2023-06-05 16:34:16

by Mirsad Todorovac

Subject: [PATCH v1 1/1] test_firmware: return ENOMEM instead of ENOSPC on failed allocation

In a couple of situations like:

name = kstrndup(buf, count, GFP_KERNEL);
if (!name)
return -ENOSPC;

the error is not actually "No space left on device", but "Out of memory".

So, it is semantically correct to return -ENOMEM in all failed kstrndup()
and kzalloc() cases in this driver, as it is not a problem with disk
space, but with kernel memory allocator.

The semantically correct should be:

name = kstrndup(buf, count, GFP_KERNEL);
if (!name)
return -ENOMEM;

Cc: Dan Carpenter <[email protected]>
Cc: Takashi Iwai <[email protected]>
Cc: Kees Cook <[email protected]>
Cc: [email protected]
Signed-off-by: Mirsad Goran Todorovac <[email protected]>
---
lib/test_firmware.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/lib/test_firmware.c b/lib/test_firmware.c
index 1d7d480b8eeb..add4699fc6cd 100644
--- a/lib/test_firmware.c
+++ b/lib/test_firmware.c
@@ -214,7 +214,7 @@ static int __kstrncpy(char **dst, const char *name, size_t count, gfp_t gfp)
{
*dst = kstrndup(name, count, gfp);
if (!*dst)
- return -ENOSPC;
+ return -ENOMEM;
return count;
}

@@ -671,7 +671,7 @@ static ssize_t trigger_request_store(struct device *dev,

name = kstrndup(buf, count, GFP_KERNEL);
if (!name)
- return -ENOSPC;
+ return -ENOMEM;

pr_info("loading '%s'\n", name);

@@ -719,7 +719,7 @@ static ssize_t trigger_request_platform_store(struct device *dev,

name = kstrndup(buf, count, GFP_KERNEL);
if (!name)
- return -ENOSPC;
+ return -ENOMEM;

pr_info("inserting test platform fw '%s'\n", name);
efi_embedded_fw.name = name;
@@ -772,7 +772,7 @@ static ssize_t trigger_async_request_store(struct device *dev,

name = kstrndup(buf, count, GFP_KERNEL);
if (!name)
- return -ENOSPC;
+ return -ENOMEM;

pr_info("loading '%s'\n", name);

@@ -817,7 +817,7 @@ static ssize_t trigger_custom_fallback_store(struct device *dev,

name = kstrndup(buf, count, GFP_KERNEL);
if (!name)
- return -ENOSPC;
+ return -ENOMEM;

pr_info("loading '%s' using custom fallback mechanism\n", name);

@@ -868,7 +868,7 @@ static int test_fw_run_batch_request(void *data)

test_buf = kzalloc(TEST_FIRMWARE_BUF_SIZE, GFP_KERNEL);
if (!test_buf)
- return -ENOSPC;
+ return -ENOMEM;

if (test_fw_config->partial)
req->rc = request_partial_firmware_into_buf
--
2.34.1

2023-06-06 05:54:23

by Dan Carpenter

Subject: Re: [PATCH v1 1/1] test_firmware: return ENOMEM instead of ENOSPC on failed allocation

On Mon, Jun 05, 2023 at 06:27:47PM +0200, Mirsad Goran Todorovac wrote:
> In a couple of situations like:
>
> name = kstrndup(buf, count, GFP_KERNEL);
> if (!name)
> return -ENOSPC;
>
> the error is not actually "No space left on device", but "Out of memory".
>
> So, it is semantically correct to return -ENOMEM in all failed kstrndup()
> and kzalloc() cases in this driver, as it is not a problem with disk
> space, but with kernel memory allocator.
>
> The semantically correct should be:
>
> name = kstrndup(buf, count, GFP_KERNEL);
> if (!name)
> return -ENOMEM;
>
> Cc: Dan Carpenter <[email protected]>
> Cc: Takashi Iwai <[email protected]>
> Cc: Kees Cook <[email protected]>
> Cc: [email protected]
> Signed-off-by: Mirsad Goran Todorovac <[email protected]>
> ---

The Cc stable might be a little bit much... No Fixes tag either. But
otherwise it looks fine.

regards,
dan carpenter

2023-06-06 07:17:46

by Mirsad Todorovac

Subject: Re: [PATCH v1 1/1] test_firmware: return ENOMEM instead of ENOSPC on failed allocation

On 6/6/23 07:37, Dan Carpenter wrote:
> On Mon, Jun 05, 2023 at 06:27:47PM +0200, Mirsad Goran Todorovac wrote:
>> In a couple of situations like:
>>
>> name = kstrndup(buf, count, GFP_KERNEL);
>> if (!name)
>> return -ENOSPC;
>>
>> the error is not actually "No space left on device", but "Out of memory".
>>
>> So, it is semantically correct to return -ENOMEM in all failed kstrndup()
>> and kzalloc() cases in this driver, as it is not a problem with disk
>> space, but with kernel memory allocator.
>>
>> The semantically correct should be:
>>
>> name = kstrndup(buf, count, GFP_KERNEL);
>> if (!name)
>> return -ENOMEM;
>>
>> Cc: Dan Carpenter <[email protected]>
>> Cc: Takashi Iwai <[email protected]>
>> Cc: Kees Cook <[email protected]>
>> Cc: [email protected]
>> Signed-off-by: Mirsad Goran Todorovac <[email protected]>
>> ---
>
> The Cc stable might be a little bit much... No Fixes tag either. But
> otherwise it looks fine.

Thank you. I will fix these and resubmit.

Best regards,
Mirsad Todorovac

> regards,
> dan carpenter