![Andy Lutomirski](/assets/img/avatar_default.png)
Intel's VMX is daft and resets the hidden TSS limit register to 0x67 on VMX reload, and the 0x67 is not configurable. KVM currently reloads TR using the LTR instruction on every exit, but this is quite slow because LTR is serializing. The 0x67 limit is entirely harmless unless ioperm() is in use, so defer the reload until a task using ioperm() is actually running. Here's some poorly done benchmarking using kvm-unit-tests: Before: cpuid 1313 vmcall 1195 mov_from_cr8 11 mov_to_cr8 17 inl_from_pmtimer 6770 inl_from_qemu 6856 inl_from_kernel 2435 outl_to_kernel 1402 After: cpuid 1291 vmcall 1181 mov_from_cr8 11 mov_to_cr8 16 inl_from_pmtimer 6457 inl_from_qemu 6209 inl_from_kernel 2339 outl_to_kernel 1391 Signed-off-by: Andy Lutomirski <luto@kernel.org> [Force-reload TR in invalidate_tss_limit. - Paolo] Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
126 lines
3.2 KiB
C
126 lines
3.2 KiB
C
/*
|
|
* This contains the io-permission bitmap code - written by obz, with changes
|
|
* by Linus. 32/64 bits code unification by Miguel Botón.
|
|
*/
|
|
|
|
#include <linux/sched.h>
|
|
#include <linux/kernel.h>
|
|
#include <linux/capability.h>
|
|
#include <linux/errno.h>
|
|
#include <linux/types.h>
|
|
#include <linux/ioport.h>
|
|
#include <linux/smp.h>
|
|
#include <linux/stddef.h>
|
|
#include <linux/slab.h>
|
|
#include <linux/thread_info.h>
|
|
#include <linux/syscalls.h>
|
|
#include <linux/bitmap.h>
|
|
#include <asm/syscalls.h>
|
|
#include <asm/desc.h>
|
|
|
|
/*
|
|
* this changes the io permissions bitmap in the current task.
|
|
*/
|
|
asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
|
|
{
|
|
struct thread_struct *t = ¤t->thread;
|
|
struct tss_struct *tss;
|
|
unsigned int i, max_long, bytes, bytes_updated;
|
|
|
|
if ((from + num <= from) || (from + num > IO_BITMAP_BITS))
|
|
return -EINVAL;
|
|
if (turn_on && !capable(CAP_SYS_RAWIO))
|
|
return -EPERM;
|
|
|
|
/*
|
|
* If it's the first ioperm() call in this thread's lifetime, set the
|
|
* IO bitmap up. ioperm() is much less timing critical than clone(),
|
|
* this is why we delay this operation until now:
|
|
*/
|
|
if (!t->io_bitmap_ptr) {
|
|
unsigned long *bitmap = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
|
|
|
|
if (!bitmap)
|
|
return -ENOMEM;
|
|
|
|
memset(bitmap, 0xff, IO_BITMAP_BYTES);
|
|
t->io_bitmap_ptr = bitmap;
|
|
set_thread_flag(TIF_IO_BITMAP);
|
|
|
|
preempt_disable();
|
|
refresh_TR();
|
|
preempt_enable();
|
|
}
|
|
|
|
/*
|
|
* do it in the per-thread copy and in the TSS ...
|
|
*
|
|
* Disable preemption via get_cpu() - we must not switch away
|
|
* because the ->io_bitmap_max value must match the bitmap
|
|
* contents:
|
|
*/
|
|
tss = &per_cpu(cpu_tss, get_cpu());
|
|
|
|
if (turn_on)
|
|
bitmap_clear(t->io_bitmap_ptr, from, num);
|
|
else
|
|
bitmap_set(t->io_bitmap_ptr, from, num);
|
|
|
|
/*
|
|
* Search for a (possibly new) maximum. This is simple and stupid,
|
|
* to keep it obviously correct:
|
|
*/
|
|
max_long = 0;
|
|
for (i = 0; i < IO_BITMAP_LONGS; i++)
|
|
if (t->io_bitmap_ptr[i] != ~0UL)
|
|
max_long = i;
|
|
|
|
bytes = (max_long + 1) * sizeof(unsigned long);
|
|
bytes_updated = max(bytes, t->io_bitmap_max);
|
|
|
|
t->io_bitmap_max = bytes;
|
|
|
|
/* Update the TSS: */
|
|
memcpy(tss->io_bitmap, t->io_bitmap_ptr, bytes_updated);
|
|
|
|
put_cpu();
|
|
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* sys_iopl has to be used when you want to access the IO ports
|
|
* beyond the 0x3ff range: to get the full 65536 ports bitmapped
|
|
* you'd need 8kB of bitmaps/process, which is a bit excessive.
|
|
*
|
|
* Here we just change the flags value on the stack: we allow
|
|
* only the super-user to do it. This depends on the stack-layout
|
|
* on system-call entry - see also fork() and the signal handling
|
|
* code.
|
|
*/
|
|
SYSCALL_DEFINE1(iopl, unsigned int, level)
|
|
{
|
|
struct pt_regs *regs = current_pt_regs();
|
|
struct thread_struct *t = ¤t->thread;
|
|
|
|
/*
|
|
* Careful: the IOPL bits in regs->flags are undefined under Xen PV
|
|
* and changing them has no effect.
|
|
*/
|
|
unsigned int old = t->iopl >> X86_EFLAGS_IOPL_BIT;
|
|
|
|
if (level > 3)
|
|
return -EINVAL;
|
|
/* Trying to gain more privileges? */
|
|
if (level > old) {
|
|
if (!capable(CAP_SYS_RAWIO))
|
|
return -EPERM;
|
|
}
|
|
regs->flags = (regs->flags & ~X86_EFLAGS_IOPL) |
|
|
(level << X86_EFLAGS_IOPL_BIT);
|
|
t->iopl = level << X86_EFLAGS_IOPL_BIT;
|
|
set_iopl_mask(t->iopl);
|
|
|
|
return 0;
|
|
}
|