^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) * Xen mmu operations
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * This file contains the various mmu fetch and update operations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * The most important job they must perform is the mapping between the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) * domain's pfn and the overall machine mfns.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) * Xen allows guests to directly update the pagetable, in a controlled
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) * fashion. In other words, the guest modifies the same pagetable
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) * that the CPU actually uses, which eliminates the overhead of having
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) * a separate shadow pagetable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) * In order to allow this, it falls on the guest domain to map its
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) * notion of a "physical" pfn - which is just a domain-local linear
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) * address - into a real "machine address" which the CPU's MMU can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) * use.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) * A pgd_t/pmd_t/pte_t will typically contain an mfn, and so can be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) * inserted directly into the pagetable. When creating a new
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) * pte/pmd/pgd, it converts the passed pfn into an mfn. Conversely,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) * when reading the content back with __(pgd|pmd|pte)_val, it converts
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) * the mfn back into a pfn.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) * The other constraint is that all pages which make up a pagetable
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) * must be mapped read-only in the guest. This prevents uncontrolled
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) * guest updates to the pagetable. Xen strictly enforces this, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) * will disallow any pagetable update which will end up mapping a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) * pagetable page RW, and will disallow using any writable page as a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) * pagetable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) * Naively, when loading %cr3 with the base of a new pagetable, Xen
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) * would need to validate the whole pagetable before going on.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) * Naturally, this is quite slow. The solution is to "pin" a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) * pagetable, which enforces all the constraints on the pagetable even
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) * when it is not actively in use. This menas that Xen can be assured
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) * that it is still valid when you do load it into %cr3, and doesn't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) * need to revalidate it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) * Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) #include <linux/sched/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) #include <linux/highmem.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) #include <linux/debugfs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) #include <linux/bug.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) #include <linux/vmalloc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) #include <linux/export.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) #include <linux/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) #include <linux/gfp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) #include <linux/memblock.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) #include <linux/seq_file.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) #include <linux/crash_dump.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) #include <linux/pgtable.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) #ifdef CONFIG_KEXEC_CORE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) #include <linux/kexec.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) #include <trace/events/xen.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) #include <asm/tlbflush.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) #include <asm/fixmap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) #include <asm/mmu_context.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) #include <asm/setup.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) #include <asm/paravirt.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) #include <asm/e820/api.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) #include <asm/linkage.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) #include <asm/page.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) #include <asm/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) #include <asm/memtype.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) #include <asm/smp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) #include <asm/tlb.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) #include <asm/xen/hypercall.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) #include <asm/xen/hypervisor.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) #include <xen/xen.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) #include <xen/page.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) #include <xen/interface/xen.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) #include <xen/interface/hvm/hvm_op.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) #include <xen/interface/version.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) #include <xen/interface/memory.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) #include <xen/hvc-console.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) #include "multicalls.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) #include "mmu.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) #include "debugfs.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) /* l3 pud for userspace vsyscall mapping */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) static pud_t level3_user_vsyscall[PTRS_PER_PUD] __page_aligned_bss;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) * Protects atomic reservation decrease/increase against concurrent increases.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) * Also protects non-atomic updates of current_pages and balloon lists.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) static DEFINE_SPINLOCK(xen_reservation_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) * Note about cr3 (pagetable base) values:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) * xen_cr3 contains the current logical cr3 value; it contains the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) * last set cr3. This may not be the current effective cr3, because
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) * its update may be being lazily deferred. However, a vcpu looking
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) * at its own cr3 can use this value knowing that it everything will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) * be self-consistent.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) * xen_current_cr3 contains the actual vcpu cr3; it is set once the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) * hypercall to set the vcpu cr3 is complete (so it may be a little
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) * out of date, but it will never be set early). If one vcpu is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) * looking at another vcpu's cr3 value, it should use this variable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) DEFINE_PER_CPU(unsigned long, xen_cr3); /* cr3 stored as physaddr */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) DEFINE_PER_CPU(unsigned long, xen_current_cr3); /* actual vcpu cr3 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) static phys_addr_t xen_pt_base, xen_pt_size __initdata;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) static DEFINE_STATIC_KEY_FALSE(xen_struct_pages_ready);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) * Just beyond the highest usermode address. STACK_TOP_MAX has a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) * redzone above it, so round it up to a PGD boundary.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) #define USER_LIMIT ((STACK_TOP_MAX + PGDIR_SIZE - 1) & PGDIR_MASK)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) void make_lowmem_page_readonly(void *vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) pte_t *pte, ptev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) unsigned long address = (unsigned long)vaddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) unsigned int level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) pte = lookup_address(address, &level);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) if (pte == NULL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) return; /* vaddr missing */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) ptev = pte_wrprotect(*pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) if (HYPERVISOR_update_va_mapping(address, ptev, 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) void make_lowmem_page_readwrite(void *vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) pte_t *pte, ptev;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) unsigned long address = (unsigned long)vaddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) unsigned int level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) pte = lookup_address(address, &level);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) if (pte == NULL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) return; /* vaddr missing */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) ptev = pte_mkwrite(*pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) if (HYPERVISOR_update_va_mapping(address, ptev, 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) * During early boot all page table pages are pinned, but we do not have struct
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) * pages, so return true until struct pages are ready.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) static bool xen_page_pinned(void *ptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) if (static_branch_likely(&xen_struct_pages_ready)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) struct page *page = virt_to_page(ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) return PagePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) static void xen_extend_mmu_update(const struct mmu_update *update)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) struct mmu_update *u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) mcs = xen_mc_extend_args(__HYPERVISOR_mmu_update, sizeof(*u));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) if (mcs.mc != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) mcs.mc->args[1]++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) mcs = __xen_mc_entry(sizeof(*u));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, DOMID_SELF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) u = mcs.args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) *u = *update;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) static void xen_extend_mmuext_op(const struct mmuext_op *op)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) struct mmuext_op *u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) mcs = xen_mc_extend_args(__HYPERVISOR_mmuext_op, sizeof(*u));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) if (mcs.mc != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) mcs.mc->args[1]++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) mcs = __xen_mc_entry(sizeof(*u));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) MULTI_mmuext_op(mcs.mc, mcs.args, 1, NULL, DOMID_SELF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) u = mcs.args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) *u = *op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) static void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) struct mmu_update u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) preempt_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) /* ptr may be ioremapped for 64-bit pagetable setup */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) u.ptr = arbitrary_virt_to_machine(ptr).maddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) u.val = pmd_val_ma(val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219) xen_extend_mmu_update(&u);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) preempt_enable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) static void xen_set_pmd(pmd_t *ptr, pmd_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) trace_xen_mmu_set_pmd(ptr, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) /* If page is not pinned, we can just update the entry
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) directly */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) if (!xen_page_pinned(ptr)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) *ptr = val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) xen_set_pmd_hyper(ptr, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) * Associate a virtual page frame with a given physical page frame
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) * and protection flags for that frame.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) void set_pte_mfn(unsigned long vaddr, unsigned long mfn, pgprot_t flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) set_pte_vaddr(vaddr, mfn_pte(mfn, flags));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) static bool xen_batched_set_pte(pte_t *ptep, pte_t pteval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) struct mmu_update u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) if (paravirt_get_lazy_mode() != PARAVIRT_LAZY_MMU)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) u.ptr = virt_to_machine(ptep).maddr | MMU_NORMAL_PT_UPDATE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) u.val = pte_val_ma(pteval);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) xen_extend_mmu_update(&u);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) static inline void __xen_set_pte(pte_t *ptep, pte_t pteval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) if (!xen_batched_set_pte(ptep, pteval)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) * Could call native_set_pte() here and trap and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) * emulate the PTE write, but a hypercall is much cheaper.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) struct mmu_update u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) u.ptr = virt_to_machine(ptep).maddr | MMU_NORMAL_PT_UPDATE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) u.val = pte_val_ma(pteval);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) static void xen_set_pte(pte_t *ptep, pte_t pteval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) trace_xen_mmu_set_pte(ptep, pteval);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) __xen_set_pte(ptep, pteval);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) pte_t xen_ptep_modify_prot_start(struct vm_area_struct *vma,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) unsigned long addr, pte_t *ptep)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) /* Just return the pte as-is. We preserve the bits on commit */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) trace_xen_mmu_ptep_modify_prot_start(vma->vm_mm, addr, ptep, *ptep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) return *ptep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) void xen_ptep_modify_prot_commit(struct vm_area_struct *vma, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) pte_t *ptep, pte_t pte)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) struct mmu_update u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) trace_xen_mmu_ptep_modify_prot_commit(vma->vm_mm, addr, ptep, pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) u.ptr = virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) u.val = pte_val_ma(pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) xen_extend_mmu_update(&u);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) /* Assume pteval_t is equivalent to all the other *val_t types. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) static pteval_t pte_mfn_to_pfn(pteval_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) if (val & _PAGE_PRESENT) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) unsigned long mfn = (val & XEN_PTE_MFN_MASK) >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) unsigned long pfn = mfn_to_pfn(mfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) pteval_t flags = val & PTE_FLAGS_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) if (unlikely(pfn == ~0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) val = flags & ~_PAGE_PRESENT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) val = ((pteval_t)pfn << PAGE_SHIFT) | flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) return val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) static pteval_t pte_pfn_to_mfn(pteval_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) if (val & _PAGE_PRESENT) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) unsigned long pfn = (val & PTE_PFN_MASK) >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) pteval_t flags = val & PTE_FLAGS_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) unsigned long mfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) mfn = __pfn_to_mfn(pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) * If there's no mfn for the pfn, then just create an
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) * empty non-present pte. Unfortunately this loses
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) * information about the original pfn, so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) * pte_mfn_to_pfn is asymmetric.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) if (unlikely(mfn == INVALID_P2M_ENTRY)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) mfn = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) mfn &= ~(FOREIGN_FRAME_BIT | IDENTITY_FRAME_BIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) val = ((pteval_t)mfn << PAGE_SHIFT) | flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) return val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) __visible pteval_t xen_pte_val(pte_t pte)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) pteval_t pteval = pte.pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) return pte_mfn_to_pfn(pteval);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) PV_CALLEE_SAVE_REGS_THUNK(xen_pte_val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) __visible pgdval_t xen_pgd_val(pgd_t pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) return pte_mfn_to_pfn(pgd.pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) PV_CALLEE_SAVE_REGS_THUNK(xen_pgd_val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) __visible pte_t xen_make_pte(pteval_t pte)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) pte = pte_pfn_to_mfn(pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) return native_make_pte(pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) PV_CALLEE_SAVE_REGS_THUNK(xen_make_pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) __visible pgd_t xen_make_pgd(pgdval_t pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) pgd = pte_pfn_to_mfn(pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) return native_make_pgd(pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) PV_CALLEE_SAVE_REGS_THUNK(xen_make_pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) __visible pmdval_t xen_pmd_val(pmd_t pmd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) return pte_mfn_to_pfn(pmd.pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) PV_CALLEE_SAVE_REGS_THUNK(xen_pmd_val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) static void xen_set_pud_hyper(pud_t *ptr, pud_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) struct mmu_update u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) preempt_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) /* ptr may be ioremapped for 64-bit pagetable setup */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) u.ptr = arbitrary_virt_to_machine(ptr).maddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) u.val = pud_val_ma(val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) xen_extend_mmu_update(&u);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) preempt_enable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) static void xen_set_pud(pud_t *ptr, pud_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) trace_xen_mmu_set_pud(ptr, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) /* If page is not pinned, we can just update the entry
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) directly */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) if (!xen_page_pinned(ptr)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) *ptr = val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) xen_set_pud_hyper(ptr, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) __visible pmd_t xen_make_pmd(pmdval_t pmd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) pmd = pte_pfn_to_mfn(pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) return native_make_pmd(pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) PV_CALLEE_SAVE_REGS_THUNK(xen_make_pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) __visible pudval_t xen_pud_val(pud_t pud)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) return pte_mfn_to_pfn(pud.pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) PV_CALLEE_SAVE_REGS_THUNK(xen_pud_val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) __visible pud_t xen_make_pud(pudval_t pud)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) pud = pte_pfn_to_mfn(pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) return native_make_pud(pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) PV_CALLEE_SAVE_REGS_THUNK(xen_make_pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) static pgd_t *xen_get_user_pgd(pgd_t *pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) pgd_t *pgd_page = (pgd_t *)(((unsigned long)pgd) & PAGE_MASK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) unsigned offset = pgd - pgd_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) pgd_t *user_ptr = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) if (offset < pgd_index(USER_LIMIT)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) struct page *page = virt_to_page(pgd_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) user_ptr = (pgd_t *)page->private;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) if (user_ptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) user_ptr += offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) return user_ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) static void __xen_set_p4d_hyper(p4d_t *ptr, p4d_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) struct mmu_update u;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) u.ptr = virt_to_machine(ptr).maddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) u.val = p4d_val_ma(val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) xen_extend_mmu_update(&u);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) * Raw hypercall-based set_p4d, intended for in early boot before
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) * there's a page structure. This implies:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) * 1. The only existing pagetable is the kernel's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) * 2. It is always pinned
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) * 3. It has no user pagetable attached to it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) static void __init xen_set_p4d_hyper(p4d_t *ptr, p4d_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) preempt_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) __xen_set_p4d_hyper(ptr, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) preempt_enable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) static void xen_set_p4d(p4d_t *ptr, p4d_t val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) pgd_t *user_ptr = xen_get_user_pgd((pgd_t *)ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) pgd_t pgd_val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) trace_xen_mmu_set_p4d(ptr, (p4d_t *)user_ptr, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) /* If page is not pinned, we can just update the entry
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495) directly */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) if (!xen_page_pinned(ptr)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) *ptr = val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) if (user_ptr) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) WARN_ON(xen_page_pinned(user_ptr));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) pgd_val.pgd = p4d_val_ma(val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) *user_ptr = pgd_val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) /* If it's pinned, then we can at least batch the kernel and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) user updates together. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) __xen_set_p4d_hyper(ptr, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) if (user_ptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) __xen_set_p4d_hyper((p4d_t *)user_ptr, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) #if CONFIG_PGTABLE_LEVELS >= 5
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) __visible p4dval_t xen_p4d_val(p4d_t p4d)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) return pte_mfn_to_pfn(p4d.p4d);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) PV_CALLEE_SAVE_REGS_THUNK(xen_p4d_val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) __visible p4d_t xen_make_p4d(p4dval_t p4d)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) p4d = pte_pfn_to_mfn(p4d);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) return native_make_p4d(p4d);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) PV_CALLEE_SAVE_REGS_THUNK(xen_make_p4d);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) #endif /* CONFIG_PGTABLE_LEVELS >= 5 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) static void xen_pmd_walk(struct mm_struct *mm, pmd_t *pmd,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) void (*func)(struct mm_struct *mm, struct page *,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) enum pt_level),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) bool last, unsigned long limit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) int i, nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) nr = last ? pmd_index(limit) + 1 : PTRS_PER_PMD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) for (i = 0; i < nr; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) if (!pmd_none(pmd[i]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) (*func)(mm, pmd_page(pmd[i]), PT_PTE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) static void xen_pud_walk(struct mm_struct *mm, pud_t *pud,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) void (*func)(struct mm_struct *mm, struct page *,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) enum pt_level),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) bool last, unsigned long limit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) int i, nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) nr = last ? pud_index(limit) + 1 : PTRS_PER_PUD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) for (i = 0; i < nr; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) if (pud_none(pud[i]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) pmd = pmd_offset(&pud[i], 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) if (PTRS_PER_PMD > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) (*func)(mm, virt_to_page(pmd), PT_PMD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) xen_pmd_walk(mm, pmd, func, last && i == nr - 1, limit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) static void xen_p4d_walk(struct mm_struct *mm, p4d_t *p4d,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) void (*func)(struct mm_struct *mm, struct page *,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) enum pt_level),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) bool last, unsigned long limit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) if (p4d_none(*p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) pud = pud_offset(p4d, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) if (PTRS_PER_PUD > 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) (*func)(mm, virt_to_page(pud), PT_PUD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) xen_pud_walk(mm, pud, func, last, limit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) * (Yet another) pagetable walker. This one is intended for pinning a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) * pagetable. This means that it walks a pagetable and calls the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) * callback function on each page it finds making up the page table,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) * at every level. It walks the entire pagetable, but it only bothers
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) * pinning pte pages which are below limit. In the normal case this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) * will be STACK_TOP_MAX, but at boot we need to pin up to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) * FIXADDR_TOP.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) * We must skip the Xen hole in the middle of the address space, just after
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) * the big x86-64 virtual hole.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) static void __xen_pgd_walk(struct mm_struct *mm, pgd_t *pgd,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) void (*func)(struct mm_struct *mm, struct page *,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) enum pt_level),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) unsigned long limit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) int i, nr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) unsigned hole_low = 0, hole_high = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) /* The limit is the last byte to be touched */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) limit--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) BUG_ON(limit >= FIXADDR_TOP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) * 64-bit has a great big hole in the middle of the address
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) * space, which contains the Xen mappings.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) hole_low = pgd_index(GUARD_HOLE_BASE_ADDR);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) hole_high = pgd_index(GUARD_HOLE_END_ADDR);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) nr = pgd_index(limit) + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) for (i = 0; i < nr; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) if (i >= hole_low && i < hole_high)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) if (pgd_none(pgd[i]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) p4d = p4d_offset(&pgd[i], 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) xen_p4d_walk(mm, p4d, func, i == nr - 1, limit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) /* Do the top level last, so that the callbacks can use it as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) a cue to do final things like tlb flushes. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) (*func)(mm, virt_to_page(pgd), PT_PGD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) static void xen_pgd_walk(struct mm_struct *mm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) void (*func)(struct mm_struct *mm, struct page *,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) enum pt_level),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) unsigned long limit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) __xen_pgd_walk(mm, mm->pgd, func, limit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) /* If we're using split pte locks, then take the page's lock and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) return a pointer to it. Otherwise return NULL. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) static spinlock_t *xen_pte_lock(struct page *page, struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) spinlock_t *ptl = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) #if USE_SPLIT_PTE_PTLOCKS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) ptl = ptlock_ptr(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) spin_lock_nest_lock(ptl, &mm->page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) return ptl;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) static void xen_pte_unlock(void *v)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) spinlock_t *ptl = v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) spin_unlock(ptl);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) static void xen_do_pin(unsigned level, unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) struct mmuext_op op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) op.cmd = level;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) op.arg1.mfn = pfn_to_mfn(pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) xen_extend_mmuext_op(&op);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) static void xen_pin_page(struct mm_struct *mm, struct page *page,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) enum pt_level level)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) unsigned pgfl = TestSetPagePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) if (!pgfl) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) void *pt = lowmem_page_address(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) unsigned long pfn = page_to_pfn(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) struct multicall_space mcs = __xen_mc_entry(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) spinlock_t *ptl;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) * We need to hold the pagetable lock between the time
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686) * we make the pagetable RO and when we actually pin
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) * it. If we don't, then other users may come in and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) * attempt to update the pagetable by writing it,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) * which will fail because the memory is RO but not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) * pinned, so Xen won't do the trap'n'emulate.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) * If we're using split pte locks, we can't hold the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) * entire pagetable's worth of locks during the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) * traverse, because we may wrap the preempt count (8
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) * bits). The solution is to mark RO and pin each PTE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) * page while holding the lock. This means the number
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) * of locks we end up holding is never more than a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) * batch size (~32 entries, at present).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) * If we're not using split pte locks, we needn't pin
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) * the PTE pages independently, because we're
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) * protected by the overall pagetable lock.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) ptl = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) if (level == PT_PTE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) ptl = xen_pte_lock(page, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) MULTI_update_va_mapping(mcs.mc, (unsigned long)pt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) pfn_pte(pfn, PAGE_KERNEL_RO),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) level == PT_PGD ? UVMF_TLB_FLUSH : 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) if (ptl) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) xen_do_pin(MMUEXT_PIN_L1_TABLE, pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) /* Queue a deferred unlock for when this batch
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) is completed. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) xen_mc_callback(xen_pte_unlock, ptl);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) /* This is called just after a mm has been created, but it has not
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) been used yet. We need to make sure that its pagetable is all
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) read-only, and can be pinned. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) static void __xen_pgd_pin(struct mm_struct *mm, pgd_t *pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) pgd_t *user_pgd = xen_get_user_pgd(pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) trace_xen_mmu_pgd_pin(mm, pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) __xen_pgd_walk(mm, pgd, xen_pin_page, USER_LIMIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) xen_do_pin(MMUEXT_PIN_L4_TABLE, PFN_DOWN(__pa(pgd)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) if (user_pgd) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) xen_pin_page(mm, virt_to_page(user_pgd), PT_PGD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) xen_do_pin(MMUEXT_PIN_L4_TABLE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) PFN_DOWN(__pa(user_pgd)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743) xen_mc_issue(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) static void xen_pgd_pin(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) __xen_pgd_pin(mm, mm->pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) * On save, we need to pin all pagetables to make sure they get their
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) * mfns turned into pfns. Search the list for any unpinned pgds and pin
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) * them (unpinned pgds are not currently in use, probably because the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) * process is under construction or destruction).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) * Expected to be called in stop_machine() ("equivalent to taking
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) * every spinlock in the system"), so the locking doesn't really
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) * matter all that much.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) void xen_mm_pin_all(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) spin_lock(&pgd_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) list_for_each_entry(page, &pgd_list, lru) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) if (!PagePinned(page)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) __xen_pgd_pin(&init_mm, (pgd_t *)page_address(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) SetPageSavePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) spin_unlock(&pgd_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) static void __init xen_mark_pinned(struct mm_struct *mm, struct page *page,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) enum pt_level level)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) SetPagePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) * The init_mm pagetable is really pinned as soon as its created, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) * that's before we have page structures to store the bits. So do all
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) * the book-keeping now once struct pages for allocated pages are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) * initialized. This happens only after memblock_free_all() is called.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) static void __init xen_after_bootmem(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) static_branch_enable(&xen_struct_pages_ready);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) SetPagePinned(virt_to_page(level3_user_vsyscall));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) xen_pgd_walk(&init_mm, xen_mark_pinned, FIXADDR_TOP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) static void xen_unpin_page(struct mm_struct *mm, struct page *page,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) enum pt_level level)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) unsigned pgfl = TestClearPagePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) if (pgfl) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) void *pt = lowmem_page_address(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) unsigned long pfn = page_to_pfn(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) spinlock_t *ptl = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) * Do the converse to pin_page. If we're using split
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) * pte locks, we must be holding the lock for while
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) * the pte page is unpinned but still RO to prevent
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) * concurrent updates from seeing it in this
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) * partially-pinned state.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) if (level == PT_PTE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) ptl = xen_pte_lock(page, mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) if (ptl)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) xen_do_pin(MMUEXT_UNPIN_TABLE, pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) mcs = __xen_mc_entry(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) MULTI_update_va_mapping(mcs.mc, (unsigned long)pt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) pfn_pte(pfn, PAGE_KERNEL),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) level == PT_PGD ? UVMF_TLB_FLUSH : 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) if (ptl) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) /* unlock when batch completed */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) xen_mc_callback(xen_pte_unlock, ptl);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) /* Release a pagetables pages back as normal RW */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) static void __xen_pgd_unpin(struct mm_struct *mm, pgd_t *pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) pgd_t *user_pgd = xen_get_user_pgd(pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) trace_xen_mmu_pgd_unpin(mm, pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) xen_do_pin(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) if (user_pgd) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) xen_do_pin(MMUEXT_UNPIN_TABLE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) PFN_DOWN(__pa(user_pgd)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) xen_unpin_page(mm, virt_to_page(user_pgd), PT_PGD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) __xen_pgd_walk(mm, pgd, xen_unpin_page, USER_LIMIT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) xen_mc_issue(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) static void xen_pgd_unpin(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) __xen_pgd_unpin(mm, mm->pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) * On resume, undo any pinning done at save, so that the rest of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) * kernel doesn't see any unexpected pinned pagetables.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) void xen_mm_unpin_all(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) spin_lock(&pgd_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) list_for_each_entry(page, &pgd_list, lru) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) if (PageSavePinned(page)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) BUG_ON(!PagePinned(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) __xen_pgd_unpin(&init_mm, (pgd_t *)page_address(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) ClearPageSavePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) spin_unlock(&pgd_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) static void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) spin_lock(&next->page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) xen_pgd_pin(next);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) spin_unlock(&next->page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) static void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) spin_lock(&mm->page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) xen_pgd_pin(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) spin_unlock(&mm->page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) static void drop_mm_ref_this_cpu(void *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) struct mm_struct *mm = info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) if (this_cpu_read(cpu_tlbstate.loaded_mm) == mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) leave_mm(smp_processor_id());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) * If this cpu still has a stale cr3 reference, then make sure
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) * it has been flushed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) if (this_cpu_read(xen_current_cr3) == __pa(mm->pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) xen_mc_flush();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) #ifdef CONFIG_SMP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) * Another cpu may still have their %cr3 pointing at the pagetable, so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) * we need to repoint it somewhere else before we can unpin it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) static void xen_drop_mm_ref(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) cpumask_var_t mask;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) unsigned cpu;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) drop_mm_ref_this_cpu(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) /* Get the "official" set of cpus referring to our pagetable. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) if (!alloc_cpumask_var(&mask, GFP_ATOMIC)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) for_each_online_cpu(cpu) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) if (per_cpu(xen_current_cr3, cpu) != __pa(mm->pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) smp_call_function_single(cpu, drop_mm_ref_this_cpu, mm, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) * It's possible that a vcpu may have a stale reference to our
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) * cr3, because its in lazy mode, and it hasn't yet flushed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) * its set of pending hypercalls yet. In this case, we can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) * look at its actual current cr3 value, and force it to flush
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) * if needed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) cpumask_clear(mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) for_each_online_cpu(cpu) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) if (per_cpu(xen_current_cr3, cpu) == __pa(mm->pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) cpumask_set_cpu(cpu, mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) smp_call_function_many(mask, drop_mm_ref_this_cpu, mm, 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) free_cpumask_var(mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) static void xen_drop_mm_ref(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) drop_mm_ref_this_cpu(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957) * While a process runs, Xen pins its pagetables, which means that the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) * hypervisor forces it to be read-only, and it controls all updates
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) * to it. This means that all pagetable updates have to go via the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) * hypervisor, which is moderately expensive.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) * Since we're pulling the pagetable down, we switch to use init_mm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) * unpin old process pagetable and mark it all read-write, which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) * allows further operations on it to be simple memory accesses.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) * The only subtle point is that another CPU may be still using the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) * pagetable because of lazy tlb flushing. This means we need need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) * switch all CPUs off this pagetable before we can unpin it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) static void xen_exit_mmap(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) get_cpu(); /* make sure we don't move around */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) xen_drop_mm_ref(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) put_cpu();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) spin_lock(&mm->page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) /* pgd may not be pinned in the error exit path of execve */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) if (xen_page_pinned(mm->pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) xen_pgd_unpin(mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) spin_unlock(&mm->page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) static void xen_post_allocator_init(void);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987) static void __init pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) struct mmuext_op op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) op.cmd = cmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) op.arg1.mfn = pfn_to_mfn(pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) if (HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) static void __init xen_cleanhighmap(unsigned long vaddr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) unsigned long vaddr_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) unsigned long kernel_end = roundup((unsigned long)_brk_end, PMD_SIZE) - 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) pmd_t *pmd = level2_kernel_pgt + pmd_index(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) /* NOTE: The loop is more greedy than the cleanup_highmap variant.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) * We include the PMD passed in on _both_ boundaries. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) for (; vaddr <= vaddr_end && (pmd < (level2_kernel_pgt + PTRS_PER_PMD));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) pmd++, vaddr += PMD_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) if (pmd_none(*pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) if (vaddr < (unsigned long) _text || vaddr > kernel_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010) set_pmd(pmd, __pmd(0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) /* In case we did something silly, we should crash in this function
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) * instead of somewhere later and be confusing. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) xen_mc_flush();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) * Make a page range writeable and free it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020) static void __init xen_free_ro_pages(unsigned long paddr, unsigned long size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) void *vaddr = __va(paddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) void *vaddr_end = vaddr + size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) for (; vaddr < vaddr_end; vaddr += PAGE_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) make_lowmem_page_readwrite(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) memblock_free(paddr, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) static void __init xen_cleanmfnmap_free_pgtbl(void *pgtbl, bool unpin)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) unsigned long pa = __pa(pgtbl) & PHYSICAL_PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) if (unpin)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(pa));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) ClearPagePinned(virt_to_page(__va(pa)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038) xen_free_ro_pages(pa, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) static void __init xen_cleanmfnmap_pmd(pmd_t *pmd, bool unpin)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) unsigned long pa;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044) pte_t *pte_tbl;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) if (pmd_large(*pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) pa = pmd_val(*pmd) & PHYSICAL_PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) xen_free_ro_pages(pa, PMD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) pte_tbl = pte_offset_kernel(pmd, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) for (i = 0; i < PTRS_PER_PTE; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) if (pte_none(pte_tbl[i]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) pa = pte_pfn(pte_tbl[i]) << PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) xen_free_ro_pages(pa, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060) set_pmd(pmd, __pmd(0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) xen_cleanmfnmap_free_pgtbl(pte_tbl, unpin);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) static void __init xen_cleanmfnmap_pud(pud_t *pud, bool unpin)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) unsigned long pa;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) pmd_t *pmd_tbl;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) if (pud_large(*pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) pa = pud_val(*pud) & PHYSICAL_PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) xen_free_ro_pages(pa, PUD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) pmd_tbl = pmd_offset(pud, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) for (i = 0; i < PTRS_PER_PMD; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) if (pmd_none(pmd_tbl[i]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) xen_cleanmfnmap_pmd(pmd_tbl + i, unpin);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) set_pud(pud, __pud(0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) xen_cleanmfnmap_free_pgtbl(pmd_tbl, unpin);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) static void __init xen_cleanmfnmap_p4d(p4d_t *p4d, bool unpin)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) unsigned long pa;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) pud_t *pud_tbl;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) if (p4d_large(*p4d)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) pa = p4d_val(*p4d) & PHYSICAL_PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) xen_free_ro_pages(pa, P4D_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) pud_tbl = pud_offset(p4d, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) for (i = 0; i < PTRS_PER_PUD; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) if (pud_none(pud_tbl[i]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) xen_cleanmfnmap_pud(pud_tbl + i, unpin);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) set_p4d(p4d, __p4d(0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) xen_cleanmfnmap_free_pgtbl(pud_tbl, unpin);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) * Since it is well isolated we can (and since it is perhaps large we should)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) * also free the page tables mapping the initial P->M table.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) static void __init xen_cleanmfnmap(unsigned long vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116) bool unpin;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) unpin = (vaddr == 2 * PGDIR_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) vaddr &= PMD_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) pgd = pgd_offset_k(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) p4d = p4d_offset(pgd, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) if (!p4d_none(*p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) xen_cleanmfnmap_p4d(p4d, unpin);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) static void __init xen_pagetable_p2m_free(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) unsigned long size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) unsigned long addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) /* No memory or already called. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) if ((unsigned long)xen_p2m_addr == xen_start_info->mfn_list)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) /* using __ka address and sticking INVALID_P2M_ENTRY! */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) memset((void *)xen_start_info->mfn_list, 0xff, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) addr = xen_start_info->mfn_list;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) * We could be in __ka space.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) * We roundup to the PMD, which means that if anybody at this stage is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) * using the __ka address of xen_start_info or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) * xen_start_info->shared_info they are in going to crash. Fortunately
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) * we have already revectored in xen_setup_kernel_pagetable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) size = roundup(size, PMD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) if (addr >= __START_KERNEL_map) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) xen_cleanhighmap(addr, addr + size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152) size = PAGE_ALIGN(xen_start_info->nr_pages *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) sizeof(unsigned long));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) memblock_free(__pa(addr), size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) xen_cleanmfnmap(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) static void __init xen_pagetable_cleanhighmap(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) unsigned long size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163) unsigned long addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) /* At this stage, cleanup_highmap has already cleaned __ka space
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166) * from _brk_limit way up to the max_pfn_mapped (which is the end of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) * the ramdisk). We continue on, erasing PMD entries that point to page
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) * tables - do note that they are accessible at this stage via __va.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) * As Xen is aligning the memory end to a 4MB boundary, for good
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) * measure we also round up to PMD_SIZE * 2 - which means that if
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) * anybody is using __ka address to the initial boot-stack - and try
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172) * to use it - they are going to crash. The xen_start_info has been
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) * taken care of already in xen_setup_kernel_pagetable. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) addr = xen_start_info->pt_base;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175) size = xen_start_info->nr_pt_frames * PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) xen_cleanhighmap(addr, roundup(addr + size, PMD_SIZE * 2));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) xen_start_info->pt_base = (unsigned long)__va(__pa(xen_start_info->pt_base));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181) static void __init xen_pagetable_p2m_setup(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) xen_vmalloc_p2m_tree();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) xen_pagetable_p2m_free();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) xen_pagetable_cleanhighmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) /* And revector! Bye bye old array */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) xen_start_info->mfn_list = (unsigned long)xen_p2m_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) static void __init xen_pagetable_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) paging_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196) xen_post_allocator_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) xen_pagetable_p2m_setup();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) /* Allocate and initialize top and mid mfn levels for p2m structure */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) xen_build_mfn_list_list();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) /* Remap memory freed due to conflicts with E820 map */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204) xen_remap_memory();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205) xen_setup_mfn_list_list();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) static void xen_write_cr2(unsigned long cr2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) this_cpu_read(xen_vcpu)->arch.cr2 = cr2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) static noinline void xen_flush_tlb(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) struct mmuext_op *op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) preempt_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) mcs = xen_mc_entry(sizeof(*op));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) op = mcs.args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) op->cmd = MMUEXT_TLB_FLUSH_LOCAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) preempt_enable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) static void xen_flush_tlb_one_user(unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) struct mmuext_op *op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235) trace_xen_mmu_flush_tlb_one_user(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) preempt_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) mcs = xen_mc_entry(sizeof(*op));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) op = mcs.args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) op->cmd = MMUEXT_INVLPG_LOCAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) op->arg1.linear_addr = addr & PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) preempt_enable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) static void xen_flush_tlb_others(const struct cpumask *cpus,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251) const struct flush_tlb_info *info)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) struct {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254) struct mmuext_op op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) DECLARE_BITMAP(mask, NR_CPUS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) } *args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) const size_t mc_entry_size = sizeof(args->op) +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) sizeof(args->mask[0]) * BITS_TO_LONGS(num_possible_cpus());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) trace_xen_mmu_flush_tlb_others(cpus, info->mm, info->start, info->end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) if (cpumask_empty(cpus))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) return; /* nothing to do */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) mcs = xen_mc_entry(mc_entry_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) args = mcs.args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) args->op.arg2.vcpumask = to_cpumask(args->mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) /* Remove us, and any offline CPUS. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) cpumask_and(to_cpumask(args->mask), cpus, cpu_online_mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272) cpumask_clear_cpu(smp_processor_id(), to_cpumask(args->mask));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) args->op.cmd = MMUEXT_TLB_FLUSH_MULTI;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275) if (info->end != TLB_FLUSH_ALL &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) (info->end - info->start) <= PAGE_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277) args->op.cmd = MMUEXT_INVLPG_MULTI;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) args->op.arg1.linear_addr = info->start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281) MULTI_mmuext_op(mcs.mc, &args->op, 1, NULL, DOMID_SELF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) static unsigned long xen_read_cr3(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) return this_cpu_read(xen_cr3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) static void set_current_cr3(void *v)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) this_cpu_write(xen_current_cr3, (unsigned long)v);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) static void __xen_write_cr3(bool kernel, unsigned long cr3)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) struct mmuext_op op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299) unsigned long mfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) trace_xen_mmu_write_cr3(kernel, cr3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303) if (cr3)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) mfn = pfn_to_mfn(PFN_DOWN(cr3));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) mfn = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) WARN_ON(mfn == 0 && kernel);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) op.cmd = kernel ? MMUEXT_NEW_BASEPTR : MMUEXT_NEW_USER_BASEPTR;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) op.arg1.mfn = mfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) xen_extend_mmuext_op(&op);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315) if (kernel) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316) this_cpu_write(xen_cr3, cr3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) /* Update xen_current_cr3 once the batch has actually
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) been submitted. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) xen_mc_callback(set_current_cr3, (void *)cr3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323) static void xen_write_cr3(unsigned long cr3)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325) pgd_t *user_pgd = xen_get_user_pgd(__va(cr3));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) BUG_ON(preemptible());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) xen_mc_batch(); /* disables interrupts */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331) /* Update while interrupts are disabled, so its atomic with
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) respect to ipis */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) this_cpu_write(xen_cr3, cr3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335) __xen_write_cr3(true, cr3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) if (user_pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338) __xen_write_cr3(false, __pa(user_pgd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) __xen_write_cr3(false, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) xen_mc_issue(PARAVIRT_LAZY_CPU); /* interrupts restored */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346) * At the start of the day - when Xen launches a guest, it has already
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) * built pagetables for the guest. We diligently look over them
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) * in xen_setup_kernel_pagetable and graft as appropriate them in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) * init_top_pgt and its friends. Then when we are happy we load
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350) * the new init_top_pgt - and continue on.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) * The generic code starts (start_kernel) and 'init_mem_mapping' sets
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) * up the rest of the pagetables. When it has completed it loads the cr3.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) * N.B. that baremetal would start at 'start_kernel' (and the early
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) * #PF handler would create bootstrap pagetables) - so we are running
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) * with the same assumptions as what to do when write_cr3 is executed
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) * at this point.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359) * Since there are no user-page tables at all, we have two variants
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) * of xen_write_cr3 - the early bootup (this one), and the late one
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361) * (xen_write_cr3). The reason we have to do that is that in 64-bit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362) * the Linux kernel and user-space are both in ring 3 while the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363) * hypervisor is in ring 0.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) static void __init xen_write_cr3_init(unsigned long cr3)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) BUG_ON(preemptible());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369) xen_mc_batch(); /* disables interrupts */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) /* Update while interrupts are disabled, so its atomic with
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) respect to ipis */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373) this_cpu_write(xen_cr3, cr3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) __xen_write_cr3(true, cr3);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377) xen_mc_issue(PARAVIRT_LAZY_CPU); /* interrupts restored */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) static int xen_pgd_alloc(struct mm_struct *mm)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) pgd_t *pgd = mm->pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383) struct page *page = virt_to_page(pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) pgd_t *user_pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) int ret = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387) BUG_ON(PagePinned(virt_to_page(pgd)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) BUG_ON(page->private != 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390) user_pgd = (pgd_t *)__get_free_page(GFP_KERNEL | __GFP_ZERO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) page->private = (unsigned long)user_pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) if (user_pgd != NULL) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394) #ifdef CONFIG_X86_VSYSCALL_EMULATION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) user_pgd[pgd_index(VSYSCALL_ADDR)] =
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) __pgd(__pa(level3_user_vsyscall) | _PAGE_TABLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398) ret = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401) BUG_ON(PagePinned(virt_to_page(xen_get_user_pgd(pgd))));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406) static void xen_pgd_free(struct mm_struct *mm, pgd_t *pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) pgd_t *user_pgd = xen_get_user_pgd(pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) if (user_pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411) free_page((unsigned long)user_pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) * Init-time set_pte while constructing initial pagetables, which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416) * doesn't allow RO page table pages to be remapped RW.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) * If there is no MFN for this PFN then this page is initially
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) * ballooned out so clear the PTE (as in decrease_reservation() in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420) * drivers/xen/balloon.c).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422) * Many of these PTE updates are done on unpinned and writable pages
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423) * and doing a hypercall for these is unnecessary and expensive. At
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) * this point it is not possible to tell if a page is pinned or not,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425) * so always write the PTE directly and rely on Xen trapping and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) * emulating any updates as necessary.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428) __visible pte_t xen_make_pte_init(pteval_t pte)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430) unsigned long pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) * Pages belonging to the initial p2m list mapped outside the default
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434) * address range must be mapped read-only. This region contains the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) * page tables for mapping the p2m list, too, and page tables MUST be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) * mapped read-only.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) pfn = (pte & PTE_PFN_MASK) >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439) if (xen_start_info->mfn_list < __START_KERNEL_map &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) pfn >= xen_start_info->first_p2m_pfn &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) pfn < xen_start_info->first_p2m_pfn + xen_start_info->nr_p2m_frames)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) pte &= ~_PAGE_RW;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444) pte = pte_pfn_to_mfn(pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) return native_make_pte(pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447) PV_CALLEE_SAVE_REGS_THUNK(xen_make_pte_init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449) static void __init xen_set_pte_init(pte_t *ptep, pte_t pte)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) __xen_set_pte(ptep, pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454) /* Early in boot, while setting up the initial pagetable, assume
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455) everything is pinned. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) static void __init xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458) #ifdef CONFIG_FLATMEM
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459) BUG_ON(mem_map); /* should only be used early */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) make_lowmem_page_readonly(__va(PFN_PHYS(pfn)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465) /* Used for pmd and pud */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) static void __init xen_alloc_pmd_init(struct mm_struct *mm, unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468) #ifdef CONFIG_FLATMEM
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) BUG_ON(mem_map); /* should only be used early */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) make_lowmem_page_readonly(__va(PFN_PHYS(pfn)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474) /* Early release_pte assumes that all pts are pinned, since there's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475) only init_mm and anything attached to that is pinned. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) static void __init xen_release_pte_init(unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478) pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479) make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) static void __init xen_release_pmd_init(unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484) make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487) static inline void __pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490) struct mmuext_op *op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492) mcs = __xen_mc_entry(sizeof(*op));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) op = mcs.args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) op->cmd = cmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495) op->arg1.mfn = pfn_to_mfn(pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) MULTI_mmuext_op(mcs.mc, mcs.args, 1, NULL, DOMID_SELF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) static inline void __set_pfn_prot(unsigned long pfn, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) unsigned long addr = (unsigned long)__va(pfn << PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) mcs = __xen_mc_entry(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) MULTI_update_va_mapping(mcs.mc, (unsigned long)addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507) pfn_pte(pfn, prot), 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510) /* This needs to make sure the new pte page is pinned iff its being
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) attached to a pinned pagetable. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) static inline void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513) unsigned level)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) bool pinned = xen_page_pinned(mm->pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) trace_xen_mmu_alloc_ptpage(mm, pfn, level, pinned);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) if (pinned) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520) struct page *page = pfn_to_page(pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) if (static_branch_likely(&xen_struct_pages_ready))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523) SetPagePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527) __set_pfn_prot(pfn, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) if (level == PT_PTE && USE_SPLIT_PTE_PTLOCKS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) __pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) static void xen_alloc_pte(struct mm_struct *mm, unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538) xen_alloc_ptpage(mm, pfn, PT_PTE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) static void xen_alloc_pmd(struct mm_struct *mm, unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) xen_alloc_ptpage(mm, pfn, PT_PMD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) /* This should never happen until we're OK to use struct page */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) static inline void xen_release_ptpage(unsigned long pfn, unsigned level)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) struct page *page = pfn_to_page(pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) bool pinned = PagePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552) trace_xen_mmu_release_ptpage(pfn, level, pinned);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) if (pinned) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) if (level == PT_PTE && USE_SPLIT_PTE_PTLOCKS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558) __pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) __set_pfn_prot(pfn, PAGE_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) ClearPagePinned(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) static void xen_release_pte(unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) xen_release_ptpage(pfn, PT_PTE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) static void xen_release_pmd(unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) xen_release_ptpage(pfn, PT_PMD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) static void xen_alloc_pud(struct mm_struct *mm, unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) xen_alloc_ptpage(mm, pfn, PT_PUD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583) static void xen_release_pud(unsigned long pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585) xen_release_ptpage(pfn, PT_PUD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589) * Like __va(), but returns address in the kernel mapping (which is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590) * all we have until the physical memory mapping has been set up.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) static void * __init __ka(phys_addr_t paddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) return (void *)(paddr + __START_KERNEL_map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597) /* Convert a machine address to physical address */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) static unsigned long __init m2p(phys_addr_t maddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600) phys_addr_t paddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602) maddr &= XEN_PTE_MFN_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) paddr = mfn_to_pfn(maddr >> PAGE_SHIFT) << PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605) return paddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608) /* Convert a machine address to kernel virtual */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) static void * __init m2v(phys_addr_t maddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) return __ka(m2p(maddr));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614) /* Set the page permissions on an identity-mapped pages */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) static void __init set_page_prot_flags(void *addr, pgprot_t prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) unsigned long flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) unsigned long pfn = __pa(addr) >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619) pte_t pte = pfn_pte(pfn, prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, flags))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) static void __init set_page_prot(void *addr, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626) return set_page_prot_flags(addr, prot, UVMF_NONE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) void __init xen_setup_machphys_mapping(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631) struct xen_machphys_mapping mapping;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634) machine_to_phys_mapping = (unsigned long *)mapping.v_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) machine_to_phys_nr = mapping.max_mfn + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637) machine_to_phys_nr = MACH2PHYS_NR_ENTRIES;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641) static void __init convert_pfn_mfn(void *v)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643) pte_t *pte = v;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646) /* All levels are converted the same way, so just treat them
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) as ptes. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648) for (i = 0; i < PTRS_PER_PTE; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649) pte[i] = xen_make_pte(pte[i].pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651) static void __init check_pt_base(unsigned long *pt_base, unsigned long *pt_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652) unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1653) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1654) if (*pt_base == PFN_DOWN(__pa(addr))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1655) set_page_prot_flags((void *)addr, PAGE_KERNEL, UVMF_INVLPG);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1656) clear_page((void *)addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1657) (*pt_base)++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1658) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1659) if (*pt_end == PFN_DOWN(__pa(addr))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1660) set_page_prot_flags((void *)addr, PAGE_KERNEL, UVMF_INVLPG);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1661) clear_page((void *)addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1662) (*pt_end)--;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1663) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1664) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1665) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1666) * Set up the initial kernel pagetable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1667) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1668) * We can construct this by grafting the Xen provided pagetable into
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1669) * head_64.S's preconstructed pagetables. We copy the Xen L2's into
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1670) * level2_ident_pgt, and level2_kernel_pgt. This means that only the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1671) * kernel has a physical mapping to start with - but that's enough to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1672) * get __va working. We need to fill in the rest of the physical
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1673) * mapping once some sort of allocator has been set up.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1674) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1675) void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1676) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1677) pud_t *l3;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1678) pmd_t *l2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1679) unsigned long addr[3];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1680) unsigned long pt_base, pt_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1681) unsigned i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1682)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1683) /* max_pfn_mapped is the last pfn mapped in the initial memory
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1684) * mappings. Considering that on Xen after the kernel mappings we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1685) * have the mappings of some pages that don't exist in pfn space, we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1686) * set max_pfn_mapped to the last real pfn mapped. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1687) if (xen_start_info->mfn_list < __START_KERNEL_map)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1688) max_pfn_mapped = xen_start_info->first_p2m_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1689) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1690) max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->mfn_list));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1691)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1692) pt_base = PFN_DOWN(__pa(xen_start_info->pt_base));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1693) pt_end = pt_base + xen_start_info->nr_pt_frames;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1694)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1695) /* Zap identity mapping */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1696) init_top_pgt[0] = __pgd(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1697)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1698) /* Pre-constructed entries are in pfn, so convert to mfn */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1699) /* L4[273] -> level3_ident_pgt */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1700) /* L4[511] -> level3_kernel_pgt */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1701) convert_pfn_mfn(init_top_pgt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1702)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1703) /* L3_i[0] -> level2_ident_pgt */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1704) convert_pfn_mfn(level3_ident_pgt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1705) /* L3_k[510] -> level2_kernel_pgt */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1706) /* L3_k[511] -> level2_fixmap_pgt */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1707) convert_pfn_mfn(level3_kernel_pgt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1708)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1709) /* L3_k[511][508-FIXMAP_PMD_NUM ... 507] -> level1_fixmap_pgt */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1710) convert_pfn_mfn(level2_fixmap_pgt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1711)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1712) /* We get [511][511] and have Xen's version of level2_kernel_pgt */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1713) l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1714) l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1715)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1716) addr[0] = (unsigned long)pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1717) addr[1] = (unsigned long)l3;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1718) addr[2] = (unsigned long)l2;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1719) /* Graft it onto L4[273][0]. Note that we creating an aliasing problem:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1720) * Both L4[273][0] and L4[511][510] have entries that point to the same
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1721) * L2 (PMD) tables. Meaning that if you modify it in __va space
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1722) * it will be also modified in the __ka space! (But if you just
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1723) * modify the PMD table to point to other PTE's or none, then you
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1724) * are OK - which is what cleanup_highmap does) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1725) copy_page(level2_ident_pgt, l2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1726) /* Graft it onto L4[511][510] */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1727) copy_page(level2_kernel_pgt, l2);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1728)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1729) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1730) * Zap execute permission from the ident map. Due to the sharing of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1731) * L1 entries we need to do this in the L2.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1732) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1733) if (__supported_pte_mask & _PAGE_NX) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1734) for (i = 0; i < PTRS_PER_PMD; ++i) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1735) if (pmd_none(level2_ident_pgt[i]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1736) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1737) level2_ident_pgt[i] = pmd_set_flags(level2_ident_pgt[i], _PAGE_NX);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1738) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1739) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1740)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1741) /* Copy the initial P->M table mappings if necessary. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1742) i = pgd_index(xen_start_info->mfn_list);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1743) if (i && i < pgd_index(__START_KERNEL_map))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1744) init_top_pgt[i] = ((pgd_t *)xen_start_info->pt_base)[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1745)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1746) /* Make pagetable pieces RO */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1747) set_page_prot(init_top_pgt, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1748) set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1749) set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1750) set_page_prot(level3_user_vsyscall, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1751) set_page_prot(level2_ident_pgt, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1752) set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1753) set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1754)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1755) for (i = 0; i < FIXMAP_PMD_NUM; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1756) set_page_prot(level1_fixmap_pgt + i * PTRS_PER_PTE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1757) PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1758) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1759)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1760) /* Pin down new L4 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1761) pin_pagetable_pfn(MMUEXT_PIN_L4_TABLE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1762) PFN_DOWN(__pa_symbol(init_top_pgt)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1763)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1764) /* Unpin Xen-provided one */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1765) pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1766)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1767) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1768) * At this stage there can be no user pgd, and no page structure to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1769) * attach it to, so make sure we just set kernel pgd.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1770) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1771) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1772) __xen_write_cr3(true, __pa(init_top_pgt));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1773) xen_mc_issue(PARAVIRT_LAZY_CPU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1774)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1775) /* We can't that easily rip out L3 and L2, as the Xen pagetables are
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1776) * set out this way: [L4], [L1], [L2], [L3], [L1], [L1] ... for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1777) * the initial domain. For guests using the toolstack, they are in:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1778) * [L4], [L3], [L2], [L1], [L1], order .. So for dom0 we can only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1779) * rip out the [L4] (pgd), but for guests we shave off three pages.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1780) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1781) for (i = 0; i < ARRAY_SIZE(addr); i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1782) check_pt_base(&pt_base, &pt_end, addr[i]);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1783)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1784) /* Our (by three pages) smaller Xen pagetable that we are using */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1785) xen_pt_base = PFN_PHYS(pt_base);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1786) xen_pt_size = (pt_end - pt_base) * PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1787) memblock_reserve(xen_pt_base, xen_pt_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1788)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1789) /* Revector the xen_start_info */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1790) xen_start_info = (struct start_info *)__va(__pa(xen_start_info));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1791) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1792)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1793) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1794) * Read a value from a physical address.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1795) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1796) static unsigned long __init xen_read_phys_ulong(phys_addr_t addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1797) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1798) unsigned long *vaddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1799) unsigned long val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1800)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1801) vaddr = early_memremap_ro(addr, sizeof(val));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1802) val = *vaddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1803) early_memunmap(vaddr, sizeof(val));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1804) return val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1805) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1806)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1807) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1808) * Translate a virtual address to a physical one without relying on mapped
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1809) * page tables. Don't rely on big pages being aligned in (guest) physical
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1810) * space!
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1811) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1812) static phys_addr_t __init xen_early_virt_to_phys(unsigned long vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1813) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1814) phys_addr_t pa;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1815) pgd_t pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1816) pud_t pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1817) pmd_t pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1818) pte_t pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1819)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1820) pa = read_cr3_pa();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1821) pgd = native_make_pgd(xen_read_phys_ulong(pa + pgd_index(vaddr) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1822) sizeof(pgd)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1823) if (!pgd_present(pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1824) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1825)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1826) pa = pgd_val(pgd) & PTE_PFN_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1827) pud = native_make_pud(xen_read_phys_ulong(pa + pud_index(vaddr) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1828) sizeof(pud)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1829) if (!pud_present(pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1830) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1831) pa = pud_val(pud) & PTE_PFN_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1832) if (pud_large(pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1833) return pa + (vaddr & ~PUD_MASK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1834)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1835) pmd = native_make_pmd(xen_read_phys_ulong(pa + pmd_index(vaddr) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1836) sizeof(pmd)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1837) if (!pmd_present(pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1838) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1839) pa = pmd_val(pmd) & PTE_PFN_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1840) if (pmd_large(pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1841) return pa + (vaddr & ~PMD_MASK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1842)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1843) pte = native_make_pte(xen_read_phys_ulong(pa + pte_index(vaddr) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1844) sizeof(pte)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1845) if (!pte_present(pte))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1846) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1847) pa = pte_pfn(pte) << PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1848)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1849) return pa | (vaddr & ~PAGE_MASK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1850) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1851)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1852) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1853) * Find a new area for the hypervisor supplied p2m list and relocate the p2m to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1854) * this area.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1855) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1856) void __init xen_relocate_p2m(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1857) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1858) phys_addr_t size, new_area, pt_phys, pmd_phys, pud_phys;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1859) unsigned long p2m_pfn, p2m_pfn_end, n_frames, pfn, pfn_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1860) int n_pte, n_pt, n_pmd, n_pud, idx_pte, idx_pt, idx_pmd, idx_pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1861) pte_t *pt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1862) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1863) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1864) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1865) unsigned long *new_p2m;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1866)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1867) size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1868) n_pte = roundup(size, PAGE_SIZE) >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1869) n_pt = roundup(size, PMD_SIZE) >> PMD_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1870) n_pmd = roundup(size, PUD_SIZE) >> PUD_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1871) n_pud = roundup(size, P4D_SIZE) >> P4D_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1872) n_frames = n_pte + n_pt + n_pmd + n_pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1873)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1874) new_area = xen_find_free_area(PFN_PHYS(n_frames));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1875) if (!new_area) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1876) xen_raw_console_write("Can't find new memory area for p2m needed due to E820 map conflict\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1877) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1878) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1879)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1880) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1881) * Setup the page tables for addressing the new p2m list.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1882) * We have asked the hypervisor to map the p2m list at the user address
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1883) * PUD_SIZE. It may have done so, or it may have used a kernel space
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1884) * address depending on the Xen version.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1885) * To avoid any possible virtual address collision, just use
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1886) * 2 * PUD_SIZE for the new area.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1887) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1888) pud_phys = new_area;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1889) pmd_phys = pud_phys + PFN_PHYS(n_pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1890) pt_phys = pmd_phys + PFN_PHYS(n_pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1891) p2m_pfn = PFN_DOWN(pt_phys) + n_pt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1892)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1893) pgd = __va(read_cr3_pa());
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1894) new_p2m = (unsigned long *)(2 * PGDIR_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1895) for (idx_pud = 0; idx_pud < n_pud; idx_pud++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1896) pud = early_memremap(pud_phys, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1897) clear_page(pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1898) for (idx_pmd = 0; idx_pmd < min(n_pmd, PTRS_PER_PUD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1899) idx_pmd++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1900) pmd = early_memremap(pmd_phys, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1901) clear_page(pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1902) for (idx_pt = 0; idx_pt < min(n_pt, PTRS_PER_PMD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1903) idx_pt++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1904) pt = early_memremap(pt_phys, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1905) clear_page(pt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1906) for (idx_pte = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1907) idx_pte < min(n_pte, PTRS_PER_PTE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1908) idx_pte++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1909) pt[idx_pte] = pfn_pte(p2m_pfn,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1910) PAGE_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1911) p2m_pfn++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1912) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1913) n_pte -= PTRS_PER_PTE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1914) early_memunmap(pt, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1915) make_lowmem_page_readonly(__va(pt_phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1916) pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1917) PFN_DOWN(pt_phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1918) pmd[idx_pt] = __pmd(_PAGE_TABLE | pt_phys);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1919) pt_phys += PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1920) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1921) n_pt -= PTRS_PER_PMD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1922) early_memunmap(pmd, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1923) make_lowmem_page_readonly(__va(pmd_phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1924) pin_pagetable_pfn(MMUEXT_PIN_L2_TABLE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1925) PFN_DOWN(pmd_phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1926) pud[idx_pmd] = __pud(_PAGE_TABLE | pmd_phys);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1927) pmd_phys += PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1928) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1929) n_pmd -= PTRS_PER_PUD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1930) early_memunmap(pud, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1931) make_lowmem_page_readonly(__va(pud_phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1932) pin_pagetable_pfn(MMUEXT_PIN_L3_TABLE, PFN_DOWN(pud_phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1933) set_pgd(pgd + 2 + idx_pud, __pgd(_PAGE_TABLE | pud_phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1934) pud_phys += PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1935) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1936)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1937) /* Now copy the old p2m info to the new area. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1938) memcpy(new_p2m, xen_p2m_addr, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1939) xen_p2m_addr = new_p2m;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1940)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1941) /* Release the old p2m list and set new list info. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1942) p2m_pfn = PFN_DOWN(xen_early_virt_to_phys(xen_start_info->mfn_list));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1943) BUG_ON(!p2m_pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1944) p2m_pfn_end = p2m_pfn + PFN_DOWN(size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1945)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1946) if (xen_start_info->mfn_list < __START_KERNEL_map) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1947) pfn = xen_start_info->first_p2m_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1948) pfn_end = xen_start_info->first_p2m_pfn +
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1949) xen_start_info->nr_p2m_frames;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1950) set_pgd(pgd + 1, __pgd(0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1951) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1952) pfn = p2m_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1953) pfn_end = p2m_pfn_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1954) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1955)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1956) memblock_free(PFN_PHYS(pfn), PAGE_SIZE * (pfn_end - pfn));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1957) while (pfn < pfn_end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1958) if (pfn == p2m_pfn) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1959) pfn = p2m_pfn_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1960) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1961) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1962) make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1963) pfn++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1964) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1965)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1966) xen_start_info->mfn_list = (unsigned long)xen_p2m_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1967) xen_start_info->first_p2m_pfn = PFN_DOWN(new_area);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1968) xen_start_info->nr_p2m_frames = n_frames;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1969) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1970)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1971) void __init xen_reserve_special_pages(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1972) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1973) phys_addr_t paddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1974)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1975) memblock_reserve(__pa(xen_start_info), PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1976) if (xen_start_info->store_mfn) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1977) paddr = PFN_PHYS(mfn_to_pfn(xen_start_info->store_mfn));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1978) memblock_reserve(paddr, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1979) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1980) if (!xen_initial_domain()) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1981) paddr = PFN_PHYS(mfn_to_pfn(xen_start_info->console.domU.mfn));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1982) memblock_reserve(paddr, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1983) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1984) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1985)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1986) void __init xen_pt_check_e820(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1987) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1988) if (xen_is_e820_reserved(xen_pt_base, xen_pt_size)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1989) xen_raw_console_write("Xen hypervisor allocated page table memory conflicts with E820 map\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1990) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1991) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1992) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1993)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1994) static unsigned char dummy_mapping[PAGE_SIZE] __page_aligned_bss;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1995)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1996) static void xen_set_fixmap(unsigned idx, phys_addr_t phys, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1997) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1998) pte_t pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1999)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2000) phys >>= PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2001)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2002) switch (idx) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2003) case FIX_BTMAP_END ... FIX_BTMAP_BEGIN:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2004) #ifdef CONFIG_X86_VSYSCALL_EMULATION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2005) case VSYSCALL_PAGE:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2006) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2007) /* All local page mappings */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2008) pte = pfn_pte(phys, prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2009) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2010)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2011) #ifdef CONFIG_X86_LOCAL_APIC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2012) case FIX_APIC_BASE: /* maps dummy local APIC */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2013) pte = pfn_pte(PFN_DOWN(__pa(dummy_mapping)), PAGE_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2014) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2015) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2016)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2017) #ifdef CONFIG_X86_IO_APIC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2018) case FIX_IO_APIC_BASE_0 ... FIX_IO_APIC_BASE_END:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2019) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2020) * We just don't map the IO APIC - all access is via
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2021) * hypercalls. Keep the address in the pte for reference.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2022) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2023) pte = pfn_pte(PFN_DOWN(__pa(dummy_mapping)), PAGE_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2024) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2025) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2026)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2027) case FIX_PARAVIRT_BOOTMAP:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2028) /* This is an MFN, but it isn't an IO mapping from the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2029) IO domain */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2030) pte = mfn_pte(phys, prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2031) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2032)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2033) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2034) /* By default, set_fixmap is used for hardware mappings */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2035) pte = mfn_pte(phys, prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2036) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2037) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2038)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2039) __native_set_fixmap(idx, pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2040)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2041) #ifdef CONFIG_X86_VSYSCALL_EMULATION
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2042) /* Replicate changes to map the vsyscall page into the user
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2043) pagetable vsyscall mapping. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2044) if (idx == VSYSCALL_PAGE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2045) unsigned long vaddr = __fix_to_virt(idx);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2046) set_pte_vaddr_pud(level3_user_vsyscall, vaddr, pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2047) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2048) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2049) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2050)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2051) static void __init xen_post_allocator_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2052) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2053) pv_ops.mmu.set_pte = xen_set_pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2054) pv_ops.mmu.set_pmd = xen_set_pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2055) pv_ops.mmu.set_pud = xen_set_pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2056) pv_ops.mmu.set_p4d = xen_set_p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2057)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2058) /* This will work as long as patching hasn't happened yet
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2059) (which it hasn't) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2060) pv_ops.mmu.alloc_pte = xen_alloc_pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2061) pv_ops.mmu.alloc_pmd = xen_alloc_pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2062) pv_ops.mmu.release_pte = xen_release_pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2063) pv_ops.mmu.release_pmd = xen_release_pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2064) pv_ops.mmu.alloc_pud = xen_alloc_pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2065) pv_ops.mmu.release_pud = xen_release_pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2066) pv_ops.mmu.make_pte = PV_CALLEE_SAVE(xen_make_pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2067)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2068) pv_ops.mmu.write_cr3 = &xen_write_cr3;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2069) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2070)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2071) static void xen_leave_lazy_mmu(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2072) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2073) preempt_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2074) xen_mc_flush();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2075) paravirt_leave_lazy_mmu();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2076) preempt_enable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2077) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2078)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2079) static const struct pv_mmu_ops xen_mmu_ops __initconst = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2080) .read_cr2 = __PV_IS_CALLEE_SAVE(xen_read_cr2),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2081) .write_cr2 = xen_write_cr2,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2082)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2083) .read_cr3 = xen_read_cr3,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2084) .write_cr3 = xen_write_cr3_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2085)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2086) .flush_tlb_user = xen_flush_tlb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2087) .flush_tlb_kernel = xen_flush_tlb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2088) .flush_tlb_one_user = xen_flush_tlb_one_user,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2089) .flush_tlb_others = xen_flush_tlb_others,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2090) .tlb_remove_table = tlb_remove_table,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2091)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2092) .pgd_alloc = xen_pgd_alloc,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2093) .pgd_free = xen_pgd_free,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2094)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2095) .alloc_pte = xen_alloc_pte_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2096) .release_pte = xen_release_pte_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2097) .alloc_pmd = xen_alloc_pmd_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2098) .release_pmd = xen_release_pmd_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2099)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2100) .set_pte = xen_set_pte_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2101) .set_pmd = xen_set_pmd_hyper,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2102)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2103) .ptep_modify_prot_start = __ptep_modify_prot_start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2104) .ptep_modify_prot_commit = __ptep_modify_prot_commit,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2106) .pte_val = PV_CALLEE_SAVE(xen_pte_val),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2107) .pgd_val = PV_CALLEE_SAVE(xen_pgd_val),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2109) .make_pte = PV_CALLEE_SAVE(xen_make_pte_init),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2110) .make_pgd = PV_CALLEE_SAVE(xen_make_pgd),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2111)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2112) .set_pud = xen_set_pud_hyper,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2113)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2114) .make_pmd = PV_CALLEE_SAVE(xen_make_pmd),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2115) .pmd_val = PV_CALLEE_SAVE(xen_pmd_val),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2117) .pud_val = PV_CALLEE_SAVE(xen_pud_val),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2118) .make_pud = PV_CALLEE_SAVE(xen_make_pud),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2119) .set_p4d = xen_set_p4d_hyper,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2120)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2121) .alloc_pud = xen_alloc_pmd_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2122) .release_pud = xen_release_pmd_init,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2123)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2124) #if CONFIG_PGTABLE_LEVELS >= 5
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2125) .p4d_val = PV_CALLEE_SAVE(xen_p4d_val),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2126) .make_p4d = PV_CALLEE_SAVE(xen_make_p4d),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2127) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2128)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2129) .activate_mm = xen_activate_mm,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2130) .dup_mmap = xen_dup_mmap,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2131) .exit_mmap = xen_exit_mmap,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2133) .lazy_mode = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2134) .enter = paravirt_enter_lazy_mmu,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2135) .leave = xen_leave_lazy_mmu,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2136) .flush = paravirt_flush_lazy_mmu,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2137) },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2138)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2139) .set_fixmap = xen_set_fixmap,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2140) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2142) void __init xen_init_mmu_ops(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2143) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2144) x86_init.paging.pagetable_init = xen_pagetable_init;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2145) x86_init.hyper.init_after_bootmem = xen_after_bootmem;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2147) pv_ops.mmu = xen_mmu_ops;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2149) memset(dummy_mapping, 0xff, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2150) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2152) /* Protected by xen_reservation_lock. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2153) #define MAX_CONTIG_ORDER 9 /* 2MB */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2154) static unsigned long discontig_frames[1<<MAX_CONTIG_ORDER];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2156) #define VOID_PTE (mfn_pte(0, __pgprot(0)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2157) static void xen_zap_pfn_range(unsigned long vaddr, unsigned int order,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2158) unsigned long *in_frames,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2159) unsigned long *out_frames)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2160) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2161) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2162) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2164) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2165) for (i = 0; i < (1UL<<order); i++, vaddr += PAGE_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2166) mcs = __xen_mc_entry(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2167)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2168) if (in_frames)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2169) in_frames[i] = virt_to_mfn(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2170)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2171) MULTI_update_va_mapping(mcs.mc, vaddr, VOID_PTE, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2172) __set_phys_to_machine(virt_to_pfn(vaddr), INVALID_P2M_ENTRY);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2173)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2174) if (out_frames)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2175) out_frames[i] = virt_to_pfn(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2176) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2177) xen_mc_issue(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2178) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2179)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2180) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2181) * Update the pfn-to-mfn mappings for a virtual address range, either to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2182) * point to an array of mfns, or contiguously from a single starting
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2183) * mfn.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2184) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2185) static void xen_remap_exchanged_ptes(unsigned long vaddr, int order,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2186) unsigned long *mfns,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2187) unsigned long first_mfn)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2188) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2189) unsigned i, limit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2190) unsigned long mfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2191)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2192) xen_mc_batch();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2193)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2194) limit = 1u << order;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2195) for (i = 0; i < limit; i++, vaddr += PAGE_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2196) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2197) unsigned flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2198)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2199) mcs = __xen_mc_entry(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2200) if (mfns)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2201) mfn = mfns[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2202) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2203) mfn = first_mfn + i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2204)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2205) if (i < (limit - 1))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2206) flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2207) else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2208) if (order == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2209) flags = UVMF_INVLPG | UVMF_ALL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2210) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2211) flags = UVMF_TLB_FLUSH | UVMF_ALL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2212) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2214) MULTI_update_va_mapping(mcs.mc, vaddr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2215) mfn_pte(mfn, PAGE_KERNEL), flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2217) set_phys_to_machine(virt_to_pfn(vaddr), mfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2218) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2219)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2220) xen_mc_issue(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2221) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2223) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2224) * Perform the hypercall to exchange a region of our pfns to point to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2225) * memory with the required contiguous alignment. Takes the pfns as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2226) * input, and populates mfns as output.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2227) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2228) * Returns a success code indicating whether the hypervisor was able to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2229) * satisfy the request or not.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2230) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2231) static int xen_exchange_memory(unsigned long extents_in, unsigned int order_in,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2232) unsigned long *pfns_in,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2233) unsigned long extents_out,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2234) unsigned int order_out,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2235) unsigned long *mfns_out,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2236) unsigned int address_bits)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2237) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2238) long rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2239) int success;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2240)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2241) struct xen_memory_exchange exchange = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2242) .in = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2243) .nr_extents = extents_in,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2244) .extent_order = order_in,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2245) .extent_start = pfns_in,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2246) .domid = DOMID_SELF
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2247) },
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2248) .out = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2249) .nr_extents = extents_out,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2250) .extent_order = order_out,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2251) .extent_start = mfns_out,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2252) .address_bits = address_bits,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2253) .domid = DOMID_SELF
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2254) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2255) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2256)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2257) BUG_ON(extents_in << order_in != extents_out << order_out);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2258)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2259) rc = HYPERVISOR_memory_op(XENMEM_exchange, &exchange);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2260) success = (exchange.nr_exchanged == extents_in);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2261)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2262) BUG_ON(!success && ((exchange.nr_exchanged != 0) || (rc == 0)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2263) BUG_ON(success && (rc != 0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2264)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2265) return success;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2266) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2267)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2268) int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2269) unsigned int address_bits,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2270) dma_addr_t *dma_handle)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2271) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2272) unsigned long *in_frames = discontig_frames, out_frame;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2273) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2274) int success;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2275) unsigned long vstart = (unsigned long)phys_to_virt(pstart);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2276)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2277) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2278) * Currently an auto-translated guest will not perform I/O, nor will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2279) * it require PAE page directories below 4GB. Therefore any calls to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2280) * this function are redundant and can be ignored.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2281) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2283) if (unlikely(order > MAX_CONTIG_ORDER))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2284) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2285)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2286) memset((void *) vstart, 0, PAGE_SIZE << order);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2287)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2288) spin_lock_irqsave(&xen_reservation_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2289)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2290) /* 1. Zap current PTEs, remembering MFNs. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2291) xen_zap_pfn_range(vstart, order, in_frames, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2292)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2293) /* 2. Get a new contiguous memory extent. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2294) out_frame = virt_to_pfn(vstart);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2295) success = xen_exchange_memory(1UL << order, 0, in_frames,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2296) 1, order, &out_frame,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2297) address_bits);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2298)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2299) /* 3. Map the new extent in place of old pages. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2300) if (success)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2301) xen_remap_exchanged_ptes(vstart, order, NULL, out_frame);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2302) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2303) xen_remap_exchanged_ptes(vstart, order, in_frames, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2304)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2305) spin_unlock_irqrestore(&xen_reservation_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2307) *dma_handle = virt_to_machine(vstart).maddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2308) return success ? 0 : -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2309) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2311) void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2312) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2313) unsigned long *out_frames = discontig_frames, in_frame;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2314) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2315) int success;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2316) unsigned long vstart;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2317)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2318) if (unlikely(order > MAX_CONTIG_ORDER))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2319) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2320)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2321) vstart = (unsigned long)phys_to_virt(pstart);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2322) memset((void *) vstart, 0, PAGE_SIZE << order);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2323)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2324) spin_lock_irqsave(&xen_reservation_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2326) /* 1. Find start MFN of contiguous extent. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2327) in_frame = virt_to_mfn(vstart);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2328)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2329) /* 2. Zap current PTEs. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2330) xen_zap_pfn_range(vstart, order, NULL, out_frames);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2331)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2332) /* 3. Do the exchange for non-contiguous MFNs. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2333) success = xen_exchange_memory(1, order, &in_frame, 1UL << order,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2334) 0, out_frames, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2335)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2336) /* 4. Map new pages in place of old pages. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2337) if (success)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2338) xen_remap_exchanged_ptes(vstart, order, out_frames, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2339) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2340) xen_remap_exchanged_ptes(vstart, order, NULL, in_frame);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2342) spin_unlock_irqrestore(&xen_reservation_lock, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2343) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2345) static noinline void xen_flush_tlb_all(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2346) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2347) struct mmuext_op *op;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2348) struct multicall_space mcs;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2349)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2350) preempt_disable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2351)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2352) mcs = xen_mc_entry(sizeof(*op));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2353)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2354) op = mcs.args;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2355) op->cmd = MMUEXT_TLB_FLUSH_ALL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2356) MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2357)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2358) xen_mc_issue(PARAVIRT_LAZY_MMU);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2359)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2360) preempt_enable();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2361) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2363) #define REMAP_BATCH_SIZE 16
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2364)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2365) struct remap_data {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2366) xen_pfn_t *pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2367) bool contiguous;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2368) bool no_translate;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2369) pgprot_t prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2370) struct mmu_update *mmu_update;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2371) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2372)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2373) static int remap_area_pfn_pte_fn(pte_t *ptep, unsigned long addr, void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2374) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2375) struct remap_data *rmd = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2376) pte_t pte = pte_mkspecial(mfn_pte(*rmd->pfn, rmd->prot));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2378) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2379) * If we have a contiguous range, just update the pfn itself,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2380) * else update pointer to be "next pfn".
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2381) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2382) if (rmd->contiguous)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2383) (*rmd->pfn)++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2384) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2385) rmd->pfn++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2386)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2387) rmd->mmu_update->ptr = virt_to_machine(ptep).maddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2388) rmd->mmu_update->ptr |= rmd->no_translate ?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2389) MMU_PT_UPDATE_NO_TRANSLATE :
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2390) MMU_NORMAL_PT_UPDATE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2391) rmd->mmu_update->val = pte_val_ma(pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2392) rmd->mmu_update++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2393)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2394) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2395) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2397) int xen_remap_pfn(struct vm_area_struct *vma, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2398) xen_pfn_t *pfn, int nr, int *err_ptr, pgprot_t prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2399) unsigned int domid, bool no_translate, struct page **pages)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2400) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2401) int err = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2402) struct remap_data rmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2403) struct mmu_update mmu_update[REMAP_BATCH_SIZE];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2404) unsigned long range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2405) int mapped = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2406)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2407) BUG_ON(!((vma->vm_flags & (VM_PFNMAP | VM_IO)) == (VM_PFNMAP | VM_IO)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2408)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2409) rmd.pfn = pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2410) rmd.prot = prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2411) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2412) * We use the err_ptr to indicate if there we are doing a contiguous
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2413) * mapping or a discontigious mapping.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2414) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2415) rmd.contiguous = !err_ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2416) rmd.no_translate = no_translate;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2417)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2418) while (nr) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2419) int index = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2420) int done = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2421) int batch = min(REMAP_BATCH_SIZE, nr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2422) int batch_left = batch;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2423)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2424) range = (unsigned long)batch << PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2425)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2426) rmd.mmu_update = mmu_update;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2427) err = apply_to_page_range(vma->vm_mm, addr, range,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2428) remap_area_pfn_pte_fn, &rmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2429) if (err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2430) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2431)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2432) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2433) * We record the error for each page that gives an error, but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2434) * continue mapping until the whole set is done
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2435) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2436) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2437) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2438)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2439) err = HYPERVISOR_mmu_update(&mmu_update[index],
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2440) batch_left, &done, domid);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2441)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2442) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2443) * @err_ptr may be the same buffer as @gfn, so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2444) * only clear it after each chunk of @gfn is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2445) * used.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2446) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2447) if (err_ptr) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2448) for (i = index; i < index + done; i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2449) err_ptr[i] = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2450) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2451) if (err < 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2452) if (!err_ptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2453) goto out;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2454) err_ptr[i] = err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2455) done++; /* Skip failed frame. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2456) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2457) mapped += done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2458) batch_left -= done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2459) index += done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2460) } while (batch_left);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2461)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2462) nr -= batch;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2463) addr += range;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2464) if (err_ptr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2465) err_ptr += batch;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2466) cond_resched();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2467) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2468) out:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2469)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2470) xen_flush_tlb_all();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2472) return err < 0 ? err : mapped;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2473) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2474) EXPORT_SYMBOL_GPL(xen_remap_pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2475)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2476) #ifdef CONFIG_KEXEC_CORE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2477) phys_addr_t paddr_vmcoreinfo_note(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2478) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2479) if (xen_pv_domain())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2480) return virt_to_machine(vmcoreinfo_note).maddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2481) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2482) return __pa(vmcoreinfo_note);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2483) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2484) #endif /* CONFIG_KEXEC_CORE */
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags