[PATCH][next] btrfs: fix memory leak of path on error return path

[PATCH][next] btrfs: fix memory leak of path on error return path

[Colin King]

From: Colin Ian King <colin.king@canonical.com>

Currently if the allocation of roots or tmp_ulist fails the error handling
does not free up the allocation of path causing a memory leak. Fix this by
freeing path with a call to btrfs_free_path before taking the error return
path.

Addresses-Coverity: ("Resource leak")
Fixes: 5911c8fe05c5 ("btrfs: fiemap: preallocate ulists for btrfs_check_shared")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
---
 fs/btrfs/extent_io.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 1eb671c16ff1..d7f37a33d597 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -4600,6 +4600,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	tmp_ulist = ulist_alloc(GFP_KERNEL);
 	if (!roots || !tmp_ulist) {
 		ret = -ENOMEM;
+		btrfs_free_path(path);
 		goto out_free_ulist;
 	}
 
-- 
2.20.1

Re: [PATCH][next] btrfs: fix memory leak of path on error return path

[David Sterba]

On Tue, Jul 02, 2019 at 03:10:28PM +0100, Colin King wrote:
> From: Colin Ian King <colin.king@canonical.com>
> 
> Currently if the allocation of roots or tmp_ulist fails the error handling
> does not free up the allocation of path causing a memory leak. Fix this by
> freeing path with a call to btrfs_free_path before taking the error return
> path.
> 
> Addresses-Coverity: ("Resource leak")

Does this have an id, that coverity uses?

> Fixes: 5911c8fe05c5 ("btrfs: fiemap: preallocate ulists for btrfs_check_shared")
> Signed-off-by: Colin Ian King <colin.king@canonical.com>
> ---
>  fs/btrfs/extent_io.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
> index 1eb671c16ff1..d7f37a33d597 100644
> --- a/fs/btrfs/extent_io.c
> +++ b/fs/btrfs/extent_io.c
> @@ -4600,6 +4600,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
>  	tmp_ulist = ulist_alloc(GFP_KERNEL);
>  	if (!roots || !tmp_ulist) {
>  		ret = -ENOMEM;
> +		btrfs_free_path(path);

This fixes only one leak, therere are more that I spotted while
reviewing this patch. The gotos from the while-loop jump to
out_free_list but that leave the path behind.

That's why the exit block is a better place for the cleanups. This
requires proper nesting of the cleanup calls, that's slightly
inconvenient in this case. The free_path is before call to
unlock_extent_cached so when the ordre is switched and free_path moved
to out_free_ulist, then all the leaks are addressed in one go.

Bummer that the leaks escaped sight of original patch author (me), 2
reviewers and now 1 fix reviewer.

Re: [PATCH][next] btrfs: fix memory leak of path on error return path

[Colin Ian King]

On 04/07/2019 17:37, David Sterba wrote:
> On Tue, Jul 02, 2019 at 03:10:28PM +0100, Colin King wrote:
>> From: Colin Ian King <colin.king@canonical.com>
>>
>> Currently if the allocation of roots or tmp_ulist fails the error handling
>> does not free up the allocation of path causing a memory leak. Fix this by
>> freeing path with a call to btrfs_free_path before taking the error return
>> path.
>>
>> Addresses-Coverity: ("Resource leak")
> 
> Does this have an id, that coverity uses?

Not a public one that I know of. These are based on private scans I run
in Canonical.

> 
>> Fixes: 5911c8fe05c5 ("btrfs: fiemap: preallocate ulists for btrfs_check_shared")
>> Signed-off-by: Colin Ian King <colin.king@canonical.com>
>> ---
>>  fs/btrfs/extent_io.c | 1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
>> index 1eb671c16ff1..d7f37a33d597 100644
>> --- a/fs/btrfs/extent_io.c
>> +++ b/fs/btrfs/extent_io.c
>> @@ -4600,6 +4600,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
>>  	tmp_ulist = ulist_alloc(GFP_KERNEL);
>>  	if (!roots || !tmp_ulist) {
>>  		ret = -ENOMEM;
>> +		btrfs_free_path(path);
> 
> This fixes only one leak, therere are more that I spotted while
> reviewing this patch. The gotos from the while-loop jump to
> out_free_list but that leave the path behind>
> That's why the exit block is a better place for the cleanups. This
> requires proper nesting of the cleanup calls, that's slightly
> inconvenient in this case. The free_path is before call to
> unlock_extent_cached so when the ordre is switched and free_path moved
> to out_free_ulist, then all the leaks are addressed in one go.

Oh, yes. Even static analysis missed that too!

> 
> Bummer that the leaks escaped sight of original patch author (me), 2
> reviewers and now 1 fix reviewer.
> 
Given that you can see more issues, I'll leave the fix in your capable
hands.

Colin

Re: [PATCH][next] btrfs: fix memory leak of path on error return path

[David Sterba]

On Thu, Jul 04, 2019 at 05:47:50PM +0100, Colin Ian King wrote:
> >>  	tmp_ulist = ulist_alloc(GFP_KERNEL);
> >>  	if (!roots || !tmp_ulist) {
> >>  		ret = -ENOMEM;
> >> +		btrfs_free_path(path);
> > 
> > This fixes only one leak, therere are more that I spotted while
> > reviewing this patch. The gotos from the while-loop jump to
> > out_free_list but that leave the path behind>
> > That's why the exit block is a better place for the cleanups. This
> > requires proper nesting of the cleanup calls, that's slightly
> > inconvenient in this case. The free_path is before call to
> > unlock_extent_cached so when the ordre is switched and free_path moved
> > to out_free_ulist, then all the leaks are addressed in one go.
> 
> Oh, yes. Even static analysis missed that too!
> 
> > Bummer that the leaks escaped sight of original patch author (me), 2
> > reviewers and now 1 fix reviewer.
> > 
> Given that you can see more issues, I'll leave the fix in your capable
> hands.

This

--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -4764,11 +4764,11 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
                ret = emit_last_fiemap_cache(fieinfo, &cache);
        free_extent_map(em);
 out:
-       btrfs_free_path(path);
        unlock_extent_cached(&BTRFS_I(inode)->io_tree, start, start + len - 1,
                             &cached_state);
 
 out_free_ulist:
+       btrfs_free_path(path);
        ulist_free(roots);
        ulist_free(tmp_ulist);
        return ret;
---

should fix it.

Re: [PATCH][next] btrfs: fix memory leak of path on error return path

[Anand Jain]

On 5/7/19 12:37 AM, David Sterba wrote:
> On Tue, Jul 02, 2019 at 03:10:28PM +0100, Colin King wrote:
>> From: Colin Ian King <colin.king@canonical.com>
>>
>> Currently if the allocation of roots or tmp_ulist fails the error handling
>> does not free up the allocation of path causing a memory leak. Fix this by
>> freeing path with a call to btrfs_free_path before taking the error return
>> path.
>>
>> Addresses-Coverity: ("Resource leak")
> 
> Does this have an id, that coverity uses?
> 
>> Fixes: 5911c8fe05c5 ("btrfs: fiemap: preallocate ulists for btrfs_check_shared")
>> Signed-off-by: Colin Ian King <colin.king@canonical.com>
>> ---
>>   fs/btrfs/extent_io.c | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
>> index 1eb671c16ff1..d7f37a33d597 100644
>> --- a/fs/btrfs/extent_io.c
>> +++ b/fs/btrfs/extent_io.c
>> @@ -4600,6 +4600,7 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
>>   	tmp_ulist = ulist_alloc(GFP_KERNEL);
>>   	if (!roots || !tmp_ulist) {
>>   		ret = -ENOMEM;
>> +		btrfs_free_path(path);
> 
> This fixes only one leak, therere are more that I spotted while
> reviewing this patch. The gotos from the while-loop jump to
> out_free_list but that leave the path behind.
> 
> That's why the exit block is a better place for the cleanups. This
> requires proper nesting of the cleanup calls, that's slightly
> inconvenient in this case. The free_path is before call to
> unlock_extent_cached so when the ordre is switched and free_path moved
> to out_free_ulist, then all the leaks are addressed in one go.
> 
> Bummer that the leaks escaped sight of original patch author (me), 2
> reviewers and now 1 fix reviewer.
> 

There is no goto out_free_ulist inside the while loop; where do you see
that? I did git pull again and checked.

4664         while (!end) {
::
4726                                 goto out_free;
::
4748                         goto out;
::
4759                         goto out_free;
::
4761         }