[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <df4ccaf7-005d-4cbe-acef-20878421ce20@linux.intel.com>
Date: Fri, 11 Jul 2025 10:01:47 -0700
From: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@...ux.intel.com>
To: James Bottomley <James.Bottomley@...senPartnership.com>,
yangge1116@....com, ardb@...nel.org
Cc: jarkko@...nel.org, ilias.apalodimas@...aro.org, jgg@...pe.ca,
linux-efi@...r.kernel.org, linux-kernel@...r.kernel.org,
stable@...r.kernel.org, liuzixing@...on.cn
Subject: Re: [PATCH V4] efi/tpm: Fix the issue where the CC platforms event
log header can't be correctly identified
On 7/11/25 7:00 AM, James Bottomley wrote:
> On Tue, 2025-07-08 at 15:54 +0800, yangge1116@....com wrote:
>> From: Ge Yang <yangge1116@....com>
>>
>> Since commit d228814b1913 ("efi/libstub: Add get_event_log() support
>> for CC platforms") reuses TPM2 support code for the CC platforms,
>> when launching a TDX virtual machine with coco measurement enabled,
>> the following error log is generated:
>>
>> [Firmware Bug]: Failed to parse event in TPM Final Events Log
>>
>> Call Trace:
>> efi_config_parse_tables()
>> efi_tpm_eventlog_init()
>> tpm2_calc_event_log_size()
>> __calc_tpm2_event_size()
>>
>> The pcr_idx value in the Intel TDX log header is 1, causing the
>> function __calc_tpm2_event_size() to fail to recognize the log
>> header, ultimately leading to the "Failed to parse event in TPM Final
>> Events Log" error.
>>
>> According to UEFI Specification 2.10, Section 38.4.1: For TDX, TPM
>> PCR 0 maps to MRTD, so the log header uses TPM PCR 1 instead. To
>> successfully parse the TDX event log header, the check for a pcr_idx
>> value of 0 must be skipped.
>>
>> According to Table 6 in Section 10.2.1 of the TCG PC Client
>> Specification, the index field does not require the PCR index to be
>> fixed at zero. Therefore, skipping the check for a pcr_idx value of
>> 0 for CC platforms is safe.
> This is wrong: the spec does not allow a header EV_ACTION to be
> recorded with anything other than pcrIndex == 0.
>
> However, the fact that Intel, who practically wrote the TPM spec, can
> get this wrong shows that others can too. So the best way to fix this
> is to remove the pcrIndex check for the first event. There's no danger
> of this causing problems because we check for the TCG_SPECID_SIG
> signature as the next thing. That means you don't need to thread
> knowledge of whether this is a CC environment and we're pre-emptively
> ready for any other spec violators who misread the spec in the same way
> Intel did.
I agree with James Bottomley's suggestion to remove the pcr_index check
without adding any replacement checks.
This check was originally introduced in the following commit to handle a
case where certain Dell platforms provided an event log without a valid
header:
commit 7dfc06a0f25b593a9f51992f540c0f80a57f3629
Author: Fabian Vogt <fvogt@...e.de>
Date: Mon Jun 15 09:16:36 2020 +0200
efi/tpm: Verify event log header before parsing
At first, I was concerned that the pcr_index check might still be important for
this fix. However, after re-reading the commit and reviewing the intent, it appears
that relying on the event_type and digest checks should be sufficient for validating
the event log header.
> Regards,
>
> James
>
--
Sathyanarayanan Kuppuswamy
Linux Kernel Developer
Powered by blists - more mailing lists