[PATCH v3] soc: qcom: mdt_loader: Add Upperbounds check for program header access

[PATCH v3] soc: qcom: mdt_loader: Add Upperbounds check for program header access

[Auditya Bhattaram]

hash_index is evaluated by looping phdrs till QCOM_MDT_TYPE_HASH
is found. Add an upperbound check to phdrs to access within elf size.

Fixes: 64fb5eb87d58 ("soc: qcom: mdt_loader: Allow hash to reside in any segment")
Cc: <stable@vger.kernel.org>
Signed-off-by: Auditya Bhattaram <quic_audityab@quicinc.com>
---
Changes in v3:
 - Corrected wrong patch versioning in the Subject.
 - Added error prints for Invalid access.
Link to v2 https://lore.kernel.org/linux-arm-msm/9773d189-c896-d5c5-804c-e086c24987b4@quicinc.com/T/#t
Link to v1 https://lore.kernel.org/linux-arm-msm/5d7a3b97-d840-4863-91a0-32c1d8e7532f@linaro.org/T/#t
---
 drivers/soc/qcom/mdt_loader.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/soc/qcom/mdt_loader.c b/drivers/soc/qcom/mdt_loader.c
index 6f177e46fa0f..61e2377cc5c3 100644
--- a/drivers/soc/qcom/mdt_loader.c
+++ b/drivers/soc/qcom/mdt_loader.c
@@ -145,6 +145,11 @@ void *qcom_mdt_read_metadata(const struct firmware *fw, size_t *data_len,
 	if (phdrs[0].p_type == PT_LOAD)
 		return ERR_PTR(-EINVAL);

+	if (((size_t)(phdrs + ehdr->e_phnum)) > ((size_t)ehdr + fw->size)) {
+		dev_err(dev, "Invalid phdrs access: %s\n", fw_name);
+		return ERR_PTR(-EINVAL);
+	}
+
 	for (i = 1; i < ehdr->e_phnum; i++) {
 		if ((phdrs[i].p_flags & QCOM_MDT_TYPE_MASK) == QCOM_MDT_TYPE_HASH) {
 			hash_segment = i;
--
2.17.1

Re: [PATCH v3] soc: qcom: mdt_loader: Add Upperbounds check for program header access

[Mukesh Ojha]

On 2/8/2024 6:05 PM, Auditya Bhattaram wrote:
> hash_index is evaluated by looping phdrs till QCOM_MDT_TYPE_HASH
> is found. Add an upperbound check to phdrs to access within elf size.
> 
> Fixes: 64fb5eb87d58 ("soc: qcom: mdt_loader: Allow hash to reside in any segment")
> Cc: <stable@vger.kernel.org>
> Signed-off-by: Auditya Bhattaram <quic_audityab@quicinc.com>
> ---
> Changes in v3:
>   - Corrected wrong patch versioning in the Subject.
>   - Added error prints for Invalid access.
> Link to v2 https://lore.kernel.org/linux-arm-msm/9773d189-c896-d5c5-804c-e086c24987b4@quicinc.com/T/#t
> Link to v1 https://lore.kernel.org/linux-arm-msm/5d7a3b97-d840-4863-91a0-32c1d8e7532f@linaro.org/T/#t
> ---
>   drivers/soc/qcom/mdt_loader.c | 5 +++++
>   1 file changed, 5 insertions(+)
> 
> diff --git a/drivers/soc/qcom/mdt_loader.c b/drivers/soc/qcom/mdt_loader.c
> index 6f177e46fa0f..61e2377cc5c3 100644
> --- a/drivers/soc/qcom/mdt_loader.c
> +++ b/drivers/soc/qcom/mdt_loader.c
> @@ -145,6 +145,11 @@ void *qcom_mdt_read_metadata(const struct firmware *fw, size_t *data_len,
>   	if (phdrs[0].p_type == PT_LOAD)
>   		return ERR_PTR(-EINVAL);
> 
> +	if (((size_t)(phdrs + ehdr->e_phnum)) > ((size_t)ehdr + fw->size)) {

This change is valid only if somehow, ehdr->e_phnum gets corrupted or 
changed via some engineering means and results in out-of-bounds access.

Acked-by: Mukesh Ojha <quic_mojha@quicinc.com>

> +		dev_err(dev, "Invalid phdrs access: %s\n", fw_name);

Should it print ehdr->e_phnum as well to be more valid?

-Mukesh

> +		return ERR_PTR(-EINVAL);
> +	}
> +
>   	for (i = 1; i < ehdr->e_phnum; i++) {
>   		if ((phdrs[i].p_flags & QCOM_MDT_TYPE_MASK) == QCOM_MDT_TYPE_HASH) {
>   			hash_segment = i;
> --
> 2.17.1
> 
> 

Re: [PATCH v3] soc: qcom: mdt_loader: Add Upperbounds check for program header access

[Auditya Bhattaram]

On 2/8/2024 7:59 PM, Mukesh Ojha wrote:
> 
> 
> On 2/8/2024 6:05 PM, Auditya Bhattaram wrote:
>> hash_index is evaluated by looping phdrs till QCOM_MDT_TYPE_HASH
>> is found. Add an upperbound check to phdrs to access within elf size.
>>
>> Fixes: 64fb5eb87d58 ("soc: qcom: mdt_loader: Allow hash to reside in 
>> any segment")
>> Cc: <stable@vger.kernel.org>
>> Signed-off-by: Auditya Bhattaram <quic_audityab@quicinc.com>
>> ---
>> Changes in v3:
>>   - Corrected wrong patch versioning in the Subject.
>>   - Added error prints for Invalid access.
>> Link to v2 
>> https://lore.kernel.org/linux-arm-msm/9773d189-c896-d5c5-804c-e086c24987b4@quicinc.com/T/#t
>> Link to v1 
>> https://lore.kernel.org/linux-arm-msm/5d7a3b97-d840-4863-91a0-32c1d8e7532f@linaro.org/T/#t
>> ---
>>   drivers/soc/qcom/mdt_loader.c | 5 +++++
>>   1 file changed, 5 insertions(+)
>>
>> diff --git a/drivers/soc/qcom/mdt_loader.c 
>> b/drivers/soc/qcom/mdt_loader.c
>> index 6f177e46fa0f..61e2377cc5c3 100644
>> --- a/drivers/soc/qcom/mdt_loader.c
>> +++ b/drivers/soc/qcom/mdt_loader.c
>> @@ -145,6 +145,11 @@ void *qcom_mdt_read_metadata(const struct 
>> firmware *fw, size_t *data_len,
>>       if (phdrs[0].p_type == PT_LOAD)
>>           return ERR_PTR(-EINVAL);
>>
>> +    if (((size_t)(phdrs + ehdr->e_phnum)) > ((size_t)ehdr + fw->size)) {
> 
> This change is valid only if somehow, ehdr->e_phnum gets corrupted or 
> changed via some engineering means and results in out-of-bounds access.
> 
> Acked-by: Mukesh Ojha <quic_mojha@quicinc.com>
> 
>> +        dev_err(dev, "Invalid phdrs access: %s\n", fw_name);
> 
> Should it print ehdr->e_phnum as well to be more valid?
> 
> -Mukesh
> 

Sure, planning to add fw->size as well.
dev_err(dev, "Invalid phdrs access for fw: %s, e_phnum: %hu, fw->size: 
%zu \n", fw_name, ehdr->e_phnum, fw->size);

>> +        return ERR_PTR(-EINVAL);
>> +    }
>> +
>>       for (i = 1; i < ehdr->e_phnum; i++) {
>>           if ((phdrs[i].p_flags & QCOM_MDT_TYPE_MASK) == 
>> QCOM_MDT_TYPE_HASH) {
>>               hash_segment = i;
>> -- 
>> 2.17.1
>>
>>