From: Jamin Lin via <qemu-devel@nongnu.org>
To: "Cédric Le Goater" <clg@kaod.org>,
"Peter Maydell" <peter.maydell@linaro.org>,
"Steven Lee" <steven_lee@aspeedtech.com>,
"Troy Lee" <leetroy@gmail.com>,
"Andrew Jeffery" <andrew@codeconstruct.com.au>,
"Joel Stanley" <joel@jms.id.au>,
"Fabiano Rosas" <farosas@suse.de>,
"Laurent Vivier" <lvivier@redhat.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"open list:ASPEED BMCs" <qemu-arm@nongnu.org>,
"open list:All patches CC here" <qemu-devel@nongnu.org>
Cc: <jamin_lin@aspeedtech.com>, <troy_lee@aspeedtech.com>
Subject: [PATCH v3 00/28] Fix incorrect hash results on AST2700
Date: Thu, 15 May 2025 16:09:32 +0800 [thread overview]
Message-ID: <20250515081008.583578-1-jamin_lin@aspeedtech.com> (raw)
v1:
1. Added support for 64-bit DMA in the HACE model
2. Refactored the do_hash operation in the HACE model
3. Fixed a crash caused by out-of-bound memory access in HACE
4. Added more trace events and implemented dumping of source hash data and
resulting digests to improve debugging
5. Refactored the HACE QTest framework to support both AST1030 and AST2700
6. Added a test case for SHA384
v2:
1. Create new helper functions
hash_get_source_addr
hash_prepare_direct_iov
hash_prepare_sg_iov
hash_get_digest_addr
hash_write_digest_and_unmap_iov
hash_execute_non_acc_mode
hash_execute_acc_mode
2. Refactor do_hash_operation
3. Fix review issue
4. Revise trace-events
5. Move register size to instance class and dynamically allocate regs
v3:
1. Split patch to introduce these routines one by one :
hash_prepare_sg_iov
hash_prepare_direct_iov
hash_execute_acc_mode
hash_execute_non_acc_mode
hash_write_digest_and_unmap_iov
2. Fix run qtest failed
This patchset resolves incorrect hash results reported on the AST2700 platform.
This update addresses the following kernel warnings and test failures related to
the crypto self-test framework:
aspeed-hmac-sha512 test failed (incorrect result)
aspeed-hmac-sha384 test failed (incorrect result)
aspeed-sha512 test failed (incorrect result)
aspeed-sha384 test failed (incorrect result)
aspeed-hmac-sha256 test failed (incorrect result)
aspeed-hmac-sha224 test failed (incorrect result)
aspeed-hmac-sha1 test failed (incorrect result)
aspeed-sha224 test failed (incorrect result)
aspeed-sha256 test failed (incorrect result)
aspeed-sha1 test failed (incorrect result)
How to test it
Use the following command to dump information about the supported digest methods
via the ast_crypto_engine hardware engine:
root@ast2700-default:~# openssl engine -pre DUMP_INFO ast_crypto_engine
Digest SHA1, NID=64, AF_ALG info: name=sha1ALG_ERR: , driver=aspeed-sha1 (hw accelerated)
Digest SHA224, NID=675, AF_ALG info: name=sha224ALG_ERR: , driver=aspeed-sha224 (hw accelerated)
Digest SHA256, NID=672, AF_ALG info: name=sha256ALG_ERR: , driver=aspeed-sha256 (hw accelerated)
Digest SHA384, NID=673, AF_ALG info: name=sha384ALG_ERR: , driver=aspeed-sha384 (hw accelerated)
Digest SHA512, NID=674, AF_ALG info: name=sha512ALG_ERR: , driver=aspeed-sha512 (hw accelerated)
The status of SHA1, SHA224, SHA256, SHA384, and SHA512 should be marked as
hw accelerated, indicating that these algorithms are supported by hardware
acceleration via the aspeed drivers.
Create a test file on the host machine and compute its HASH value as the
expected result
Create a 256MB test file
$ dd if=/dev/random of=/tmp/256M bs=1M count=256
Generate Hash Values Using SHA1, SHA224, SHA256, SHA384, and SHA512
Use the following commands to generate HASH values for a 256MB file using
different SHA algorithms:
$ sha1sum /tmp/256M
7fc628811a31ab87b0502dab3ed8d3ef07565885 /tmp/256M
$ sha224sum /tmp/256M
2d261c11ba05b3a62e0efeab51c307d9933426c7e18204683ef3da54 /tmp/256M
$ sha256sum /tmp/256M
5716d1700ee35c92ca5ca5b466639e9c36eed3f1447c1aec27f16d0fe113f94d /tmp/256M
$ sha384sum /tmp/256M
fb6bc62afa1096dcd3b870e7d2546b7a5a177b5f2bbd5c9759218182454709e0c504a2d9c26404e04aa8010a291b7f1c /tmp/256M
$ sha512sum /tmp/256M
fbceda7be34836fe857781656318ecd5b457a833a24c8736d5b8ef8d07e1950eebcdb140eebe4f12b5ff59586f7eb1c64fa95869c63dd9e4703d91261093c5c9 /tmp/256M
Generate HASH Values Using the Hardware Engine
Use the following commands to generate HASH values for a 256MB file using
various SHA algorithms with the ast_crypto_engine hardware engine:
root@ast2700-default:~# openssl dgst -sha1 -engine ast_crypto_engine /tmp/256M
Engine "ast_crypto_engine" set.
SHA1(/tmp/256M)= 7fc628811a31ab87b0502dab3ed8d3ef07565885
root@ast2700-default:~# openssl dgst -sha224 -engine ast_crypto_engine /tmp/256M
Engine "ast_crypto_engine" set.
SHA2-224(/tmp/256M)= 2d261c11ba05b3a62e0efeab51c307d9933426c7e18204683ef3da54
root@ast2700-default:~# openssl dgst -sha256 -engine ast_crypto_engine /tmp/256M
Engine "ast_crypto_engine" set.
SHA2-256(/tmp/256M)= 5716d1700ee35c92ca5ca5b466639e9c36eed3f1447c1aec27f16d0fe113f94d
root@ast2700-default:~# openssl dgst -sha384 -engine ast_crypto_engine /tmp/256M
Engine "ast_crypto_engine" set.
SHA2-384(/tmp/256M)= fb6bc62afa1096dcd3b870e7d2546b7a5a177b5f2bbd5c9759218182454709e0c504a2d9c26404e04aa8010a291b7f1c
root@ast2700-default:~# openssl dgst -sha512 -engine ast_crypto_engine /tmp/256M
Engine "ast_crypto_engine" set.
SHA2-512(/tmp/256M)= fbceda7be34836fe857781656318ecd5b457a833a24c8736d5b8ef8d07e1950eebcdb140eebe4f12b5ff59586f7eb1c64fa95869c63dd9e4703d91261093c5c9
The HASH values generated here should exactly match those computed on the host
machine using sha shell commands, verifying both the correctness of the
hardware-accelerated results and the functionality of the ast_crypto_engine.
Jamin Lin (28):
hw/misc/aspeed_hace: Remove unused code for better readability
hw/misc/aspeed_hace: Improve readability and consistency in variable
naming
hw/misc/aspeed_hace: Ensure HASH_IRQ is always set to prevent firmware
hang
hw/misc/aspeed_hace: Extract direct mode hash buffer setup into helper
function
hw/misc/aspeed_hace: Extract SG-mode hash buffer setup into helper
function
hw/misc/aspeed_hace: Extract digest write and iov unmap into helper
function
hw/misc/aspeed_hace: Extract non-accumulation hash execution into
helper function
hw/misc/aspeed_hace: Extract accumulation-mode hash execution into
helper function
hw/misc/aspeed_hace: Introduce 64-bit hash source address helper
function
hw/misc/aspeed_hace: Rename R_HASH_DEST to R_HASH_DIGEST and introduce
64-bit hash digest address helper
hw/misc/aspeed_hace: Support accumulative mode for direct access mode
hw/misc/aspeed_hace: Move register size to instance class and
dynamically allocate regs
hw/misc/aspeed_hace: Add support for source, digest, key buffer 64 bit
addresses
hw/misc/aspeed_hace: Support DMA 64 bits dram address
hw/misc/aspeed_hace: Add trace-events for better debugging
hw/misc/aspeed_hace: Support to dump plaintext and digest for better
debugging
tests/qtest: Reorder aspeed test list
test/qtest: Introduce a new aspeed-hace-utils.c to place common
testcases
test/qtest/hace: Specify explicit array sizes for test vectors and
hash results
test/qtest/hace: Adjust test address range for AST1030 due to SRAM
limitations
test/qtest/hace: Add SHA-384 test cases for ASPEED HACE model
test/qtest/hace: Add SHA-384 tests for AST2600
test/qtest/hace: Add tests for AST1030
test/qtest/hace: Update source data and digest data type to 64-bit
test/qtest/hace: Support 64-bit source and digest addresses for
AST2700
test/qtest/hace: Support to test upper 32 bits of digest and source
addresses
test/qtest/hace: Support to validate 64-bit hmac key buffer addresses
test/qtest/hace: Add tests for AST2700
include/hw/misc/aspeed_hace.h | 11 +-
tests/qtest/aspeed-hace-utils.h | 84 +++++
hw/misc/aspeed_hace.c | 479 +++++++++++++++--------
tests/qtest/aspeed-hace-utils.c | 646 ++++++++++++++++++++++++++++++++
tests/qtest/aspeed_hace-test.c | 577 +++++-----------------------
tests/qtest/ast2700-hace-test.c | 98 +++++
hw/misc/trace-events | 8 +
tests/qtest/meson.build | 13 +-
8 files changed, 1279 insertions(+), 637 deletions(-)
create mode 100644 tests/qtest/aspeed-hace-utils.h
create mode 100644 tests/qtest/aspeed-hace-utils.c
create mode 100644 tests/qtest/ast2700-hace-test.c
--
2.43.0
next reply other threads:[~2025-05-15 8:16 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-15 8:09 Jamin Lin via [this message]
2025-05-15 8:09 ` [PATCH v3 01/28] hw/misc/aspeed_hace: Remove unused code for better readability Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 02/28] hw/misc/aspeed_hace: Improve readability and consistency in variable naming Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 03/28] hw/misc/aspeed_hace: Ensure HASH_IRQ is always set to prevent firmware hang Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 04/28] hw/misc/aspeed_hace: Extract direct mode hash buffer setup into helper function Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 05/28] hw/misc/aspeed_hace: Extract SG-mode " Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 06/28] hw/misc/aspeed_hace: Extract digest write and iov unmap " Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 07/28] hw/misc/aspeed_hace: Extract non-accumulation hash execution " Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 08/28] hw/misc/aspeed_hace: Extract accumulation-mode " Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 09/28] hw/misc/aspeed_hace: Introduce 64-bit hash source address " Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 10/28] hw/misc/aspeed_hace: Rename R_HASH_DEST to R_HASH_DIGEST and introduce 64-bit hash digest address helper Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 11/28] hw/misc/aspeed_hace: Support accumulative mode for direct access mode Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 12/28] hw/misc/aspeed_hace: Move register size to instance class and dynamically allocate regs Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 13/28] hw/misc/aspeed_hace: Add support for source, digest, key buffer 64 bit addresses Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 14/28] hw/misc/aspeed_hace: Support DMA 64 bits dram address Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 15/28] hw/misc/aspeed_hace: Add trace-events for better debugging Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 16/28] hw/misc/aspeed_hace: Support to dump plaintext and digest " Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 17/28] tests/qtest: Reorder aspeed test list Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 18/28] test/qtest: Introduce a new aspeed-hace-utils.c to place common testcases Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 19/28] test/qtest/hace: Specify explicit array sizes for test vectors and hash results Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 20/28] test/qtest/hace: Adjust test address range for AST1030 due to SRAM limitations Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 21/28] test/qtest/hace: Add SHA-384 test cases for ASPEED HACE model Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 22/28] test/qtest/hace: Add SHA-384 tests for AST2600 Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 23/28] test/qtest/hace: Add tests for AST1030 Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 24/28] test/qtest/hace: Update source data and digest data type to 64-bit Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 25/28] test/qtest/hace: Support 64-bit source and digest addresses for AST2700 Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 26/28] test/qtest/hace: Support to test upper 32 bits of digest and source addresses Jamin Lin via
2025-05-15 8:09 ` [PATCH v3 27/28] test/qtest/hace: Support to validate 64-bit hmac key buffer addresses Jamin Lin via
2025-05-15 8:10 ` [PATCH v3 28/28] test/qtest/hace: Add tests for AST2700 Jamin Lin via
2025-05-20 14:58 ` [PATCH v3 00/28] Fix incorrect hash results on AST2700 Fabiano Rosas
2025-05-23 7:17 ` Cédric Le Goater
2025-05-23 8:10 ` Cédric Le Goater
2025-05-29 7:29 ` Michael Tokarev
2025-05-29 7:38 ` Cédric Le Goater
2025-05-29 7:40 ` Jamin Lin
2025-05-29 7:45 ` Michael Tokarev
2025-05-29 12:17 ` Cédric Le Goater
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250515081008.583578-1-jamin_lin@aspeedtech.com \
--to=qemu-devel@nongnu.org \
--cc=andrew@codeconstruct.com.au \
--cc=clg@kaod.org \
--cc=farosas@suse.de \
--cc=jamin_lin@aspeedtech.com \
--cc=joel@jms.id.au \
--cc=leetroy@gmail.com \
--cc=lvivier@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-arm@nongnu.org \
--cc=steven_lee@aspeedtech.com \
--cc=troy_lee@aspeedtech.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).