[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] Re: [PATCH 5/5] x86-64: Add user_64bit_mode paravirt op



On Wed, Jul 27, 2011 at 1:24 PM, Jeremy Fitzhardinge <jeremy@xxxxxxxx> wrote:
> On 07/26/2011 08:20 PM, Andy Lutomirski wrote:
>> Three places in the kernel assume that the only long mode CPL 3
>> selector is __USER_CS.  This is not true on Xen -- Xen's sysretq
>> changes cs to the magic value 0xe033.
>>
>> Two of the places are corner cases, but as of "x86-64: Improve
>> vsyscall emulation CS and RIP handling"
>> (c9712944b2a12373cb6ff8059afcfb7e826a6c54), vsyscalls will segfault
>> if called with Xen's extra CS selector.  This causes a panic when
>> older init builds die.
>>
>> It seems impossible to make Xen use __USER_CS reliably without
>> taking a performance hit on every system call, so this fixes the
>> tests instead with a new paravirt op.  It's a little ugly because
>> ptrace.h can't include paravirt.h.
>>
>> Signed-off-by: Andy Lutomirski <luto@xxxxxxx>
>> Reported-by: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
>> ---
>>  arch/x86/include/asm/desc.h           |    4 ++--
>>  arch/x86/include/asm/paravirt_types.h |    6 ++++++
>>  arch/x86/include/asm/ptrace.h         |   19 +++++++++++++++++++
>>  arch/x86/kernel/paravirt.c            |    4 ++++
>>  arch/x86/kernel/step.c                |    2 +-
>>  arch/x86/kernel/vsyscall_64.c         |    6 +-----
>>  arch/x86/mm/fault.c                   |    2 +-
>>  arch/x86/xen/enlighten.c              |    1 +
>>  8 files changed, 35 insertions(+), 9 deletions(-)
>>
>> diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h
>> index 7b439d9..41935fa 100644
>> --- a/arch/x86/include/asm/desc.h
>> +++ b/arch/x86/include/asm/desc.h
>> @@ -27,8 +27,8 @@ static inline void fill_ldt(struct desc_struct *desc, 
>> const struct user_desc *in
>>
>>       desc->base2             = (info->base_addr & 0xff000000) >> 24;
>>       /*
>> -      * Don't allow setting of the lm bit. It is useless anyway
>> -      * because 64bit system calls require __USER_CS:
>> +      * Don't allow setting of the lm bit. It would confuse
>> +      * user_64bit_mode and would get overridden by sysret anyway.
>>        */
>>       desc->l                 = 0;
>>  }
>> diff --git a/arch/x86/include/asm/paravirt_types.h 
>> b/arch/x86/include/asm/paravirt_types.h
>> index 2c76521..8e8b9a4 100644
>> --- a/arch/x86/include/asm/paravirt_types.h
>> +++ b/arch/x86/include/asm/paravirt_types.h
>> @@ -41,6 +41,7 @@
>>
>>  #include <asm/desc_defs.h>
>>  #include <asm/kmap_types.h>
>> +#include <asm/pgtable_types.h>
>>
>>  struct page;
>>  struct thread_struct;
>> @@ -63,6 +64,11 @@ struct paravirt_callee_save {
>>  struct pv_info {
>>       unsigned int kernel_rpl;
>>       int shared_kernel_pmd;
>> +
>> +#ifdef CONFIG_X86_64
>> +     u16 extra_user_64bit_cs;  /* __USER_CS if none */
>> +#endif
>> +
>>       int paravirt_enabled;
>>       const char *name;
>>  };
>> diff --git a/arch/x86/include/asm/ptrace.h b/arch/x86/include/asm/ptrace.h
>> index 94e7618..3566454 100644
>> --- a/arch/x86/include/asm/ptrace.h
>> +++ b/arch/x86/include/asm/ptrace.h
>> @@ -131,6 +131,9 @@ struct pt_regs {
>>  #ifdef __KERNEL__
>>
>>  #include <linux/init.h>
>> +#ifdef CONFIG_PARAVIRT
>> +#include <asm/paravirt_types.h>
>> +#endif
>>
>>  struct cpuinfo_x86;
>>  struct task_struct;
>> @@ -187,6 +190,22 @@ static inline int v8086_mode(struct pt_regs *regs)
>>  #endif
>>  }
>>
>> +#ifdef CONFIG_X86_64
>> +static inline bool user_64bit_mode(struct pt_regs *regs)
>> +{
>> +#ifndef CONFIG_PARAVIRT
>> +     /*
>> +      * On non-paravirt systems, this is the only long mode CPL 3
>> +      * selector.  We do not allow long mode selectors in the LDT.
>> +      */
>> +     return regs->cs == __USER_CS;
>> +#else
>> +     /* Headers are too twisted for this to go in paravirt.h. */
>> +     return regs->cs == __USER_CS || regs->cs == 
>> pv_info.extra_user_64bit_cs;
>
> Is this necessary because usermode may sometimes be on __USER_CS or
> sometimes on Xen's?  Could we just commit to one or the other and make
> it a simple comparison?

Currently (from memory), brand new threads start out on __USER_CS.
Also, there might be software out there that thunks back and forth
between 32-bit and 64-bit mode and hardcodes CS=51 as the 32->64 bit
jump target.

It is said that 32-64 bit thunking is impossible, but this is
empirically untrue -- I've done it in the intcc32 vsyscall test I
wrote, and if you remove the actual intcc instruction, it will survive
the thunk in both directions.  My code didn't hardcode the assumption.

>
> What if __USER_CS were a variable?

That sounds a little evil :)  It will also make the FIXUP_TOP_OF_STACK
macro a little uglier than it is.

But maybe it's not so bad.  We could even remove the legacy 64-bit
selector, and presumably everything would still work.

What do you think?  I'll benchmark removing VCGF_in_syscall later tonight.

--Andy

>
>    J
>> +#endif
>> +}
>> +#endif
>> +
>>  /*
>>   * X86_32 CPUs don't save ss and esp if the CPU is already in kernel mode
>>   * when it traps.  The previous stack will be directly underneath the saved
>> diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
>> index 613a793..d90272e 100644
>> --- a/arch/x86/kernel/paravirt.c
>> +++ b/arch/x86/kernel/paravirt.c
>> @@ -307,6 +307,10 @@ struct pv_info pv_info = {
>>       .paravirt_enabled = 0,
>>       .kernel_rpl = 0,
>>       .shared_kernel_pmd = 1, /* Only used when CONFIG_X86_PAE is set */
>> +
>> +#ifdef CONFIG_X86_64
>> +     .extra_user_64bit_cs = __USER_CS,
>> +#endif
>>  };
>>
>>  struct pv_init_ops pv_init_ops = {
>> diff --git a/arch/x86/kernel/step.c b/arch/x86/kernel/step.c
>> index 7977f0c..c346d11 100644
>> --- a/arch/x86/kernel/step.c
>> +++ b/arch/x86/kernel/step.c
>> @@ -74,7 +74,7 @@ static int is_setting_trap_flag(struct task_struct *child, 
>> struct pt_regs *regs)
>>
>>  #ifdef CONFIG_X86_64
>>               case 0x40 ... 0x4f:
>> -                     if (regs->cs != __USER_CS)
>> +                     if (!user_64bit_mode(regs))
>>                               /* 32-bit mode: register increment */
>>                               return 0;
>>                       /* 64-bit mode: REX prefix */
>> diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
>> index dda7dff..1725930 100644
>> --- a/arch/x86/kernel/vsyscall_64.c
>> +++ b/arch/x86/kernel/vsyscall_64.c
>> @@ -127,11 +127,7 @@ void dotraplinkage do_emulate_vsyscall(struct pt_regs 
>> *regs, long error_code)
>>
>>       local_irq_enable();
>>
>> -     /*
>> -      * Real 64-bit user mode code has cs == __USER_CS.  Anything else
>> -      * is bogus.
>> -      */
>> -     if (regs->cs != __USER_CS) {
>> +     if (!user_64bit_mode(regs)) {
>>               /*
>>                * If we trapped from kernel mode, we might as well OOPS now
>>                * instead of returning to some random address and OOPSing
>> diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
>> index 4d09df0..decd51a 100644
>> --- a/arch/x86/mm/fault.c
>> +++ b/arch/x86/mm/fault.c
>> @@ -105,7 +105,7 @@ check_prefetch_opcode(struct pt_regs *regs, unsigned 
>> char *instr,
>>                * but for now it's good enough to assume that long
>>                * mode only uses well known segments or kernel.
>>                */
>> -             return (!user_mode(regs)) || (regs->cs == __USER_CS);
>> +             return (!user_mode(regs) || user_64bit_mode(regs));
>>  #endif
>>       case 0x60:
>>               /* 0x64 thru 0x67 are valid prefixes in all modes. */
>> diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
>> index 974a528..a9c710a 100644
>> --- a/arch/x86/xen/enlighten.c
>> +++ b/arch/x86/xen/enlighten.c
>> @@ -950,6 +950,7 @@ static unsigned xen_patch(u8 type, u16 clobbers, void 
>> *insnbuf,
>>  static const struct pv_info xen_info __initconst = {
>>       .paravirt_enabled = 1,
>>       .shared_kernel_pmd = 0,
>> +     .extra_user_64bit_cs = FLAT_USER_CS64,
>>
>>       .name = "Xen",
>>  };
>
>

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.