Re: [PATCH AUTOSEL 4.14 05/14] binder: take read mode of mmap_sem in binder_alloc_free_page()

[Tyler Hicks <tyhicks@canonical.com>]

Hello, Sasha!

On 2019-08-29 06:50:34, Sasha Levin wrote:
> From: Tyler Hicks <tyhicks@canonical.com>
> 
> [ Upstream commit 60d4885710836595192c42d3e04b27551d30ec91 ]
> 
> Restore the behavior of locking mmap_sem for reading in
> binder_alloc_free_page(), as was first done in commit 3013bf62b67a
> ("binder: reduce mmap_sem write-side lock"). That change was
> inadvertently reverted by commit 5cec2d2e5839 ("binder: fix race between
> munmap() and direct reclaim").
> 
> In addition, change the name of the label for the error path to
> accurately reflect that we're taking the lock for reading.
> 
> Backporting note: This fix is only needed when *both* of the commits
> mentioned above are applied. That's an unlikely situation since they
> both landed during the development of v5.1 but only one of them is
> targeted for stable.

This patch isn't meant to be applied to 4.14 since commit 3013bf62b67a
("binder: reduce mmap_sem write-side lock") was never brought back to
4.14.

My backporting note above isn't helpful for AUTOSEL purposes. Do you
have a suggestion for what I could have done in the patch tags to convey
that guidance to AUTOSEL?

Tyler

> 
> Fixes: 5cec2d2e5839 ("binder: fix race between munmap() and direct reclaim")
> Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
> Acked-by: Todd Kjos <tkjos@android.com>
> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Signed-off-by: Sasha Levin <sashal@kernel.org>
> ---
>  drivers/android/binder_alloc.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c
> index e0b0399ff7ec8..81c67459259ec 100644
> --- a/drivers/android/binder_alloc.c
> +++ b/drivers/android/binder_alloc.c
> @@ -949,8 +949,8 @@ enum lru_status binder_alloc_free_page(struct list_head *item,
>  	mm = alloc->vma_vm_mm;
>  	if (!mmget_not_zero(mm))
>  		goto err_mmget;
> -	if (!down_write_trylock(&mm->mmap_sem))
> -		goto err_down_write_mmap_sem_failed;
> +	if (!down_read_trylock(&mm->mmap_sem))
> +		goto err_down_read_mmap_sem_failed;
>  	vma = binder_alloc_get_vma(alloc);
>  
>  	list_lru_isolate(lru, item);
> @@ -965,7 +965,7 @@ enum lru_status binder_alloc_free_page(struct list_head *item,
>  
>  		trace_binder_unmap_user_end(alloc, index);
>  	}
> -	up_write(&mm->mmap_sem);
> +	up_read(&mm->mmap_sem);
>  	mmput(mm);
>  
>  	trace_binder_unmap_kernel_start(alloc, index);
> @@ -980,7 +980,7 @@ enum lru_status binder_alloc_free_page(struct list_head *item,
>  	mutex_unlock(&alloc->mutex);
>  	return LRU_REMOVED_RETRY;
>  
> -err_down_write_mmap_sem_failed:
> +err_down_read_mmap_sem_failed:
>  	mmput_async(mm);
>  err_mmget:
>  err_page_already_freed:
> -- 
> 2.20.1
>