Message-ID: <002601cb3926$edaf24d0$66f8800a@maildom.okisemi.com>
From: "Masayuki Ohtake" <masa-korg@dsn.okisemi.com>
To: "Greg KH" <gregkh@suse.de>
Cc: <meego-dev@meego.com>, "LKML" <linux-kernel@vger.kernel.org>,
        <qi.wang@intel.com>, <yong.y.wang@intel.com>,
        <andrew.chih.howe.khor@intel.com>, <arjan@linux.intel.com>
References: <4C612AA4.5040100@dsn.okisemi.com> <20100810171343.GE21220@suse.de>
Subject: Re: [MeeGo-Dev][PATCH] Topcliff: Update PCH_IEEE1588 driver to 2.6.35
Date: Wed, 11 Aug 2010 16:29:25 +0900
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 5834
Lines: 171

Hi Greg,

> Do they all have to be ioctls?  What exactly are they doing?
I think using ioctl is common for this patch.
Do you think that using ioctl is NOT appropriate for this patch?
Let me know your intension in more detail.

>
> And are they 32/64bit safe?
Only 32bit support.

Thanks, Ohtake
----- Original Message ----- 
From: "Greg KH" <gregkh@suse.de>
To: "Masayuki Ohtak" <masa-korg@dsn.okisemi.com>
Cc: <meego-dev@meego.com>; "LKML" <linux-kernel@vger.kernel.org>; <qi.wang@intel.com>; <yong.y.wang@intel.com>;
<andrew.chih.howe.khor@intel.com>; <arjan@linux.intel.com>
Sent: Wednesday, August 11, 2010 2:13 AM
Subject: Re: [MeeGo-Dev][PATCH] Topcliff: Update PCH_IEEE1588 driver to 2.6.35


> On Tue, Aug 10, 2010 at 07:32:04PM +0900, Masayuki Ohtak wrote:
> > IEEE1588 driver of Topcliff PCH
> >
> > Topcliff PCH is the platform controller hub that is going to be used in
> > Intel's upcoming general embedded platform. All IO peripherals in
> > Topcliff PCH are actually devices sitting on AMBA bus.
> > Topcliff PCH has IEEE1588 I/F. This driver enables IEEE1588 function on CAN or
> > Ethernet communications.
>
> So it's a CAN or Ethernet driver?  If so, why isn't this a network
> driver like the CAN and Ethernet drivers?
>
> > +#ifdef __GNUC__
> > +#define UNUSED __attribute__ ((unused))
> > +#define UNUSED_ARG(x)
> > +#else
> > +#define UNUSED
> > +#define UNUSED_ARG(x) (void) x
> > +#endif
> > +
> > +
> > +#define TRUE 1
> > +#define FALSE 0
>
> These lines should not be needed at all.
>
> > +#define IOC_1588_BASE 0xf8
>
> Do you have documentation for these new ioctls you are creating?  It's a
> lot of them:
>
> > +#define IOCTL_1588_PORT_CONFIG_SET \
> > + _IOW(IOC_1588_BASE, 0, struct pch_portcfg_ioctl)
> > +
> > +#define IOCTL_1588_PORT_CONFIG_GET \
> > + _IOWR(IOC_1588_BASE, 1, struct pch_portcfg_ioctl)
> > +
> > +#define IOCTL_1588_RX_POLL _IOWR(IOC_1588_BASE, 2, struct pch_rxtxpoll_ioctl)
> > +
> > +#define IOCTL_1588_TX_POLL _IOWR(IOC_1588_BASE, 3, struct pch_rxtxpoll_ioctl)
> > +
> > +#define IOCTL_1588_CAN_POLL _IOWR(IOC_1588_BASE, 4, struct pch_canpoll_ioctl)
> > +
> > +#define IOCTL_1588_SYS_TIME_GET _IOR(IOC_1588_BASE, 5, struct pch_tim_val)
> > +
> > +#define IOCTL_1588_SYS_TIME_SET _IOW(IOC_1588_BASE, 6, struct pch_tim_val)
> > +
> > +#define IOCTL_1588_TICK_RATE_SET _IOW(IOC_1588_BASE, 7, __u32)
> > +
> > +#define IOCTL_1588_TICK_RATE_GET _IOR(IOC_1588_BASE, 8, __u32)
> > +
> > +#define IOCTL_1588_TARG_TIME_INTRPT_ENABLE _IO(IOC_1588_BASE, 9)
> > +
> > +#define IOCTL_1588_TARG_TIME_INTRPT_DISABLE _IO(IOC_1588_BASE, 10)
> > +
> > +#define IOCTL_1588_TARG_TIME_POLL \
> > + _IOR(IOC_1588_BASE, 11, struct pch_timepoll_ioctl)
> > +
> > +#define IOCTL_1588_TARG_TIME_SET \
> > + _IOW(IOC_1588_BASE, 12, struct pch_tim_val)
> > +
> > +#define IOCTL_1588_TARG_TIME_GET \
> > + _IOR(IOC_1588_BASE, 13, struct pch_tim_val)
> > +
> > +#define IOCTL_1588_AUX_TIME_INTRPT_ENABLE _IOW(IOC_1588_BASE, 14,\
> > + enum pch_auxmode)
> > +
> > +#define IOCTL_1588_AUX_TIME_INTRPT_DISABLE _IOW(IOC_1588_BASE, 15,\
> > + enum pch_auxmode)
> > +
> > +#define IOCTL_1588_AUX_TIME_POLL \
> > + _IOWR(IOC_1588_BASE, 16, struct pch_timepoll_ioctl)
> > +
> > +#define IOCTL_1588_RESET _IO(IOC_1588_BASE, 17)
> > +
> > +#define IOCTL_1588_CHNL_RESET _IOW(IOC_1588_BASE, 18, enum pch_ptpport)
> > +
> > +#define IOCTL_1588_STATS_GET _IOR(IOC_1588_BASE, 19, struct pch_stats)
> > +
> > +#define IOCTL_1588_STATS_RESET _IO(IOC_1588_BASE, 20)
> > +
> > +#define IOCTL_1588_SHOW_ALL _IO(IOC_1588_BASE, 21)
> > +
> > +#define IOCTL_1588_AUX_TARG_TIME_INTRPT_ENABLE _IO(IOC_1588_BASE, 22)
> > +
> > +#define IOCTL_1588_AUX_TARG_TIME_INTRPT_DISABLE _IO(IOC_1588_BASE, 23)
> > +
> > +#define IOCTL_1588_AUX_TARG_TIME_POLL \
> > + _IOR(IOC_1588_BASE, 24, struct pch_timepoll_ioctl)
> > +
> > +#define IOCTL_1588_AUX_TARG_TIME_SET \
> > + _IOW(IOC_1588_BASE, 25, struct pch_tim_val)
> > +
> > +#define IOCTL_1588_AUX_TARG_TIME_GET \
> > + _IOR(IOC_1588_BASE, 26, struct pch_tim_val)
> > +
> > +#define IOCTL_1588_PULSE_PER_SEC_INTRPT_ENABLE _IO(IOC_1588_BASE, 27)
> > +
> > +#define IOCTL_1588_PULSE_PER_SEC_INTRPT_DISABLE _IO(IOC_1588_BASE, 28)
> > +
> > +#define IOCTL_1588_TARG_TIME_NOTIFY \
> > + _IOR(IOC_1588_BASE, 29, struct pch_tim_val)
> > +
> > +#define IOCTL_1588_AUX_TIME_NOTIFY \
> > + _IOR(IOC_1588_BASE, 30, struct pch_auxtimeioctl)
> > +
> > +#define IOCTL_1588_AUX_TARG_TIME_NOTIFY \
> > + _IOR(IOC_1588_BASE, 31, struct pch_tim_val)
> > +
> > +#define IOCTL_1588_PULSE_PER_SEC_NOTIFY _IOR(IOC_1588_BASE, 32, __u32)
> > +
> > +#define IOCTL_1588_TARG_TIME_CLR_NOTIFY _IO(IOC_1588_BASE, 33)
> > +
> > +#define IOCTL_1588_AUX_TIME_CLR_NOTIFY _IO(IOC_1588_BASE, 34)
> > +
> > +#define IOCTL_1588_AUX_TARG_TIME_CLR_NOTIFY _IO(IOC_1588_BASE, 35)
> > +
> > +#define IOCTL_1588_PULSE_PER_SEC_CLR_NOTIFY _IO(IOC_1588_BASE, 36)
> > +
> > +#define IOCTL_1588_PULSE_PER_SEC_TIME_GET _IOR(IOC_1588_BASE, 37, __u32)
> > +
> > +#define IOCTL_1588_PULSE_PER_SEC_TIME_SET _IOW(IOC_1588_BASE, 38, __u32)
> > +
> > +#define IOCTL_1588_PORT_VERSION_SET \
> > + _IOW(IOC_1588_BASE, 39, struct pch_versionioctl)
> > +
> > +#define IOCTL_1588_PORT_VERSION_GET \
> > + _IOWR(IOC_1588_BASE, 40, struct pch_versionioctl)
> > +
> > +#define IOCTL_1588_PORT_OPERATION_MODE_SET \
> > + _IOW(IOC_1588_BASE, 41, struct pch_opemode_ioctl)
> > +
> > +#define IOCTL_1588_PORT_OPERATION_MODE_GET \
> > + _IOWR(IOC_1588_BASE, 42, struct pch_opemode_ioctl)
>
> Do they all have to be ioctls?  What exactly are they doing?
>
> And are they 32/64bit safe?
>
> thanks,
>
> greg k-h
>


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/