From: Pat Gefre <pfg@sgi.com>

Remove __ASSEMBLY__ tags from shubio.h

From: Pat Gefre <pfg@sgi.com>

Some more minor clean up

From: Pat Gefre <pfg@sgi.com>

A few small clean ups

From: Pat Gefre <pfg@sgi.com>

Moved code out of pciio and into its own file - snia_if.c and renamed the
functions

pciio clean up

From: Pat Gefre <pfg@sgi.com>

Some small bte code clean ups

From: Pat Gefre <pfg@sgi.com>

hcl.c clean up for init failures and OOM

From: Pat Gefre <pfg@sgi.com>

Fix a couple of compiler warnings

From: Pat Gefre <pfg@sgi.com>

Mostly code pruning - a couple of adds due to the clean up

From: Pat Gefre <pfg@sgi.com>

irq clean up and update

From: Pat Gefre <pfg@sgi.com>

General xtalk.c clean up

From: Pat Gefre <pfg@sgi.com>

shubio.c cleanup

From: Pat Gefre <pfg@sgi.com>

General module.c clean up

From: Pat Gefre <pfg@sgi.com>

More klgraph.c clean up
Remove is_specified

From: Pat Gefre <pfg@sgi.com>

klgraph.c clean up

From: Pat Gefre <pfg@sgi.com>

New code for Opus and CGbrick
More minor clean up

From: Pat Gefre <pfg@sgi.com>

Fix the last patch - missed an IS_PIC_SOFT and needed the CG definition

Header file clean up
Added some __init definitions for functions that were init only
Removed some 'porting' code - replaced it with linux calls/defintions
Cleaned up some of the types we were using (ones that had been added)
Fixed include files that had the wrong path in their ifdef'd names

From: Pat Gefre <pfg@sgi.com>

Fix the last patch - missed an IS_PIC_SOFT and needed the CG definition

From: Pat Gefre <pfg@sgi.com>

Remove the bridge and xbridge code - everything not PIC

From: Pat Gefre <pfg@sgi.com>

General clean up of xbow.c

From: Pat Gefre <pfg@sgi.com>

machvec/pci.c clean up
Other clean up related to above

From: Pat Gefre <pfg@sgi.com>

General code clean up of sn/io/io.c

From: Pat Gefre <pfg@sgi.com>

Delete invent.h
Delete sgi_if.c
Cleaned up some of the NEW/DEL calls

From: Pat Gefre <pfg@sgi.com>

Kill the arcs/*.h files
Some general clean up on klconfig.h

From: Pat Gefre <pfg@sgi.com>

Kill nag.h

From: Pat Gefre <pfg@sgi.com>
      Jesse Barnes <jbarnes@tomahawk.engr.sgi.com>

From: Pat Gefre <pfg@sgi.com>

remove unused enum

From: Pat Gefre <pfg@sgi.com>

kill $Id$

From: Pat Gefre <pfg@sgi.com>

kill big endian stuff

From: Pat Gefre <pfg@sgi.com>

namespace cleanup: ioerror_dump->sn_ioerror_dump

From: Pat Gefre <pfg@sgi.com>

copyright update

From: Pat Gefre <pfg@sgi.com>

Some hwgraph code clean up

From: "H. Peter Anvin" <hpa@zytor.com>

Declaring a function to return a const scalar value is pretty meaningless. 
These functions are really trying to say that they don't alter any external
state.

Fix that up by using __attribute__((const)), if the compiler supports that.

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

Fix endless loop in get_user_pages() on s390.  It happens only on s/390
because pte_dirty always returns 0.  For all other architectures this is an
optimization.

In the case of "write && !pte_dirty(pte)" follow_page() returns NULL.  On all
architectures except s390 handle_pte_fault() will then create a pte with
pte_dirty(pte)==1 because write_access==1.  In the following, second call to
follow_page() all is fine.  With the physical dirty bit patch pte_dirty() is
always 0 for s/390 because the dirty bit doesn't live in the pte.

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

while searching for a s390 tlb flush problem I noticed some superflous tlb
flushes.  One in zeromap_page_range, one in remap_page_range, and another one
in filemap_sync.  The patch just adds comments but I think these three
flush_tlb_range calls can be removed.

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

Ok, its just the revert of the page_cache_release delta.

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

while working on my mm patch for s390 I played with rmap a bit, adding
BUG statements and the like. While doing so I noticed some room for
improvement in rmap. Its minor stuff but anyway... 

The first observation is that the pte chain array doesn't have holes,
meaning that from the pte_chain_idx() of the first array every slot of
all following pte chain arrays are full. That is there can't be NULL
pointers. The "if (!pte_paddr)" check in try_to_unmap() can be removed
and if the loop in page_referenced() is started from pte_chain_idx(pc)
then the "if (!pte_paddr)" in page_referenced() can be removed as well.

The second observation is that the first pte array of a pte chain has
at least one entry. Empty pte chain arrays are always freed immediatly
after the last entry was removed. Because of that victim_i can be
calculated in a simpler way. Instead of setting victim_i to -1 and then
check in each loop iteration against -1 victim_i can just be set to
the pte_chain_idx of the first pte chain array.

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

I think I found a potential race in install_page/install_file_pte. The
inline function zap_pte releases pages by calling page_remove_rmap and
page_cache_release.  If this was the last user of a page it can get
purged from the page cache and then get immediatly reused. But there
might still be a tlb for this page on another cpu. The tlb is removed
in the callers of zap_pte, install_page and install_file_pte, but this
is too late. I admit that its a very unlikely race but never the less..

I fixed this by using the new ptep_clear_flush function that is introduced
with the tlb flush optimization patch for s/390.

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

this is another s/390 related mm patch. It introduces the concept of
physical dirty and referenced bits into the common mm code. I always
had the nagging feeling that the pte functions for setting/clearing
the dirty and referenced bits are not appropriate for s/390. It works
but it is a bit of a hack. 
After the wake of rmap it is now possible to put a much better solution
into place. The idea is simple: since there are not dirty/referenced
bits in the pte make these function nops on s/390 and add operations
on the physical page to the appropriate places. For the referenced bit
this is the page_referenced() function. For the dirty bit there are
two relevant spots: in page_remove_rmap after the last user of the
page removed its reverse mapping and in try_to_unmap after the last
user was unmapped. There are two new functions to accomplish this:

 * page_test_and_clear_dirty: Test and clear the dirty bit of a
   physical page. This function is analog to ptep_test_and_clear_dirty
   but gets a struct page as argument instead of a pte_t pointer.

 * page_test_and_clear_young: Test and clear the referenced bit
   of a physical page. This function is analog to ptep_test_and_clear_young
   but gets a struct page as argument instead of a pte_t pointer.

Its pretty straightforward and with it the s/390 mm makes much more
sense. You'll need the tls flush optimization patch for the patch.
Comments ?

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

On the s/390 architecture we still have the issue with tlb flushing and the
ipte instruction.  We can optimize the tlb flushing a lot with some minor
interface changes between the arch backend and the memory management core. 
In the end the whole thing is about the Invalidate Page Table Entry (ipte)
instruction.  The instruction sets the invalid bit in the pte and removes the
tlb for the page on all cpus for the virtual to physical mapping of the page
in a particular address space.  The nice thing is that only the tlb for this
page gets removed, all the other tlbs stay valid.  The reason we can't use
ipte to implement flush_tlb_page() is one of the requirements of the
instruction: the pte that should get flushed needs to be *valid*.

I'd like to add the following four functions to the mm interface:

  * ptep_establish: Establish a new mapping. This sets a pte entry to a
    page table and flushes the tlb of the old entry on all cpus if it
    exists. This is more or less what establish_pte in mm/memory.c does
    right now but without the update_mmu_cache call.

  * ptep_test_and_clear_and_flush_young. Do what ptep_test_and_clear_young
    does and flush the tlb.

  * ptep_test_and_clear_and_flush_dirty. Do what ptep_test_and_clear_dirty
    does and flush the tlb.

  * ptep_get_and_clear_and_flush: Do what ptep_get_and_clear does and
    flush the tlb.

The s/390 specific functions in include/pgtable.h define their own optimized
version of these four functions by use of the ipte.

I avoid the definition of these function for every architecture I added them
to include/asm-generic/pgtable.h.  Since i386/x86 and others don't include
this header yet and define their own version of the functions found there I
#ifdef'd all functions in include/asm-generic/pgtable.h to be able to pick
the ones that are needed for each architecture (see patch for details).

With the new functions in place it is easy to do the optimization, e.g.  the
sequence

         ptep_get_and_clear(ptep);
         flush_tlb_page(vma, address);

gets replace by

         ptep_get_and_clear_and_flush(vma, address, ptep);

The old sequence still works but it is suboptimal on s/390.

From: Martin Schwidefsky <schwidefsky@de.ibm.com>

- audit all 32 bit pointer accesses and make them use compat_ioctl(),
  because of the necessary conversion on s390
- introduce ULONG_IOCTL() which is used instead of COMPATIBLE_IOCTL()
  for all ioctls that have their argument encoded in 'arg' instead
  of the memory pointed to by arg. Same reason as above.
- remove most #ifdefs in <linux/compat_ioctl.h>: They don't make
  any sense if the respective handlers in fs/compat_ioctl.c are
  not disabled as well and they are potentially harmful (the
  CONFIG_BLK_DEV_DM e.g. was insufficient).
- comment out  COMPATIBLE_IOCTL(SIOCSIFBR) and
  COMPATIBLE_IOCTL(SIOCGIFBR), they appear to require a handler
- implement copy_in_user for s390, as needed for many handlers in
  fs/compat_ioctl.c
- get rid of all duplicate stuff in arch/s390/kernel/compat_ioctl.c
  that is also in fs/compat_ioctl.c