Add minimum requirement in arm when deploying TiFlash (#12386) (#12406)

pingcap · Feb 7, 2023 · 7042fd6 · 7042fd6
1 parent 4f13623
commit 7042fd6
Show file tree

Hide file tree

Showing 4 changed files with 4 additions and 4 deletions.
diff --git a/hardware-and-software-requirements.md b/hardware-and-software-requirements.md
@@ -108,7 +108,7 @@ You can deploy and run TiDB on the 64-bit generic hardware server platform in th
 > - For the TiKV server, it is recommended to use NVMe SSDs to ensure faster reads and writes.
 > - If you only want to test and verify the features, follow [Quick Start Guide for TiDB](/quick-start-with-tidb.md) to deploy TiDB on a single machine.
 > - The TiDB server uses the disk to store server logs, so there are no special requirements for the disk type and capacity in the test environment.
-> - Starting from v6.3.0, to deploy TiFlash under the Linux AMD64 architecture, the CPU must support AVX2 instruction sets. Use `cat /proc/cpuinfo | grep avx2` to confirm that there is output. By using such CPU instruction sets, TiFlash's vectorization engine can deliver better performance.
+> - Starting from v6.3.0, to deploy TiFlash under the Linux AMD64 architecture, the CPU must support the AVX2 instruction set. Ensure that `cat /proc/cpuinfo | grep avx2` has output. To deploy TiFlash under the Linux ARM64 architecture, the CPU must support the ARMv8 instruction set architecture. Ensure that `cat /proc/cpuinfo | grep 'crc32' | grep 'asimd'` has output. By using the instruction set extensions, TiFlash's vectorization engine can deliver better performance.
 
 ### Production environment
 

diff --git a/releases/release-6.3.0.md b/releases/release-6.3.0.md
@@ -256,7 +256,7 @@ In v6.3.0-DMR, the key new features and improvements are as follows:
 * Log backup supports GCS and Azure Blob Storage as backup storage.
 * Log backup is now compatible with the `exchange partition` DDL.
 * The SQL statement `ALTER TABLE ...SET TiFLASH MODE ...` previously used for enabling [fastscan](/develop/dev-guide-use-fastscan.md) is deprecated, and replaced by the system variable [`tiflash_fastscan`](/system-variables.md#tiflash_fastscan-new-in-v630). When you upgrade from v6.2.0 to v6.3.0, all FastScan settings in v6.2.0 will become invalid, but will not affect the normal reading of data. In this case, you need to configure the variable [`tiflash_fastscan`](/system-variables.md#tiflash_fastscan-new-in-v630) to enable or disable FastScan. When you upgrade from an earlier version to v6.3.0, the FastScan feature is not enabled by default for all sessions to keep data consistent.
-* To deploy TiFlash under the Linux AMD64 architecture, the CPU must support AVX2 instruction sets. Use `cat /proc/cpuinfo | grep avx2` to confirm that there is output. By using such CPU instruction sets, TiFlash's vectorization engine can deliver better performance.
+* To deploy TiFlash under the Linux AMD64 architecture, the CPU must support the AVX2 instruction set. Ensure that `cat /proc/cpuinfo | grep avx2` has output. To deploy TiFlash under the Linux ARM64 architecture, the CPU must support the ARMv8 instruction set architecture. Ensure that `cat /proc/cpuinfo | grep 'crc32' | grep 'asimd'` has output. By using the instruction set extensions, TiFlash's vectorization engine can deliver better performance.
 * The minimum version of HAProxy that works with TiDB is now v1.5. HAProxy versions between v1.5 and v2.1 now require the `post-41` configuration option to be set in `mysql-check`. It is recommended to use HAProxy v2.2 or newer.
 
 ## Removed feature

diff --git a/tiflash/tiflash-overview.md b/tiflash/tiflash-overview.md
@@ -25,7 +25,7 @@ TiFlash provides the columnar storage, with a layer of coprocessors efficiently
 
 TiFlash conducts real-time replication of data in the TiKV nodes at a low cost that does not block writes in TiKV. Meanwhile, it provides the same read consistency as in TiKV and ensures that the latest data is read. The Region replica in TiFlash is logically identical to those in TiKV, and is split and merged along with the Leader replica in TiKV at the same time.
 
-To deploy TiFlash under the Linux AMD64 architecture, the CPU must support AVX2 instruction sets. Use `cat /proc/cpuinfo | grep avx2` to confirm that there is output. By using such CPU instruction sets, TiFlash's vectorization engine can deliver better performance.
+To deploy TiFlash under the Linux AMD64 architecture, the CPU must support the AVX2 instruction set. Ensure that `cat /proc/cpuinfo | grep avx2` has output. To deploy TiFlash under the Linux ARM64 architecture, the CPU must support the ARMv8 instruction set architecture. Ensure that `cat /proc/cpuinfo | grep 'crc32' | grep 'asimd'` has output. By using the instruction set extensions, TiFlash's vectorization engine can deliver better performance.
 
 <CustomContent platform="tidb">
 

diff --git a/upgrade-tidb-using-tiup.md b/upgrade-tidb-using-tiup.md
@@ -36,7 +36,7 @@ This document is targeted for the following upgrade paths:
     3. Update the 3.0 version to 4.0 according to [Upgrade TiDB Using TiUP (v4.0)](https://docs.pingcap.com/tidb/v4.0/upgrade-tidb-using-tiup#import-tidb-ansible-and-the-inventoryini-configuration-to-tiup).
     4. Upgrade the cluster to v6.5.0 according to this document.
 - Support upgrading the versions of TiDB Binlog, TiCDC, TiFlash, and other components.
-- When upgrading TiFlash deployed under the Linux AMD64 architecture from versions earlier than v6.3.0 to v6.3.0 and later versions, ensure that the CPU supports AVX2 instruction sets. For details, see the description in [v6.3.0 Release Notes](/releases/release-6.3.0.md#others).
+- When upgrading TiFlash from versions earlier than v6.3.0 to v6.3.0 and later versions, note that the CPU must support the AVX2 instruction set under the Linux AMD64 architecture and the ARMv8 instruction set architecture under the Linux ARM64 architecture. For details, see the description in [v6.3.0 Release Notes](/releases/release-6.3.0.md#others).
 - For detailed compatibility changes of different versions, see the [Release Notes](/releases/release-notes.md) of each version. Modify your cluster configuration according to the "Compatibility Changes" section of the corresponding release notes.
 - For clusters that upgrade from versions earlier than v5.3 to v5.3 or later versions, the default deployed Prometheus will upgrade from v2.8.1 to v2.27.1. Prometheus v2.27.1 provides more features and fixes a security issue. Compared with v2.8.1, alert time representation in v2.27.1 is changed. For more details, see [Prometheus commit](https://github.com/prometheus/prometheus/commit/7646cbca328278585be15fa615e22f2a50b47d06) for more details.