[PATCH 1/1] taging: fbtft: fix memory leak

[PATCH 1/1] taging: fbtft: fix memory leak

[Xidong Wang]

From: Xidong Wang <2711406067@qq.com>

In function fbtft_framebuffer_alloc(), the memory allocated by
framebuffer_alloc() is not released on the error path that txbuflen > 0
and txbuf, which holds the return value of devm_kzalloc(), is NULL.
This will result in a memory leak bug.

Signed-off-by: Xidong Wang <2711406067@qq.com>
---
 drivers/staging/fbtft/fbtft-core.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/staging/fbtft/fbtft-core.c b/drivers/staging/fbtft/fbtft-core.c
index 0e36b66..169e9dc 100644
--- a/drivers/staging/fbtft/fbtft-core.c
+++ b/drivers/staging/fbtft/fbtft-core.c
@@ -819,7 +819,7 @@ struct fb_info *fbtft_framebuffer_alloc(struct fbtft_display *display,
 	if (par->gamma.curves && gamma) {
 		if (fbtft_gamma_parse_str(par,
 			par->gamma.curves, gamma, strlen(gamma)))
-			goto alloc_fail;
+			goto err_info;
 	}
 
 	/* Transmit buffer */
@@ -872,6 +872,9 @@ struct fb_info *fbtft_framebuffer_alloc(struct fbtft_display *display,
 
 	return info;
 
+err_info:
+	framebuffer_release(info);
+
 alloc_fail:
 	vfree(vmem);
 
-- 
2.7.4

Re: [PATCH 1/1] taging: fbtft: fix memory leak

[Greg Kroah-Hartman]

On Tue, Apr 03, 2018 at 09:14:28PM +0800, Xidong Wang wrote:
> From: Xidong Wang <2711406067@qq.com>
> 
> In function fbtft_framebuffer_alloc(), the memory allocated by
> framebuffer_alloc() is not released on the error path that txbuflen > 0
> and txbuf, which holds the return value of devm_kzalloc(), is NULL.
> This will result in a memory leak bug.
> 
> Signed-off-by: Xidong Wang <2711406067@qq.com>
> ---
>  drivers/staging/fbtft/fbtft-core.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)

<formletter>

This is not the correct way to submit patches for inclusion in the
stable kernel tree.  Please read:
    https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
for how to do this properly.

</formletter>

Re: [PATCH 1/1] taging: fbtft: fix memory leak

[Dan Carpenter]

There is a typo in the subject.  It should be "Staging" instead of
"taging:".

On Tue, Apr 03, 2018 at 09:14:28PM +0800, Xidong Wang wrote:
> From: Xidong Wang <2711406067@qq.com>
> 
> In function fbtft_framebuffer_alloc(), the memory allocated by
> framebuffer_alloc() is not released on the error path that txbuflen > 0
> and txbuf, which holds the return value of devm_kzalloc(), is NULL.
> This will result in a memory leak bug.


The txbuf chunk seems to have been dropped.  You're right that it needs
to be fixed as well.

        if (txbuflen > 0) {
                txbuf = devm_kzalloc(par->info->device, txbuflen, GFP_KERNEL);
                if (!txbuf)
-                        goto alloc_fail;
+			 goto err_info;

regards,
dan carpenter

[PATCH 1/1] taging: fbtft: fix memory leak

[Xidong Wang]

From: Xidong Wang <2711406067@qq.com>

In function fbtft_framebuffer_alloc(), the memory allocated by
framebuffer_alloc() is not released on the error path that txbuflen > 0
and txbuf, which holds the return value of devm_kzalloc(), is NULL.
This will result in a memory leak bug.

Signed-off-by: Xidong Wang <2711406067@qq.com>
---
 drivers/staging/fbtft/fbtft-core.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/staging/fbtft/fbtft-core.c b/drivers/staging/fbtft/fbtft-core.c
index 0e36b66..e92771e 100644
--- a/drivers/staging/fbtft/fbtft-core.c
+++ b/drivers/staging/fbtft/fbtft-core.c
@@ -836,7 +836,7 @@ struct fb_info *fbtft_framebuffer_alloc(struct fbtft_display *display,
 	if (txbuflen > 0) {
 		txbuf = devm_kzalloc(par->info->device, txbuflen, GFP_KERNEL);
 		if (!txbuf)
-			goto alloc_fail;
+			goto err_info;
 		par->txbuf.buf = txbuf;
 		par->txbuf.len = txbuflen;
 	}
@@ -872,6 +872,9 @@ struct fb_info *fbtft_framebuffer_alloc(struct fbtft_display *display,
 
 	return info;
 
+err_info:
+	framebuffer_release(info);
+
 alloc_fail:
 	vfree(vmem);
 
-- 
2.7.4

Re: [PATCH 1/1] taging: fbtft: fix memory leak

[Joe Perches]

On Tue, 2018-04-03 at 21:33 +0800, Xidong Wang wrote:
> From: Xidong Wang <2711406067@qq.com>
> 
> In function fbtft_framebuffer_alloc(), the memory allocated by
> framebuffer_alloc() is not released on the error path that txbuflen > 0
> and txbuf, which holds the return value of devm_kzalloc(), is NULL.
> This will result in a memory leak bug.
[]
> diff --git a/drivers/staging/fbtft/fbtft-core.c b/drivers/staging/fbtft/fbtft-core.c
[]
> @@ -836,7 +836,7 @@ struct fb_info *fbtft_framebuffer_alloc(struct fbtft_display *display,
>  	if (txbuflen > 0) {
>  		txbuf = devm_kzalloc(par->info->device, txbuflen, GFP_KERNEL);
>  		if (!txbuf)
> -			goto alloc_fail;
> +			goto err_info;
>  		par->txbuf.buf = txbuf;
>  		par->txbuf.len = txbuflen;
>  	}
> @@ -872,6 +872,9 @@ struct fb_info *fbtft_framebuffer_alloc(struct fbtft_display *display,
>  
>  	return info;
>  
> +err_info:
> +	framebuffer_release(info);
> +
>  alloc_fail:
>  	vfree(vmem);

What about the

	if (par->gamma.curves && gamma) {
		if (fbtft_gamma_parse_str(par,
			par->gamma.curves, gamma, strlen(gamma)))
			goto alloc_fail;
	}

a little above this?

Presumable then it should goto err_info too.