Variable-size structures are helpful, because you can allocate a structure containing an array of an unkown (at compile time) size at the end. For example, a “nuda_conf” structure containing informations about an MD Array and a “dev_info” structure per disks. The number of disks is unknown at compile time, so the only solution is to use a kalloc\/vmalloc. The classical solution is to do :<\/p>\n
[code lang=”c”]<\/p>\n
struct dev_info {\r\nint whatever;\r\nstruct rdev* rdev;\r\n}\r\n\r\nstruct nuda_conf {\r\nchar* name;\r\nint whotever;\r\nstruct dev_info disks*\r\n}\r\n\r\n...\r\n\r\nstruct nuda_conf* conf;\r\nconf = kzalloc(sizeof(struct nuda_conf));\r\nconf->disks =\u00a0kzalloc(sizeof(struct dev_info) * NUMER_OF_DISKS);<\/pre>\n[\/code]<\/p>\n
But variable-size structures allows to do only one allocation :<\/p>\n
[code lang=”c”]<\/p>\n
struct nuda_conf {\r\nchar* name;\r\nint whotever;\r\nstruct dev_info disks[0];\r\n}\r\n\r\nstruct nuda_conf* conf;\r\n\r\nconf = kzalloc(sizeof(struct nuda_conf) + sizeof(struct dev_info) * NUMBER_OF_DISKS);<\/pre>\n[\/code]<\/p>\n
This will allocate the memory for the char* and the int of nuda_conf, and the memory for a number of “dev_info”. To access them, one can do\u00a0conf->disks[N] <\/em>where N is the disk index.<\/p>\nBut, this is limited to one variable-size array. <\/strong>If you do :<\/p>\n[code lang=”c”]<\/p>\n
struct nuda_conf {\r\nchar* name;\r\nint whotever;\r\nstruct dev_info disks[0];\r\nstruct indirection_line indirection_table[0];\r\n}<\/pre>\n[\/code]<\/p>\n
The “indirection_table” and the “disks” pointer will point to the same location ! Even if you allocate size for both ! The solution is either to use a classical pointer as in the first example, or handle the memory access yourself :<\/p>\n
[code lang=”c”]<\/p>\n
struct nuda_conf {\r\nchar* name;\r\nint whotever;\r\n}\r\n...\r\nconf = kzalloc (sizeof(struct nuda_conf) + sizeof(struct dev_info) * NUMBER_OF_DISKS + sizeof(struct indirection_line)*NUMBER_OF_CHUNKS);\r\nstruct dev_info* disks = (struct dev_info*)(conf + 1);\r\nstruct indirection_line* indirection_table = (struct dev_info*)(disks + NUMBER_OF_DISKS);<\/pre>\n[\/code]<\/p>\n
But there is no interest in doing so, and both disks and indirection_table pointers have to be manually computed each time you need them… The use for that is really really really specific (dynamic metadata in packets for example).<\/p>\n","protected":false},"excerpt":{"rendered":"
Variable-size structures are helpful, because you can allocate a structure containing an array of an unkown (at compile time) size at the end. For example, a “nuda_conf” structure containing informations about an MD Array and a “dev_info” structure per disks. The number of disks is unknown at compile time, so the only solution is to … <\/p>\n