Re: Re: A question about _thread_ of Linux

["cheng long" <kevinclcn@xxxxxxxxx>]

> Hi, I did a quick look at the code of the NPTL. I found that the
> pthread_self does not return the tid but the address of the struct
> pthread which is allocated in the stack (at least on i386 and possibly
> on other Arch too). Struct pthread contains various information used
> by the pthreaded calls. It created in the stack during pthread
> creation. pthread_self just finds the structure from stack and return
> its address.
Yes! It is right. I "strace"d the _pid_ file compiled from pid.c, I get
the result as follows(just a snapshot).

clone(child_stack=0xb7f7d4d4,

    flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|

        CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|

        CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID,

    parent_tidptr=0xb7f7dbe8,

    {entry_number:6, base_addr:0xb7f7dba0, limit:1048575,

         seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1,

         seg_not_present:0, useable:1},

   child_tidptr=0xb7f7dbe8) = 21945

clone(child_stack=0xb757c4d4,

     flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|

         CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|

         CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID,

     parent_tidptr=0xb757cbe8,

     {entry_number:6, base_addr:0xb757cba0, limit:1048575,

         seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1,

         seg_not_present:0, useable:1}, child_tidptr=0xb757cbe8) = 21946

futex(0xb7f7dbe8, FUTEX_WAIT, 21945, NULL

     In the thread 0:posix tid 3086474144
     In the thread 0:pid 21944
     In the thread 0:tid 21945) = 0

futex(0xb757cbe8, FUTEX_WAIT, 21946, NULL

     In the thread 1:posix tid 3075984288
     In the thread 1:pid 21944
     In the thread 1:tid 21946) = 0

 

As we see, the blue number == the blue number,

while the red number == the red number

 

Best Regards,

Cheng