[Skiboot] [PATCH V6 21/21] core/pldm: Get file handle and file length

Christophe Lombard clombard at linux.vnet.ibm.com
Wed Apr 19 01:02:54 AEST 2023



Le 13/04/2023 à 15:28, Frederic Barrat a écrit :
>
>
> On 13/09/2022 12:27, Christophe Lombard wrote:
>> diff --git a/core/pldm/pldm-file-io-requests.c 
>> b/core/pldm/pldm-file-io-requests.c
>> index 0665a679..a1191b89 100644
>> --- a/core/pldm/pldm-file-io-requests.c
>> +++ b/core/pldm/pldm-file-io-requests.c
>> @@ -34,6 +34,82 @@ static void file_io_init_complete(bool success)
>>       file_io_ready = true;
>>   }
>>   +#define CHKSUM_PADDING 8
>> +
>> +/*
>> + * Retrieve the file handle and file length from the file attribute
>> + * table.
>> + */
>> +static int find_file_handle_by_lid_id(const char *lid_id,
>> +                      uint32_t *file_handle,
>> +                      uint32_t *file_length)
>> +{
>> +    const struct pldm_file_attr_table_entry *file_entry;
>> +    char *startptr, *endptr;
>> +    uint16_t file_name_length;
>> +
>> +    if ((file_attr_table == NULL) || (file_attr_length == 0))
>> +        return OPAL_PARAMETER;
>> +
>> +    startptr = (char *)file_attr_table;
>> +    endptr = startptr + file_attr_length - CHKSUM_PADDING;
>> +    *file_handle = 0;
>> +    *file_length = 0;
>> +
>> +    while (startptr < endptr) {
>> +        /* file entry:
>> +         *   4 Bytes: file handle
>> +         *   2 Bytes: file name length
>> +         *   <file name length> Bytes: file name
>> +         *   4 Bytes: file length
>> +         */
>> +        file_entry = (struct pldm_file_attr_table_entry *)startptr;
>> +
>> +        *file_handle = le32_to_cpu(file_entry->file_handle);
>> +        startptr += sizeof(uint32_t);
>> +
>> +        file_name_length = le16_to_cpu(file_entry->file_name_length);
>> +        startptr += sizeof(file_name_length);
>> +
>> +        if (!strncmp(startptr, lid_id, strlen(lid_id))) {
>> +            startptr += file_name_length;
>> +            *file_length = le32_to_cpu(*(uint32_t *)startptr);
>> +            break;
>> +        }
>> +        startptr += file_name_length;
>> +        startptr += sizeof(uint32_t);
>> +        startptr += sizeof(bitfield32_t);
>> +    }
>> +
>> +    if (*file_length == 0) {
>> +        prlog(PR_ERR, "%s - No file handle found, lid_id: %s\n",
>> +                  __func__, lid_id);
>> +        *file_handle = 0xff;
>> +        *file_length = 0;
>> +        return OPAL_PARAMETER;
>> +    }
>
>
> (sorry for the late comment, but I'm re-reading this as it's being 
> used by a patch in a later series)
>
> So 0xff is an invalid file handle (should probably be a macro). IIUC, 
> in such a case, we'll still add the struct pldm_lid to the list of 
> LIDs (lid_files in core/pldm/pldm-lid-files.c). Though we don't really 
> expect it, one could try reading/writing such a lid!
>
> - handles are apparently 16-bits and seem to be mostly sequential, but 
> wouldn't it be safer to choose 0xffff as an invalid handle?
>

right. 0xffff seems to be a better value. Thanks

> - we should detect operations on such invalid handles later (in 
> read_file_req/write_file_req/...)
>
>
>   Fred
>
>



More information about the Skiboot mailing list