[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH] x86/fred: Send an EVENT_CHECK IPI on exit from NMI
- To: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
- From: Jan Beulich <jbeulich@xxxxxxxx>
- Date: Wed, 24 Jun 2026 17:20:09 +0200
- Authentication-results: eu.smtp.expurgate.cloud; dkim=pass header.s=google header.d=suse.com header.i="@suse.com" header.h="Content-Transfer-Encoding:In-Reply-To:Autocrypt:From:Content-Language:References:Cc:To:Subject:User-Agent:MIME-Version:Date:Message-ID"
- Autocrypt: addr=jbeulich@xxxxxxxx; keydata= xsDiBFk3nEQRBADAEaSw6zC/EJkiwGPXbWtPxl2xCdSoeepS07jW8UgcHNurfHvUzogEq5xk hu507c3BarVjyWCJOylMNR98Yd8VqD9UfmX0Hb8/BrA+Hl6/DB/eqGptrf4BSRwcZQM32aZK 7Pj2XbGWIUrZrd70x1eAP9QE3P79Y2oLrsCgbZJfEwCgvz9JjGmQqQkRiTVzlZVCJYcyGGsD /0tbFCzD2h20ahe8rC1gbb3K3qk+LpBtvjBu1RY9drYk0NymiGbJWZgab6t1jM7sk2vuf0Py O9Hf9XBmK0uE9IgMaiCpc32XV9oASz6UJebwkX+zF2jG5I1BfnO9g7KlotcA/v5ClMjgo6Gl MDY4HxoSRu3i1cqqSDtVlt+AOVBJBACrZcnHAUSuCXBPy0jOlBhxPqRWv6ND4c9PH1xjQ3NP nxJuMBS8rnNg22uyfAgmBKNLpLgAGVRMZGaGoJObGf72s6TeIqKJo/LtggAS9qAUiuKVnygo 3wjfkS9A3DRO+SpU7JqWdsveeIQyeyEJ/8PTowmSQLakF+3fote9ybzd880fSmFuIEJldWxp Y2ggPGpiZXVsaWNoQHN1c2UuY29tPsJgBBMRAgAgBQJZN5xEAhsDBgsJCAcDAgQVAggDBBYC AwECHgECF4AACgkQoDSui/t3IH4J+wCfQ5jHdEjCRHj23O/5ttg9r9OIruwAn3103WUITZee e7Sbg12UgcQ5lv7SzsFNBFk3nEQQCACCuTjCjFOUdi5Nm244F+78kLghRcin/awv+IrTcIWF hUpSs1Y91iQQ7KItirz5uwCPlwejSJDQJLIS+QtJHaXDXeV6NI0Uef1hP20+y8qydDiVkv6l IreXjTb7DvksRgJNvCkWtYnlS3mYvQ9NzS9PhyALWbXnH6sIJd2O9lKS1Mrfq+y0IXCP10eS FFGg+Av3IQeFatkJAyju0PPthyTqxSI4lZYuJVPknzgaeuJv/2NccrPvmeDg6Coe7ZIeQ8Yj t0ARxu2xytAkkLCel1Lz1WLmwLstV30g80nkgZf/wr+/BXJW/oIvRlonUkxv+IbBM3dX2OV8 AmRv1ySWPTP7AAMFB/9PQK/VtlNUJvg8GXj9ootzrteGfVZVVT4XBJkfwBcpC/XcPzldjv+3 HYudvpdNK3lLujXeA5fLOH+Z/G9WBc5pFVSMocI71I8bT8lIAzreg0WvkWg5V2WZsUMlnDL9 mpwIGFhlbM3gfDMs7MPMu8YQRFVdUvtSpaAs8OFfGQ0ia3LGZcjA6Ik2+xcqscEJzNH+qh8V m5jjp28yZgaqTaRbg3M/+MTbMpicpZuqF4rnB0AQD12/3BNWDR6bmh+EkYSMcEIpQmBM51qM EKYTQGybRCjpnKHGOxG0rfFY1085mBDZCH5Kx0cl0HVJuQKC+dV2ZY5AqjcKwAxpE75MLFkr wkkEGBECAAkFAlk3nEQCGwwACgkQoDSui/t3IH7nnwCfcJWUDUFKdCsBH/E5d+0ZnMQi+G0A nAuWpQkjM1ASeQwSHEeAWPgskBQL
- Cc: Roger Pau Monné <roger.pau@xxxxxxxxxx>, Teddy Astie <teddy.astie@xxxxxxxxxx>, Oleksii Kurochko <oleksii.kurochko@xxxxxxxxx>, Xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxxx>
- Delivery-date: Wed, 24 Jun 2026 15:20:20 +0000
- List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
On 24.06.2026 17:09, Andrew Cooper wrote:
> On 24/06/2026 3:47 pm, Jan Beulich wrote:
>> On 24.06.2026 16:23, Andrew Cooper wrote:
>>> --- a/xen/arch/x86/x86_64/entry-fred.S
>>> +++ b/xen/arch/x86/x86_64/entry-fred.S
>>> @@ -20,6 +20,12 @@ FUNC(entry_FRED_R3, 4096)
>>> GET_STACK_END(14)
>>> movq STACK_CPUINFO_FIELD(current_vcpu)(%r14), %rbx
>>>
>>> + /* NMIs need special handling on return to guest. */
>>> + movzbl UREGS_ss + 6(%rsp), %eax
>>> + and $0xf, %eax
>>> + cmp $X86_ET_NMI, %al
>>> + je nmi_exit_to_guest
>>> +
>>> jmp test_all_events
>> Actually, how about shrinking this to just
>>
>> test %al, %al
>> jnz nmi_exit_to_guest
>>
>> by having entry_from_pv() return a boolean?
>
> I considered that, and dismissed it.
>
> It involve changing large chunks of traps.c (simply to compile) and puts
> far more than 4 instructions of logic onto the common path.
Does it? Code size overall may grow, but since every return there is that
of a constant (false in all but one case), every individual path would go
down from 4 insns (with a memory access, admittedly a cache-hot one) to
just 3 (the two here and the MOV or XOR on the return paths).
Jan
|
