[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1817268.LulUJvKFVv@silver>
Date: Thu, 21 Apr 2022 13:36:14 +0200
From: Christian Schoenebeck <linux_oss@...debyte.com>
To: asmadeus@...ewreck.org, David Howells <dhowells@...hat.com>
Cc: dhowells@...hat.com, David Kahurani <k.kahurani@...il.com>,
davem@...emloft.net, ericvh@...il.com, kuba@...nel.org,
linux-kernel@...r.kernel.org, lucho@...kov.net,
netdev@...r.kernel.org, v9fs-developer@...ts.sourceforge.net,
Greg Kurz <groug@...d.org>
Subject: Re: 9p EBADF with cache enabled (Was: 9p fs-cache tests/benchmark (was: 9p
fscache Duplicate cookie detected))
On Donnerstag, 21. April 2022 12:36:12 CEST David Howells wrote:
> asmadeus@...ewreck.org wrote:
> > int fd = open(argv[1], O_WRONLY|O_APPEND);
> > if (fd < 0)
> >
> > return 1;
> >
> > if (write(fd, "test\n", 5) < 0)
>
> I think I need to implement the ability to store writes in non-uptodate
> pages without needing to read from the server as NFS does. This may fix
> the performance drop also.
>
> David
I hope this does not sound harsh, wouldn't it make sense to revert
eb497943fa215897f2f60fd28aa6fe52da27ca6c for now until those issues are sorted
out? My concern is that it might take a long time to address them, and these
are not minor issues.
Best regards,
Christian Schoenebeck
Powered by blists - more mailing lists