freebsd-nq/sys/powerpc/aim/slb.c
Nathan Whitehorn 3b4b38304e Improve hash coverage for kernel page table entries by modifying the kernel
ESID -> VSID map function. This makes ZFS run stably on PowerPC under
heavy loads (repeated simultaneous SVN checkouts and updates).
2010-07-31 21:35:15 +00:00

297 lines
7.2 KiB
C

/*-
* Copyright (c) 2010 Nathan Whitehorn
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* $FreeBSD$
*/
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/systm.h>
#include <sys/tree.h>
#include <vm/vm.h>
#include <vm/pmap.h>
#include <vm/uma.h>
#include <vm/vm_map.h>
#include <machine/md_var.h>
#include <machine/pmap.h>
#include <machine/vmparam.h>
uintptr_t moea64_get_unique_vsid(void);
void moea64_release_vsid(uint64_t vsid);
struct slbcontainer {
struct slb slb;
SPLAY_ENTRY(slbcontainer) slb_node;
};
static int slb_compare(struct slbcontainer *a, struct slbcontainer *b);
static void slb_zone_init(void *);
SPLAY_PROTOTYPE(slb_tree, slbcontainer, slb_node, slb_compare);
SPLAY_GENERATE(slb_tree, slbcontainer, slb_node, slb_compare);
uma_zone_t slb_zone;
uma_zone_t slb_cache_zone;
SYSINIT(slb_zone_init, SI_SUB_KMEM, SI_ORDER_ANY, slb_zone_init, NULL);
int
va_to_slb_entry(pmap_t pm, vm_offset_t va, struct slb *slb)
{
struct slbcontainer cont, *found;
uint64_t esid;
esid = (uintptr_t)va >> ADDR_SR_SHFT;
slb->slbe = (esid << SLBE_ESID_SHIFT) | SLBE_VALID;
if (pm == kernel_pmap) {
/* Set kernel VSID to deterministic value */
slb->slbv = va_to_vsid(kernel_pmap, va) << SLBV_VSID_SHIFT;
/* Figure out if this is a large-page mapping */
if (hw_direct_map && va < VM_MIN_KERNEL_ADDRESS) {
/*
* XXX: If we have set up a direct map, assumes
* all physical memory is mapped with large pages.
*/
if (mem_valid(va, 0) == 0)
slb->slbv |= SLBV_L;
}
return (0);
}
PMAP_LOCK_ASSERT(pm, MA_OWNED);
cont.slb.slbe = slb->slbe;
found = SPLAY_FIND(slb_tree, &pm->pm_slbtree, &cont);
if (found == NULL)
return (-1);
slb->slbv = found->slb.slbv;
return (0);
}
uint64_t
va_to_vsid(pmap_t pm, vm_offset_t va)
{
struct slb entry;
/* Shortcut kernel case */
if (pm == kernel_pmap)
return (KERNEL_VSID((uintptr_t)va >> ADDR_SR_SHFT));
/*
* If there is no vsid for this VA, we need to add a new entry
* to the PMAP's segment table.
*/
if (va_to_slb_entry(pm, va, &entry) != 0)
return (allocate_vsid(pm, (uintptr_t)va >> ADDR_SR_SHFT, 0));
return ((entry.slbv & SLBV_VSID_MASK) >> SLBV_VSID_SHIFT);
}
uint64_t
allocate_vsid(pmap_t pm, uint64_t esid, int large)
{
uint64_t vsid;
struct slbcontainer *slb_entry, kern_entry;
struct slb *prespill;
prespill = NULL;
if (pm == kernel_pmap) {
vsid = va_to_vsid(pm, esid << ADDR_SR_SHFT);
slb_entry = &kern_entry;
prespill = PCPU_GET(slb);
} else {
vsid = moea64_get_unique_vsid();
slb_entry = uma_zalloc(slb_zone, M_NOWAIT);
if (slb_entry == NULL)
panic("Could not allocate SLB mapping!");
prespill = pm->pm_slb;
}
slb_entry->slb.slbe = (esid << SLBE_ESID_SHIFT) | SLBE_VALID;
slb_entry->slb.slbv = vsid << SLBV_VSID_SHIFT;
if (large)
slb_entry->slb.slbv |= SLBV_L;
if (pm != kernel_pmap) {
PMAP_LOCK_ASSERT(pm, MA_OWNED);
SPLAY_INSERT(slb_tree, &pm->pm_slbtree, slb_entry);
}
/*
* Someone probably wants this soon, and it may be a wired
* SLB mapping, so pre-spill this entry.
*/
if (prespill != NULL)
slb_insert(pm, prespill, &slb_entry->slb);
return (vsid);
}
/* Lock entries mapping kernel text and stacks */
#define SLB_SPILLABLE(slbe) \
(((slbe & SLBE_ESID_MASK) < VM_MIN_KERNEL_ADDRESS && \
(slbe & SLBE_ESID_MASK) > 16*SEGMENT_LENGTH) || \
(slbe & SLBE_ESID_MASK) > VM_MAX_KERNEL_ADDRESS)
void
slb_insert(pmap_t pm, struct slb *slbcache, struct slb *slb_entry)
{
uint64_t slbe, slbv;
int i, j, to_spill;
/* We don't want to be preempted while modifying the kernel map */
critical_enter();
to_spill = -1;
slbv = slb_entry->slbv;
slbe = slb_entry->slbe;
/* Hunt for a likely candidate */
for (i = mftb() % 64, j = 0; j < 64; j++, i = (i+1) % 64) {
if (pm == kernel_pmap && i == USER_SR)
continue;
if (!(slbcache[i].slbe & SLBE_VALID)) {
to_spill = i;
break;
}
if (to_spill < 0 && (pm != kernel_pmap ||
SLB_SPILLABLE(slbcache[i].slbe)))
to_spill = i;
}
if (to_spill < 0)
panic("SLB spill on ESID %#lx, but no available candidates!\n",
(slbe & SLBE_ESID_MASK) >> SLBE_ESID_SHIFT);
if (slbcache[to_spill].slbe & SLBE_VALID) {
/* Invalidate this first to avoid races */
slbcache[to_spill].slbe = 0;
mb();
}
slbcache[to_spill].slbv = slbv;
slbcache[to_spill].slbe = slbe | (uint64_t)to_spill;
/* If it is for this CPU, put it in the SLB right away */
if (pm == kernel_pmap && pmap_bootstrapped) {
/* slbie not required */
__asm __volatile ("slbmte %0, %1" ::
"r"(slbcache[to_spill].slbv),
"r"(slbcache[to_spill].slbe));
}
critical_exit();
}
int
vsid_to_esid(pmap_t pm, uint64_t vsid, uint64_t *esid)
{
uint64_t slbv;
struct slbcontainer *entry;
#ifdef INVARIANTS
if (pm == kernel_pmap)
panic("vsid_to_esid only works on user pmaps");
PMAP_LOCK_ASSERT(pm, MA_OWNED);
#endif
slbv = vsid << SLBV_VSID_SHIFT;
SPLAY_FOREACH(entry, slb_tree, &pm->pm_slbtree) {
if (slbv == entry->slb.slbv) {
*esid = entry->slb.slbe >> SLBE_ESID_SHIFT;
return (0);
}
}
return (-1);
}
void
free_vsids(pmap_t pm)
{
struct slbcontainer *entry;
while (!SPLAY_EMPTY(&pm->pm_slbtree)) {
entry = SPLAY_MIN(slb_tree, &pm->pm_slbtree);
SPLAY_REMOVE(slb_tree, &pm->pm_slbtree, entry);
moea64_release_vsid(entry->slb.slbv >> SLBV_VSID_SHIFT);
uma_zfree(slb_zone, entry);
}
}
static int
slb_compare(struct slbcontainer *a, struct slbcontainer *b)
{
if (a->slb.slbe == b->slb.slbe)
return (0);
else if (a->slb.slbe < b->slb.slbe)
return (-1);
else
return (1);
}
static void
slb_zone_init(void *dummy)
{
slb_zone = uma_zcreate("SLB segment", sizeof(struct slbcontainer),
NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, UMA_ZONE_VM);
slb_cache_zone = uma_zcreate("SLB cache", 64*sizeof(struct slb),
NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, UMA_ZONE_VM);
}
struct slb *
slb_alloc_user_cache(void)
{
return (uma_zalloc(slb_cache_zone, M_ZERO));
}
void
slb_free_user_cache(struct slb *slb)
{
uma_zfree(slb_cache_zone, slb);
}