[Bug testsuite/14615] New: singlejmp-reverse.exp assumes "b main; run" stops at *main, not after the prologue

[Bug testsuite/14615] New: singlejmp-reverse.exp assumes "b main; run" stops at *main, not after the prologue

[dje at google dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=14615

             Bug #: 14615
           Summary: singlejmp-reverse.exp assumes "b main; run" stops at
                    *main, not after the prologue
           Product: gdb
           Version: HEAD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: testsuite
        AssignedTo: unassigned@sourceware.org
        ReportedBy: dje@google.com
    Classification: Unclassified


If gdb is smart enough to recognize "sub $0x8,%rsp" as being part of the
prologue this will fail:

FAIL: gdb.reverse/singlejmp-reverse.exp: next to v = 1

There are more failures, but at least some are due to the first.

The test does this:

if ![runto_main] {
    return -1
}

[...]

gdb_test "next" {v = 1;} "next to v = 1"

Here's the original C source:

int
main (void)
{
  v = 1;

which in the assembler version just has "sub $0x8,%rsp" for the prologue.
Current gdb doesn't recognize it as being part of the prologue, so "b main"
doesn't skip it, and thus the need to "next to v = 1".  But a gdb that does
recognize it as being part of the prologue will fail.

One solution would be to see where "b main; run" stops, and only "next" to "v =
1;" if necessary.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug testsuite/14615] singlejmp-reverse.exp assumes "b main; run" stops at *main, not after the prologue

[jan.kratochvil at redhat dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=14615

Jan Kratochvil <jan.kratochvil at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
         AssignedTo|unassigned at sourceware    |jan.kratochvil at redhat
                   |dot org                     |dot com

--- Comment #1 from Jan Kratochvil <jan.kratochvil at redhat dot com> 2012-09-25 05:52:36 UTC ---
Created attachment 6651
  --> http://sourceware.org/bugzilla/attachment.cgi?id=6651
Testcase fix.

This testcase is unrelated to the issue of whether to skip or not to skip.
OK with check-in?

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug testsuite/14615] singlejmp-reverse.exp assumes "b main; run" stops at *main, not after the prologue

[jan.kratochvil at redhat dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=14615

Jan Kratochvil <jan.kratochvil at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jan.kratochvil at redhat
                   |                            |dot com

--- Comment #2 from Jan Kratochvil <jan.kratochvil at redhat dot com> 2012-09-25 05:59:04 UTC ---
[patch] testsuite: singlejmp-reverse.exp compatibility with Doug's GDB (PR
14615)
http://sourceware.org/ml/gdb-patches/2012-09/msg00533.html

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.