Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Stan Shebs <stan@codesourcery.com>
To: gdb-patches@sourceware.org
Subject: [PATCH] Disallow non-scalars as rvalues in agent expressions
Date: Thu, 18 Mar 2010 05:00:00 -0000	[thread overview]
Message-ID: <4BA1B365.1070608@codesourcery.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 759 bytes --]

Bad Stuff(tm) happens when you try to compile an agent expression that 
wants to use a non-scalar object like a struct.  (We're talking about 
the struct itself, not a pointer to it.) Now, it actually works in GDB 
to say "print (int)bigstruct" - the first 32 bits of the struct get made 
into an integer, and the rest is discarded.  While in theory the 
bytecode compiler could do likewise, and perhaps someone will make that 
happen someday (it's not as simple as taking address, struct might be 
partly in registers), in the meantime we need to prevent non-scalar 
types from triggering internal errors all over the place.  Committed to 
trunk.

Stan

2010-03-17 Stan Shebs  <stan@codesourcery.com>

    * ax-gdb.c (require_rvalue): Disallow non-scalars.


[-- Attachment #2: nonscalar-patch-1 --]
[-- Type: text/plain, Size: 908 bytes --]

Index: ax-gdb.c
===================================================================
RCS file: /cvs/src/src/gdb/ax-gdb.c,v
retrieving revision 1.69
diff -p -r1.69 ax-gdb.c
*** ax-gdb.c	17 Mar 2010 22:04:43 -0000	1.69
--- ax-gdb.c	18 Mar 2010 01:44:44 -0000
*************** gen_int_literal (struct agent_expr *ax, 
*** 745,750 ****
--- 745,759 ----
  static void
  require_rvalue (struct agent_expr *ax, struct axs_value *value)
  {
+   /* Only deal with scalars, structs and such may be too large
+      to fit in a stack entry.  */
+   value->type = check_typedef (value->type);
+   if (TYPE_CODE (value->type) == TYPE_CODE_ARRAY
+       || TYPE_CODE (value->type) == TYPE_CODE_STRUCT
+       || TYPE_CODE (value->type) == TYPE_CODE_UNION
+       || TYPE_CODE (value->type) == TYPE_CODE_FUNC)
+     error ("Value not scalar: cannot be an rvalue.");
+ 
    switch (value->kind)
      {
      case axs_rvalue:

             reply	other threads:[~2010-03-18  5:00 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-03-18  5:00 Stan Shebs [this message]
2010-03-18  7:11 ` Eli Zaretskii
2010-03-18 16:02   ` Stan Shebs

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=4BA1B365.1070608@codesourcery.com \
    --to=stan@codesourcery.com \
    --cc=gdb-patches@sourceware.org \
    /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