Date: Fri, 5 Jul 2019 15:27:56 +0000 From: Rick Macklem <rmacklem@uoguelph.ca> To: "freebsd-current@FreeBSD.org" <freebsd-current@FreeBSD.org> Cc: Jilles Tjoelker <jilles@FreeBSD.org>, Alan Somers <asomers@freebsd.org>, "kib@freebsd.org" <kib@FreeBSD.org> Subject: Re: [Differential] D20584: add a linux compatible copy_file_range(2) syscall Message-ID: <YTXPR01MB02854D6BDAF9EB943DEC3F1EDDF50@YTXPR01MB0285.CANPRD01.PROD.OUTLOOK.COM> In-Reply-To: <f9a25b0afb51a92b4fd28f57a18b0d73@localhost.localdomain> References: <differential-rev-PHID-DREV-ki3nyojxtg6yf3c3i7o3-req@reviews.freebsd.org>, <f9a25b0afb51a92b4fd28f57a18b0d73@localhost.localdomain>
next in thread | previous in thread | raw e-mail | index | archive | help
jilles wrote in copy_file_range.2:99 > The Linux man page (from http://man7.org/linux/man->pages/man2/copy_file_range.2.html ) says that a non-zero flags argument will cause >the call to return an [EINVAL] error. I think that is better than ignoring the argument >completely since it allows adding flags more safely (since there will not be existing >applications that pass in, for example, uninitialized data as flags). The fun part is that the Linux folks are already discussing adding flags. I don't know if they are already in Linux-next (or whatever they call their next release), but it sounded like they were headed that way. As such, I thought ignoring "flags" would be easier than returning EINVAL for code that works on Linux. However, I can see the counter argument, which is "returning EINVAL will indicate that the Linux flag isn't used on FreeBSD", so that developers will become aware of that. What do others think w.r.t. which is the better approach? rick
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?YTXPR01MB02854D6BDAF9EB943DEC3F1EDDF50>