Re: [PATCH 2/2] x86: Rewrite ret_from_fork() in C

[H. Peter Anvin]

On June 22, 2023 9:04:03 AM PDT, Brian Gerst <brgerst@xxxxxxxxx> wrote:
>On Thu, Jun 22, 2023 at 9:29 AM Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
>>
>> On Thu, Jun 22, 2023 at 08:07:50AM -0400, Brian Gerst wrote:
>> > When kCFI is enabled, special handling is needed for the indirect call
>> > to the kernel thread function.  Rewrite the ret_from_fork() function in
>> > C so that the compiler can properly handle the indirect call.
>> >
>> > Suggested-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
>> > Signed-off-by: Brian Gerst <brgerst@xxxxxxxxx>
>>
>> This is much nicer indeed. I'll take these patches into my series and
>> repost later today if you don't mind.
>
>Yes, that's fine.
>
>> One little niggle below..
>>
>> > ---
>>
>> > diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
>> > index f31e286c2977..5ee32e7e29e8 100644
>> > --- a/arch/x86/entry/entry_64.S
>> > +++ b/arch/x86/entry/entry_64.S
>> > @@ -284,36 +284,21 @@ SYM_FUNC_END(__switch_to_asm)
>> >   * r12: kernel thread arg
>> >   */
>> >  .pushsection .text, "ax"
>> > +SYM_CODE_START(ret_from_fork_asm)
>> >       UNWIND_HINT_END_OF_STACK
>> >       ANNOTATE_NOENDBR // copy_thread
>> >       CALL_DEPTH_ACCOUNT
>> >
>> > +     /* return address for the stack unwinder */
>> > +     pushq   $swapgs_restore_regs_and_return_to_usermode
>> > +     UNWIND_HINT_FUNC
>> >
>> > +     movq    %rax, %rdi              /* prev */
>> > +     movq    %rsp, %rsi              /* regs */
>> > +     movq    %rbx, %rdx              /* fn */
>> > +     movq    %r12, %rcx              /* fn_arg */
>> > +     jmp     ret_from_fork
>> > +SYM_CODE_END(ret_from_fork_asm)
>> >  .popsection
>> >
>> >  .macro DEBUG_ENTRY_ASSERT_IRQS_OFF
>>
>> > diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
>> > index dac41a0072ea..f5dbfebac076 100644
>> > --- a/arch/x86/kernel/process.c
>> > +++ b/arch/x86/kernel/process.c
>> > @@ -28,6 +28,7 @@
>> >  #include <linux/static_call.h>
>> >  #include <trace/events/power.h>
>> >  #include <linux/hw_breakpoint.h>
>> > +#include <linux/entry-common.h>
>> >  #include <asm/cpu.h>
>> >  #include <asm/apic.h>
>> >  #include <linux/uaccess.h>
>> > @@ -134,6 +135,25 @@ static int set_new_tls(struct task_struct *p, unsigned long tls)
>> >               return do_set_thread_area_64(p, ARCH_SET_FS, tls);
>> >  }
>> >
>> > +__visible noinstr void ret_from_fork(struct task_struct *prev, struct pt_regs *regs,
>> > +                                  int (*fn)(void *), void *fn_arg)
>>
>> So I had noinstr in my initial patch, but it leads to objtool
>> complaints. I suppose we can actually handle tracing and all the other
>> gunk at this point, so I've removed it.
>
>I'm not an expert on noinstr usage, but looking at the other syscall
>functions, instrumentation needs to be disabled before
>syscall_exit_to_user_mode() is called.  Perhaps adding an
>instrumentation_begin()/instrumentation_end() pair to this function is
>needed?
>
>Brian Gerst
>

I don't have the code in front of me right now, but how does this affect FRED enabling? In the case of FRED, the exit path is much simpler; in the FRED enabling patchset we simply deal with it by alternatives-patching the terminal jump after resetting the stack pointer to the standard FRED user space exit stub (which simply pops the user space registers and executes ERETU.)

I'm assuming this is still valid/possible after your patches, since resetting the stack pointer isn't possible in C, but I wanted to double check.