Re: [Qemu-devel] [PATCH 3/5] ide/atapi: Use table instead of switch for commands

[Kevin Wolf <kwolf@redhat.com>]

Am 19.04.2011 15:20, schrieb Amit Shah:
> On (Tue) 19 Apr 2011 [14:36:41], Kevin Wolf wrote:
> 
>> +struct {
>> +    void (*handler)(IDEState *s, uint8_t *buf);
>> +    int flags;
>> +} atapi_cmd_table[0x100] = {
>> +    [ 0x00 ] = { cmd_test_unit_ready,               0 },
>> +    [ 0x03 ] = { cmd_request_sense,                 ALLOW_UA },
>> +    [ 0x12 ] = { cmd_inquiry,                       ALLOW_UA },
>> +    [ 0x1a ] = { cmd_mode_sense, /* (6) */          0 },
>> +    [ 0x1b ] = { cmd_start_stop_unit,               0 },
>> +    [ 0x1e ] = { cmd_prevent_allow_medium_removal,  0 },
>> +    [ 0x25 ] = { cmd_read_cdvd_capacity,            0 },
>> +    [ 0x28 ] = { cmd_read, /* (10) */               0 },
>> +    [ 0x2b ] = { cmd_seek,                          0 },
>> +    [ 0x43 ] = { cmd_read_toc_pma_atip,             0 },
>> +    [ 0x46 ] = { cmd_get_configuration,             ALLOW_UA },
>> +    [ 0x4a ] = { cmd_get_event_status_notification, ALLOW_UA },
>> +    [ 0x5a ] = { cmd_mode_sense, /* (10) */         0 },
>> +    [ 0xa8 ] = { cmd_read, /* (12) */               0 },
>> +    [ 0xad ] = { cmd_read_dvd_structure,            0 },
>> +    [ 0xbb ] = { cmd_set_speed,                     0 },
>> +    [ 0xbd ] = { cmd_mechanism_status,              0 },
>> +    [ 0xbe ] = { cmd_read_cd,                       0 },
>> +};
> 
> I'd prefer to use the GPCMD_TEST_UNIT_READY, etc., defines we already
> have in internal.h instead of using the command numbers here.

In fact, I was considering to remove the GPCMD_* definitions because
most of them are unused now (only cmd_mode_sense/cmd_read still use them
to distinguish the variants).

What I like about this table is that you have all information about a
command directly visible: Its opcode, its name and any flags. Using the
constants would hide the opcode again, duplicate the name and probably
make the lines longer than 80 characters.

Kevin