lkl / linux

Linux kernel source tree

Home Page:https://lkl.github.io/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Disable raid6 by default?

AkihiroSuda opened this issue · comments

raid6 seems causing delay in booting kernel.

  [    0.000000] Linux version 5.3.0+ (suda@suda-ws01) (gcc version 9.2.1 20191008 (Ubuntu 9.2.1-9ubuntu2)) #1 Fri Feb 21 00:43:33 JST 2020
  [    0.000000] memblock address range: 0x7fbaab000000 - 0x7fbaabfff000
  [    0.000000] On node 0 totalpages: 4095
  [    0.000000]   Normal zone: 56 pages used for memmap
  [    0.000000]   Normal zone: 0 pages reserved
  [    0.000000]   Normal zone: 4095 pages, LIFO batch:0
  [    0.000000] pcpu-alloc: s0 r0 d32768 u32768 alloc=1*32768
  [    0.000000] pcpu-alloc: [0] 0 
  [    0.000000] Built 1 zonelists, mobility grouping off.  Total pages: 4039
  [    0.000000] Kernel command line: mem=16M loglevel=8
  [    0.000000] Dentry cache hash table entries: 2048 (order: 2, 16384 bytes, linear)
  [    0.000000] Inode-cache hash table entries: 1024 (order: 1, 8192 bytes, linear)
  [    0.000000] mem auto-init: stack:off, heap alloc:off, heap free:off
  [    0.000000] Memory available: 16092k/16380k RAM
  [    0.000000] SLUB: HWalign=32, Order=0-3, MinObjects=0, CPUs=1, Nodes=1
  [    0.000000] NR_IRQS: 4096
  [    0.000000] lkl: irqs initialized
  [    0.000000] clocksource: lkl: mask: 0xffffffffffffffff max_cycles: 0x1cd42e4dffb, max_idle_ns: 881590591483 ns
  [    0.000001] lkl: time and timers initialized (irq1)
  [    0.000007] pid_max: default: 4096 minimum: 301
  [    0.000038] Mount-cache hash table entries: 512 (order: 0, 4096 bytes, linear)
  [    0.000043] Mountpoint-cache hash table entries: 512 (order: 0, 4096 bytes, linear)
  [    0.007013] random: get_random_bytes called from _etext+0xbd5c/0x14b86 with crng_init=0
  [    0.007052] printk: console [lkl_console0] enabled
  [    0.007078] clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 19112604462750000 ns
  [    0.007083] xor: automatically using best checksumming function   8regs     
  [    0.007209] NET: Registered protocol family 16
  [    0.174770] raid6: int64x8  gen()  3650 MB/s
  [    0.347313] raid6: int64x8  xor()  2756 MB/s
  [    0.520721] raid6: int64x4  gen()  4022 MB/s
  [    0.694763] raid6: int64x4  xor()  2849 MB/s
  [    0.868448] raid6: int64x2  gen()  5657 MB/s
  [    1.042517] raid6: int64x2  xor()  3401 MB/s
  [    1.216471] raid6: int64x1  gen()  4648 MB/s
  [    1.390256] raid6: int64x1  xor()  2601 MB/s
  [    1.390260] raid6: using algorithm int64x2 gen() 5657 MB/s
  [    1.390261] raid6: .... xor() 3401 MB/s, rmw enabled
  [    1.390262] raid6: using intx1 recovery algorithm
  [    1.390429] clocksource: Switched to clocksource lkl
  [    1.390568] NET: Registered protocol family 2
  [    1.390688] tcp_listen_portaddr_hash hash table entries: 256 (order: 0, 4096 bytes, linear)
  [    1.390692] TCP established hash table entries: 512 (order: 0, 4096 bytes, linear)
  [    1.390698] TCP bind hash table entries: 512 (order: 0, 4096 bytes, linear)
  [    1.390701] TCP: Hash tables configured (established 512 bind 512)
  [    1.390768] UDP hash table entries: 128 (order: 0, 4096 bytes, linear)
  [    1.390771] UDP-Lite hash table entries: 128 (order: 0, 4096 bytes, linear)
  [    1.391412] workingset: timestamp_bits=62 max_order=13 bucket_order=0
  [    1.392560] SGI XFS with ACLs, security attributes, no debug enabled
  [    1.393542] io scheduler mq-deadline registered
  [    1.393544] io scheduler kyber registered
  [    1.396620] NET: Registered protocol family 10
  [    1.397247] Segment Routing with IPv6
  [    1.397262] sit: IPv6, IPv4 and MPLS over IPv4 tunneling driver
  [    1.397735] Btrfs loaded, crc32c=crc32c-generic
  [    1.398000] Warning: unable to open an initial console.
  [    1.398014] This architecture does not have kernel memory protection.
  [    1.398016] Run /init as init process

Any chance to disable raid6 by default?

thanks, it's long standing issue in LKL and this can be work-arounded with !RAID6_PQ_BENCHMARK instead of disabling the entire option.

https://patchwork.ozlabs.org/patch/1209040/

Thanks, seems fixed in 16a701b