RE: [PATCH 1/3] OMAP: DSS2: Zorder enum in display.h

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> -----Original Message-----
> From: Taneja, Archit 
> Sent: Thursday, July 22, 2010 9:42 AM
> To: Premi, Sanjeev; tomi.valkeinen@xxxxxxxxx
> Cc: linux-omap@xxxxxxxxxxxxxxx; Semwal, Sumit; Nilofer, Samreen
> Subject: RE: [PATCH 1/3] OMAP: DSS2: Zorder enum in display.h
> 
>  
> 
> > -----Original Message-----
> > From: Premi, Sanjeev 
> > Sent: Wednesday, July 21, 2010 7:47 PM
> > To: Taneja, Archit; tomi.valkeinen@xxxxxxxxx
> > Cc: linux-omap@xxxxxxxxxxxxxxx; Semwal, Sumit; Nilofer, 
> > Samreen; Taneja, Archit
> > Subject: RE: [PATCH 1/3] OMAP: DSS2: Zorder enum in display.h
> > 
> > > -----Original Message-----
> > > From: linux-omap-owner@xxxxxxxxxxxxxxx 
> > > [mailto:linux-omap-owner@xxxxxxxxxxxxxxx] On Behalf Of 
> Archit Taneja
> > > Sent: Monday, July 19, 2010 5:40 PM
> > > To: tomi.valkeinen@xxxxxxxxx
> > > Cc: linux-omap@xxxxxxxxxxxxxxx; Semwal, Sumit; Nilofer, Samreen; 
> > > Taneja, Archit
> > > Subject: [PATCH 1/3] OMAP: DSS2: Zorder enum in display.h
> > > 
> > > From: Sumit Semwal <sumit.semwal@xxxxxx>
> > > 
> > > Add Zorder enum in display.h
> > > 
> > 
> > Patches 1 and 2 in the series can easily be conbined into one.
> > Separating few line changes in header file from 
> > implementation across 2 patches isn't useful.
> 
> [archit] We are introducing a new DSS feature for OMAP4 in every
> patch series. In order to clearly explain the feature introduced,
> the first patch of every series makes changes on in the display.h
> header which is central to the DSS2 code.
> 
> If this is not a accepted norm or a strong enough reason to have a
> separate small patch, I can rework these series, I would need more
> comments from others though.

[sp] I understand the festure intoduction, but spliting patches across
     headers and implementation doesn't seem to be logical.

~sanjeev

> 
> <snap>
> 
> > > +enum omap_overlay_zorder {
> > > +	OMAP_DSS_OVL_ZORDER_0	= 0x0,
> > > +	OMAP_DSS_OVL_ZORDER_1	= 0x1,
> > > +	OMAP_DSS_OVL_ZORDER_2	= 0x2,
> > > +	OMAP_DSS_OVL_ZORDER_3	= 0x3,
> > > +};
> > 
> > Is "_DSS_" really needed in these emums? considering that 
> > enum itself doesn't contain "_dss_" in its name.
> > 
> 
> [archit] I agree with this, but the present header is inconsistent
> with the point you have made, there are other enums which don't have
> "_dss_" but have "_DSS_" in its enum members. We should try to make
> this uniform (unless there is a motive behind it).
> 
> > ~sanjeev
> > 
> 
> Regards,
> 
> Archit
> --
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [Linux Arm (vger)]     [ARM Kernel]     [ARM MSM]     [Linux Tegra]     [Linux WPAN Networking]     [Linux Wireless Networking]     [Maemo Users]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux