linux-kernel - RE: [char-misc-next 4/4 V2] mei: bus: enable non-blocking RX

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 for Android: free password hash cracker in your pocket

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

Message-ID: <5B8DA87D05A7694D9FA63FD143655C1B54333D99@hasmsx108.ger.corp.intel.com>
Date:   Tue, 29 Nov 2016 20:09:38 +0000
From:   "Winkler, Tomas" <tomas.winkler@...el.com>
To:     'Greg Kroah-Hartman' <gregkh@...uxfoundation.org>
CC:     "Usyskin, Alexander" <alexander.usyskin@...el.com>,
        "'linux-kernel@...r.kernel.org'" <linux-kernel@...r.kernel.org>
Subject: RE: [char-misc-next 4/4 V2] mei: bus: enable non-blocking RX


> On Mon, Nov 28, 2016 at 11:03:20PM +0000, Winkler, Tomas wrote:
> > >
> > >
> > > >
> > > > On Sat, Nov 19, 2016 at 02:16:11PM +0200, Tomas Winkler wrote:
> > > > > From: Alexander Usyskin <alexander.usyskin@...el.com>
> > > > >
> > > > > Enable non-blocking receive for drivers on mei bus, this allows
> > > > > checking for data availability by mei client drivers. This is
> > > > > most effective for fixed address clients, that lacks flow control.
> > > > >
> > > > > This function adds new API function mei_cldev_recv_nonblock(),
> > > > > it retuns -EGAIN if function will block.
> > > > >
> > > > > Signed-off-by: Alexander Usyskin <alexander.usyskin@...el.com>
> > > > > Signed-off-by: Tomas Winkler <tomas.winkler@...el.com>
> > > > > ---
> > > > > V2: use _nonblock() function suffix instead of NONBLOCK flag
> > > > >
> > > > >  drivers/misc/mei/bus-fixup.c |  4 ++--
> > > > >  drivers/misc/mei/bus.c       | 31 +++++++++++++++++++++++++++++--
> > > > >  drivers/misc/mei/mei_dev.h   |  7 ++++++-
> > > > >  include/linux/mei_cl_bus.h   |  6 ++++--
> > > > >  4 files changed, 41 insertions(+), 7 deletions(-)
> > > > >
> > > > > diff --git a/drivers/misc/mei/bus-fixup.c
> > > > > b/drivers/misc/mei/bus-fixup.c index 7f2cef9011ae..18e05ca7584f
> > > > > 100644
> > > > > --- a/drivers/misc/mei/bus-fixup.c
> > > > > +++ b/drivers/misc/mei/bus-fixup.c
> > > > > @@ -141,7 +141,7 @@ static int mei_osver(struct mei_cl_device
> *cldev)
> > > > >  	if (ret < 0)
> > > > >  		return ret;
> > > > >
> > > > > -	ret = __mei_cl_recv(cldev->cl, buf, length);
> > > > > +	ret = __mei_cl_recv(cldev->cl, buf, length, 0);
> > > >
> > > > What is 0 here?  Again, mode...
> > >
> > > Yes,  it means no change in behavior,  but this is an internal function.
> 
> So, it makes no sense, are you not going to have to read this code again in 10
> years?  New developers?  Make the code make sense please.


Sorry Greg, the code does make sense to me, the whole kernel passes nonblock around as flag starting from the syscall (O_NONBLOCK)
 it doesn't make sense to write two functions that differ in one 'if' statement.
I understand that you  are in some crusade against flags, but you are  not proposing a concrete solution and I don't have one either.
I can solve it in the external wrapper, but internally it's just a same function.

Thanks
Tomas