[PATCHv6,02/10] gdb: create_breakpoint: asserts relating to extra_string/parse_extra

Message ID adeb0daaee1757989f7bc91fe3f0ff626dd5c187.1701513409.git.aburgess@redhat.com
State New
Headers
Series [PATCHv6,01/10] gdb: create_breakpoint: add asserts and additional comments |

Checks

Context Check Description
linaro-tcwg-bot/tcwg_gdb_build--master-aarch64 success Testing passed
linaro-tcwg-bot/tcwg_gdb_build--master-arm success Testing passed
linaro-tcwg-bot/tcwg_gdb_check--master-aarch64 success Testing passed

Commit Message

Andrew Burgess Dec. 2, 2023, 10:42 a.m. UTC
  The goal of this commit is to better define the API for
create_breakpoint especially around the use of extra_string and
parse_extra.  This will be useful in the next commit when I plan to
make some changes to create_breakpoint.

This commit makes one possibly breaking change: until this commit it
was possible to create thread-specific dprintf breakpoint like this:

  (gdb) dprintf call_me, thread 1 "%s", "hello"
  Dprintf 2 at 0x401152: file /tmp/hello.c, line 8.
  (gdb) info breakpoints
  Num     Type           Disp Enb Address            What
  2       dprintf        keep y   0x0000000000401152 in call_me at /tmp/hello.c:8 thread 1
          stop only in thread 1
          printf "%s", "hello"
  (gdb)

this feature of dprintf was not documented, was not tested, and is
slightly different in syntax to how we create thread specific
breakpoints and/or watchpoints -- the thread condition appears after
the first ','.

I believe that this worked at all was simply by luck.  We happen to
pass the parse_extra flag as true from dprintf_command to
create_breakpoint.

So in this commit I made the choice change this.  We now pass
parse_extra as false from dprintf_command to create_breakpoint.  With
this done it is assumed that the only thing in the extra_string is the
dprintf format and arguments.

Beyond this change I've updated the comment on create_breakpoint in
breakpoint.h, and I've then added some asserts into
create_breakpoint as well as moving around some of the error
handling.

 - We now assert on the incoming argument values,

 - I've moved an error check to sit after the call to
   find_condition_and_thread_for_sals, this ensures the extra_string
   was parsed correctly,

In dprintf_command:

 - We now throw an error if there is no format string after the
   dprintf location.  This error was already being thrown, but was
   being caught later in the process.  With this change we catch the
   missing string earlier,

 - And, as mentioned earlier, we pass parse_extra as false when
   calling create_breakpoint,

In create_tracepoint_from_upload:

 - We now throw an error if the parsed location doesn't completely
   consume the addr_str variable.  This error has now effectively
   moved out of create_breakpoint.
---
 gdb/breakpoint.c | 42 +++++++++++++++++++++++++++++-------------
 gdb/breakpoint.h | 44 +++++++++++++++++++++++++++-----------------
 2 files changed, 56 insertions(+), 30 deletions(-)
  

Comments

Tankut Baris Aktemur Dec. 4, 2023, 7:40 p.m. UTC | #1
On Saturday, December 2, 2023 11:42 AM, Andrew Burgess wrote:
> The goal of this commit is to better define the API for
> create_breakpoint especially around the use of extra_string and
> parse_extra.  This will be useful in the next commit when I plan to
> make some changes to create_breakpoint.
> 
> This commit makes one possibly breaking change: until this commit it
> was possible to create thread-specific dprintf breakpoint like this:
> 
>   (gdb) dprintf call_me, thread 1 "%s", "hello"
>   Dprintf 2 at 0x401152: file /tmp/hello.c, line 8.
>   (gdb) info breakpoints
>   Num     Type           Disp Enb Address            What
>   2       dprintf        keep y   0x0000000000401152 in call_me at /tmp/hello.c:8 thread 1
>           stop only in thread 1
>           printf "%s", "hello"
>   (gdb)
> 
> this feature of dprintf was not documented, was not tested, and is

this -> This

> slightly different in syntax to how we create thread specific
> breakpoints and/or watchpoints -- the thread condition appears after
> the first ','.
> 
> I believe that this worked at all was simply by luck.  We happen to
> pass the parse_extra flag as true from dprintf_command to
> create_breakpoint.
> 
> So in this commit I made the choice change this.  We now pass
> parse_extra as false from dprintf_command to create_breakpoint.  With
> this done it is assumed that the only thing in the extra_string is the
> dprintf format and arguments.
> 
> Beyond this change I've updated the comment on create_breakpoint in
> breakpoint.h, and I've then added some asserts into
> create_breakpoint as well as moving around some of the error
> handling.
> 
>  - We now assert on the incoming argument values,
> 
>  - I've moved an error check to sit after the call to
>    find_condition_and_thread_for_sals, this ensures the extra_string
>    was parsed correctly,
> 
> In dprintf_command:
> 
>  - We now throw an error if there is no format string after the
>    dprintf location.  This error was already being thrown, but was
>    being caught later in the process.  With this change we catch the
>    missing string earlier,
> 
>  - And, as mentioned earlier, we pass parse_extra as false when
>    calling create_breakpoint,
> 
> In create_tracepoint_from_upload:
> 
>  - We now throw an error if the parsed location doesn't completely
>    consume the addr_str variable.  This error has now effectively
>    moved out of create_breakpoint.
> ---
>  gdb/breakpoint.c | 42 +++++++++++++++++++++++++++++-------------
>  gdb/breakpoint.h | 44 +++++++++++++++++++++++++++-----------------
>  2 files changed, 56 insertions(+), 30 deletions(-)
> 
> diff --git a/gdb/breakpoint.c b/gdb/breakpoint.c
> index dd415ff42f0..bd28236ce7d 100644
> --- a/gdb/breakpoint.c
> +++ b/gdb/breakpoint.c
> @@ -9240,6 +9240,17 @@ create_breakpoint (struct gdbarch *gdbarch,
>    if (extra_string != NULL && *extra_string == '\0')
>      extra_string = NULL;
> 
> +  /* A bp_dprintf must always have an accompanying EXTRA_STRING containing
> +     the dprintf format and arguments -- PARSE_EXTRA should always be false
> +     in this case.
> +
> +     For all other breakpoint types, EXTRA_STRING should be nullptr unless
> +     PARSE_EXTRA is true.  */
> +  gdb_assert ((type_wanted == bp_dprintf
> +	       && extra_string != nullptr && !parse_extra)
> +	      || (type_wanted != bp_dprintf
> +		  && (extra_string == nullptr || parse_extra)));
> +

This could be a personal taste, but I'd find it easier to read if
it was written 

  if (type_wanted == bp_dprintf)
    gdb_assert ((extra_string != nullptr) && !parse_extra);
  else
    gdb_assert ((extra_string == nullptr) || parse_extra);

or

  gdb_assert ((type_wanted == bp_dprintf)
		 ? (extra_string != nullptr) && !parse_extra
		 : (extra_string == nullptr) || parse_extra));

Thanks
-Baris



Intel Deutschland GmbH
Registered Address: Am Campeon 10, 85579 Neubiberg, Germany
Tel: +49 89 99 8853-0, www.intel.de <http://www.intel.de>
Managing Directors: Christin Eisenschmid, Sharon Heck, Tiffany Doon Silva  
Chairperson of the Supervisory Board: Nicole Lau
Registered Office: Munich
Commercial Register: Amtsgericht Muenchen HRB 186928
  

Patch

diff --git a/gdb/breakpoint.c b/gdb/breakpoint.c
index dd415ff42f0..bd28236ce7d 100644
--- a/gdb/breakpoint.c
+++ b/gdb/breakpoint.c
@@ -9240,6 +9240,17 @@  create_breakpoint (struct gdbarch *gdbarch,
   if (extra_string != NULL && *extra_string == '\0')
     extra_string = NULL;
 
+  /* A bp_dprintf must always have an accompanying EXTRA_STRING containing
+     the dprintf format and arguments -- PARSE_EXTRA should always be false
+     in this case.
+
+     For all other breakpoint types, EXTRA_STRING should be nullptr unless
+     PARSE_EXTRA is true.  */
+  gdb_assert ((type_wanted == bp_dprintf
+	       && extra_string != nullptr && !parse_extra)
+	      || (type_wanted != bp_dprintf
+		  && (extra_string == nullptr || parse_extra)));
+
   try
     {
       ops->create_sals_from_location_spec (locspec, &canonical);
@@ -9303,6 +9314,8 @@  create_breakpoint (struct gdbarch *gdbarch,
 
       if (parse_extra)
 	{
+	  gdb_assert (type_wanted != bp_dprintf);
+
 	  gdb::unique_xmalloc_ptr<char> rest;
 	  gdb::unique_xmalloc_ptr<char> cond;
 
@@ -9311,15 +9324,15 @@  create_breakpoint (struct gdbarch *gdbarch,
 	  find_condition_and_thread_for_sals (lsal.sals, extra_string,
 					      &cond, &thread, &inferior,
 					      &task, &rest);
+
+	  if (rest.get () != nullptr && *(rest.get ()) != '\0')
+	    error (_("Garbage '%s' at end of command"), rest.get ());
+
 	  cond_string_copy = std::move (cond);
 	  extra_string_copy = std::move (rest);
 	}
       else
 	{
-	  if (type_wanted != bp_dprintf
-	      && extra_string != NULL && *extra_string != '\0')
-		error (_("Garbage '%s' at end of location"), extra_string);
-
 	  /* Check the validity of the condition.  We should error out
 	     if the condition is invalid at all of the locations and
 	     if it is not forced.  In the PARSE_EXTRA case above, this
@@ -9530,21 +9543,18 @@  dprintf_command (const char *arg, int from_tty)
 
   /* If non-NULL, ARG should have been advanced past the location;
      the next character must be ','.  */
-  if (arg != NULL)
+  if (arg == nullptr || arg[0] != ',' || arg[1] == '\0')
+    error (_("Format string required"));
+  else
     {
-      if (arg[0] != ',' || arg[1] == '\0')
-	error (_("Format string required"));
-      else
-	{
-	  /* Skip the comma.  */
-	  ++arg;
-	}
+      /* Skip the comma.  */
+      ++arg;
     }
 
   create_breakpoint (get_current_arch (),
 		     locspec.get (),
 		     NULL, -1, -1,
-		     arg, false, 1 /* parse arg */,
+		     arg, false, 0 /* parse arg */,
 		     0, bp_dprintf,
 		     0 /* Ignore count */,
 		     pending_break_support,
@@ -14149,6 +14159,12 @@  create_tracepoint_from_upload (struct uploaded_tp *utp)
 
   location_spec_up locspec = string_to_location_spec (&addr_str,
 						      current_language);
+
+
+  gdb_assert (addr_str != nullptr);
+  if (*addr_str != '\0')
+    error (_("Garbage '%s' at end of location"), addr_str);
+
   if (!create_breakpoint (get_current_arch (),
 			  locspec.get (),
 			  utp->cond_string.get (), -1, -1, addr_str,
diff --git a/gdb/breakpoint.h b/gdb/breakpoint.h
index 4abf6d0762c..95f98b59e41 100644
--- a/gdb/breakpoint.h
+++ b/gdb/breakpoint.h
@@ -1585,32 +1585,42 @@  enum breakpoint_create_flags
    functions for setting a breakpoint at LOCSPEC.
 
    This function has two major modes of operations, selected by the
-   PARSE_EXTRA parameter.
+   PARSE_EXTRA and WANTED_TYPE parameters.
 
-   If PARSE_EXTRA is zero, LOCSPEC is just the breakpoint's location
-   spec, with condition, thread, and extra string specified by the
-   COND_STRING, THREAD, and EXTRA_STRING parameters.
+   When WANTED_TYPE is not bp_dprintf the following rules apply:
 
-   If PARSE_EXTRA is non-zero, this function will attempt to extract
-   the condition, thread, and extra string from EXTRA_STRING, ignoring
-   the similarly named parameters.
+     If PARSE_EXTRA is zero, LOCSPEC is just the breakpoint's location
+     spec, with condition, thread, and extra string specified by the
+     COND_STRING, THREAD, and EXTRA_STRING parameters.
 
-   If FORCE_CONDITION is true, the condition is accepted even when it is
-   invalid at all of the locations.  However, if PARSE_EXTRA is non-zero,
-   the FORCE_CONDITION parameter is ignored and the corresponding argument
-   is parsed from EXTRA_STRING.
+     If PARSE_EXTRA is non-zero, this function will attempt to extract the
+     condition, thread, and extra string from EXTRA_STRING, ignoring the
+     similarly named parameters.
+
+   When WANTED_TYPE is bp_dprintf the following rules apply:
+
+     PARSE_EXTRA must always be zero, LOCSPEC is just the breakpoint's
+     location spec, with condition, thread, and extra string (which
+     contains the dprintf format and arguments) specified by the
+     COND_STRING, THREAD, and EXTRA_STRING parameters.
+
+   If FORCE_CONDITION is true, the condition (in COND_STRING) is accepted
+   even when it is invalid at all of the locations.  However, if
+   PARSE_EXTRA is non-zero and WANTED_TYPE is not bp_dprintf, the
+   FORCE_CONDITION parameter is ignored and the corresponding argument is
+   parsed from EXTRA_STRING.
 
    The THREAD should be a global thread number, the created breakpoint will
    only apply for that thread.  If the breakpoint should apply for all
-   threads then pass -1.  However, if PARSE_EXTRA is non-zero then the
-   THREAD parameter is ignored and an optional thread number will be parsed
-   from EXTRA_STRING.
+   threads then pass -1.  However, if PARSE_EXTRA is non-zero and
+   WANTED_TYPE is not bp_dprintf, then the THREAD parameter is ignored and
+   an optional thread number will be parsed from EXTRA_STRING.
 
    The INFERIOR should be a global inferior number, the created breakpoint
    will only apply for that inferior.  If the breakpoint should apply for
-   all inferiors then pass -1.  However, if PARSE_EXTRA is non-zero then
-   the INFERIOR parameter is ignored and an optional inferior number will
-   be parsed from EXTRA_STRING.
+   all inferiors then pass -1.  However, if PARSE_EXTRA is non-zero and
+   WANTED_TYPE is not bp_dprintf, then the INFERIOR parameter is ignored
+   and an optional inferior number will be parsed from EXTRA_STRING.
 
    At most one of THREAD and INFERIOR should be set to a value other than
    -1; breakpoints can be thread specific, or inferior specific, but not