netdev - Re: [PATCH v3 08/11] Input: hyperv-keyboard: Make ringbuffer at least take two pages

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20200914102239.GB45838@debian-boqun.qqnc3lrjykvubdpftowmye0fmh.lx.internal.cloudapp.net>
Date:   Mon, 14 Sep 2020 18:22:39 +0800
From:   Boqun Feng <boqun.feng@...il.com>
To:     Wei Liu <wei.liu@...nel.org>
Cc:     linux-hyperv@...r.kernel.org, linux-input@...r.kernel.org,
        linux-kernel@...r.kernel.org, netdev@...r.kernel.org,
        linux-scsi@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
        "K. Y. Srinivasan" <kys@...rosoft.com>,
        Haiyang Zhang <haiyangz@...rosoft.com>,
        Stephen Hemminger <sthemmin@...rosoft.com>,
        Jiri Kosina <jikos@...nel.org>,
        Benjamin Tissoires <benjamin.tissoires@...hat.com>,
        Dmitry Torokhov <dmitry.torokhov@...il.com>,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        "James E.J. Bottomley" <jejb@...ux.ibm.com>,
        "Martin K. Petersen" <martin.petersen@...cle.com>,
        Michael Kelley <mikelley@...rosoft.com>, will@...nel.org,
        ardb@...nel.org, arnd@...db.de, catalin.marinas@....com,
        mark.rutland@....com, maz@...nel.org
Subject: Re: [PATCH v3 08/11] Input: hyperv-keyboard: Make ringbuffer at
 least take two pages

On Mon, Sep 14, 2020 at 09:30:16AM +0000, Wei Liu wrote:
> On Mon, Sep 14, 2020 at 04:46:00PM +0800, Boqun Feng wrote:
> > On Thu, Sep 10, 2020 at 10:34:52PM +0800, Boqun Feng wrote:
> > > When PAGE_SIZE > HV_HYP_PAGE_SIZE, we need the ringbuffer size to be at
> > > least 2 * PAGE_SIZE: one page for the header and at least one page of
> > > the data part (because of the alignment requirement for double mapping).
> > > 
> > > So make sure the ringbuffer sizes to be at least 2 * PAGE_SIZE when
> > > using vmbus_open() to establish the vmbus connection.
> > > 
> > > Signed-off-by: Boqun Feng <boqun.feng@...il.com>
> > > ---
> > >  drivers/input/serio/hyperv-keyboard.c | 4 ++--
> > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/drivers/input/serio/hyperv-keyboard.c b/drivers/input/serio/hyperv-keyboard.c
> > > index df4e9f6f4529..6ebc61e2db3f 100644
> > > --- a/drivers/input/serio/hyperv-keyboard.c
> > > +++ b/drivers/input/serio/hyperv-keyboard.c
> > > @@ -75,8 +75,8 @@ struct synth_kbd_keystroke {
> > >  
> > >  #define HK_MAXIMUM_MESSAGE_SIZE 256
> > >  
> > > -#define KBD_VSC_SEND_RING_BUFFER_SIZE		(40 * 1024)
> > > -#define KBD_VSC_RECV_RING_BUFFER_SIZE		(40 * 1024)
> > > +#define KBD_VSC_SEND_RING_BUFFER_SIZE	max(40 * 1024, (int)(2 * PAGE_SIZE))
> > > +#define KBD_VSC_RECV_RING_BUFFER_SIZE	max(40 * 1024, (int)(2 * PAGE_SIZE))
> > >  
> > 
> > Hmm.. just realized there is a problem here, if PAGE_SIZE = 16k, then
> > 40 * 1024 > 2 * PAGE_SIZE, however in the ring buffer size should also
> > be page aligned, otherwise vmbus_open() will fail.
> > 
> > I plan to modify this as
> > 
> > in linux/hyperv.h:
> > 
> > #define VMBUS_RING_SIZE(payload_sz) PAGE_ALIGN(sizeof(struct hv_ring_buffer) + (playload_sz))
> > 
> > and here:
> > 
> > #define KBD_VSC_SEND_RING_BUFFER_SIZE VMBUS_RING_SIZE(36 * 1024)
> > #define KBD_VSC_RECV_RING_BUFFER_SIZE VMBUS_RING_SIZE(36 * 1024)
> > 
> > and the similar change for patch #9.
> 
> OOI why do you reduce the size by 4k here?
> 

To keep the total ring buffer size unchanged (still 40k) when
PAGE_SIZE=4k. Because in VMBUS_RING_SIZE() (which I plan to rename as
HV_RING_SIZE()), the hv_ring_buffer size is already added.

Regards,
Boqun

> Wei.
> 
> > 
> > Thoughts?
> > 
> > Regards,
> > Boqun
> > 
> > >  #define XTKBD_EMUL0     0xe0
> > >  #define XTKBD_EMUL1     0xe1
> > > -- 
> > > 2.28.0
> > >