From: Yongseok Koh <yskoh@mellanox.com>
To: Honnappa Nagarahalli <Honnappa.Nagarahalli@arm.com>
Cc: "jerinj@marvell.com" <jerinj@marvell.com>,
"bruce.richardson@intel.com" <bruce.richardson@intel.com>,
Pavan Nikhilesh Bhagavatula <pbhagavatula@marvell.com>,
Shahaf Shuler <shahafs@mellanox.com>,
"dev@dpdk.org" <dev@dpdk.org>,
Thomas Monjalon <thomas@monjalon.net>,
"Gavin Hu (Arm Technology China)" <Gavin.Hu@arm.com>,
nd <nd@arm.com>
Subject: Re: [dpdk-dev] [EXT] [PATCH 5/6] build: add option for armv8 crypto extension
Date: Thu, 2 May 2019 01:54:35 +0000 [thread overview]
Message-ID: <0B9F2C32-4530-4A7E-AE63-657361771001@mellanox.com> (raw)
In-Reply-To: <VE1PR08MB514979EA9CDF07C6810A7183983A0@VE1PR08MB5149.eurprd08.prod.outlook.com>
> On Apr 29, 2019, at 8:33 PM, Honnappa Nagarahalli <Honnappa.Nagarahalli@arm.com> wrote:
>
>> On Apr 15, 2019, at 1:13 PM, Honnappa Nagarahalli
>> <Honnappa.Nagarahalli@arm.com> wrote:
>>
>>>>>> Subject: [EXT] [PATCH 5/6] build: add option for armv8 crypto
>>>>>> extension
>>>>>>
>>>>>> CONFIG_RTE_MACHINE="armv8a"
>>>>>> +CONFIG_RTE_ENABLE_ARMV8_CRYPTO=y
>>>>>
>>>>> This approach is not scalable. Even, it is not good for BlueField as
>>>>> you you need to maintain two images.
>>>>>
>>>>> Unlike other CPU flags, arm64's crypto cpu flag is really _optional_.
>>>>> Access to crypto instructions is always at under runtime check.
>>>>> See the following in rte_armv8_pmd.c
>>>>>
>>>>>
>>>>> /* Check CPU for support for AES instruction set */
>>>>> if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_AES)) {
>>>>> ARMV8_CRYPTO_LOG_ERR(
>>>>> "AES instructions not supported by CPU");
>>>>> return -EFAULT;
>>>>> }
>>>>>
>>>>> /* Check CPU for support for SHA instruction set */
>>>>> if (!rte_cpu_get_flag_enabled(RTE_CPUFLAG_SHA1) ||
>>>>> !rte_cpu_get_flag_enabled(RTE_CPUFLAG_SHA2)) {
>>>>> ARMV8_CRYPTO_LOG_ERR(
>>>>> "SHA1/SHA2 instructions not supported by CPU");
>>>>> return -EFAULT;
>>>>> }
>>>>>
>>>>> So In order to avoid one more config flags specific to armv8 in
>>>>> meson and makefile build infra And avoid the need for 6/6 patch.
>>>>> IMO, # Introduce optional CPU flag scheme in eal. Treat armv8 crypto
>>>>> as optional flag # Skip the eal init check for optional flag.
>>>>>
>>>>> Do you see any issues with that approach?
>>>>
>>>> I also thought about that approach and that was my number 1 priority.
>>>> But, I had one question came to my mind. Maybe, arm people can
>>>> confirm it. Is it 100% guaranteed that compiler never makes use of
>>>> any of crypto instructions even if there's no specific asm/intrinsic
>>>> code? The crypto extension has aes, pmull,
>>>> sha1 and sha2. In case of rte_memcpy() for x86, for example, compiler
>>>> may optimize code using avx512f instructions even though it is
>>>> written specifically with avx2 intrinsics (__mm256_*) unless avx512f is
>> disabled.
>>>>
>>>> If a complier expert in arm (or anyone else) confirm it is completely
>>>> **optional**, then I'd love to take that approach for sure.
>>>>
>>>> Copied dpdk-on-arm ML.
>>>>
>>> I do not know the answer, will have to check with the compiler team. I will get
>> back on this.
>>
>> Any update yet?
> Currently, enabling 'crypto' flag will generate the crypto instructions only when crypto intrinsics are used. However, when 'sha3' (part of 8.2 crypto) flag is enabled, compiler can generate 3-way exclusive OR instructions beyond the intrinsics. Compiler team cannot provide a guarantee that other crypto instructions will not be used beyond the intrinsics.
>
> The current suggestion is to use GNU indirect function [1] or similar. I am not sure on GNU indirect function portability.
>
> [1] https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwillnewton.name%2F2013%2F07%2F02%2Fusing-gnu-indirect-functions%2F&data=02%7C01%7Cyskoh%40mellanox.com%7Ce8738c4f725a4ca608ea08d6cd1cac03%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C0%7C636921920373635167&sdata=kuq6dbpTBfRgokrv2L%2FV4BIM0q1k%2FiL1JaMqCHUc2c0%3D&reserved=0
Thanks for the update,
Then, I think the original patch to have build config is currently okay.
Will submit it again.
thanks
Yongseok
next prev parent reply other threads:[~2019-05-02 1:54 UTC|newest]
Thread overview: 120+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-12 23:24 [dpdk-dev] [PATCH 0/6] build: fix build for arm64 Yongseok Koh
2019-04-12 23:24 ` Yongseok Koh
2019-04-12 23:24 ` [dpdk-dev] [PATCH 1/6] meson: disable octeontx for buggy compilers on arm64 Yongseok Koh
2019-04-12 23:24 ` Yongseok Koh
2019-04-13 5:52 ` [dpdk-dev] [EXT] " Pavan Nikhilesh Bhagavatula
2019-04-13 5:52 ` Pavan Nikhilesh Bhagavatula
2019-04-15 18:16 ` Yongseok Koh
2019-04-15 18:16 ` Yongseok Koh
2019-04-12 23:24 ` [dpdk-dev] [PATCH 2/6] meson: change default cache line size for cortex-a72 Yongseok Koh
2019-04-12 23:24 ` Yongseok Koh
2019-04-13 6:43 ` [dpdk-dev] [EXT] " Jerin Jacob Kollanukkaran
2019-04-13 6:43 ` Jerin Jacob Kollanukkaran
2019-04-15 4:35 ` Honnappa Nagarahalli
2019-04-15 4:35 ` Honnappa Nagarahalli
2019-04-15 13:40 ` Honnappa Nagarahalli
2019-04-15 13:40 ` Honnappa Nagarahalli
2019-04-15 20:40 ` Yongseok Koh
2019-04-15 20:40 ` Yongseok Koh
2019-04-15 20:44 ` Honnappa Nagarahalli
2019-04-15 20:44 ` Honnappa Nagarahalli
2019-04-12 23:24 ` [dpdk-dev] [PATCH 3/6] net/mlx: fix library search in meson build Yongseok Koh
2019-04-12 23:24 ` Yongseok Koh
2019-04-15 9:19 ` Bruce Richardson
2019-04-15 9:19 ` Bruce Richardson
2019-04-15 19:48 ` Yongseok Koh
2019-04-15 19:48 ` Yongseok Koh
2019-04-15 10:12 ` Luca Boccassi
2019-04-15 10:12 ` Luca Boccassi
2019-04-15 19:48 ` Yongseok Koh
2019-04-15 19:48 ` Yongseok Koh
2019-04-18 9:25 ` Luca Boccassi
2019-04-18 9:25 ` Luca Boccassi
2019-04-18 10:14 ` Bruce Richardson
2019-04-18 10:14 ` Bruce Richardson
2019-04-18 11:25 ` Yongseok Koh
2019-04-18 11:25 ` Yongseok Koh
2019-04-12 23:24 ` [dpdk-dev] [PATCH 4/6] meson: add Mellanox BlueField cross-compile config Yongseok Koh
2019-04-12 23:24 ` Yongseok Koh
2019-04-13 7:04 ` [dpdk-dev] [EXT] " Jerin Jacob Kollanukkaran
2019-04-13 7:04 ` Jerin Jacob Kollanukkaran
2019-04-12 23:24 ` [dpdk-dev] [PATCH 5/6] build: add option for armv8 crypto extension Yongseok Koh
2019-04-12 23:24 ` Yongseok Koh
2019-04-13 7:22 ` [dpdk-dev] [EXT] " Jerin Jacob Kollanukkaran
2019-04-13 7:22 ` Jerin Jacob Kollanukkaran
2019-04-15 4:52 ` Honnappa Nagarahalli
2019-04-15 4:52 ` Honnappa Nagarahalli
2019-04-15 18:43 ` Yongseok Koh
2019-04-15 18:43 ` Yongseok Koh
2019-04-15 20:13 ` Honnappa Nagarahalli
2019-04-15 20:13 ` Honnappa Nagarahalli
2019-04-17 16:28 ` Yongseok Koh
2019-04-17 16:28 ` Yongseok Koh
2019-04-30 3:33 ` Honnappa Nagarahalli
2019-04-30 3:33 ` Honnappa Nagarahalli
2019-05-02 1:54 ` Yongseok Koh [this message]
2019-05-02 1:54 ` Yongseok Koh
2019-05-02 10:13 ` Jerin Jacob Kollanukkaran
2019-05-02 10:13 ` Jerin Jacob Kollanukkaran
2019-05-02 23:08 ` Yongseok Koh
2019-05-02 23:08 ` Yongseok Koh
2019-05-02 23:33 ` Yongseok Koh
2019-05-02 23:33 ` Yongseok Koh
2019-05-03 10:28 ` Jerin Jacob Kollanukkaran
2019-05-03 10:28 ` Jerin Jacob Kollanukkaran
2019-05-03 3:54 ` Honnappa Nagarahalli
2019-05-03 3:54 ` Honnappa Nagarahalli
2019-05-03 9:49 ` Yongseok Koh
2019-05-03 9:49 ` Yongseok Koh
2019-05-03 14:21 ` Honnappa Nagarahalli
2019-05-03 14:21 ` Honnappa Nagarahalli
2019-04-12 23:24 ` [dpdk-dev] [PATCH 6/6] mk: disable armv8 crypto extension for Mellanox BlueField Yongseok Koh
2019-04-12 23:24 ` Yongseok Koh
2019-04-13 7:12 ` [dpdk-dev] [EXT] [PATCH 0/6] build: fix build for arm64 Jerin Jacob Kollanukkaran
2019-04-13 7:12 ` Jerin Jacob Kollanukkaran
2019-04-15 20:56 ` Yongseok Koh
2019-04-15 20:56 ` Yongseok Koh
2019-04-16 5:57 ` Jerin Jacob Kollanukkaran
2019-04-16 5:57 ` Jerin Jacob Kollanukkaran
2019-04-17 20:06 ` [dpdk-dev] " Thomas Monjalon
2019-04-17 20:06 ` Thomas Monjalon
2019-04-17 20:24 ` Honnappa Nagarahalli
2019-04-17 20:24 ` Honnappa Nagarahalli
2019-04-17 22:14 ` Yongseok Koh
2019-04-17 22:14 ` Yongseok Koh
2019-04-18 1:47 ` [dpdk-dev] [PATCH v2 1/4] meson: disable octeontx for buggy compilers on arm64 Yongseok Koh
2019-04-18 1:47 ` Yongseok Koh
2019-04-18 1:47 ` [dpdk-dev] [PATCH v2 2/4] meson: change default cache line size for armv8 Yongseok Koh
2019-04-18 1:47 ` Yongseok Koh
2019-04-18 5:00 ` Honnappa Nagarahalli
2019-04-18 5:00 ` Honnappa Nagarahalli
2019-04-18 8:23 ` [dpdk-dev] [EXT] " Hemant Agrawal
2019-04-18 8:23 ` Hemant Agrawal
2019-04-18 11:32 ` Yongseok Koh
2019-04-18 11:32 ` Yongseok Koh
2019-04-18 1:47 ` [dpdk-dev] [PATCH v2 3/4] net/mlx: fix library search in meson build Yongseok Koh
2019-04-18 1:47 ` Yongseok Koh
2019-04-18 1:47 ` [dpdk-dev] [PATCH v2 4/4] meson: add Mellanox BlueField cross-compile config Yongseok Koh
2019-04-18 1:47 ` Yongseok Koh
2019-04-18 7:21 ` [dpdk-dev] [EXT] [PATCH v2 1/4] meson: disable octeontx for buggy compilers on arm64 Jerin Jacob Kollanukkaran
2019-04-18 7:21 ` Jerin Jacob Kollanukkaran
2019-04-18 10:41 ` Yongseok Koh
2019-04-18 10:41 ` Yongseok Koh
2019-04-18 11:04 ` Thomas Monjalon
2019-04-18 11:04 ` Thomas Monjalon
2019-04-18 11:10 ` Yongseok Koh
2019-04-18 11:10 ` Yongseok Koh
2019-04-18 11:49 ` [dpdk-dev] [PATCH v3 1/4] drivers/event: " Yongseok Koh
2019-04-18 11:49 ` Yongseok Koh
2019-04-18 11:49 ` [dpdk-dev] [PATCH v3 2/4] meson: change default config for armv8 Yongseok Koh
2019-04-18 11:49 ` Yongseok Koh
2019-04-18 14:25 ` Honnappa Nagarahalli
2019-04-18 14:25 ` Honnappa Nagarahalli
2019-04-18 11:49 ` [dpdk-dev] [PATCH v3 3/4] net/mlx: fix library search in meson build Yongseok Koh
2019-04-18 11:49 ` Yongseok Koh
2019-04-18 12:02 ` Luca Boccassi
2019-04-18 12:02 ` Luca Boccassi
2019-04-18 11:49 ` [dpdk-dev] [PATCH v3 4/4] meson: add Mellanox BlueField cross-compile config Yongseok Koh
2019-04-18 11:49 ` Yongseok Koh
2019-04-18 16:23 ` Thomas Monjalon
2019-04-18 16:23 ` Thomas Monjalon
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=0B9F2C32-4530-4A7E-AE63-657361771001@mellanox.com \
--to=yskoh@mellanox.com \
--cc=Gavin.Hu@arm.com \
--cc=Honnappa.Nagarahalli@arm.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=jerinj@marvell.com \
--cc=nd@arm.com \
--cc=pbhagavatula@marvell.com \
--cc=shahafs@mellanox.com \
--cc=thomas@monjalon.net \
/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).