From: Julien Grall <jgrall@amazon.com>

Currently, we are ignoring any error from perform_gunzip() and replacing
the compressed kernel with the "uncompressed" kernel.

If there is a gzip failure, then it means that the output buffer may
contain garbagge. So it can result to various sort of behavior that may
be difficult to root cause.

In case of failure, free the output buffer and propagate the error.
We also need to adjust the return check for kernel_compress() as
perform_gunzip() may return a positive value.

Take the opportunity to adjust the code style for the check.

Signed-off-by: Julien Grall <jgrall@amazon.com>

---
    Changes in v2:
        - Fix build
---
 xen/arch/arm/kernel.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/xen/arch/arm/kernel.c b/xen/arch/arm/kernel.c
index ab78689ed2a6..8f43caa1866d 100644
--- a/xen/arch/arm/kernel.c
+++ b/xen/arch/arm/kernel.c
@@ -292,6 +292,12 @@ static __init int kernel_decompress(struct bootmodule *mod)
     iounmap(input);
     vunmap(output);
 
+    if ( rc )
+    {
+        free_domheap_pages(pages, kernel_order_out);
+        return rc;
+    }
+
     mod->start = page_to_maddr(pages);
     mod->size = output_size;
 
@@ -503,7 +509,7 @@ int __init kernel_probe(struct kernel_info *info,
 
     /* if it is a gzip'ed image, 32bit or 64bit, uncompress it */
     rc = kernel_decompress(mod);
-    if (rc < 0 && rc != -EINVAL)
+    if ( rc && rc != -EINVAL )
         return rc;
 
 #ifdef CONFIG_ARM_64
-- 
2.17.1

On 06.04.2021 21:15, Julien Grall wrote:
> From: Julien Grall <jgrall@amazon.com>
> 
> Currently, we are ignoring any error from perform_gunzip() and replacing
> the compressed kernel with the "uncompressed" kernel.
> 
> If there is a gzip failure, then it means that the output buffer may
> contain garbagge. So it can result to various sort of behavior that may
> be difficult to root cause.
> 
> In case of failure, free the output buffer and propagate the error.
> We also need to adjust the return check for kernel_compress() as
> perform_gunzip() may return a positive value.
> 
> Take the opportunity to adjust the code style for the check.
> 
> Signed-off-by: Julien Grall <jgrall@amazon.com>
> 
> ---
>     Changes in v2:
>         - Fix build
> ---

Reviewed-by: Michal Orzel <michal.orzel@arm.com>

Hi Michal,

On 12/04/2021 07:45, Michal Orzel wrote:
> On 06.04.2021 21:15, Julien Grall wrote:
>> From: Julien Grall <jgrall@amazon.com>
>>
>> Currently, we are ignoring any error from perform_gunzip() and replacing
>> the compressed kernel with the "uncompressed" kernel.
>>
>> If there is a gzip failure, then it means that the output buffer may
>> contain garbagge. So it can result to various sort of behavior that may
>> be difficult to root cause.
>>
>> In case of failure, free the output buffer and propagate the error.
>> We also need to adjust the return check for kernel_compress() as
>> perform_gunzip() may return a positive value.
>>
>> Take the opportunity to adjust the code style for the check.
>>
>> Signed-off-by: Julien Grall <jgrall@amazon.com>
>>
>> ---
>>      Changes in v2:
>>          - Fix build
>> ---
> 
> Reviewed-by: Michal Orzel <michal.orzel@arm.com>

Thanks! @Stefano, can I get your acked-by?

Cheers,

-- 
Julien Grall

On 18/04/2021 19:26, Julien Grall wrote:
> On 12/04/2021 07:45, Michal Orzel wrote:
>> On 06.04.2021 21:15, Julien Grall wrote:
>>> From: Julien Grall <jgrall@amazon.com>
>>>
>>> Currently, we are ignoring any error from perform_gunzip() and replacing
>>> the compressed kernel with the "uncompressed" kernel.
>>>
>>> If there is a gzip failure, then it means that the output buffer may
>>> contain garbagge. So it can result to various sort of behavior that may
>>> be difficult to root cause.
>>>
>>> In case of failure, free the output buffer and propagate the error.
>>> We also need to adjust the return check for kernel_compress() as
>>> perform_gunzip() may return a positive value.
>>>
>>> Take the opportunity to adjust the code style for the check.
>>>
>>> Signed-off-by: Julien Grall <jgrall@amazon.com>
>>>
>>> ---
>>>      Changes in v2:
>>>          - Fix build
>>> ---
>>
>> Reviewed-by: Michal Orzel <michal.orzel@arm.com>
> 
> Thanks! @Stefano, can I get your acked-by?

Ping? I intend to commit it on Wednesday unless I hear otherwise.

Cheers,

-- 
Julien Grall