Buffer busy waits -- p3=1012 and 1016

[Diego Cutrone]

Hi list, I'm analyzing parameter "p3" from trace 10046 level 8 and I've got some doubts about it. What is the difference between p3=1012 and p3=1016? According to some docs I've read, 1012="A modification is happening on a XCUR (or SCUR) buffer and it has not yet completed" (and I think that the session suffering the wait needs the block in CURRENT mode, thats because it has to wait) 1016="The session wants the block in SCUR or XCUR mode" (And I think that the session that wants the block has to wait because the block has been modified by another session) Aren't these two p3 values the same? The only difference seems to be that in the first case the block is *currently* being modified by another session, while in the second case it has already been modified. am I right? Please someone correct me if I'm wrong. TIA PD: Some examples below: SOURCE P1,P2,P3                OWNER      SEGMENT_NAME                  SEGMENT_TYPE p1=2 p2=1887 p3=1012           SYS        R10                           R p1=2 p2=1887 p3=1016           SYS        R10                           R p1=2 p2=19697 p3=1016          SYS        R08                          R p1=42 p2=16853 p3=0            GL        GL_JE_LINES_N1                 I p1=42 p2=22492 p3=0            GL        GL_JE_LINES_N1                 I p1=43 p2=68119 p3=0            GL         GL_JE_LINES                   T What about this P3=0 case ? I think it means that the block wanted by the session. A is being read by another session (say B) from disk to the SGA. So session A has to wait for the read to complete. What can I do to eliminate  (or at least improve) this wait? (may be to cache these table blocks in the SGA??) thanks