NetBSD-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: kern/52769



The following reply was made to PR kern/52769; it has been noted by GNATS.

From: Sevan Janiyan <venture37%geeklan.co.uk@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: kern/52769
Date: Sun, 4 Mar 2018 17:28:32 +0000

 I do not have an nvme based system but I frequently experience deadlocks
 on i386. The deadlocks are pretty catastrophic in that it clobbers the
 file system.
 
 At first I thought I was issue with the HDD, the disk I was originally
 running with was the 64Gb SATA disk my X60s came with. I swapped that
 out for a shiny 60GB sandisk SSD, issue continued.
 The file system was mounted with the log option on both disks.
 I switched to discard recently for TRIM on the SSD, system crashes.
 
 System is running with coreboot, thought it may be because of that, I
 had some oddities with DTrace previously which resolved itself after a
 coreboot update, still could be this bug reports make me think that
 issue lies elsewhere and I only experience these deadlocks when running CVS.
 
 In my case, the last time it happened, I was unable to enter DDB, it
 just sat there waiting. The console was responsive though.
 
 The issue in my case happens when there is a lot of CVS activity, that
 is triggered from updating from an old check out. I updated from
 Decembers checkout and that took out a large part of my checkout (the
 files in CVS directories get filled with snippets of other files).
 The recent GCC import triggered it.
 Wipping src/external, updating to regain the directory then updating
 once more sometimes triggers it.
 
 will try running / without any additional options beynod rw and see have
 I go.
 
 Sevan
 


Home | Main Index | Thread Index | Old Index