-/*
+/*
* Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com)
* Licensed under the GPL
*/
#include "asm/a.out.h"
#include "asm/current.h"
#include "asm/irq.h"
+#include "sysdep/sigcontext.h"
#include "user_util.h"
#include "kern_util.h"
#include "kern.h"
#include "chan_kern.h"
#include "mconsole_kern.h"
-#include "2_5compat.h"
#include "mem.h"
#include "mem_kern.h"
-
+#include "sysdep/sigcontext.h"
+#include "sysdep/ptrace.h"
+#include "os.h"
+#ifdef CONFIG_MODE_SKAS
+#include "skas.h"
+#endif
+#include "os.h"
+
+/* Note this is constrained to return 0, -EFAULT, -EACCESS, -ENOMEM by segv(). */
int handle_page_fault(unsigned long address, unsigned long ip,
int is_write, int is_user, int *code_out)
{
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
- unsigned long page;
int err = -EFAULT;
*code_out = SEGV_MAPERR;
+
+ /* If the fault was during atomic operation, don't take the fault, just
+ * fail. */
+ if (in_atomic())
+ goto out_nosemaphore;
+
down_read(&mm->mmap_sem);
vma = find_vma(mm, address);
if(!vma)
goto good_area;
else if(!(vma->vm_flags & VM_GROWSDOWN))
goto out;
- else if(!ARCH_IS_STACKGROW(address))
+ else if(is_user && !ARCH_IS_STACKGROW(address))
goto out;
else if(expand_stack(vma, address))
goto out;
- good_area:
+good_area:
*code_out = SEGV_ACCERR;
if(is_write && !(vma->vm_flags & VM_WRITE))
goto out;
- page = address & PAGE_MASK;
- pgd = pgd_offset(mm, page);
- pud = pud_offset(pgd, page);
- pmd = pmd_offset(pud, page);
+
+ /* Don't require VM_READ|VM_EXEC for write faults! */
+ if(!is_write && !(vma->vm_flags & (VM_READ | VM_EXEC)))
+ goto out;
+
do {
- survive:
+survive:
switch (handle_mm_fault(mm, vma, address, is_write)){
case VM_FAULT_MINOR:
current->min_flt++;
default:
BUG();
}
- pgd = pgd_offset(mm, page);
- pud = pud_offset(pgd, page);
- pmd = pmd_offset(pud, page);
- pte = pte_offset_kernel(pmd, page);
+ pgd = pgd_offset(mm, address);
+ pud = pud_offset(pgd, address);
+ pmd = pmd_offset(pud, address);
+ pte = pte_offset_kernel(pmd, address);
} while(!pte_present(*pte));
err = 0;
- *pte = pte_mkyoung(*pte);
- if(pte_write(*pte)) *pte = pte_mkdirty(*pte);
- flush_tlb_page(vma, page);
- out:
+ /* The below warning was added in place of
+ * pte_mkyoung(); if (is_write) pte_mkdirty();
+ * If it's triggered, we'd see normally a hang here (a clean pte is
+ * marked read-only to emulate the dirty bit).
+ * However, the generic code can mark a PTE writable but clean on a
+ * concurrent read fault, triggering this harmlessly. So comment it out.
+ */
+#if 0
+ WARN_ON(!pte_young(*pte) || (is_write && !pte_dirty(*pte)));
+#endif
+ flush_tlb_page(vma, address);
+out:
up_read(&mm->mmap_sem);
+out_nosemaphore:
return(err);
/*
goto out;
}
-LIST_HEAD(physmem_remappers);
-
-void register_remapper(struct remapper *info)
-{
- list_add(&info->list, &physmem_remappers);
-}
-
-static int check_remapped_addr(unsigned long address, int is_write)
+void segv_handler(int sig, union uml_pt_regs *regs)
{
- struct remapper *remapper;
- struct list_head *ele;
- __u64 offset;
- int fd;
-
- fd = phys_mapping(__pa(address), &offset);
- if(fd == -1)
- return(0);
+ struct faultinfo * fi = UPT_FAULTINFO(regs);
- list_for_each(ele, &physmem_remappers){
- remapper = list_entry(ele, struct remapper, list);
- if((*remapper->proc)(fd, address, is_write, offset))
- return(1);
+ if(UPT_IS_USER(regs) && !SEGV_IS_FIXABLE(fi)){
+ bad_segv(*fi, UPT_IP(regs));
+ return;
}
-
- return(0);
+ segv(*fi, UPT_IP(regs), UPT_IS_USER(regs), regs);
}
-unsigned long segv(unsigned long address, unsigned long ip, int is_write,
- int is_user, void *sc)
+struct kern_handlers handlinfo_kern = {
+ .relay_signal = relay_signal,
+ .winch = winch,
+ .bus_handler = relay_signal,
+ .page_fault = segv_handler,
+ .sigio_handler = sigio_handler,
+ .timer_handler = timer_handler
+};
+/*
+ * We give a *copy* of the faultinfo in the regs to segv.
+ * This must be done, since nesting SEGVs could overwrite
+ * the info in the regs. A pointer to the info then would
+ * give us bad data!
+ */
+unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, void *sc)
{
struct siginfo si;
void *catcher;
int err;
+ int is_write = FAULT_WRITE(fi);
+ unsigned long address = FAULT_ADDRESS(fi);
if(!is_user && (address >= start_vm) && (address < end_vm)){
flush_tlb_kernel_vm();
return(0);
}
- else if(check_remapped_addr(address & PAGE_MASK, is_write))
- return(0);
else if(current->mm == NULL)
panic("Segfault with no mm");
- err = handle_page_fault(address, ip, is_write, is_user, &si.si_code);
+
+ if (SEGV_IS_FIXABLE(&fi) || SEGV_MAYBE_FIXABLE(&fi))
+ err = handle_page_fault(address, ip, is_write, is_user, &si.si_code);
+ else {
+ err = -EFAULT;
+ /* A thread accessed NULL, we get a fault, but CR2 is invalid.
+ * This code is used in __do_copy_from_user() of TT mode. */
+ address = 0;
+ }
catcher = current->thread.fault_catcher;
if(!err)
}
else if(current->thread.fault_addr != NULL)
panic("fault_addr set but no fault catcher");
- else if(arch_fixup(ip, sc))
+ else if(!is_user && arch_fixup(ip, sc))
return(0);
if(!is_user)
panic("Kernel mode fault at addr 0x%lx, ip 0x%lx",
address, ip);
- if(err == -EACCES){
+ if (err == -EACCES) {
si.si_signo = SIGBUS;
si.si_errno = 0;
si.si_code = BUS_ADRERR;
- si.si_addr = (void *)address;
+ si.si_addr = (void __user *)address;
+ current->thread.arch.faultinfo = fi;
force_sig_info(SIGBUS, &si, current);
- }
- else if(err == -ENOMEM){
+ } else if (err == -ENOMEM) {
printk("VM: killing process %s\n", current->comm);
do_exit(SIGKILL);
- }
- else {
+ } else {
+ BUG_ON(err != -EFAULT);
si.si_signo = SIGSEGV;
- si.si_addr = (void *) address;
- current->thread.cr2 = address;
- current->thread.err = is_write;
+ si.si_addr = (void __user *) address;
+ current->thread.arch.faultinfo = fi;
force_sig_info(SIGSEGV, &si, current);
}
return(0);
}
-void bad_segv(unsigned long address, unsigned long ip, int is_write)
+void bad_segv(struct faultinfo fi, unsigned long ip)
{
struct siginfo si;
si.si_signo = SIGSEGV;
si.si_code = SEGV_ACCERR;
- si.si_addr = (void *) address;
- current->thread.cr2 = address;
- current->thread.err = is_write;
+ si.si_addr = (void __user *) FAULT_ADDRESS(fi);
+ current->thread.arch.faultinfo = fi;
force_sig_info(SIGSEGV, &si, current);
}
if(arch_handle_signal(sig, regs)) return;
if(!UPT_IS_USER(regs))
panic("Kernel mode signal %d", sig);
+ current->thread.arch.faultinfo = *UPT_FAULTINFO(regs);
force_sig(sig, current);
}
void trap_init(void)
{
}
-
-DEFINE_SPINLOCK(trap_lock);
-
-static int trap_index = 0;
-
-int next_trap_index(int limit)
-{
- int ret;
-
- spin_lock(&trap_lock);
- ret = trap_index;
- if(++trap_index == limit)
- trap_index = 0;
- spin_unlock(&trap_lock);
- return(ret);
-}
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only. This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
git://git.onelab.eu / linux-2.6.git / blobdiff