Re: [PATCH 10/10] sparc64: use iommu_num_pages function in IOMMU code

!MAILaRCHIVE_VOTE_RePLACE
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Joerg Roedel <joerg.roedel@...>
Cc: David Miller <davem@...>, <akpm@...>, <mingo@...>, <tglx@...>, <hpa@...>, <rth@...>, <paulus@...>, <fujita.tomonori@...>, <muli@...>, <linux-kernel@...>
Date: Tuesday, August 5, 2008 - 4:23 am

> I didn't want to change multiple architectures with a single patch so I

I wouldn't say there's a strict rule. It's a balance to find between
the complexity of the patch, the risk of breaking archs, etc...

In your case, the added churn of going through a temporary renamed
helper represents more, imho, than the risk of breakage of what is
a fairly trivial patch in the first place. Also, it only affects a few
archs. So it's simpler or everybody to deal with a single patch even if
that means it -might- need a rev or two in case it does indeed break
something...

Ben.


--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[PATCH 0/10] introduce generic iommu_num_pages function, Joerg Roedel, (Mon Aug 4, 12:04 pm)
Re: [PATCH 0/10] introduce generic iommu_num_pages function, Benjamin Herrenschmidt, (Mon Aug 4, 9:08 pm)
Re: [PATCH 10/10] sparc64: use iommu_num_pages function in I..., Benjamin Herrenschmidt, (Tue Aug 5, 4:23 am)
Re: [PATCH 10/10] sparc64: use iommu_num_pages function in I..., Benjamin Herrenschmidt, (Mon Aug 4, 9:09 pm)