d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

o Use uint64_t for page number as it doesn't fit uint32_t.

Browse Source 
o Implement growkernel bits for L1 level of pagetables.

This allows us to boot with 128GB of physical memory.

Sponsored by:	DARPA, AFRL
Sponsored by:	HEIF5
This commit is contained in:
Ruslan Bukin 2016-02-26 14:04:00 +00:00
parent 461a7671c7
commit 14232d424d
2 changed files with 16 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
sys/riscv/include/pte.h
View File

@ -41,6 +41,7 @@
#ifndef LOCORE
typedef uint64_t pd_entry_t; /* page directory entry */
typedef uint64_t pt_entry_t; /* page table entry */
typedef uint64_t pn_t; /* page number */
#endif
/* Level 0 table, 512GiB per entry */

29
sys/riscv/riscv/pmap.c
View File

@ -450,7 +450,7 @@ pmap_bootstrap_dmap(vm_offset_t l1pt, vm_paddr_t kernstart)
pd_entry_t *l1;
u_int l1_slot;
pt_entry_t entry;
u_int pn;
pn_t pn;
pa = kernstart & ~L1_OFFSET;
va = DMAP_MIN_ADDRESS;
@ -924,7 +924,7 @@ pmap_qenter(vm_offset_t sva, vm_page_t *ma, int count)
vm_offset_t va;
vm_page_t m;
pt_entry_t entry;
u_int pn;
pn_t pn;
int i;
va = sva;
@ -1143,7 +1143,7 @@ _pmap_alloc_l3(pmap_t pmap, vm_pindex_t ptepindex, struct rwlock **lockp)
vm_page_t m, /*pdppg, */pdpg;
pt_entry_t entry;
vm_paddr_t phys;
int pn;
pn_t pn;
PMAP_LOCK_ASSERT(pmap, MA_OWNED);
@ -1323,7 +1323,7 @@ pmap_growkernel(vm_offset_t addr)
vm_page_t nkpg;
pd_entry_t *l1, *l2;
pt_entry_t entry;
int pn;
pn_t pn;
mtx_assert(&kernel_map->system_mtx, MA_OWNED);
@ -1343,10 +1343,11 @@ pmap_growkernel(vm_offset_t addr)
pmap_zero_page(nkpg);
paddr = VM_PAGE_TO_PHYS(nkpg);
panic("%s: implement grow l1\n", __func__);
#if 0
pmap_load_store(l1, paddr | L1_TABLE);
#endif
pn = (paddr / PAGE_SIZE);
entry = (PTE_VALID | (PTE_TYPE_PTR << PTE_TYPE_S));
entry |= (pn << PTE_PPN0_S);
pmap_load_store(l1, entry);
PTE_SYNC(l1);
continue; /* try again */
}
@ -1933,9 +1934,9 @@ pmap_enter(pmap_t pmap, vm_offset_t va, vm_page_t m, vm_prot_t prot,
vm_page_t mpte, om, l2_m, l3_m;
boolean_t nosleep;
pt_entry_t entry;
int l2_pn;
int l3_pn;
int pn;
pn_t l2_pn;
pn_t l3_pn;
pn_t pn;
va = trunc_page(va);
if ((m->oflags & VPO_UNMANAGED) == 0 && !vm_page_xbusied(m))
@ -2211,7 +2212,7 @@ pmap_enter_quick_locked(pmap_t pmap, vm_offset_t va, vm_page_t m,
pt_entry_t *l3;
vm_paddr_t pa;
pt_entry_t entry;
int pn;
pn_t pn;
KASSERT(va < kmi.clean_sva || va >= kmi.clean_eva ||
(m->oflags & VPO_UNMANAGED) != 0,
@ -3084,8 +3085,8 @@ pmap_mincore(pmap_t pmap, vm_offset_t addr, vm_paddr_t *locked_pa)
void
pmap_activate(struct thread *td)
{
uint64_t entry;
uint64_t pn;
pt_entry_t entry;
pn_t pn;
pmap_t pmap;
critical_enter();