Kill the 4MB kernel limit dead. [I hope :-)].

For UP, we were using $tmp_stk as a stack from the data section.  If the
kernel text section grew beyond ~3MB, the data section would be pushed
beyond the temporary 4MB P==V mapping.  This would cause the trampoline
up to high memory to fault.  The hack workaround I did was to use all of
the page table pages that we already have while preparing the initial
P==V mapping, instead of just the first one.
For SMP, the AP bootstrap process suffered the same sort of problem and
got the same treatment.

MFC candidate - this breaks on 4.x just the same..

Thanks to:	Richard Todd <rmtodd@ichotolot.servalan.com>

6 files changed, 32 insertions, 13 deletions

diff --git a/sys/amd64/amd64/locore.S b/sys/amd64/amd64/locore.S
index 68927d5..379af45 100644
--- a/sys/amd64/amd64/locore.S
+++ b/sys/amd64/amd64/locore.S
@@ -889,7 +889,7 @@ map_read_write:
 /* install a pde for temporary double map of bottom of VA */
 	movl	R(KPTphys), %eax
 	xorl	%ebx, %ebx
-	movl	$1, %ecx
+	movl	$NKPT, %ecx
 	fillkpt(R(IdlePTD), $PG_RW)
 
 /* install pde's for pt's */
diff --git a/sys/amd64/amd64/locore.s b/sys/amd64/amd64/locore.s
index 68927d5..379af45 100644
--- a/sys/amd64/amd64/locore.s
+++ b/sys/amd64/amd64/locore.s
@@ -889,7 +889,7 @@ map_read_write:
 /* install a pde for temporary double map of bottom of VA */
 	movl	R(KPTphys), %eax
 	xorl	%ebx, %ebx
-	movl	$1, %ecx
+	movl	$NKPT, %ecx
 	fillkpt(R(IdlePTD), $PG_RW)
 
 /* install pde's for pt's */
diff --git a/sys/amd64/amd64/mp_machdep.c b/sys/amd64/amd64/mp_machdep.c
index f6d23dd..9797905 100644
--- a/sys/amd64/amd64/mp_machdep.c
+++ b/sys/amd64/amd64/mp_machdep.c
@@ -1943,6 +1943,7 @@ start_all_aps(u_int boot_addr)
 	u_long  mpbioswarmvec;
 	struct globaldata *gd;
 	char *stack;
+	uintptr_t kptbase;
 
 	POSTCODE(START_ALL_APS_POST);
 
@@ -1964,8 +1965,12 @@ start_all_aps(u_int boot_addr)
 	/* record BSP in CPU map */
 	all_cpus = 1;
 
-	/* set up 0 -> 4MB P==V mapping for AP boot */
-	*(int *)PTD = PG_V | PG_RW | ((uintptr_t)(void *)KPTphys & PG_FRAME);
+	/* set up temporary P==V mapping for AP boot */
+	/* XXX this is a hack, we should boot the AP on its own stack/PTD */
+	kptbase = (uintptr_t)(void *)KPTphys;
+	for (x = 0; x < NKPT; x++)
+		PTD[x] = (pd_entry_t)(PG_V | PG_RW |
+		    ((kptbase + x * PAGE_SIZE) & PG_FRAME));
 	invltlb();
 
 	/* start each AP */
@@ -2047,7 +2052,8 @@ start_all_aps(u_int boot_addr)
 		SMPpt[1 + i] = (pt_entry_t)
 		    (PG_V | PG_RW | vtophys(PAGE_SIZE * i + stack));
 
-	*(int *)PTD = 0;
+	for (x = 0; x < NKPT; x++)
+		PTD[x] = 0;
 	pmap_set_opt();
 
 	/* number of APs actually started */
diff --git a/sys/amd64/amd64/mptable.c b/sys/amd64/amd64/mptable.c
index f6d23dd..9797905 100644
--- a/sys/amd64/amd64/mptable.c
+++ b/sys/amd64/amd64/mptable.c
@@ -1943,6 +1943,7 @@ start_all_aps(u_int boot_addr)
 	u_long  mpbioswarmvec;
 	struct globaldata *gd;
 	char *stack;
+	uintptr_t kptbase;
 
 	POSTCODE(START_ALL_APS_POST);
 
@@ -1964,8 +1965,12 @@ start_all_aps(u_int boot_addr)
 	/* record BSP in CPU map */
 	all_cpus = 1;
 
-	/* set up 0 -> 4MB P==V mapping for AP boot */
-	*(int *)PTD = PG_V | PG_RW | ((uintptr_t)(void *)KPTphys & PG_FRAME);
+	/* set up temporary P==V mapping for AP boot */
+	/* XXX this is a hack, we should boot the AP on its own stack/PTD */
+	kptbase = (uintptr_t)(void *)KPTphys;
+	for (x = 0; x < NKPT; x++)
+		PTD[x] = (pd_entry_t)(PG_V | PG_RW |
+		    ((kptbase + x * PAGE_SIZE) & PG_FRAME));
 	invltlb();
 
 	/* start each AP */
@@ -2047,7 +2052,8 @@ start_all_aps(u_int boot_addr)
 		SMPpt[1 + i] = (pt_entry_t)
 		    (PG_V | PG_RW | vtophys(PAGE_SIZE * i + stack));
 
-	*(int *)PTD = 0;
+	for (x = 0; x < NKPT; x++)
+		PTD[x] = 0;
 	pmap_set_opt();
 
 	/* number of APs actually started */
diff --git a/sys/amd64/amd64/pmap.c b/sys/amd64/amd64/pmap.c
index 4bf0ace..9360328 100644
--- a/sys/amd64/amd64/pmap.c
+++ b/sys/amd64/amd64/pmap.c
@@ -282,6 +282,7 @@ pmap_bootstrap(firstaddr, loadaddr)
 {
 	vm_offset_t va;
 	pt_entry_t *pte;
+	int i;
 
 	avail_start = firstaddr;
 
@@ -360,8 +361,8 @@ pmap_bootstrap(firstaddr, loadaddr)
 	virtual_avail = va;
 
 	*(int *) CMAP1 = *(int *) CMAP2 = 0;
-	*(int *) PTD = 0;
-
+	for (i = 0; i < NKPT; i++)
+		PTD[i] = 0;
 
 	pgeflag = 0;
 #if !defined(SMP)			/* XXX - see also mp_machdep.c */
diff --git a/sys/amd64/include/mptable.h b/sys/amd64/include/mptable.h
index f6d23dd..9797905 100644
--- a/sys/amd64/include/mptable.h
+++ b/sys/amd64/include/mptable.h
@@ -1943,6 +1943,7 @@ start_all_aps(u_int boot_addr)
 	u_long  mpbioswarmvec;
 	struct globaldata *gd;
 	char *stack;
+	uintptr_t kptbase;
 
 	POSTCODE(START_ALL_APS_POST);
 
@@ -1964,8 +1965,12 @@ start_all_aps(u_int boot_addr)
 	/* record BSP in CPU map */
 	all_cpus = 1;
 
-	/* set up 0 -> 4MB P==V mapping for AP boot */
-	*(int *)PTD = PG_V | PG_RW | ((uintptr_t)(void *)KPTphys & PG_FRAME);
+	/* set up temporary P==V mapping for AP boot */
+	/* XXX this is a hack, we should boot the AP on its own stack/PTD */
+	kptbase = (uintptr_t)(void *)KPTphys;
+	for (x = 0; x < NKPT; x++)
+		PTD[x] = (pd_entry_t)(PG_V | PG_RW |
+		    ((kptbase + x * PAGE_SIZE) & PG_FRAME));
 	invltlb();
 
 	/* start each AP */
@@ -2047,7 +2052,8 @@ start_all_aps(u_int boot_addr)
 		SMPpt[1 + i] = (pt_entry_t)
 		    (PG_V | PG_RW | vtophys(PAGE_SIZE * i + stack));
 
-	*(int *)PTD = 0;
+	for (x = 0; x < NKPT; x++)
+		PTD[x] = 0;
 	pmap_set_opt();
 
 	/* number of APs actually started */