linux-kernel - Re: [PATCH 6/7] platform/x86: fujitsu-laptop: More accurately represent the hotkey ring buffer managed by firmware

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

Date:   Sun, 4 Mar 2018 20:57:28 +0100
From:   Michał Kępień <kernel@...pniu.pl>
To:     Andy Shevchenko <andy.shevchenko@...il.com>
Cc:     Jonathan Woithe <jwoithe@...t42.net>,
        Darren Hart <dvhart@...radead.org>,
        Andy Shevchenko <andy@...radead.org>,
        Platform Driver <platform-driver-x86@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 6/7] platform/x86: fujitsu-laptop: More accurately
 represent the hotkey ring buffer managed by firmware

> On Tue, Feb 27, 2018 at 11:15 PM, Michał Kępień <kernel@...pniu.pl> wrote:
> > The MAX_HOTKEY_RINGBUFFER_SIZE constant is set to 100, which allows up
> > to 100 hotkey events to be drained from the firmware ring buffer upon
> > module load.  However, no known firmware is capable of holding more than
> > 16 hotkey events in its internal ring buffer:
> 
> > The RINGBUFFERSIZE constant is set to 40, which allows the module to
> > queue up to 40 hotkey events for delivery to user space.  As this value
> > seems arbitrarily chosen and 16 should be more than enough for any
> > practical use case, merge the two aforementioned constants into one
> > (HOTKEY_RINGBUFFER_SIZE) in order to simplify code and more accurately
> > represent the underlying data structure.
> 
> > +#define HOTKEY_RINGBUFFER_SIZE         16
> 
> This need the comment or a
> 
> BUILD_BUG_ON(!is_power_of_2(...));

Okay, I will probably take the comment route in v2.

> > -       ret = kfifo_alloc(&priv->fifo, RINGBUFFERSIZE * sizeof(int),
> > +       ret = kfifo_alloc(&priv->fifo, HOTKEY_RINGBUFFER_SIZE * sizeof(int),
> >                           GFP_KERNEL);
> 
> >         while (call_fext_func(device, FUNC_BUTTONS, OP_GET_EVENTS,
> >                               0x0, 0x0) != 0 &&
> > -              i++ < MAX_HOTKEY_RINGBUFFER_SIZE)
> > +              i++ < HOTKEY_RINGBUFFER_SIZE)
> >                 ; /* No action, result is discarded */
> 
> This looks horrible.

It sure does!  Hence patch 7/7, which does the following:

-	while (call_fext_func(device, FUNC_BUTTONS, OP_GET_EVENTS,
-			      0x0, 0x0) != 0 &&
+	while (fext_buttons(device, OP_GET_EVENTS, 0x0, 0x0) != 0 &&
	       i++ < HOTKEY_RINGBUFFER_SIZE)

In other words, patch 6/7 is just a stopover on the way to shorten
current module code:

-	while (call_fext_func(device, FUNC_BUTTONS, 0x1, 0x0, 0x0) != 0 &&
-	       i++ < MAX_HOTKEY_RINGBUFFER_SIZE)
+	while (fext_buttons(device, OP_GET_EVENTS, 0x0, 0x0) != 0 &&
+	       i++ < HOTKEY_RINGBUFFER_SIZE)

> >         while ((irb = call_fext_func(device, FUNC_BUTTONS, OP_GET_EVENTS,
> >                                      0x0, 0x0)) != 0 &&
> > -              i++ < MAX_HOTKEY_RINGBUFFER_SIZE) {
> > +              i++ < HOTKEY_RINGBUFFER_SIZE) {
> 
> Ditto.

Similarly, patch 7/7 does the following:

-	while ((irb = call_fext_func(device, FUNC_BUTTONS, OP_GET_EVENTS,
-				     0x0, 0x0)) != 0 &&
+	while ((irb = fext_buttons(device, OP_GET_EVENTS, 0x0, 0x0)) != 0 &&
	       i++ < HOTKEY_RINGBUFFER_SIZE) {

The diff against current module code is thus:

-	while ((irb = call_fext_func(device,
-				     FUNC_BUTTONS, 0x1, 0x0, 0x0)) != 0 &&
-	       i++ < MAX_HOTKEY_RINGBUFFER_SIZE) {
+	while ((irb = fext_buttons(device, OP_GET_EVENTS, 0x0, 0x0)) != 0 &&
+	       i++ < HOTKEY_RINGBUFFER_SIZE) {

-- 
Best regards,
Michał Kępień