Re: kern/55658: rumpvfs:t_etfs triggers bug in ufs_balloc_range() for 16KB page

[Rin Okuyama <rokuyama.rk%gmail.com@localhost>]

The following reply was made to PR kern/55658; it has been noted by GNATS.

From: Rin Okuyama <rokuyama.rk%gmail.com@localhost>
To: Chuck Silvers <chuq%chuq.com@localhost>
Cc: gnats-bugs%netbsd.org@localhost
Subject: Re: kern/55658: rumpvfs:t_etfs triggers bug in ufs_balloc_range() for
 16KB page
Date: Thu, 1 Oct 2020 14:11:54 +0900

 On 2020/09/21 16:42, Chuck Silvers wrote:
 > On Wed, Sep 16, 2020 at 01:09:23AM +0900, Rin Okuyama wrote:
 >> Does this patch seem OK to you? Is there anything else to be checked
 >> before commit?
 > 
 > The one thing that I can think of to be concerned about is that
 > now that we track page dirtiness via the radix tree tag,
 > we have to make sure that a clean page is not mapped writable,
 > so it would be good to have an assertion in ubc_fault_page() like:
 > 
 > 	KASSERT((access_type & VM_PROT_WRITE) == 0 ||
 > 		uvm_pagegetdirty(pg) != UVM_PAGE_STATUS_CLEAN);
 > 
 
 Thank you very much for your comment, and I'm so sorry for the late reply.
 
 I've added this KASSERT to ubc_fault_page(), and confirmed that it does
 not fire during full ATF runs at least for amd64, powerpc/oea, and
 powerpc/ibm4xx.
 
 I will commit it soon, probably in this weekend.
 
 Thanks,
 rin