public inbox for systemtap@sourceware.org
 help / color / mirror / Atom feed
* [Bug translator/11815] New: SDT_V2 sdt.h broken for modern gcc on x86-64, cxxclass.exp
@ 2010-07-13 18:59 fche at redhat dot com
  2010-07-13 19:03 ` [Bug translator/11815] " fche at redhat dot com
  0 siblings, 1 reply; 2+ messages in thread
From: fche at redhat dot com @ 2010-07-13 18:59 UTC (permalink / raw)
  To: systemtap

With gcc 4.5ish on x86-64, SDT_V2 gets super optimized,
and gives us this ditty in the .probes section assembly code:

3:
        .asciz "$.LC0 $64 $24"  #,,
        .balign 8

where $.LC0 is a reference to a string literal.  This symbol
does not get propagated past the assembler, so stap has no
hope of resolving this particular operand.

Secondarily, the regexp magic in tapsets.cxx:4651ish

      // this pattern matches 0xD(%R) | (%R) | %R                              
                                                           
      const char *pattern =
"\\([0x]*-*[0-9]*\\)\\([(]*\\)\\(%*[0-9a-z][0-9a-z]*\\)\\([)]*\\)";            
                               

this unfortunately appears to match each of the above quoted operands, and
leads to wrong code generation (fetch_register(0)).

-- 
           Summary: SDT_V2 sdt.h broken for modern gcc on x86-64,
                    cxxclass.exp
           Product: systemtap
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: translator
        AssignedTo: systemtap at sources dot redhat dot com
        ReportedBy: fche at redhat dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=11815

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

^ permalink raw reply	[flat|nested] 2+ messages in thread

* [Bug translator/11815] SDT_V2 sdt.h broken for modern gcc on x86-64, cxxclass.exp
  2010-07-13 18:59 [Bug translator/11815] New: SDT_V2 sdt.h broken for modern gcc on x86-64, cxxclass.exp fche at redhat dot com
@ 2010-07-13 19:03 ` fche at redhat dot com
  0 siblings, 0 replies; 2+ messages in thread
From: fche at redhat dot com @ 2010-07-13 19:03 UTC (permalink / raw)
  To: systemtap


------- Additional Comments From fche at redhat dot com  2010-07-13 19:02 -------
re. problem#1, jakub advises that passing -L to the assembler may preserve
those symbols into the final object files, at the cost of considerable
extra storage.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|systemtap at sources dot    |fche at redhat dot com
                   |redhat dot com              |
             Status|NEW                         |ASSIGNED


http://sourceware.org/bugzilla/show_bug.cgi?id=11815

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2010-07-13 19:03 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-07-13 18:59 [Bug translator/11815] New: SDT_V2 sdt.h broken for modern gcc on x86-64, cxxclass.exp fche at redhat dot com
2010-07-13 19:03 ` [Bug translator/11815] " fche at redhat dot com

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).