Re: [patch 02/38] x86/cpu: Use native_wrmsrl() in load_percpu_segment()

From: Thomas Gleixner
Date: Sun Jul 17 2022 - 16:08:24 EST


On Sun, Jul 17 2022 at 21:08, Thomas Gleixner wrote:
> On Sun, Jul 17 2022 at 00:22, Andrew Cooper wrote:
>>  #ifdef CONFIG_X86_32
>>  /* The 32-bit entry code needs to find cpu_entry_area. */
>>  DEFINE_PER_CPU(struct cpu_entry_area *, cpu_entry_area);
>> @@ -742,12 +732,15 @@ EXPORT_SYMBOL_GPL(load_fixmap_gdt);
>>   * Current gdt points %fs at the "master" per-cpu area: after this,
>>   * it's on the real one.
>>   */
>> -void switch_to_new_gdt(int cpu)
>> +void __noinstr switch_to_new_gdt(int cpu)
>>  {
>>         /* Load the original GDT */
>>         load_direct_gdt(cpu);
>> +
>> +#ifdef CONFIG_X86_32
>>         /* Reload the per-cpu base */
>> -       load_percpu_segment(cpu);
>> +       loadsegment(fs, __KERNEL_PERCPU);
>> +#endif
>>  }
>>  
>>  static const struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
>>
>>
>> It's only 32bit where the percpu pointer is tied to the GDT.  On 64bit,
>> gsbase is good before this, and remains good after.
>>
>> With this change,
>>
>> # Make sure load_percpu_segment has no stackprotector
>> CFLAGS_common.o         := -fno-stack-protector
>>
>> comes up for re-evaluation too.
>
> Good point. Let me stare at it some more.

If it only would be that simple :)

loadsegment_simple() was a red herring. The gs segment is already zero.

So what explodes here is the early boot when switching from early per
CPU to the real per CPU area.

start_kernel()
.....
setup_per_cpu_areas();
smp_prepare_boot_cpu()
switch_to_new_gdt()
load_direct_gdt(cpu);
load_percpu_segment(cpu)
GS: 0
GS_BASE: 0xffffffff829d0000 (early PERCPU)
wrmsrl()
GS_BASE: 0xffff888237c00000 (real PERCPU)

So the explosion happens when accessing a per CPU variable after loading
the GDT and before GS_BASE is fixed up.

That's the only case AFAICT where this matters. In all other invocations
GS_BASE is already correct.

Let me fix this proper.

Thanks,

tglx