linux-kernel - Re: [PATCH] ufs: core: add caps UFSHCD_CAP_MCQ

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20250422072226.844-1-tanghuan@vivo.com>
Date: Tue, 22 Apr 2025 15:22:26 +0800
From: Huan Tang <tanghuan@...o.com>
To: bvanassche@....org
Cc: James.Bottomley@...senPartnership.com,
	alim.akhtar@...sung.com,
	avri.altman@....com,
	ebiggers@...gle.com,
	linux-kernel@...r.kernel.org,
	linux-scsi@...r.kernel.org,
	manivannan.sadhasivam@...aro.org,
	martin.petersen@...cle.com,
	minwoo.im@...sung.com,
	opensource.kernel@...o.com,
	peter.wang@...iatek.com,
	quic_nguyenb@...cinc.com,
	tanghuan@...o.com
Subject: Re: [PATCH] ufs: core: add caps UFSHCD_CAP_MCQ_EN

> What is the root-cause of the high latency? If it is the time spent in
> UFS interrupt handlers, please try to switch to threaded interrupt
> handlers instead of disabling MCQ. See also
> https://lore.kernel.org/linux-scsi/20250407-topic-ufs-use-threaded-irq-v3-0-08bee980f71e@linaro.org/

Hi bart sir,

Thanks for your reply!

Thank you for your advice on the direction of interrupt handlers. 

The reason has not been found yet, but a comparative experiment was done, 
and the results are as follows:

The failure probability is 7/400 when MCQ is turned on, and 0/400
when MCQ is turned off; the definition of failure here is: the maximum
latency is greater than 5s.



Thanks
Huan