throtl/{002,003}: update max_sectors setting

[yizhanglinux]

On 64k ppc64le kernel, the max_sectors should be set by (64k/512), or it will trigger below WARNING:

[ 299.201428] ------------[ cut here ]------------ [ 299.201453] WARNING: CPU: 3 PID: 4680 at block/blk-settings.c:286 blk_validate_limits+0x440/0x540 [ 299.201471] Modules linked in: null_blk bonding tls rfkill ibmveth sg pseries_rng vmx_crypto fuse loop nfnetlink xfs sd_mod ibmvscsi scsi_transport_srp dm_mirror dm_region_hash dm_log dm_mod [ 299.201508] CPU: 3 UID: 0 PID: 4680 Comm: bash Kdump: loaded Not tainted 6.13.0-0.rc2.22.eln144.ppc64le #1 [ 299.201517] Hardware name: IBM,9009-22A POWER9 (architected) 0x4e0202 0xf000005 of:IBM,FW910.00 (VL910_062) hv:phyp pSeries [ 299.201523] NIP: c00000000098ccc0 LR: c000000000987280 CTR: 0000000000000000
[ 299.201529] REGS: c0000000502372a0 TRAP: 0700 Not tainted (6.13.0-0.rc2.22.eln144.ppc64le)
[ 299.201536] MSR: 8000000000029033 <SF,EE,ME,IR,DR,RI,LE> CR: 84228480 XER: 20040000
[ 299.201551] CFAR: c00000000098c984 IRQMASK: 0
[ 299.201551] GPR00: c000000000987264 c000000050237540 c000000001e0a000 c0000000502377c8
[ 299.201551] GPR04: 0000000000000200 00000000f0647558 0000000000001152 0000000000001151
[ 299.201551] GPR08: 0000000000000000 0000000000000000 0000000000000200 0000000000008000
[ 299.201551] GPR12: c0000000009a2a50 c00000000f8ccb00 0000000000000000 0000000000000000
[ 299.201551] GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
[ 299.201551] GPR20: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
[ 299.201551] GPR24: c000000002c2acf8 c00000000303de60 0000000000000000 c0000000502377c8
[ 299.201551] GPR28: c008000006ec07c8 0000000000000200 0000000000000008 c0000000502377c8
[ 299.201618] NIP [c00000000098ccc0] blk_validate_limits+0x440/0x540
[ 299.201625] LR [c000000000987280] blk_alloc_queue+0xc0/0x2f0
[ 299.201634] Call Trace:
[ 299.201637] [c000000050237540] [c0000000587564e0] 0xc0000000587564e0 (unreliable)
[ 299.201651] [c0000000502375c0] [c000000000987264] blk_alloc_queue+0xa4/0x2f0
[ 299.201659] [c000000050237620] [c0000000009a22e8] blk_mq_alloc_queue+0x88/0x130
[ 299.201667] [c000000050237720] [c0000000009a2a80] __blk_mq_alloc_disk+0x30/0xc4
[ 299.201674] [c000000050237760] [c008000006e025bc] null_add_dev+0x2a4/0x6d0 [null_blk]
[ 299.201687] [c0000000502378c0] [c008000006e03e0c] nullb_device_power_store+0x194/0x200 [null_blk]
[ 299.201700] [c000000050237920] [c000000000853128] configfs_write_iter+0x118/0x210
[ 299.201712] [c000000050237970] [c000000000731244] vfs_write+0x304/0x500
[ 299.201720] [c000000050237a20] [c000000000731638] ksys_write+0x88/0x150
[ 299.201728] [c000000050237a70] [c000000000031288] system_call_exception+0x138/0x2d0
[ 299.201737] [c000000050237e50] [c00000000000d05c] system_call_vectored_common+0x15c/0x2ec

[kawasaki]

@YuKuai-huawei FYI.
@yizhanglinux Thanks for finding this issue. I think the change is reasonable. I will do some test runs for confirmation.

[kawasaki]

My test run results were positive. I will wait for a few more days in case Yu Kuai has some comments on the changes.

[YuKuai-huawei]

Hi, 在 2024/12/16 14:04, kawasaki 写道:

My test run results were positive. I will wait for a few more days in case Yu Kuai has some comments on the changes.

The changes looks good to me. This is the correct way to set a minimal max_sectors. Reviewed-by: Yu Kuai ***@***.***>

…

— Reply to this email directly, view it on GitHub <#155 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BMMPCHZVQ5UHTAS6UWYPZML2FZUNFAVCNFSM6AAAAABTRJND6KVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNBUGY3TQMZVGU>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[kawasaki]

@YuKuai-huawei Thanks for the positive comment. I merged the commit with the link to this PR, so that the review history can be referred to.