Re: [PATCH v2] x86: reduce dependencies on x86_emulate/x86_emulate.h

[Jan Beulich <jbeulich@xxxxxxxx>]

On 12.08.2026 15:39, Andrew Cooper wrote:
> On 12/08/2026 2:24 pm, Jan Beulich wrote:
>> Split out struct x86_event to an entirely separate header, and move a few
>> other items describing the architecture to a new x86-types.h. With a few
>> forward decls of structures and with a fair number of new #include-s in
>> .c files, the inclusion of x86_emulate.h (and hence
>> x86_emulate/x86_emulate.h) can be dropped from all header files except
>> hvm/emulate.h; it needs additionally adding to hvm/ioreq.h though.
>>
>> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> 
> Reviewed-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>

Thanks.

>> --- /dev/null
>> +++ b/xen/arch/x86/include/asm/x86-event.h
>> @@ -0,0 +1,31 @@
>> +/* SPDX-License-Identifier: GPL-2.0-or-later */
>> +/*
>> + * x86-event.h
>> + *
>> + * Helper definitions for event handling, which aren't prescribed by the
>> + * architecture itself.
>> + */
>> +
>> +#ifndef X86_X86_EVENT_H
>> +#define X86_X86_EVENT_H
>> +
>> +#ifdef __XEN__
>> +# include <xen/types.h>
>> +#else
>> +# include <stdint.h>
>> +#endif
>> +
>> +#define X86_EVENT_NO_EC (-1)        /* No error code. */
>> +
>> +struct x86_event {
>> +    int16_t       vector;
>> +    uint8_t       type;         /* X86_ET_* */
>> +    uint8_t       insn_len;     /* Instruction length */
>> +    int32_t       error_code;   /* X86_EVENT_NO_EC if n/a */
>> +    union {
>> +        unsigned long cr2;         /* #PF */
>> +        unsigned long pending_dbg; /* #DB (new DR6 bits, positive polarity) 
>> */
> 
> With the advent of FRED, this probably wants to become event_data (or
> just data) and drop the union.
> 
> It's also the XFD_ERR mask for #NM, and the NMI Source Bitmap (on
> capable hardware), and I think we're better off pointing to the FRED
> spec than keeping an out-of-date list of what's in it.

Yes, perhaps that's going to be better.

Jan