev] [PATCH v3] f2fs crypto: allocate buffer for decrypting filename
Message-ID<q9i8q-1Fl-5@gated-at.bofh.it>
In reply to#1225442
Hi Jaegeuk,

> -----Original Message-----
> From: Jaegeuk Kim [mailto:jaegeuk@kernel.org]
> Sent: Wednesday, September 16, 2015 12:52 AM
> To: linux-kernel@vger.kernel.org; linux-fsdevel@vger.kernel.org;
> linux-f2fs-devel@lists.sourceforge.net
> Cc: stable@vger.kernel.org
> Subject: Re: [f2fs-dev] [PATCH v3] f2fs crypto: allocate buffer for decrypting filename
> 
> Change log from v1:
>   o fix wrong pointer assignment
> 
> Chang log from v2:
>   o add one more missing call path: f2fs_encrypted_follow_link
> 
> >From 84574dd5c3e8ed9ca9fdfcbd251b354cdbc5ecab Mon Sep 17 00:00:00 2001
> From: Jaegeuk Kim <jaegeuk@kernel.org>
> Date: Thu, 3 Sep 2015 13:38:23 -0700
> Subject: [PATCH 1/3] f2fs crypto: allocate buffer for decrypting filename
> 
> We got dentry pages from high_mem, and its address space directly goes into the
> decryption path via f2fs_fname_disk_to_usr.
> But, sg_init_one assumes the address is not from high_mem, so we can get this
> panic since it doesn't call kmap_high but kunmap_high is triggered at the end.
> 
> kernel BUG at ../../../../../../kernel/mm/highmem.c:290!
> Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
> ...
>  (kunmap_high+0xb0/0xb8) from [<c0114534>] (__kunmap_atomic+0xa0/0xa4)
>  (__kunmap_atomic+0xa0/0xa4) from [<c035f028>] (blkcipher_walk_done+0x128/0x1ec)
>  (blkcipher_walk_done+0x128/0x1ec) from [<c0366c24>] (crypto_cbc_decrypt+0xc0/0x170)
>  (crypto_cbc_decrypt+0xc0/0x170) from [<c0367148>] (crypto_cts_decrypt+0xc0/0x114)
>  (crypto_cts_decrypt+0xc0/0x114) from [<c035ea98>] (async_decrypt+0x40/0x48)
>  (async_decrypt+0x40/0x48) from [<c032ca34>] (f2fs_fname_disk_to_usr+0x124/0x304)
>  (f2fs_fname_disk_to_usr+0x124/0x304) from [<c03056fc>] (f2fs_fill_dentries+0xac/0x188)
>  (f2fs_fill_dentries+0xac/0x188) from [<c03059c8>] (f2fs_readdir+0x1f0/0x300)
>  (f2fs_readdir+0x1f0/0x300) from [<c0218054>] (vfs_readdir+0x90/0xb4)
>  (vfs_readdir+0x90/0xb4) from [<c0218418>] (SyS_getdents64+0x64/0xcc)
>  (SyS_getdents64+0x64/0xcc) from [<c0105ba0>] (ret_fast_syscall+0x0/0x30)
> 
> Cc: <stable@vger.kernel.org>
> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
> ---
>  fs/f2fs/dir.c   | 14 +++++++++++---
>  fs/f2fs/namei.c |  8 +++++++-
>  2 files changed, 18 insertions(+), 4 deletions(-)
> 
> diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
> index 8f15fc1..cce512c 100644
> --- a/fs/f2fs/dir.c
> +++ b/fs/f2fs/dir.c
> @@ -773,6 +773,7 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr
> *d,
>  	unsigned int bit_pos;
>  	struct f2fs_dir_entry *de = NULL;
>  	struct f2fs_str de_name = FSTR_INIT(NULL, 0);
> +	char *name = NULL;
> 
>  	bit_pos = ((unsigned long)ctx->pos % d->max);
> 
> @@ -788,8 +789,10 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr
> *d,
>  			d_type = DT_UNKNOWN;
> 
>  		/* encrypted case */
> -		de_name.name = d->filename[bit_pos];
>  		de_name.len = le16_to_cpu(de->name_len);
> +		name = kmalloc(de_name.len, GFP_NOFS);

How do you think of handling the failure of kmalloc with GFP_NOFS?

> +		memcpy(name, d->filename[bit_pos], de_name.len);

If current inode is not encrypted, our kmalloc & memcpy will be overhead,
How about changing our codes to avoid that?

Thanks,


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1226332 — Re: [f2fs-dev] [PATCH v4] f2fs crypto: allocate buffer for decrypting filename

FromJaegeuk Kim <jaegeuk@kernel.org>
Date2015-09-16 20:00 +0200
SubjectRe: [f2fs-dev] [PATCH v4] f2fs crypto: allocate buffer for decrypting filename
Message-ID<q9p9V-3ca-37@gated-at.bofh.it>
In reply to#1225961
Thanks Chao,

Change log from v3:
  o enhance the code with proper error handling

Change log from v1:
  o fix wrong pointer assignment
 
Chang log from v2:
  o add one more missing call path: f2fs_encrypted_follow_link

From efdbbbb4c9bb6519e3c10ec5081a2dcdbaca81d2 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Thu, 3 Sep 2015 13:38:23 -0700
Subject: [PATCH] f2fs crypto: allocate buffer for decrypting filename

We got dentry pages from high_mem, and its address space directly goes into the
decryption path via f2fs_fname_disk_to_usr.
But, sg_init_one assumes the address is not from high_mem, so we can get this
panic since it doesn't call kmap_high but kunmap_high is triggered at the end.

kernel BUG at ../../../../../../kernel/mm/highmem.c:290!
Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
...
 (kunmap_high+0xb0/0xb8) from [<c0114534>] (__kunmap_atomic+0xa0/0xa4)
 (__kunmap_atomic+0xa0/0xa4) from [<c035f028>] (blkcipher_walk_done+0x128/0x1ec)
 (blkcipher_walk_done+0x128/0x1ec) from [<c0366c24>] (crypto_cbc_decrypt+0xc0/0x170)
 (crypto_cbc_decrypt+0xc0/0x170) from [<c0367148>] (crypto_cts_decrypt+0xc0/0x114)
 (crypto_cts_decrypt+0xc0/0x114) from [<c035ea98>] (async_decrypt+0x40/0x48)
 (async_decrypt+0x40/0x48) from [<c032ca34>] (f2fs_fname_disk_to_usr+0x124/0x304)
 (f2fs_fname_disk_to_usr+0x124/0x304) from [<c03056fc>] (f2fs_fill_dentries+0xac/0x188)
 (f2fs_fill_dentries+0xac/0x188) from [<c03059c8>] (f2fs_readdir+0x1f0/0x300)
 (f2fs_readdir+0x1f0/0x300) from [<c0218054>] (vfs_readdir+0x90/0xb4)
 (vfs_readdir+0x90/0xb4) from [<c0218418>] (SyS_getdents64+0x64/0xcc)
 (SyS_getdents64+0x64/0xcc) from [<c0105ba0>] (ret_fast_syscall+0x0/0x30)

Cc: <stable@vger.kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/dir.c   | 13 ++++++++++---
 fs/f2fs/namei.c | 10 +++++++++-
 2 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 8f15fc1..6726c4a 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -787,7 +787,6 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 		else
 			d_type = DT_UNKNOWN;
 
-		/* encrypted case */
 		de_name.name = d->filename[bit_pos];
 		de_name.len = le16_to_cpu(de->name_len);
 
@@ -795,12 +794,20 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 			int save_len = fstr->len;
 			int ret;
 
+			de_name.name = kmalloc(de_name.len, GFP_NOFS);
+			if (!de_name.name)
+				return false;
+
+			memcpy(de_name.name, d->filename[bit_pos], de_name.len);
+
 			ret = f2fs_fname_disk_to_usr(d->inode, &de->hash_code,
 							&de_name, fstr);
-			de_name = *fstr;
-			fstr->len = save_len;
+			kfree(de_name.name);
 			if (ret < 0)
 				return true;
+
+			de_name = *fstr;
+			fstr->len = save_len;
 		}
 
 		if (!dir_emit(ctx, de_name.name, de_name.len,
diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index a680bf3..dfa01c8 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -947,8 +947,13 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 
 	/* Symlink is encrypted */
 	sd = (struct f2fs_encrypted_symlink_data *)caddr;
-	cstr.name = sd->encrypted_path;
 	cstr.len = le16_to_cpu(sd->len);
+	cstr.name = kmalloc(cstr.len, GFP_NOFS);
+	if (!cstr.name) {
+		res = -ENOMEM;
+		goto errout;
+	}
+	memcpy(cstr.name, sd->encrypted_path, cstr.len);
 
 	/* this is broken symlink case */
 	if (cstr.name[0] == 0 && cstr.len == 0) {
@@ -970,6 +975,8 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 	if (res < 0)
 		goto errout;
 
+	kfree(cstr.name);
+
 	paddr = pstr.name;
 
 	/* Null-terminate the name */
@@ -979,6 +986,7 @@ static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cook
 	page_cache_release(cpage);
 	return *cookie = paddr;
 errout:
+	kfree(cstr.name);
 	f2fs_fname_crypto_free_buffer(&pstr);
 	kunmap(cpage);
 	page_cache_release(cpage);
-- 
2.1.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1226960 — RE: [f2fs-dev] [PATCH v4] f2fs crypto: allocate buffer for decrypting filename

FromChao Yu <chao2.yu@samsung.com>
Date2015-09-17 14:30 +0200
SubjectRE: [f2fs-dev] [PATCH v4] f2fs crypto: allocate buffer for decrypting filename
Message-ID<q9Gu6-3rD-7@gated-at.bofh.it>
In reply to#1226332
> -----Original Message-----
> From: Jaegeuk Kim [mailto:jaegeuk@kernel.org]
> Sent: Thursday, September 17, 2015 1:57 AM
> To: Chao Yu
> Cc: linux-kernel@vger.kernel.org; linux-fsdevel@vger.kernel.org;
> linux-f2fs-devel@lists.sourceforge.net; stable@vger.kernel.org
> Subject: Re: [f2fs-dev] [PATCH v4] f2fs crypto: allocate buffer for decrypting filename
> 
> Thanks Chao,
> 
> Change log from v3:
>   o enhance the code with proper error handling
> 
> Change log from v1:
>   o fix wrong pointer assignment
> 
> Chang log from v2:
>   o add one more missing call path: f2fs_encrypted_follow_link
> 
> From efdbbbb4c9bb6519e3c10ec5081a2dcdbaca81d2 Mon Sep 17 00:00:00 2001
> From: Jaegeuk Kim <jaegeuk@kernel.org>
> Date: Thu, 3 Sep 2015 13:38:23 -0700
> Subject: [PATCH] f2fs crypto: allocate buffer for decrypting filename
> 
> We got dentry pages from high_mem, and its address space directly goes into the
> decryption path via f2fs_fname_disk_to_usr.
> But, sg_init_one assumes the address is not from high_mem, so we can get this
> panic since it doesn't call kmap_high but kunmap_high is triggered at the end.
> 
> kernel BUG at ../../../../../../kernel/mm/highmem.c:290!
> Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
> ...
>  (kunmap_high+0xb0/0xb8) from [<c0114534>] (__kunmap_atomic+0xa0/0xa4)
>  (__kunmap_atomic+0xa0/0xa4) from [<c035f028>] (blkcipher_walk_done+0x128/0x1ec)
>  (blkcipher_walk_done+0x128/0x1ec) from [<c0366c24>] (crypto_cbc_decrypt+0xc0/0x170)
>  (crypto_cbc_decrypt+0xc0/0x170) from [<c0367148>] (crypto_cts_decrypt+0xc0/0x114)
>  (crypto_cts_decrypt+0xc0/0x114) from [<c035ea98>] (async_decrypt+0x40/0x48)
>  (async_decrypt+0x40/0x48) from [<c032ca34>] (f2fs_fname_disk_to_usr+0x124/0x304)
>  (f2fs_fname_disk_to_usr+0x124/0x304) from [<c03056fc>] (f2fs_fill_dentries+0xac/0x188)
>  (f2fs_fill_dentries+0xac/0x188) from [<c03059c8>] (f2fs_readdir+0x1f0/0x300)
>  (f2fs_readdir+0x1f0/0x300) from [<c0218054>] (vfs_readdir+0x90/0xb4)
>  (vfs_readdir+0x90/0xb4) from [<c0218418>] (SyS_getdents64+0x64/0xcc)
>  (SyS_getdents64+0x64/0xcc) from [<c0105ba0>] (ret_fast_syscall+0x0/0x30)
> 
> Cc: <stable@vger.kernel.org>
> Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

Reviewed-by: Chao Yu <chao2.yu@samsung.com>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web