kevin/linux
1
Fork 0
Code
47db360328
linux/arch/ppc64/kernel/pSeries_lpar.c
Benjamin Herrenschmidt 61b1a94254 [PATCH] ppc64: Store virtual address in TLB flush batches 
This patch slightly change the TLB flush batch mecanism so that we
store the full vaddr (including vsid) when adding an entry to the
batch so that the flush part doesn't have to get to the context.

This cleans it a bit, and paves the way to future updates like
dynamic vsids.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
2005-09-21 19:21:07 +10:00

518 lines
12 KiB
C
Raw Blame History

/*
* pSeries_lpar.c
* Copyright (C) 2001 Todd Inglett, IBM Corporation
*
* pSeries LPAR support.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
#define DEBUG
#include <linux/config.h>
#include <linux/kernel.h>
#include <linux/dma-mapping.h>
#include <asm/processor.h>
#include <asm/mmu.h>
#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/machdep.h>
#include <asm/abs_addr.h>
#include <asm/mmu_context.h>
#include <asm/ppcdebug.h>
#include <asm/iommu.h>
#include <asm/tlbflush.h>
#include <asm/tlb.h>
#include <asm/prom.h>
#include <asm/abs_addr.h>
#include <asm/cputable.h>
#include <asm/plpar_wrappers.h>
#ifdef DEBUG
#define DBG(fmt...) udbg_printf(fmt)
#else
#define DBG(fmt...)
#endif
/* in pSeries_hvCall.S */
EXPORT_SYMBOL(plpar_hcall);
EXPORT_SYMBOL(plpar_hcall_4out);
EXPORT_SYMBOL(plpar_hcall_norets);
EXPORT_SYMBOL(plpar_hcall_8arg_2ret);
extern void pSeries_find_serial_port(void);
int vtermno; /* virtual terminal# for udbg */
#define __ALIGNED__ __attribute__((__aligned__(sizeof(long))))
static void udbg_hvsi_putc(unsigned char c)
{
/* packet's seqno isn't used anyways */
uint8_t packet[] __ALIGNED__ = { 0xff, 5, 0, 0, c };
int rc;
if (c == '\n')
udbg_hvsi_putc('\r');
do {
rc = plpar_put_term_char(vtermno, sizeof(packet), packet);
} while (rc == H_Busy);
}
static long hvsi_udbg_buf_len;
static uint8_t hvsi_udbg_buf[256];
static int udbg_hvsi_getc_poll(void)
{
unsigned char ch;
int rc, i;
if (hvsi_udbg_buf_len == 0) {
rc = plpar_get_term_char(vtermno, &hvsi_udbg_buf_len, hvsi_udbg_buf);
if (rc != H_Success || hvsi_udbg_buf[0] != 0xff) {
/* bad read or non-data packet */
hvsi_udbg_buf_len = 0;
} else {
/* remove the packet header */
for (i = 4; i < hvsi_udbg_buf_len; i++)
hvsi_udbg_buf[i-4] = hvsi_udbg_buf[i];
hvsi_udbg_buf_len -= 4;
}
}
if (hvsi_udbg_buf_len <= 0 || hvsi_udbg_buf_len > 256) {
/* no data ready */
hvsi_udbg_buf_len = 0;
return -1;
}
ch = hvsi_udbg_buf[0];
/* shift remaining data down */
for (i = 1; i < hvsi_udbg_buf_len; i++) {
hvsi_udbg_buf[i-1] = hvsi_udbg_buf[i];
}
hvsi_udbg_buf_len--;
return ch;
}
static unsigned char udbg_hvsi_getc(void)
{
int ch;
for (;;) {
ch = udbg_hvsi_getc_poll();
if (ch == -1) {
/* This shouldn't be needed...but... */
volatile unsigned long delay;
for (delay=0; delay < 2000000; delay++)
;
} else {
return ch;
}
}
}
static void udbg_putcLP(unsigned char c)
{
char buf[16];
unsigned long rc;
if (c == '\n')
udbg_putcLP('\r');
buf[0] = c;
do {
rc = plpar_put_term_char(vtermno, 1, buf);
} while(rc == H_Busy);
}
/* Buffered chars getc */
static long inbuflen;
static long inbuf[2]; /* must be 2 longs */
static int udbg_getc_pollLP(void)
{
/* The interface is tricky because it may return up to 16 chars.
* We save them statically for future calls to udbg_getc().
*/
char ch, *buf = (char *)inbuf;
int i;
long rc;
if (inbuflen == 0) {
/* get some more chars. */
inbuflen = 0;
rc = plpar_get_term_char(vtermno, &inbuflen, buf);
if (rc != H_Success)
inbuflen = 0; /* otherwise inbuflen is garbage */
}
if (inbuflen <= 0 || inbuflen > 16) {
/* Catch error case as well as other oddities (corruption) */
inbuflen = 0;
return -1;
}
ch = buf[0];
for (i = 1; i < inbuflen; i++) /* shuffle them down. */
buf[i-1] = buf[i];
inbuflen--;
return ch;
}
static unsigned char udbg_getcLP(void)
{
int ch;
for (;;) {
ch = udbg_getc_pollLP();
if (ch == -1) {
/* This shouldn't be needed...but... */
volatile unsigned long delay;
for (delay=0; delay < 2000000; delay++)
;
} else {
return ch;
}
}
}
/* call this from early_init() for a working debug console on
* vterm capable LPAR machines
*/
void udbg_init_debug_lpar(void)
{
vtermno = 0;
udbg_putc = udbg_putcLP;
udbg_getc = udbg_getcLP;
udbg_getc_poll = udbg_getc_pollLP;
}
/* returns 0 if couldn't find or use /chosen/stdout as console */
int find_udbg_vterm(void)
{
struct device_node *stdout_node;
u32 *termno;
char *name;
int found = 0;
/* find the boot console from /chosen/stdout */
if (!of_chosen)
return 0;
name = (char *)get_property(of_chosen, "linux,stdout-path", NULL);
if (name == NULL)
return 0;
stdout_node = of_find_node_by_path(name);
if (!stdout_node)
return 0;
/* now we have the stdout node; figure out what type of device it is. */
name = (char *)get_property(stdout_node, "name", NULL);
if (!name) {
printk(KERN_WARNING "stdout node missing 'name' property!\n");
goto out;
}
if (strncmp(name, "vty", 3) == 0) {
if (device_is_compatible(stdout_node, "hvterm1")) {
termno = (u32 *)get_property(stdout_node, "reg", NULL);
if (termno) {
vtermno = termno[0];
udbg_putc = udbg_putcLP;
udbg_getc = udbg_getcLP;
udbg_getc_poll = udbg_getc_pollLP;
found = 1;
}
} else if (device_is_compatible(stdout_node, "hvterm-protocol")) {
termno = (u32 *)get_property(stdout_node, "reg", NULL);
if (termno) {
vtermno = termno[0];
udbg_putc = udbg_hvsi_putc;
udbg_getc = udbg_hvsi_getc;
udbg_getc_poll = udbg_hvsi_getc_poll;
found = 1;
}
}
} else if (strncmp(name, "serial", 6)) {
/* XXX fix ISA serial console */
printk(KERN_WARNING "serial stdout on LPAR ('%s')! "
"can't print udbg messages\n",
stdout_node->full_name);
} else {
printk(KERN_WARNING "don't know how to print to stdout '%s'\n",
stdout_node->full_name);
}
out:
of_node_put(stdout_node);
return found;
}
void vpa_init(int cpu)
{
int hwcpu = get_hard_smp_processor_id(cpu);
unsigned long vpa = (unsigned long)&(paca[cpu].lppaca);
long ret;
unsigned long flags;
/* Register the Virtual Processor Area (VPA) */
flags = 1UL << (63 - 18);
if (cpu_has_feature(CPU_FTR_ALTIVEC))
paca[cpu].lppaca.vmxregs_in_use = 1;
ret = register_vpa(flags, hwcpu, __pa(vpa));
if (ret)
printk(KERN_ERR "WARNING: vpa_init: VPA registration for "
"cpu %d (hw %d) of area %lx returns %ld\n",
cpu, hwcpu, __pa(vpa), ret);
}
long pSeries_lpar_hpte_insert(unsigned long hpte_group,
unsigned long va, unsigned long prpn,
unsigned long vflags, unsigned long rflags)
{
unsigned long lpar_rc;
unsigned long flags;
unsigned long slot;
unsigned long hpte_v, hpte_r;
unsigned long dummy0, dummy1;
hpte_v = ((va >> 23) << HPTE_V_AVPN_SHIFT) | vflags | HPTE_V_VALID;
if (vflags & HPTE_V_LARGE)
hpte_v &= ~(1UL << HPTE_V_AVPN_SHIFT);
hpte_r = (prpn << HPTE_R_RPN_SHIFT) | rflags;
/* Now fill in the actual HPTE */
/* Set CEC cookie to 0 */
/* Zero page = 0 */
/* I-cache Invalidate = 0 */
/* I-cache synchronize = 0 */
/* Exact = 0 */
flags = 0;
/* XXX why is this here? - Anton */
if (rflags & (_PAGE_GUARDED|_PAGE_NO_CACHE))
hpte_r &= ~_PAGE_COHERENT;
lpar_rc = plpar_hcall(H_ENTER, flags, hpte_group, hpte_v,
hpte_r, &slot, &dummy0, &dummy1);
if (unlikely(lpar_rc == H_PTEG_Full))
return -1;
/*
* Since we try and ioremap PHBs we don't own, the pte insert
* will fail. However we must catch the failure in hash_page
* or we will loop forever, so return -2 in this case.
*/
if (unlikely(lpar_rc != H_Success))
return -2;
/* Because of iSeries, we have to pass down the secondary
* bucket bit here as well
*/
return (slot & 7) | (!!(vflags & HPTE_V_SECONDARY) << 3);
}
static DEFINE_SPINLOCK(pSeries_lpar_tlbie_lock);
static long pSeries_lpar_hpte_remove(unsigned long hpte_group)
{
unsigned long slot_offset;
unsigned long lpar_rc;
int i;
unsigned long dummy1, dummy2;
/* pick a random slot to start at */
slot_offset = mftb() & 0x7;
for (i = 0; i < HPTES_PER_GROUP; i++) {
/* don't remove a bolted entry */
lpar_rc = plpar_pte_remove(H_ANDCOND, hpte_group + slot_offset,
(0x1UL << 4), &dummy1, &dummy2);
if (lpar_rc == H_Success)
return i;
BUG_ON(lpar_rc != H_Not_Found);
slot_offset++;
slot_offset &= 0x7;
}
return -1;
}
static void pSeries_lpar_hptab_clear(void)
{
unsigned long size_bytes = 1UL << ppc64_pft_size;
unsigned long hpte_count = size_bytes >> 4;
unsigned long dummy1, dummy2;
int i;
/* TODO: Use bulk call */
for (i = 0; i < hpte_count; i++)
plpar_pte_remove(0, i, 0, &dummy1, &dummy2);
}
/*
* NOTE: for updatepp ops we are fortunate that the linux "newpp" bits and
* the low 3 bits of flags happen to line up. So no transform is needed.
* We can probably optimize here and assume the high bits of newpp are
* already zero. For now I am paranoid.
*/
static long pSeries_lpar_hpte_updatepp(unsigned long slot, unsigned long newpp,
unsigned long va, int large, int local)
{
unsigned long lpar_rc;
unsigned long flags = (newpp & 7) | H_AVPN;
unsigned long avpn = va >> 23;
if (large)
avpn &= ~0x1UL;
lpar_rc = plpar_pte_protect(flags, slot, (avpn << 7));
if (lpar_rc == H_Not_Found)
return -1;
BUG_ON(lpar_rc != H_Success);
return 0;
}
static unsigned long pSeries_lpar_hpte_getword0(unsigned long slot)
{
unsigned long dword0;
unsigned long lpar_rc;
unsigned long dummy_word1;
unsigned long flags;
/* Read 1 pte at a time */
/* Do not need RPN to logical page translation */
/* No cross CEC PFT access */
flags = 0;
lpar_rc = plpar_pte_read(flags, slot, &dword0, &dummy_word1);
BUG_ON(lpar_rc != H_Success);
return dword0;
}
static long pSeries_lpar_hpte_find(unsigned long vpn)
{
unsigned long hash;
unsigned long i, j;
long slot;
unsigned long hpte_v;
hash = hpt_hash(vpn, 0);
for (j = 0; j < 2; j++) {
slot = (hash & htab_hash_mask) * HPTES_PER_GROUP;
for (i = 0; i < HPTES_PER_GROUP; i++) {
hpte_v = pSeries_lpar_hpte_getword0(slot);
if ((HPTE_V_AVPN_VAL(hpte_v) == (vpn >> 11))
&& (hpte_v & HPTE_V_VALID)
&& (!!(hpte_v & HPTE_V_SECONDARY) == j)) {
/* HPTE matches */
if (j)
slot = -slot;
return slot;
}
++slot;
}
hash = ~hash;
}
return -1;
}
static void pSeries_lpar_hpte_updateboltedpp(unsigned long newpp,
unsigned long ea)
{
unsigned long lpar_rc;
unsigned long vsid, va, vpn, flags;
long slot;
vsid = get_kernel_vsid(ea);
va = (vsid << 28) | (ea & 0x0fffffff);
vpn = va >> PAGE_SHIFT;
slot = pSeries_lpar_hpte_find(vpn);
BUG_ON(slot == -1);
flags = newpp & 7;
lpar_rc = plpar_pte_protect(flags, slot, 0);
BUG_ON(lpar_rc != H_Success);
}
static void pSeries_lpar_hpte_invalidate(unsigned long slot, unsigned long va,
int large, int local)
{
unsigned long avpn = va >> 23;
unsigned long lpar_rc;
unsigned long dummy1, dummy2;
if (large)
avpn &= ~0x1UL;
lpar_rc = plpar_pte_remove(H_AVPN, slot, (avpn << 7), &dummy1,
&dummy2);
if (lpar_rc == H_Not_Found)
return;
BUG_ON(lpar_rc != H_Success);
}
/*
* Take a spinlock around flushes to avoid bouncing the hypervisor tlbie
* lock.
*/
void pSeries_lpar_flush_hash_range(unsigned long number, int local)
{
int i;
unsigned long flags = 0;
struct ppc64_tlb_batch *batch = &__get_cpu_var(ppc64_tlb_batch);
int lock_tlbie = !cpu_has_feature(CPU_FTR_LOCKLESS_TLBIE);
if (lock_tlbie)
spin_lock_irqsave(&pSeries_lpar_tlbie_lock, flags);
for (i = 0; i < number; i++)
flush_hash_page(batch->vaddr[i], batch->pte[i], local);
if (lock_tlbie)
spin_unlock_irqrestore(&pSeries_lpar_tlbie_lock, flags);
}
void hpte_init_lpar(void)
{
ppc_md.hpte_invalidate = pSeries_lpar_hpte_invalidate;
ppc_md.hpte_updatepp = pSeries_lpar_hpte_updatepp;
ppc_md.hpte_updateboltedpp = pSeries_lpar_hpte_updateboltedpp;
ppc_md.hpte_insert = pSeries_lpar_hpte_insert;
ppc_md.hpte_remove = pSeries_lpar_hpte_remove;
ppc_md.flush_hash_range = pSeries_lpar_flush_hash_range;
ppc_md.hpte_clear_all = pSeries_lpar_hptab_clear;
htab_finish_init();
}
View Git Blame Copy Permalink