> On 3 Jul 2025, at 7:22 PM, Roy Hopkins <roy.hopkins@randomman.co.uk> wrote:
>
> On Fri, 2025-06-27 at 16:58 +0530, Ani Sinha wrote:
>> On Fri, Jun 13, 2025 at 8:52 PM Roy Hopkins <roy.hopkins@randomman.co.uk> wrote:
>>>
>>> The initialization sections in IGVM files contain configuration that
>>> should be applied to the guest platform before it is started. This
>>> includes guest policy and other information that can affect the security
>>> level and the startup measurement of a guest.
>>>
>>> This commit introduces handling of the initialization sections during
>>> processing of the IGVM file.
>>>
>>> Signed-off-by: Roy Hopkins <roy.hopkins@randomman.co.uk>
>>> Acked-by: Michael S. Tsirkin <mst@redhat.com>
>>> Acked-by: Gerd Hoffman <kraxel@redhat.com>
>>> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
>>> ---
>>> backends/igvm.c | 21 +++++++++++++++++++++
>>> 1 file changed, 21 insertions(+)
>>>
>>> diff --git a/backends/igvm.c b/backends/igvm.c
>>> index 2a31021d44..ebdb4594d1 100644
>>> --- a/backends/igvm.c
>>> +++ b/backends/igvm.c
>>> @@ -786,6 +786,27 @@ int qigvm_process_file(IgvmCfg *cfg, ConfidentialGuestSupport *cgs,
>>> }
>>> }
>>>
>>> + header_count =
>>> + igvm_header_count(ctx.file, IGVM_HEADER_SECTION_INITIALIZATION);
>>> + if (header_count < 0) {
>>> + error_setg(
>>> + errp,
>>> + "Invalid initialization header count in IGVM file. Error code: %X",
>>> + header_count);
>>> + goto cleanup_parameters;
>>> + }
>>> +
>>> + for (ctx.current_header_index = 0;
>>> + ctx.current_header_index < (unsigned)header_count;
>>> + ctx.current_header_index++) {
>>> + IgvmVariableHeaderType type =
>>> + igvm_get_header_type(ctx.file, IGVM_HEADER_SECTION_INITIALIZATION,
>>> + ctx.current_header_index);
>>> + if (qigvm_handler(&ctx, type, errp) < 0) {
>>
>> So the next patch, patch #14 actually defines the handler.
>>
>> @@ -92,6 +135,10 @@ static struct QIGVMHandler handlers[] = {
>> qigvm_directive_environment_info },
>> { IGVM_VHT_REQUIRED_MEMORY, IGVM_HEADER_SECTION_DIRECTIVE,
>> qigvm_directive_required_memory },
>> + { IGVM_VHT_SNP_ID_BLOCK, IGVM_HEADER_SECTION_DIRECTIVE,
>> + qigvm_directive_snp_id_block },
>> + { IGVM_VHT_GUEST_POLICY, IGVM_HEADER_SECTION_INITIALIZATION,
>> + qigvm_initialization_guest_policy },
>> };
>>
>>
>> So I think patch #14 should come before this patch in the series.
>>
>
> This was deliberately placed before #14 as it introduces the mechanism for processing
> initialization sections, allowing future patches to then add sections as necessary.
> Patch #14 then uses this new capability to handle guest policy.
My concern was that qigvm_handler() would return -1 and throw "IGVM: Unknown header type encountered when processing file:” without patch 14.
> This patch does
> compile successfully without patch #14 being applied.
I see. IGVM_HEADER_SECTION_INITIALIZATION actually comes from igvm.h header
enum IgvmHeaderSection {
IGVM_HEADER_SECTION_PLATFORM,
IGVM_HEADER_SECTION_INITIALIZATION,
IGVM_HEADER_SECTION_DIRECTIVE,
};
So that explains successful compilation.
>
>>> + goto cleanup_parameters;
>>> + }
>>> + }
>>> +
>>> /*
>>> * Contiguous pages of data with compatible flags are grouped together in
>>> * order to reduce the number of memory regions we create. Make sure the
>>> --
>>> 2.43.0