[v1] crypto: aspeed: add error handling if dmam_alloc_coherent() failed

[PATCH] crypto: aspeed: add error handling if dmam_alloc_coherent() failed

Posted by Neal Liu 2 years, 6 months ago

Since the acry_dev->buf_addr may be NULL, add error handling to
prevent any additional access to avoid potential issues.

Signed-off-by: Neal Liu <neal_liu@aspeedtech.com>
---
 drivers/crypto/aspeed/aspeed-acry.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/crypto/aspeed/aspeed-acry.c b/drivers/crypto/aspeed/aspeed-acry.c
index 1f77ebd73489..85733e418c9e 100644
--- a/drivers/crypto/aspeed/aspeed-acry.c
+++ b/drivers/crypto/aspeed/aspeed-acry.c
@@ -782,6 +782,11 @@ static int aspeed_acry_probe(struct platform_device *pdev)
 	acry_dev->buf_addr = dmam_alloc_coherent(dev, ASPEED_ACRY_BUFF_SIZE,
 						 &acry_dev->buf_dma_addr,
 						 GFP_KERNEL);
+	if (!acry_dev->buf_addr) {
+		rc = -ENOMEM;
+		goto err_engine_rsa_start;
+	}
+
 	memzero_explicit(acry_dev->buf_addr, ASPEED_ACRY_BUFF_SIZE);
 
 	aspeed_acry_register(acry_dev);
-- 
2.25.1

Re: [PATCH] crypto: aspeed: add error handling if dmam_alloc_coherent() failed

Posted by Herbert Xu 2 years, 6 months ago

On Thu, Feb 23, 2023 at 04:58:30PM +0800, Neal Liu wrote:
> Since the acry_dev->buf_addr may be NULL, add error handling to
> prevent any additional access to avoid potential issues.
> 
> Signed-off-by: Neal Liu <neal_liu@aspeedtech.com>
> ---
>  drivers/crypto/aspeed/aspeed-acry.c | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/drivers/crypto/aspeed/aspeed-acry.c b/drivers/crypto/aspeed/aspeed-acry.c
> index 1f77ebd73489..85733e418c9e 100644
> --- a/drivers/crypto/aspeed/aspeed-acry.c
> +++ b/drivers/crypto/aspeed/aspeed-acry.c
> @@ -782,6 +782,11 @@ static int aspeed_acry_probe(struct platform_device *pdev)
>  	acry_dev->buf_addr = dmam_alloc_coherent(dev, ASPEED_ACRY_BUFF_SIZE,
>  						 &acry_dev->buf_dma_addr,
>  						 GFP_KERNEL);
> +	if (!acry_dev->buf_addr) {
> +		rc = -ENOMEM;
> +		goto err_engine_rsa_start;
> +	}
> +
>  	memzero_explicit(acry_dev->buf_addr, ASPEED_ACRY_BUFF_SIZE);

Please remove this memzero in a follow-up patch as
dmam_alloc_coherent returns memory that's already zeroed.

Thanks,
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

RE: [PATCH] crypto: aspeed: add error handling if dmam_alloc_coherent() failed

Posted by Neal Liu 2 years, 6 months ago

> On Thu, Feb 23, 2023 at 04:58:30PM +0800, Neal Liu wrote:
> > Since the acry_dev->buf_addr may be NULL, add error handling to
> > prevent any additional access to avoid potential issues.
> >
> > Signed-off-by: Neal Liu <neal_liu@aspeedtech.com>
> > ---
> >  drivers/crypto/aspeed/aspeed-acry.c | 5 +++++
> >  1 file changed, 5 insertions(+)
> >
> > diff --git a/drivers/crypto/aspeed/aspeed-acry.c
> > b/drivers/crypto/aspeed/aspeed-acry.c
> > index 1f77ebd73489..85733e418c9e 100644
> > --- a/drivers/crypto/aspeed/aspeed-acry.c
> > +++ b/drivers/crypto/aspeed/aspeed-acry.c
> > @@ -782,6 +782,11 @@ static int aspeed_acry_probe(struct
> platform_device *pdev)
> >  	acry_dev->buf_addr = dmam_alloc_coherent(dev,
> ASPEED_ACRY_BUFF_SIZE,
> >  						 &acry_dev->buf_dma_addr,
> >  						 GFP_KERNEL);
> > +	if (!acry_dev->buf_addr) {
> > +		rc = -ENOMEM;
> > +		goto err_engine_rsa_start;
> > +	}
> > +
> >  	memzero_explicit(acry_dev->buf_addr, ASPEED_ACRY_BUFF_SIZE);
> 
> Please remove this memzero in a follow-up patch as dmam_alloc_coherent
> returns memory that's already zeroed.
> 
> Thanks,

Okay, thanks for the information.