On Mon, Dec 8, 2014 at 7:24 PM, Arnd Bergmann <a...@arndb.de> wrote: > On Monday 08 December 2014 17:39:27 Alexandre Courbot wrote: >> There doesn't seem to be any valid reason to allocate the pages array >> with the same flags as the buffer itself. Doing so can eventually lead >> to the following safeguard in mm/slab.c to be hit: >> >> BUG_ON(flags & GFP_SLAB_BUG_MASK); >> >> This happens when buffers are allocated with __GFP_DMA32 or >> __GFP_HIGHMEM. >> >> Fix this by allocating the pages array with GFP_KERNEL to follow what is >> done elsewhere in this file. Using GFP_KERNEL in __iommu_alloc_buffer() >> is safe because atomic allocations are handled by __iommu_alloc_atomic(). >> > > I think you need to carry over the GFP_ATOMIC flag if that is set by the > caller, but not the GFP_HIGHMEM or GFP_DMA32. Not sure if it's better > to mask out flags from the caller mask, or to start with GFP_KERNEL > and adding in extra bits.
I thought the issue of atomicity is already handled by __iommu_alloc_buffer's caller (arm_iommu_alloc_attrs): if (!(gfp & __GFP_WAIT)) return __iommu_alloc_atomic(dev, size, handle); .... pages = __iommu_alloc_buffer(dev, size, gfp, attrs); Isn't the interesting property about GFP_ATOMIC that it does not include __GFP_WAIT? I may very well misunderstand the issue, sorry if that's the case. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/