* along with this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
+/*P:450 This file contains the x86-specific lguest code. It used to be all
+ * mixed in with drivers/lguest/core.c but several foolhardy code slashers
+ * wrestled most of the dependencies out to here in preparation for porting
+ * lguest to other architectures (see what I mean by foolhardy?).
+ *
+ * This also contains a couple of non-obvious setup and teardown pieces which
+ * were implemented after days of debugging pain. :*/
#include <linux/kernel.h>
#include <linux/start_kernel.h>
#include <linux/string.h>
* also simplify copy_in_guest_info(). Note that we'd still need to restore
* things when we exit to Launcher userspace, but that's fairly easy.
*
+ * We could also try using this hooks for PGE, but that might be too expensive.
+ *
* The hooks were designed for KVM, but we can also put them to good use. :*/
/*H:040 This is the i386-specific code to setup and run the Guest. Interrupts
* we set it now, so we can trap and pass that trap to the Guest if it
* uses the FPU. */
if (cpu->ts)
- lguest_set_ts();
+ unlazy_fpu(current);
/* SYSENTER is an optimized way of doing system calls. We can't allow
* it because it always jumps to privilege level 0. A normal Guest
* was doing. */
run_guest_once(cpu, lguest_pages(raw_smp_processor_id()));
- /* Note that the "regs" pointer contains two extra entries which are
+ /* Note that the "regs" structure contains two extra entries which are
* not really registers: a trap number which says what interrupt or
* trap made the switcher code come back, and an error code which some
* traps set. */
+ /* Restore SYSENTER if it's supposed to be on. */
+ if (boot_cpu_has(X86_FEATURE_SEP))
+ wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
+
/* If the Guest page faulted, then the cr2 register will tell us the
* bad virtual address. We have to grab this now, because once we
* re-enable interrupts an interrupt could fault and thus overwrite
if (cpu->regs->trapnum == 14)
cpu->arch.last_pagefault = read_cr2();
/* Similarly, if we took a trap because the Guest used the FPU,
- * we have to restore the FPU it expects to see. */
+ * we have to restore the FPU it expects to see.
+ * math_state_restore() may sleep and we may even move off to
+ * a different CPU. So all the critical stuff should be done
+ * before this. */
else if (cpu->regs->trapnum == 7)
math_state_restore();
-
- /* Restore SYSENTER if it's supposed to be on. */
- if (boot_cpu_has(X86_FEATURE_SEP))
- wrmsr(MSR_IA32_SYSENTER_CS, __KERNEL_CS, 0);
}
/*H:130 Now we've examined the hypercall code; our Guest can make requests.
break;
case 14: /* We've intercepted a Page Fault. */
/* The Guest accessed a virtual address that wasn't mapped.
- * This happens a lot: we don't actually set up most of the
- * page tables for the Guest at all when we start: as it runs
- * it asks for more and more, and we set them up as
- * required. In this case, we don't even tell the Guest that
- * the fault happened.
+ * This happens a lot: we don't actually set up most of the page
+ * tables for the Guest at all when we start: as it runs it asks
+ * for more and more, and we set them up as required. In this
+ * case, we don't even tell the Guest that the fault happened.
*
* The errcode tells whether this was a read or a write, and
* whether kernel or userspace code. */
if (!deliver_trap(cpu, cpu->regs->trapnum))
/* If the Guest doesn't have a handler (either it hasn't
* registered any yet, or it's one of the faults we don't let
- * it handle), it dies with a cryptic error message. */
+ * it handle), it dies with this cryptic error message. */
kill_guest(cpu, "unhandled trap %li at %#lx (%#lx)",
cpu->regs->trapnum, cpu->regs->eip,
cpu->regs->trapnum == 14 ? cpu->arch.last_pagefault
* The only exception is the interrupt handlers in switcher.S: their
* addresses are placed in a table (default_idt_entries), so we need to
* update the table with the new addresses. switcher_offset() is a
- * convenience function which returns the distance between the builtin
- * switcher code and the high-mapped copy we just made. */
+ * convenience function which returns the distance between the
+ * compiled-in switcher code and the high-mapped copy we just made. */
for (i = 0; i < IDT_ENTRIES; i++)
default_idt_entries[i] += switcher_offset();
state->guest_gdt_desc.address = (long)&state->guest_gdt;
/* We know where we want the stack to be when the Guest enters
- * the switcher: in pages->regs. The stack grows upwards, so
+ * the Switcher: in pages->regs. The stack grows upwards, so
* we start it at the end of that structure. */
state->guest_tss.sp0 = (long)(&pages->regs + 1);
/* And this is the GDT entry to use for the stack: we keep a
cpu_had_pge = 1;
/* adjust_pge is a helper function which sets or unsets the PGE
* bit on its CPU, depending on the argument (0 == unset). */
- on_each_cpu(adjust_pge, (void *)0, 0, 1);
+ on_each_cpu(adjust_pge, (void *)0, 1);
/* Turn off the feature in the global feature set. */
clear_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability);
}
if (cpu_had_pge) {
set_bit(X86_FEATURE_PGE, boot_cpu_data.x86_capability);
/* adjust_pge's argument "1" means set PGE. */
- on_each_cpu(adjust_pge, (void *)1, 0, 1);
+ on_each_cpu(adjust_pge, (void *)1, 1);
}
put_online_cpus();
}
{
u32 tsc_speed;
- /* The pointer to the Guest's "struct lguest_data" is the only
- * argument. We check that address now. */
+ /* The pointer to the Guest's "struct lguest_data" is the only argument.
+ * We check that address now. */
if (!lguest_address_ok(cpu->lg, cpu->hcall->arg1,
sizeof(*cpu->lg->lguest_data)))
return -EFAULT;
return 0;
}
+/*:*/
/*L:030 lguest_arch_setup_regs()
*