summaryrefslogtreecommitdiff
path: root/arch/x86/mm/fault.c
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/mm/fault.c')
-rw-r--r--arch/x86/mm/fault.c244
1 files changed, 148 insertions, 96 deletions
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 71d4b9d4d43f..2ff25ad33233 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -27,6 +27,7 @@
#include <asm/vm86.h> /* struct vm86 */
#include <asm/mmu_context.h> /* vma_pkey() */
#include <asm/efi.h> /* efi_recover_from_page_fault()*/
+#include <asm/desc.h> /* store_idt(), ... */
#define CREATE_TRACE_POINTS
#include <asm/trace/exceptions.h>
@@ -571,10 +572,55 @@ static int is_f00f_bug(struct pt_regs *regs, unsigned long address)
return 0;
}
+static void show_ldttss(const struct desc_ptr *gdt, const char *name, u16 index)
+{
+ u32 offset = (index >> 3) * sizeof(struct desc_struct);
+ unsigned long addr;
+ struct ldttss_desc desc;
+
+ if (index == 0) {
+ pr_alert("%s: NULL\n", name);
+ return;
+ }
+
+ if (offset + sizeof(struct ldttss_desc) >= gdt->size) {
+ pr_alert("%s: 0x%hx -- out of bounds\n", name, index);
+ return;
+ }
+
+ if (probe_kernel_read(&desc, (void *)(gdt->address + offset),
+ sizeof(struct ldttss_desc))) {
+ pr_alert("%s: 0x%hx -- GDT entry is not readable\n",
+ name, index);
+ return;
+ }
+
+ addr = desc.base0 | (desc.base1 << 16) | (desc.base2 << 24);
+#ifdef CONFIG_X86_64
+ addr |= ((u64)desc.base3 << 32);
+#endif
+ pr_alert("%s: 0x%hx -- base=0x%lx limit=0x%x\n",
+ name, index, addr, (desc.limit0 | (desc.limit1 << 16)));
+}
+
+/*
+ * This helper function transforms the #PF error_code bits into
+ * "[PROT] [USER]" type of descriptive, almost human-readable error strings:
+ */
+static void err_str_append(unsigned long error_code, char *buf, unsigned long mask, const char *txt)
+{
+ if (error_code & mask) {
+ if (buf[0])
+ strcat(buf, " ");
+ strcat(buf, txt);
+ }
+}
+
static void
-show_fault_oops(struct pt_regs *regs, unsigned long error_code,
- unsigned long address)
+show_fault_oops(struct pt_regs *regs, unsigned long error_code, unsigned long address)
{
+ char err_txt[64];
+
if (!oops_may_print())
return;
@@ -602,6 +648,52 @@ show_fault_oops(struct pt_regs *regs, unsigned long error_code,
address < PAGE_SIZE ? "NULL pointer dereference" : "paging request",
(void *)address);
+ err_txt[0] = 0;
+
+ /*
+ * Note: length of these appended strings including the separation space and the
+ * zero delimiter must fit into err_txt[].
+ */
+ err_str_append(error_code, err_txt, X86_PF_PROT, "[PROT]" );
+ err_str_append(error_code, err_txt, X86_PF_WRITE, "[WRITE]");
+ err_str_append(error_code, err_txt, X86_PF_USER, "[USER]" );
+ err_str_append(error_code, err_txt, X86_PF_RSVD, "[RSVD]" );
+ err_str_append(error_code, err_txt, X86_PF_INSTR, "[INSTR]");
+ err_str_append(error_code, err_txt, X86_PF_PK, "[PK]" );
+
+ pr_alert("#PF error: %s\n", error_code ? err_txt : "[normal kernel read fault]");
+
+ if (!(error_code & X86_PF_USER) && user_mode(regs)) {
+ struct desc_ptr idt, gdt;
+ u16 ldtr, tr;
+
+ pr_alert("This was a system access from user code\n");
+
+ /*
+ * This can happen for quite a few reasons. The more obvious
+ * ones are faults accessing the GDT, or LDT. Perhaps
+ * surprisingly, if the CPU tries to deliver a benign or
+ * contributory exception from user code and gets a page fault
+ * during delivery, the page fault can be delivered as though
+ * it originated directly from user code. This could happen
+ * due to wrong permissions on the IDT, GDT, LDT, TSS, or
+ * kernel or IST stack.
+ */
+ store_idt(&idt);
+
+ /* Usable even on Xen PV -- it's just slow. */
+ native_store_gdt(&gdt);
+
+ pr_alert("IDT: 0x%lx (limit=0x%hx) GDT: 0x%lx (limit=0x%hx)\n",
+ idt.address, idt.size, gdt.address, gdt.size);
+
+ store_ldt(ldtr);
+ show_ldttss(&gdt, "LDTR", ldtr);
+
+ store_tr(tr);
+ show_ldttss(&gdt, "TR", tr);
+ }
+
dump_pagetable(address);
}
@@ -621,16 +713,30 @@ pgtable_bad(struct pt_regs *regs, unsigned long error_code,
tsk->comm, address);
dump_pagetable(address);
- tsk->thread.cr2 = address;
- tsk->thread.trap_nr = X86_TRAP_PF;
- tsk->thread.error_code = error_code;
-
if (__die("Bad pagetable", regs, error_code))
sig = 0;
oops_end(flags, regs, sig);
}
+static void set_signal_archinfo(unsigned long address,
+ unsigned long error_code)
+{
+ struct task_struct *tsk = current;
+
+ /*
+ * To avoid leaking information about the kernel page
+ * table layout, pretend that user-mode accesses to
+ * kernel addresses are always protection faults.
+ */
+ if (address >= TASK_SIZE_MAX)
+ error_code |= X86_PF_PROT;
+
+ tsk->thread.trap_nr = X86_TRAP_PF;
+ tsk->thread.error_code = error_code | X86_PF_USER;
+ tsk->thread.cr2 = address;
+}
+
static noinline void
no_context(struct pt_regs *regs, unsigned long error_code,
unsigned long address, int signal, int si_code)
@@ -639,6 +745,15 @@ no_context(struct pt_regs *regs, unsigned long error_code,
unsigned long flags;
int sig;
+ if (user_mode(regs)) {
+ /*
+ * This is an implicit supervisor-mode access from user
+ * mode. Bypass all the kernel-mode recovery code and just
+ * OOPS.
+ */
+ goto oops;
+ }
+
/* Are we prepared to handle this kernel fault? */
if (fixup_exception(regs, X86_TRAP_PF, error_code, address)) {
/*
@@ -656,9 +771,7 @@ no_context(struct pt_regs *regs, unsigned long error_code,
* faulting through the emulate_vsyscall() logic.
*/
if (current->thread.sig_on_uaccess_err && signal) {
- tsk->thread.trap_nr = X86_TRAP_PF;
- tsk->thread.error_code = error_code | X86_PF_USER;
- tsk->thread.cr2 = address;
+ set_signal_archinfo(address, error_code);
/* XXX: hwpoison faults will set the wrong code. */
force_sig_fault(signal, si_code, (void __user *)address,
@@ -726,6 +839,7 @@ no_context(struct pt_regs *regs, unsigned long error_code,
if (IS_ENABLED(CONFIG_EFI))
efi_recover_from_page_fault(address);
+oops:
/*
* Oops. The kernel tried to access some bad page. We'll have to
* terminate things with extreme prejudice:
@@ -737,10 +851,6 @@ no_context(struct pt_regs *regs, unsigned long error_code,
if (task_stack_end_corrupted(tsk))
printk(KERN_EMERG "Thread overran stack, or stack corrupted\n");
- tsk->thread.cr2 = address;
- tsk->thread.trap_nr = X86_TRAP_PF;
- tsk->thread.error_code = error_code;
-
sig = SIGKILL;
if (__die("Oops", regs, error_code))
sig = 0;
@@ -794,7 +904,7 @@ __bad_area_nosemaphore(struct pt_regs *regs, unsigned long error_code,
struct task_struct *tsk = current;
/* User mode accesses just cause a SIGSEGV */
- if (error_code & X86_PF_USER) {
+ if (user_mode(regs) && (error_code & X86_PF_USER)) {
/*
* It's possible to have interrupts off here:
*/
@@ -821,9 +931,7 @@ __bad_area_nosemaphore(struct pt_regs *regs, unsigned long error_code,
if (likely(show_unhandled_signals))
show_signal_msg(regs, error_code, address, tsk);
- tsk->thread.cr2 = address;
- tsk->thread.error_code = error_code;
- tsk->thread.trap_nr = X86_TRAP_PF;
+ set_signal_archinfo(address, error_code);
if (si_code == SEGV_PKUERR)
force_sig_pkuerr((void __user *)address, pkey);
@@ -937,9 +1045,7 @@ do_sigbus(struct pt_regs *regs, unsigned long error_code, unsigned long address,
if (is_prefetch(regs, error_code, address))
return;
- tsk->thread.cr2 = address;
- tsk->thread.error_code = error_code;
- tsk->thread.trap_nr = X86_TRAP_PF;
+ set_signal_archinfo(address, error_code);
#ifdef CONFIG_MEMORY_FAILURE
if (fault & (VM_FAULT_HWPOISON|VM_FAULT_HWPOISON_LARGE)) {
@@ -1148,23 +1254,6 @@ static int fault_in_kernel_space(unsigned long address)
return address >= TASK_SIZE_MAX;
}
-static inline bool smap_violation(int error_code, struct pt_regs *regs)
-{
- if (!IS_ENABLED(CONFIG_X86_SMAP))
- return false;
-
- if (!static_cpu_has(X86_FEATURE_SMAP))
- return false;
-
- if (error_code & X86_PF_USER)
- return false;
-
- if (!user_mode(regs) && (regs->flags & X86_EFLAGS_AC))
- return false;
-
- return true;
-}
-
/*
* Called for all faults where 'address' is part of the kernel address
* space. Might get called for faults that originate from *code* that
@@ -1230,7 +1319,6 @@ void do_user_addr_fault(struct pt_regs *regs,
unsigned long hw_error_code,
unsigned long address)
{
- unsigned long sw_error_code;
struct vm_area_struct *vma;
struct task_struct *tsk;
struct mm_struct *mm;
@@ -1252,10 +1340,16 @@ void do_user_addr_fault(struct pt_regs *regs,
pgtable_bad(regs, hw_error_code, address);
/*
- * Check for invalid kernel (supervisor) access to user
- * pages in the user address space.
+ * If SMAP is on, check for invalid kernel (supervisor) access to user
+ * pages in the user address space. The odd case here is WRUSS,
+ * which, according to the preliminary documentation, does not respect
+ * SMAP and will have the USER bit set so, in all cases, SMAP
+ * enforcement appears to be consistent with the USER bit.
*/
- if (unlikely(smap_violation(hw_error_code, regs))) {
+ if (unlikely(cpu_feature_enabled(X86_FEATURE_SMAP) &&
+ !(hw_error_code & X86_PF_USER) &&
+ !(regs->flags & X86_EFLAGS_AC)))
+ {
bad_area_nosemaphore(regs, hw_error_code, address);
return;
}
@@ -1270,13 +1364,6 @@ void do_user_addr_fault(struct pt_regs *regs,
}
/*
- * hw_error_code is literally the "page fault error code" passed to
- * the kernel directly from the hardware. But, we will shortly be
- * modifying it in software, so give it a new name.
- */
- sw_error_code = hw_error_code;
-
- /*
* It's safe to allow irq's after cr2 has been saved and the
* vmalloc fault has been handled.
*
@@ -1285,26 +1372,6 @@ void do_user_addr_fault(struct pt_regs *regs,
*/
if (user_mode(regs)) {
local_irq_enable();
- /*
- * Up to this point, X86_PF_USER set in hw_error_code
- * indicated a user-mode access. But, after this,
- * X86_PF_USER in sw_error_code will indicate either
- * that, *or* an implicit kernel(supervisor)-mode access
- * which originated from user mode.
- */
- if (!(hw_error_code & X86_PF_USER)) {
- /*
- * The CPU was in user mode, but the CPU says
- * the fault was not a user-mode access.
- * Must be an implicit kernel-mode access,
- * which we do not expect to happen in the
- * user address space.
- */
- pr_warn_once("kernel-mode error from user-mode: %lx\n",
- hw_error_code);
-
- sw_error_code |= X86_PF_USER;
- }
flags |= FAULT_FLAG_USER;
} else {
if (regs->flags & X86_EFLAGS_IF)
@@ -1313,9 +1380,9 @@ void do_user_addr_fault(struct pt_regs *regs,
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
- if (sw_error_code & X86_PF_WRITE)
+ if (hw_error_code & X86_PF_WRITE)
flags |= FAULT_FLAG_WRITE;
- if (sw_error_code & X86_PF_INSTR)
+ if (hw_error_code & X86_PF_INSTR)
flags |= FAULT_FLAG_INSTRUCTION;
#ifdef CONFIG_X86_64
@@ -1328,7 +1395,7 @@ void do_user_addr_fault(struct pt_regs *regs,
* The vsyscall page does not have a "real" VMA, so do this
* emulation before we go searching for VMAs.
*/
- if ((sw_error_code & X86_PF_INSTR) && is_vsyscall_vaddr(address)) {
+ if ((hw_error_code & X86_PF_INSTR) && is_vsyscall_vaddr(address)) {
if (emulate_vsyscall(regs, address))
return;
}
@@ -1344,18 +1411,15 @@ void do_user_addr_fault(struct pt_regs *regs,
* Only do the expensive exception table search when we might be at
* risk of a deadlock. This happens if we
* 1. Failed to acquire mmap_sem, and
- * 2. The access did not originate in userspace. Note: either the
- * hardware or earlier page fault code may set X86_PF_USER
- * in sw_error_code.
+ * 2. The access did not originate in userspace.
*/
if (unlikely(!down_read_trylock(&mm->mmap_sem))) {
- if (!(sw_error_code & X86_PF_USER) &&
- !search_exception_tables(regs->ip)) {
+ if (!user_mode(regs) && !search_exception_tables(regs->ip)) {
/*
* Fault from code in kernel from
* which we do not expect faults.
*/
- bad_area_nosemaphore(regs, sw_error_code, address);
+ bad_area_nosemaphore(regs, hw_error_code, address);
return;
}
retry:
@@ -1371,29 +1435,17 @@ retry:
vma = find_vma(mm, address);
if (unlikely(!vma)) {
- bad_area(regs, sw_error_code, address);
+ bad_area(regs, hw_error_code, address);
return;
}
if (likely(vma->vm_start <= address))
goto good_area;
if (unlikely(!(vma->vm_flags & VM_GROWSDOWN))) {
- bad_area(regs, sw_error_code, address);
+ bad_area(regs, hw_error_code, address);
return;
}
- if (sw_error_code & X86_PF_USER) {
- /*
- * Accessing the stack below %sp is always a bug.
- * The large cushion allows instructions like enter
- * and pusha to work. ("enter $65535, $31" pushes
- * 32 pointers and then decrements %sp by 65535.)
- */
- if (unlikely(address + 65536 + 32 * sizeof(unsigned long) < regs->sp)) {
- bad_area(regs, sw_error_code, address);
- return;
- }
- }
if (unlikely(expand_stack(vma, address))) {
- bad_area(regs, sw_error_code, address);
+ bad_area(regs, hw_error_code, address);
return;
}
@@ -1402,8 +1454,8 @@ retry:
* we can handle it..
*/
good_area:
- if (unlikely(access_error(sw_error_code, vma))) {
- bad_area_access_error(regs, sw_error_code, address, vma);
+ if (unlikely(access_error(hw_error_code, vma))) {
+ bad_area_access_error(regs, hw_error_code, address, vma);
return;
}
@@ -1442,13 +1494,13 @@ good_area:
return;
/* Not returning to user mode? Handle exceptions or die: */
- no_context(regs, sw_error_code, address, SIGBUS, BUS_ADRERR);
+ no_context(regs, hw_error_code, address, SIGBUS, BUS_ADRERR);
return;
}
up_read(&mm->mmap_sem);
if (unlikely(fault & VM_FAULT_ERROR)) {
- mm_fault_error(regs, sw_error_code, address, fault);
+ mm_fault_error(regs, hw_error_code, address, fault);
return;
}