... so it could fall back to normal numa and we'd reduce the impact of the
NUMAQ subarch.

NUMAQ depends on GENERICARCH
also decouple genericarch numa from acpi.
also make it fall back to bigsmp if apicid > 8.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

a multi-socket test-system with 3 or 4 ioapics, when 4 dualcore cpus or
2 quadcore cpus installed, needs to switch to bigsmp or physflat.

CPU apic id is [4,11] instead of [0,7], and we need to check max apic
id instead of cpu numbers.

also add check for 32 bit when acpi is not compiled in or acpi=off.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

don't assume we can use RAM near the end of every node.
Esp systems that have few memory and they could have
kva address and kva RAM all below max_low_pfn.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Now we are using register_e820_active_regions() instead of
add_active_range() directly. So end_pfn could be different between the
value in early_node_map to node_end_pfn.

So we need to make shrink_active_range() smarter.

shrink_active_range() is a generic MM function in mm/page_alloc.c but
it is only used on 32-bit x86. Should we move it back to some file in
arch/x86?
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

this way we keep the early_node_map all right.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

This reverts commit ea57a5a6, a better
fix will be merged.

we were adding reserved BIOS ranges as general memory as well => not good.

solves this crash:

[   20.068075] hostname used greatest stack depth: 6464 bytes left
[   20.121404] BUG: unable to handle kernel <1>BUG: unable to handle kernel NULL pointer dereference at 00000b8c
[   20.121404] IP: [<c01160ae>] kmap_atomic_prot+0x2d/0x1c3
[   20.121404] *pdpt = 00000000367eb001 *pde = 0000000000000000
[   20.121404] Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
[   20.121404]
[   20.121404] Pid: 2061, comm: rc.sysinit Not tainted (2.6.26-rc3 #2440)
[   20.121404] EIP: 0060:[<c01160ae>] EFLAGS: 00010206 CPU: 0
[   20.121404] EIP is at kmap_atomic_prot+0x2d/0x1c3
Signed-off-by: Ingo Molnar <mingo@elte.hu>

This patch adds linked list of struct setup_data supported for i386.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: andi@firstfloor.org
Cc: mingo@redhat.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

This patch extracts the common part of head32.c and head64.c into head.c.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: andi@firstfloor.org
Cc: mingo@redhat.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

This patch reserves the EFI memory map with reserve_early(). Because EFI
memory map is allocated by bootloader, if it is not reserved by
reserved_early(), it may be overwritten through address returned by
find_e820_area().
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: andi@firstfloor.org
Cc: mingo@redhat.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

This patch makes early reserved highmem pages become reserved
pages. This can be used for highmem pages allocated by bootloader such
as EFI memory map, linked list of setup_data, etc.
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: andi@firstfloor.org
Cc: mingo@redhat.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

This patch clean up reserve_early() family functions by extracting the
common part of reserve_early(), free_early() and bad_addr() into
find_overlapped_early().
Signed-off-by: Huang Ying <ying.huang@intel.com>
Cc: andi@firstfloor.org
Cc: mingo@redhat.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

should use right shift
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

this way 32-bit is more similar to 64-bit, and smarter e820 and numa.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

to prepare 32-bit to use it.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

when 1/3 user/kernel split is used, and less memory is installed, or if
we have a big hole below 4g, max_low_pfn is still using 3g-128m

try to go down from max_low_pfn until we get it. otherwise will panic.

need to make 32-bit code to use register_e820_active_regions ... later.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

so we could fall back to one node numa.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

add memory_present() calls for sparse and non-numa.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

fix:

 drivers/built-in.o: In function `acpi_pci_irq_enable':
 : undefined reference to `mp_config_acpi_gsi'
Signed-off-by: Ingo Molnar <mingo@elte.hu>

on 64-bit we only get valid max_pfn_mapped after init_memory_mapping().
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

on 32-bit in head_32.S after initial page table is done, we get initial
max_pfn_mapped, and then kernel_physical_mapping_init will give us
a final one.

We need to use that to make sure find_e820_area will get valid addresses
for boot_map and for NODE_DATA(0) on numa32.

XEN PV and lguest may need to assign max_pfn_mapped too.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

use find_e820_area to get address space...
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

make mptable to be consistent with acpi routing, so we could:

1. kexec kernel with acpi=off
2. work around BIOSes where acpi routing is working, but mptable is
   not right, so can use kernel/kexec to start other OSes that don't have
   good acpi support.

command line: update_mptable
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

also fix the print out of node_remap_end_vaddr
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

we don't need to call memory_present that early.
numa and sparse will call memory_present later and might
even fail, it will call memory_present for the full range.

also for sparse it will call alloc_bootmem ... before we set up bootmem.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

... otherwise alloc_remap will not get node_mem_map from kva area, and
alloc_node_mem_map has to alloc_bootmem_node to get mem_map.
It will use two low address copies ...
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

so every element will represent 64M instead of 256M.

AMD opteron could have HW memory hole remapping, so could have
[0, 8g + 64M) on node0. Reduce element size to 64M to keep that on node 0

Later we need to use find_e820_area() to allocate memory_node_map like
on 64-bit. But need to move memory_present out of populate_mem_map...
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

fix this build bug:

 drivers/acpi/pci_irq.c: In function 'acpi_pci_irq_enable':
 drivers/acpi/pci_irq.c:574: error: implicit declaration of function 'mp_config_acpi_gsi'
Signed-off-by: Ingo Molnar <mingo@elte.hu>

we already have them in e820.h
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

on two node system (16g RAM) with numa config I got this crash:

get_memcfg_from_srat: assigning address to rsdp
RSD PTR  v0 [ACPIAM]
ACPI: Too big length in RSDT: 92
failed to get NUMA memory information from SRAT table
NUMA - single node, flat memory mode
Node: 0, start_pfn: 0, end_pfn: 153
 Setting physnode_map array to node 0 for pfns:
 0
...
Pid: 0, comm: swapper Not tainted 2.6.26-rc4 #4
 [<80b41289>] hlt_loop+0x0/0x3
 [<8011efa0>] ? alloc_remap+0x50/0x70
 [<8079e32e>] alloc_node_mem_map+0x5e/0xa0
 [<8012e77b>] ? printk+0x1b/0x20
 [<80b590f6>] free_area_init_node+0xc6/0x470
 [<80b588fc>] ? __alloc_bootmem_node+0x2c/0x50
 [<80b58ad8>] ? find_min_pfn_for_node+0x38/0x70
 [<8012e77b>] ? printk+0x1b/0x20
 [<80b597c4>] free_area_init_nodes+0x254/0x2d0
 [<80b544d7>] zone_sizes_init+0x97/0xa0
 [<80b48a03>] setup_arch+0x383/0x530
 [<8012e77b>] ? printk+0x1b/0x20
 [<80b41aa4>] start_kernel+0x64/0x350
 [<80b412d8>] i386_start_kernel+0x8/0x10
 =======================

this patch increases the acpi table limit to 32.
Also match early_ioremap() with early_iounmap().
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

reserve early numa kva, so it will not clash with new RAMDISK
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

reserve_early pgdata for 32bit numa
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

introduce init_pg_table_start, so xen PV could specify the value.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

remove extra -1 in reseve_early calling
    panic if can not find space for new RAMDISK
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

UP builds with LOCAL_APIC=y and IO_APIC=n fail with a missing
reference to mp_bus_not_pci. Distangle the mpparse code some more and
move the ioapic specific bus check into a separate function.

This code needs sume urgent un#ifdef surgery all over the place.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>