On Aug 24, 2012, at 5:50 AM, Shaohui Xie wrote: > PowerPC platform only supports ZONE_DMA zone for 64bit kernel, so all the > memory will be put into this zone. If the memory size is greater than > the device's DMA capability and device uses dma_alloc_coherent to allocate > memory, it will get an address which is over the device's DMA addressing, > the device will fail. > > So we split the memory to two zones: zone ZONE_DMA32 & ZONE_NORMAL, since > we already allocate PCICSRBAR/PEXCSRBAR right below the 4G boundary (if the > lowest PCI address is above 4G), so we constrain the DMA zone ZONE_DMA32 > to 2GB, also, we clear flag __GFP_DMA & __GFP_DMA32 and set __GFP_DMA32 only > if the device's dma_mask < total memory size. By doing this, devices which > cannot DMA all the memory will be limited to ZONE_DMA32, but devices which > can DMA all the memory will not be affected by this limitation. > > Signed-off-by: Shaohui Xie <shaohui....@freescale.com> > Signed-off-by: Mingkai Hu <mingkai...@freescale.com> > Signed-off-by: Chen Yuanquan <b41...@freescale.com> > --- > changes for v2: > 1. use a config option for using two zones (ZONE_DMA32 & ZONE_NORMAL) in > freescale 64 bit kernel. > > arch/powerpc/Kconfig | 3 +++ > arch/powerpc/kernel/dma.c | 15 +++++++++++++++ > arch/powerpc/mm/mem.c | 4 ++++ > 3 files changed, 22 insertions(+), 0 deletions(-)
Ben, What's the feeling of doing this on ppc64 always? - k > > diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig > index 352f416..a96fbbb 100644 > --- a/arch/powerpc/Kconfig > +++ b/arch/powerpc/Kconfig > @@ -629,6 +629,9 @@ config ZONE_DMA > bool > default y > > +config ZONE_DMA32 > + def_bool (PPC64 && PPC_FSL_BOOK3E) > + > config NEED_DMA_MAP_STATE > def_bool (PPC64 || NOT_COHERENT_CACHE) > > diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c > index 355b9d8..cbf5ac1 100644 > --- a/arch/powerpc/kernel/dma.c > +++ b/arch/powerpc/kernel/dma.c > @@ -41,9 +41,24 @@ void *dma_direct_alloc_coherent(struct device *dev, size_t > size, > #else > struct page *page; > int node = dev_to_node(dev); > +#ifdef CONFIG_ZONE_DMA32 > + phys_addr_t top_ram_pfn = memblock_end_of_DRAM(); > > + /* > + * check for crappy device which has dma_mask < ZONE_DMA, and > + * we are not going to support it, just warn and fail. > + */ > + if (*dev->dma_mask < DMA_BIT_MASK(31)) { > + dev_err(dev, "Unsupported dma_mask 0x%llx\n", *dev->dma_mask); > + return NULL; > + } > /* ignore region specifiers */ > + flag &= ~(__GFP_HIGHMEM | __GFP_DMA | __GFP_DMA32); > + if (*dev->dma_mask < top_ram_pfn - 1) > + flag |= __GFP_DMA32; > +#else > flag &= ~(__GFP_HIGHMEM); > +#endif > > page = alloc_pages_node(node, flag, get_order(size)); > if (page == NULL) > diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c > index baaafde..2a11e49 100644 > --- a/arch/powerpc/mm/mem.c > +++ b/arch/powerpc/mm/mem.c > @@ -280,6 +280,10 @@ void __init paging_init(void) > #ifdef CONFIG_HIGHMEM > max_zone_pfns[ZONE_DMA] = lowmem_end_addr >> PAGE_SHIFT; > max_zone_pfns[ZONE_HIGHMEM] = top_of_ram >> PAGE_SHIFT; > +#elif defined CONFIG_ZONE_DMA32 > + max_zone_pfns[ZONE_DMA32] = min_t(phys_addr_t, top_of_ram, > + 1ull << 31) >> PAGE_SHIFT; > + max_zone_pfns[ZONE_NORMAL] = top_of_ram >> PAGE_SHIFT; > #else > max_zone_pfns[ZONE_DMA] = top_of_ram >> PAGE_SHIFT; > #endif > -- > 1.6.4 > > > _______________________________________________ > Linuxppc-dev mailing list > Linuxppc-dev@lists.ozlabs.org > https://lists.ozlabs.org/listinfo/linuxppc-dev _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev