PERFORCE change 136488 for review

[Randall R. Stewart]

http://perforce.freebsd.org/chv.cgi?CH=136488

Change 136488 by rrs at rrs-mips2-jnpr on 2008/02/29 09:39:51

	More notes.

Affected files ...

.. //depot/projects/mips2-jnpr/src/sys/TODO#17 edit

Differences ...

==== //depot/projects/mips2-jnpr/src/sys/TODO#17 (text+ko) ====

@@ -15,15 +15,16 @@
   into locore.S, and some code in sigreturn in pm_machdep.c.  Not sure where
   things are in libc.
 o Don't forget John Baldwin's comments on the user switch stuff. XXX add more
-  here XXX  This is mostly down to AST things.
+  here XXX  This is mostly down to AST things. - Done RRS.
 o What's defined in assym.s and param.h varies somewhat between the two
   code strains.  Figure out why and how to resolve.
 o machdep getmemsize() really is OBE.  Except for the mips_wired_tlb
   variable setting, which is related to something that was omitted from
   the juniper-mips dump.
-o cpu init ??? DONE ????
+o cpu init ??? DONE ???? - Now it is - RRS
 o we have a lot of memory from 0x80001000-0x81000000 that's currently
-  wasted.  We should feed it to the vm system so as not to waste it.
+  wasted.  We should feed it to the vm system so as not to waste it. - Fixed
+  for Cavium RRS.
 o Double check last merge from mips2 -> cavium code.  Something seems wrong
   in one of the edits.  Gonzo suggests looking at DO_AST macro.
 o Look at what the macro to set the per cpu data pointer will be like, and also
@@ -39,9 +40,9 @@
 o inb and readb are the same damn thing.  Really.  We should shoot them both
   and all their ilk.  This would allow us to have only one bus_tag, which we
   need anyway to do the indirection bit right.
-o We're not parsing the BIOS info quite right.
+o We're not parsing the BIOS info quite right. - Fixed RRS for Cavium
 o We're not getting memory size at all from the BIOS, but rather are getting
-  it from the constant of 96...
+  it from the constant of 96... - Fixed RRS
 o RAMDISK support
 o Network driver works only on the 36XX sim.  It dies in various ways on the
   different models.  Do we wait for a better code drop from Cavium or fix
@@ -52,8 +53,16 @@
 ************ RRS (below) *********
 o obio.c is probably not a good name for the IQ80321 driver
   -- obio is On Board I/O.  The IQ80321 is a StrongARM eval board...
-o the locations of the octeon uart stuff is probably not
-  good (in mips/octeon32)
 o Still need to update some of the file locations and s9indent it.
-o Still need to figure out where the packet units and enet driver
-  should go.
+o Still need to go in to the rgmii driver and fix some of its
+  issues. It seems to have its own spin_locks() which I think
+  should be fixed. It works.. but it needs some closer attention.
+o The PG_G bit is being set on user processe pages. I have hacked
+  it so that we basically avoid this.. put a else remove the
+  PG_G opposite the if where we did the add PG_G for kernel
+  mode. Need to trace down the root cause of the pte's having
+  this bit set.
+o Serial int's are not quite right. We stop getting them
+  in m-user.
+
+