Re: [PATCH fwctl] pds_fwctl: fix type and endian complaints

[Dave Jiang <dave.jiang@intel.com>]

On 4/2/25 10:41 AM, Nelson, Shannon wrote:
> On 4/2/2025 10:11 AM, Dave Jiang wrote:
>> On 4/2/25 9:56 AM, Shannon Nelson wrote:
>>> Fix a number of type and endian complaints from the sparse checker.
>>>
>>> Reported-by: kernel test robot <lkp@intel.com>
>>> Closes: https://lore.kernel.org/oe-kbuild-all/202504020246.Dfbhxoo9-lkp@intel.com/
>>> Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
>>> ---
>>>   drivers/fwctl/pds/main.c | 33 ++++++++++++++++++++-------------
>>>   1 file changed, 20 insertions(+), 13 deletions(-)
>>>
>>> diff --git a/drivers/fwctl/pds/main.c b/drivers/fwctl/pds/main.c
>>> index 284c4165fdd4..9b9d1f6b5556 100644
>>> --- a/drivers/fwctl/pds/main.c
>>> +++ b/drivers/fwctl/pds/main.c
>>> @@ -105,12 +105,14 @@ static int pdsfc_identify(struct pdsfc_dev *pdsfc)
>>>   static void pdsfc_free_endpoints(struct pdsfc_dev *pdsfc)
>>>   {
>>>        struct device *dev = &pdsfc->fwctl.dev;
>>> +     u32 num_endpoints;
>>>        int i;
>>>
>>>        if (!pdsfc->endpoints)
>>>                return;
>>>
>>> -     for (i = 0; pdsfc->endpoint_info && i < pdsfc->endpoints->num_entries; i++)
>>> +     num_endpoints = le32_to_cpu(pdsfc->endpoints->num_entries);
>>> +     for (i = 0; pdsfc->endpoint_info && i < num_endpoints; i++)
>>>                mutex_destroy(&pdsfc->endpoint_info[i].lock);
>>>        vfree(pdsfc->endpoint_info);
>>>        pdsfc->endpoint_info = NULL;
>>> @@ -199,7 +201,7 @@ static int pdsfc_init_endpoints(struct pdsfc_dev *pdsfc)
>>>        ep_entry = (struct pds_fwctl_query_data_endpoint *)pdsfc->endpoints->entries;
>>>        for (i = 0; i < num_endpoints; i++) {
>>>                mutex_init(&pdsfc->endpoint_info[i].lock);
>>> -             pdsfc->endpoint_info[i].endpoint = ep_entry[i].id;
>>> +             pdsfc->endpoint_info[i].endpoint = le32_to_cpu(ep_entry[i].id);
>>>        }
>>>
>>>        return 0;
>>> @@ -214,6 +216,7 @@ static struct pds_fwctl_query_data *pdsfc_get_operations(struct pdsfc_dev *pdsfc
>>>        struct pds_fwctl_query_data *data;
>>>        union pds_core_adminq_cmd cmd;
>>>        dma_addr_t data_pa;
>>> +     u32 num_entries;
>>>        int err;
>>>        int i;
>>>
>>> @@ -246,8 +249,9 @@ static struct pds_fwctl_query_data *pdsfc_get_operations(struct pdsfc_dev *pdsfc
>>>        *pa = data_pa;
>>>
>>>        entries = (struct pds_fwctl_query_data_operation *)data->entries;
>>> -     dev_dbg(dev, "num_entries %d\n", data->num_entries);
>>> -     for (i = 0; i < data->num_entries; i++) {
>>> +     num_entries = le32_to_cpu(data->num_entries);
>>> +     dev_dbg(dev, "num_entries %d\n", num_entries);
>>> +     for (i = 0; i < num_entries; i++) {
>>>
>>>                /* Translate FW command attribute to fwctl scope */
>>>                switch (entries[i].scope) {
>>> @@ -267,7 +271,7 @@ static struct pds_fwctl_query_data *pdsfc_get_operations(struct pdsfc_dev *pdsfc
>>>                        break;
>>>                }
>>>                dev_dbg(dev, "endpoint %d operation: id %x scope %d\n",
>>> -                     ep, entries[i].id, entries[i].scope);
>>> +                     ep, le32_to_cpu(entries[i].id), entries[i].scope);
>>>        }
>>>
>>>        return data;
>>> @@ -280,24 +284,26 @@ static int pdsfc_validate_rpc(struct pdsfc_dev *pdsfc,
>>>        struct pds_fwctl_query_data_operation *op_entry;
>>>        struct pdsfc_rpc_endpoint_info *ep_info = NULL;
>>>        struct device *dev = &pdsfc->fwctl.dev;
>>> +     u32 num_entries;
>>>        int i;
>>>
>>>        /* validate rpc in_len & out_len based
>>>         * on ident.max_req_sz & max_resp_sz
>>>         */
>>> -     if (rpc->in.len > pdsfc->ident.max_req_sz) {
>>> +     if (rpc->in.len > le32_to_cpu(pdsfc->ident.max_req_sz)) {
>>>                dev_dbg(dev, "Invalid request size %u, max %u\n",
>>> -                     rpc->in.len, pdsfc->ident.max_req_sz);
>>> +                     rpc->in.len, le32_to_cpu(pdsfc->ident.max_req_sz));
>>
>> Maybe use a local var for max_req_sz. I'm seeing that getting converted more than once in the same function.
>>
>>>                return -EINVAL;
>>>        }
>>>
>>> -     if (rpc->out.len > pdsfc->ident.max_resp_sz) {
>>> +     if (rpc->out.len > le32_to_cpu(pdsfc->ident.max_resp_sz)) {
>>>                dev_dbg(dev, "Invalid response size %u, max %u\n",
>>> -                     rpc->out.len, pdsfc->ident.max_resp_sz);
>>> +                     rpc->out.len, le32_to_cpu(pdsfc->ident.max_resp_sz));
>>
>> Same for max_res_sz.
> 
> I went back and forth on that, then left them this way in anticipation of possibly removing the dev_dbg() uses.  I could go either way on this.

Up to you.

Reviewed-by: Dave Jiang <dave.jiang@intel.com>

> 
> sln
> 
>>
>> DJ
>>
>>>                return -EINVAL;
>>>        }
>>>
>>> -     for (i = 0; i < pdsfc->endpoints->num_entries; i++) {
>>> +     num_entries = le32_to_cpu(pdsfc->endpoints->num_entries);
>>> +     for (i = 0; i < num_entries; i++) {
>>>                if (pdsfc->endpoint_info[i].endpoint == rpc->in.ep) {
>>>                        ep_info = &pdsfc->endpoint_info[i];
>>>                        break;
>>> @@ -326,8 +332,9 @@ static int pdsfc_validate_rpc(struct pdsfc_dev *pdsfc,
>>>
>>>        /* reject unsupported and/or out of scope commands */
>>>        op_entry = (struct pds_fwctl_query_data_operation *)ep_info->operations->entries;
>>> -     for (i = 0; i < ep_info->operations->num_entries; i++) {
>>> -             if (PDS_FWCTL_RPC_OPCODE_CMP(rpc->in.op, op_entry[i].id)) {
>>> +     num_entries = le32_to_cpu(ep_info->operations->num_entries);
>>> +     for (i = 0; i < num_entries; i++) {
>>> +             if (PDS_FWCTL_RPC_OPCODE_CMP(rpc->in.op, le32_to_cpu(op_entry[i].id))) {
>>>                        if (scope < op_entry[i].scope)
>>>                                return -EPERM;
>>>                        return 0;
>>> @@ -402,7 +409,7 @@ static void *pdsfc_fw_rpc(struct fwctl_uctx *uctx, enum fwctl_rpc_scope scope,
>>>        cmd = (union pds_core_adminq_cmd) {
>>>                .fwctl_rpc = {
>>>                        .opcode = PDS_FWCTL_CMD_RPC,
>>> -                     .flags = PDS_FWCTL_RPC_IND_REQ | PDS_FWCTL_RPC_IND_RESP,
>>> +                     .flags = cpu_to_le16(PDS_FWCTL_RPC_IND_REQ | PDS_FWCTL_RPC_IND_RESP),
>>>                        .ep = cpu_to_le32(rpc->in.ep),
>>>                        .op = cpu_to_le32(rpc->in.op),
>>>                        .req_pa = cpu_to_le64(in_payload_dma_addr),
>>
>