Your patch broke dump file printing
Daniel Berlin
dberlin@dberlin.org
Fri May 4 14:26:00 GMT 2007
Hi Simon, the following patch:
+2007-04-24 Simon Martin <simartin@users.sourceforge.net>
+
+ PR diagnostic/25923
+ * tree-pass.h (TDF_DIAGNOSTIC): New dump control to specify that a
+ diagnostic message is being built.
+ * tree-pretty-print.c (dump_generic_node): Only write the formatted text
+ into BUFFER's stream if we are not building a diagnostic message.
+ * toplev.c (default_tree_printer): Pass TDF_DIAGNOSTIC to
+ dump_generic_node.
+ * Makefile.in (toplev.o): Depend on tree-pass.h.
+
Broke dump file printing.
In particular, it looks like we don't output at the right point.
The PRE dump now looks like this:
Created value for
Created value for vuses: ()
Created value for
Created value for vuses: ()
Created value for vuses: ()
Created value for
Created value for vuses: ()
Created value for
exp_gen[0] := { }
tmp_gen[0] := {
VH.8scD.1643_1(D)VH.9*VH.8SMT.4D.1671_5(D)VH.10txq_descD.1640VH.11VH.9.VH.10SMT.4D.1671_5(D)VH.12VH.11[0]SMT.4D.1671_5(D)VH.13ed_cmdstsD.1638VH.14VH.12.VH.13SMT.4D.1671_5(D)VH.15VH.14
& 2147483647sc_1(D) (VH.8) }
Note that the calls to print_generic_expr are getting output *well*
after where they should be.
With your patch reverted, it looks like:
Created value VH.8 for scD.1643_1(D)
Created value VH.9 for *VH.8 vuses: (SMT.4D.1671_5(D))
Created value VH.10 for txq_descD.1640
Created value VH.11 for VH.9.VH.10 vuses: (SMT.4D.1671_5(D))
Created value VH.12 for VH.11[0] vuses: (SMT.4D.1671_5(D))
Created value VH.13 for ed_cmdstsD.1638
Created value VH.14 for VH.12.VH.13 vuses: (SMT.4D.1671_5(D))
Created value VH.15 for VH.14 & 2147483647
exp_gen[0] := { }
tmp_gen[0] := { sc_1(D) (VH.8) }avail_out[0] := { sc_1(D) (VH.8)
}exp_gen[2] := { sc_1(D) (VH.8) , *VH.8 (VH.9) , txq_desc (VH.10) ,
VH.9.VH.10 (VH.11) , VH.11[0] (VH.12) , ed_cmdsts (VH.13) ,
VH.12.VH.13 (VH.14) , VH.14 & 2147483647 (VH.15) }
Which is correct
Your patch makes the dump files worthless for debugging.
Please fix this or revert the patch.
Thanks,
Dan
More information about the Gcc-patches
mailing list