关于group_info的疑问(用于描述进程组)

fineamy

#define NGROUPS_SMALL                32
#define NGROUPS_PER_BLOCK        ((int)(PAGE_SIZE / sizeof(gid_t)))
struct group_info {
        int ngroups;
        atomic_t usage;
        gid_t small_block[NGROUPS_SMALL];
        int nblocks;
        gid_t *blocks[0];
};

/*
* get_group_info() must be called with the owning task locked (via task_lock())
* when task != current.  The reason being that the vast majority of callers are
* looking at current->group_info, which can not be changed except by the
* current task.  Changing current->group_info requires the task lock, too.
*/
#define get_group_info(group_info) do { \
        atomic_inc(&(group_info)->usage); \
} while (0)

#define put_group_info(group_info) do { \
        if (atomic_dec_and_test(&(group_info)->usage)) \
                groups_free(group_info); \
} while (0)

请问大虾

struct group_info

的用法,其中定义的字段如
ngroups,            //进程组数目,
usage,
small_block,
blocks,
何意,能否解释一下.

> > On Mon, Sep 29, 2003 at 03:43:43PM -0700, Tim Hockin wrote:
> > > My version uses a struct group_info which has an array of pages. The groups
>
> I'm definitely happier about this one.
>
> Not that I'm any more thrilled about users using thousands of groups. But
> this looks a bit saner.

OK, then. Here's a brushed up version against 2.6.0-test6. Linus, if you
like this, I can get it into a public BK for you. If not, please indicate
the issues.

Summary: Get rid of the NGROUPS hard limit.

This patch removes all fixed-size arrays which depend on NGROUPS, and
replaces them with struct group_info, which is refcounted, and holds an
array of pages in which to store groups. groups_alloc() and groups_free()
are used to allocate and free struct group_info, and set_group_info is used
to actually put a group_info into a task. Groups are sorted and b-searched
for efficiency. Because groups are stored in a 2-D array, the GRP_AT()
macro was added to allow simple 1-D style indexing.

不过还是不太明白,希望解释一下