From: FUJITA Tomonori This patch converts PPC's IOMMU to use the IOMMU helper functions. The IOMMU doesn't allocate a memory area spanning LLD's segment boundary anymore. iseries_hv_alloc and iseries_hv_map don't have proper device struct. 4GB boundary is used for them. Signed-off-by: FUJITA Tomonori Cc: Jeff Garzik Cc: James Bottomley Cc: Jens Axboe Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Signed-off-by: Andrew Morton --- arch/powerpc/Kconfig | 3 + arch/powerpc/kernel/dma_64.c | 6 +- arch/powerpc/kernel/iommu.c | 65 +++++++++++------------ arch/powerpc/platforms/iseries/iommu.c | 4 - include/asm-powerpc/iommu.h | 10 +-- 5 files changed, 45 insertions(+), 43 deletions(-) diff -puN arch/powerpc/Kconfig~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper arch/powerpc/Kconfig --- a/arch/powerpc/Kconfig~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper +++ a/arch/powerpc/Kconfig @@ -242,6 +242,9 @@ config IOMMU_VMERGE Most drivers don't have this problem; it is safe to say Y here. +config IOMMU_HELPER + def_bool PPC64 + config HOTPLUG_CPU bool "Support for enabling/disabling CPUs" depends on SMP && HOTPLUG && EXPERIMENTAL && (PPC_PSERIES || PPC_PMAC) diff -puN arch/powerpc/kernel/dma_64.c~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper arch/powerpc/kernel/dma_64.c --- a/arch/powerpc/kernel/dma_64.c~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper +++ a/arch/powerpc/kernel/dma_64.c @@ -31,8 +31,8 @@ static inline unsigned long device_to_ma static void *dma_iommu_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag) { - return iommu_alloc_coherent(dev->archdata.dma_data, size, dma_handle, - device_to_mask(dev), flag, + return iommu_alloc_coherent(dev, dev->archdata.dma_data, size, + dma_handle, device_to_mask(dev), flag, dev->archdata.numa_node); } @@ -52,7 +52,7 @@ static dma_addr_t dma_iommu_map_single(s size_t size, enum dma_data_direction direction) { - return iommu_map_single(dev->archdata.dma_data, vaddr, size, + return iommu_map_single(dev, dev->archdata.dma_data, vaddr, size, device_to_mask(dev), direction); } diff -puN arch/powerpc/kernel/iommu.c~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper arch/powerpc/kernel/iommu.c --- a/arch/powerpc/kernel/iommu.c~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper +++ a/arch/powerpc/kernel/iommu.c @@ -31,6 +31,7 @@ #include #include #include +#include #include #include #include @@ -81,17 +82,19 @@ static int __init setup_iommu(char *str) __setup("protect4gb=", setup_protect4gb); __setup("iommu=", setup_iommu); -static unsigned long iommu_range_alloc(struct iommu_table *tbl, +static unsigned long iommu_range_alloc(struct device *dev, + struct iommu_table *tbl, unsigned long npages, unsigned long *handle, unsigned long mask, unsigned int align_order) { - unsigned long n, end, i, start; + unsigned long n, end, start; unsigned long limit; int largealloc = npages > 15; int pass = 0; unsigned long align_mask; + unsigned long boundary_size; align_mask = 0xffffffffffffffffl >> (64 - align_order); @@ -136,14 +139,17 @@ static unsigned long iommu_range_alloc(s start &= mask; } - n = find_next_zero_bit(tbl->it_map, limit, start); - - /* Align allocation */ - n = (n + align_mask) & ~align_mask; - - end = n + npages; + if (dev) + boundary_size = ALIGN(dma_get_seg_boundary(dev) + 1, + 1 << IOMMU_PAGE_SHIFT); + else + boundary_size = ALIGN(1UL << 32, 1 << IOMMU_PAGE_SHIFT); + /* 4GB boundary for iseries_hv_alloc and iseries_hv_map */ - if (unlikely(end >= limit)) { + n = iommu_area_alloc(tbl->it_map, limit, start, npages, + tbl->it_offset, boundary_size >> IOMMU_PAGE_SHIFT, + align_mask); + if (n == -1) { if (likely(pass < 2)) { /* First failure, just rescan the half of the table. * Second failure, rescan the other half of the table. @@ -158,14 +164,7 @@ static unsigned long iommu_range_alloc(s } } - for (i = n; i < end; i++) - if (test_bit(i, tbl->it_map)) { - start = i+1; - goto again; - } - - for (i = n; i < end; i++) - __set_bit(i, tbl->it_map); + end = n + npages; /* Bump the hint to a new block for small allocs. */ if (largealloc) { @@ -184,16 +183,17 @@ static unsigned long iommu_range_alloc(s return n; } -static dma_addr_t iommu_alloc(struct iommu_table *tbl, void *page, - unsigned int npages, enum dma_data_direction direction, - unsigned long mask, unsigned int align_order) +static dma_addr_t iommu_alloc(struct device *dev, struct iommu_table *tbl, + void *page, unsigned int npages, + enum dma_data_direction direction, + unsigned long mask, unsigned int align_order) { unsigned long entry, flags; dma_addr_t ret = DMA_ERROR_CODE; spin_lock_irqsave(&(tbl->it_lock), flags); - entry = iommu_range_alloc(tbl, npages, NULL, mask, align_order); + entry = iommu_range_alloc(dev, tbl, npages, NULL, mask, align_order); if (unlikely(entry == DMA_ERROR_CODE)) { spin_unlock_irqrestore(&(tbl->it_lock), flags); @@ -224,7 +224,6 @@ static void __iommu_free(struct iommu_ta unsigned int npages) { unsigned long entry, free_entry; - unsigned long i; entry = dma_addr >> IOMMU_PAGE_SHIFT; free_entry = entry - tbl->it_offset; @@ -246,9 +245,7 @@ static void __iommu_free(struct iommu_ta } ppc_md.tce_free(tbl, entry, npages); - - for (i = 0; i < npages; i++) - __clear_bit(free_entry+i, tbl->it_map); + iommu_area_free(tbl->it_map, free_entry, npages); } static void iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr, @@ -312,7 +309,8 @@ int iommu_map_sg(struct device *dev, str /* Allocate iommu entries for that segment */ vaddr = (unsigned long) sg_virt(s); npages = iommu_num_pages(vaddr, slen); - entry = iommu_range_alloc(tbl, npages, &handle, mask >> IOMMU_PAGE_SHIFT, 0); + entry = iommu_range_alloc(dev, tbl, npages, &handle, + mask >> IOMMU_PAGE_SHIFT, 0); DBG(" - vaddr: %lx, size: %lx\n", vaddr, slen); @@ -568,9 +566,9 @@ void iommu_free_table(struct iommu_table * need not be page aligned, the dma_addr_t returned will point to the same * byte within the page as vaddr. */ -dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr, - size_t size, unsigned long mask, - enum dma_data_direction direction) +dma_addr_t iommu_map_single(struct device *dev, struct iommu_table *tbl, + void *vaddr, size_t size, unsigned long mask, + enum dma_data_direction direction) { dma_addr_t dma_handle = DMA_ERROR_CODE; unsigned long uaddr; @@ -582,7 +580,7 @@ dma_addr_t iommu_map_single(struct iommu npages = iommu_num_pages(uaddr, size); if (tbl) { - dma_handle = iommu_alloc(tbl, vaddr, npages, direction, + dma_handle = iommu_alloc(dev, tbl, vaddr, npages, direction, mask >> IOMMU_PAGE_SHIFT, 0); if (dma_handle == DMA_ERROR_CODE) { if (printk_ratelimit()) { @@ -614,8 +612,9 @@ void iommu_unmap_single(struct iommu_tab * Returns the virtual address of the buffer and sets dma_handle * to the dma address (mapping) of the first page. */ -void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size, - dma_addr_t *dma_handle, unsigned long mask, gfp_t flag, int node) +void *iommu_alloc_coherent(struct device *dev, struct iommu_table *tbl, + size_t size, dma_addr_t *dma_handle, + unsigned long mask, gfp_t flag, int node) { void *ret = NULL; dma_addr_t mapping; @@ -649,7 +648,7 @@ void *iommu_alloc_coherent(struct iommu_ /* Set up tces to cover the allocated range */ nio_pages = size >> IOMMU_PAGE_SHIFT; io_order = get_iommu_order(size); - mapping = iommu_alloc(tbl, ret, nio_pages, DMA_BIDIRECTIONAL, + mapping = iommu_alloc(dev, tbl, ret, nio_pages, DMA_BIDIRECTIONAL, mask >> IOMMU_PAGE_SHIFT, io_order); if (mapping == DMA_ERROR_CODE) { free_pages((unsigned long)ret, order); diff -puN arch/powerpc/platforms/iseries/iommu.c~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper arch/powerpc/platforms/iseries/iommu.c --- a/arch/powerpc/platforms/iseries/iommu.c~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper +++ a/arch/powerpc/platforms/iseries/iommu.c @@ -199,7 +199,7 @@ static struct iommu_table vio_iommu_tabl void *iseries_hv_alloc(size_t size, dma_addr_t *dma_handle, gfp_t flag) { - return iommu_alloc_coherent(&vio_iommu_table, size, dma_handle, + return iommu_alloc_coherent(NULL, &vio_iommu_table, size, dma_handle, DMA_32BIT_MASK, flag, -1); } EXPORT_SYMBOL_GPL(iseries_hv_alloc); @@ -213,7 +213,7 @@ EXPORT_SYMBOL_GPL(iseries_hv_free); dma_addr_t iseries_hv_map(void *vaddr, size_t size, enum dma_data_direction direction) { - return iommu_map_single(&vio_iommu_table, vaddr, size, + return iommu_map_single(NULL, &vio_iommu_table, vaddr, size, DMA_32BIT_MASK, direction); } diff -puN include/asm-powerpc/iommu.h~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper include/asm-powerpc/iommu.h --- a/include/asm-powerpc/iommu.h~iommu-sg-powerpc-convert-iommu-to-use-the-iommu-helper +++ a/include/asm-powerpc/iommu.h @@ -85,13 +85,13 @@ extern int iommu_map_sg(struct device *d extern void iommu_unmap_sg(struct iommu_table *tbl, struct scatterlist *sglist, int nelems, enum dma_data_direction direction); -extern void *iommu_alloc_coherent(struct iommu_table *tbl, size_t size, - dma_addr_t *dma_handle, unsigned long mask, - gfp_t flag, int node); +extern void *iommu_alloc_coherent(struct device *dev, struct iommu_table *tbl, + size_t size, dma_addr_t *dma_handle, + unsigned long mask, gfp_t flag, int node); extern void iommu_free_coherent(struct iommu_table *tbl, size_t size, void *vaddr, dma_addr_t dma_handle); -extern dma_addr_t iommu_map_single(struct iommu_table *tbl, void *vaddr, - size_t size, unsigned long mask, +extern dma_addr_t iommu_map_single(struct device *dev, struct iommu_table *tbl, + void *vaddr, size_t size, unsigned long mask, enum dma_data_direction direction); extern void iommu_unmap_single(struct iommu_table *tbl, dma_addr_t dma_handle, size_t size, enum dma_data_direction direction); _