public inbox for rda@sourceware.org
 help / color / mirror / Atom feed
From: Jim Blandy <jimb@redhat.com>
To: rda@sources.redhat.com
Subject: RFA: Use td_symbol_list to request symbols if present
Date: Tue, 19 Oct 2004 22:27:00 -0000	[thread overview]
Message-ID: <vt2fz4ajqff.fsf@zenia.home> (raw)


The presence of the td_symbol_list function in recent LinuxThreads and
NPTL libthread_db libraries means that RDA can reliably use the
current GDB remote protocol, unchanged, to satisfy all libthread_db
symbol lookups.

2004-10-19  Jim Blandy  <jimb@redhat.com>

	Use the libthread_db 'td_symbol_list' function to look up all the
	symbols libthread_db will need when invited to by GDB.
	* thread-db.c (add_symbol_to_list): Make 'name' argument const.
	(td_symbol_list_p): New libthread_db function pointer.
	(thread_db_dlopen): Initialize it.

Index: rda/unix/thread-db.c
===================================================================
RCS file: /cvs/src/src/rda/unix/thread-db.c,v
retrieving revision 1.6
diff -c -r1.6 thread-db.c
*** rda/unix/thread-db.c	19 Oct 2004 21:20:08 -0000	1.6
--- rda/unix/thread-db.c	19 Oct 2004 22:21:50 -0000
***************
*** 70,76 ****
     for those that have already been defined by the debugger. */
  
  static void
! add_symbol_to_list (char *name, paddr_t value, int defined_p)
  {
    struct symbol_cache *tmp;
  
--- 70,76 ----
     for those that have already been defined by the debugger. */
  
  static void
! add_symbol_to_list (const char *name, paddr_t value, int defined_p)
  {
    struct symbol_cache *tmp;
  
***************
*** 374,379 ****
--- 374,381 ----
  					  void *xregs);
  static td_err_e (*td_thr_event_enable_p) (const td_thrhandle_t *th, 
  					  int event);
+ static const char **(*td_symbol_list_p)  (void);
+ 
  
  /* Function: thread_db_state_str
     Convert a thread_db state code to a string.
***************
*** 825,830 ****
--- 827,833 ----
    td_thr_getxregsize_p  = dlsym (dlhandle, "td_thr_getxregsize");
    td_thr_getxregs_p     = dlsym (dlhandle, "td_thr_getxregs");
    td_thr_setxregs_p     = dlsym (dlhandle, "td_thr_setxregs");
+   td_symbol_list_p      = dlsym (dlhandle, "td_symbol_list");
  
    return 0;		/* success */
  }
***************
*** 2284,2297 ****
    else
      fprintf (stderr, "< ERROR attach: GDB will not read thread regs. >>>\n");
  
!   /* KLUDGE: Insert some magic symbols into the cached symbol list,
!      to be looked up later.  This is badly wrong -- we should be 
!      obtaining these values thru the thread_db interface.  Their names
!      should not be hard-coded here <sob>. */
!   add_symbol_to_list ("__pthread_sig_restart",   0, UNDEFINED);
!   add_symbol_to_list ("__pthread_sig_cancel",    0, UNDEFINED);
!   add_symbol_to_list ("__pthread_sig_debug",     0, UNDEFINED);
!   add_symbol_to_list ("__pthread_threads_debug", 0, UNDEFINED);
  
    /* Attempt to open the thread_db interface.  This attempt will 
       most likely fail (unles the child is statically linked). */
--- 2307,2334 ----
    else
      fprintf (stderr, "< ERROR attach: GDB will not read thread regs. >>>\n");
  
!   if (td_symbol_list_p)
!     {
!       /* Take all the symbol names libthread_db might try to look up
! 	 and place them in our cached symbol list, to be looked up
! 	 when invited by GDB.  */
!       const char **symbol_list = td_symbol_list_p ();
!       int i;
! 
!       for (i = 0; symbol_list[i]; i++)
! 	add_symbol_to_list (symbol_list[i], 0, UNDEFINED);
!     }
!   else
!     {
!       /* KLUDGE: Insert some magic symbols into the cached symbol list,
! 	 to be looked up later.  This is badly wrong -- we should be 
! 	 obtaining these values thru the thread_db interface.  Their names
! 	 should not be hard-coded here <sob>. */
!       add_symbol_to_list ("__pthread_sig_restart",   0, UNDEFINED);
!       add_symbol_to_list ("__pthread_sig_cancel",    0, UNDEFINED);
!       add_symbol_to_list ("__pthread_sig_debug",     0, UNDEFINED);
!       add_symbol_to_list ("__pthread_threads_debug", 0, UNDEFINED);
!     }
  
    /* Attempt to open the thread_db interface.  This attempt will 
       most likely fail (unles the child is statically linked). */

             reply	other threads:[~2004-10-19 22:27 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-10-19 22:27 Jim Blandy [this message]
2004-10-20 15:39 ` Kevin Buettner
2004-10-20 19:29   ` Jim Blandy

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=vt2fz4ajqff.fsf@zenia.home \
    --to=jimb@redhat.com \
    --cc=rda@sources.redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).