From cec08e7a948326b01555be6311480aa08e637de2 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 30 Apr 2008 15:41:48 +1000 Subject: [POWERPC] vmemmap fixes to use smaller pages This changes vmemmap to use a different region (region 0xf) of the address space, and to configure the page size of that region dynamically at boot. The problem with the current approach of always using 16M pages is that it's not well suited to machines that have small amounts of memory such as small partitions on pseries, or PS3's. In fact, on the PS3, failure to allocate the 16M page backing vmmemmap tends to prevent hotplugging the HV's "additional" memory, thus limiting the available memory even more, from my experience down to something like 80M total, which makes it really not very useable. The logic used by my match to choose the vmemmap page size is: - If 16M pages are available and there's 1G or more RAM at boot, use that size. - Else if 64K pages are available, use that - Else use 4K pages I've tested on a POWER6 (16M pages) and on an iSeries POWER3 (4K pages) and it seems to work fine. Note that I intend to change the way we organize the kernel regions & SLBs so the actual region will change from 0xf back to something else at one point, as I simplify the SLB miss handler, but that will be for a later patch. Signed-off-by: Paul Mackerras --- arch/powerpc/mm/hash_utils_64.c | 28 ++++++++++++++++++++++++++-- 1 file changed, 26 insertions(+), 2 deletions(-) (limited to 'arch/powerpc/mm/hash_utils_64.c') diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index 2b5a399f6fa..0f2d239d94c 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -94,6 +94,9 @@ unsigned long htab_hash_mask; int mmu_linear_psize = MMU_PAGE_4K; int mmu_virtual_psize = MMU_PAGE_4K; int mmu_vmalloc_psize = MMU_PAGE_4K; +#ifdef CONFIG_SPARSEMEM_VMEMMAP +int mmu_vmemmap_psize = MMU_PAGE_4K; +#endif int mmu_io_psize = MMU_PAGE_4K; int mmu_kernel_ssize = MMU_SEGSIZE_256M; int mmu_highuser_ssize = MMU_SEGSIZE_256M; @@ -387,11 +390,32 @@ static void __init htab_init_page_sizes(void) } #endif /* CONFIG_PPC_64K_PAGES */ +#ifdef CONFIG_SPARSEMEM_VMEMMAP + /* We try to use 16M pages for vmemmap if that is supported + * and we have at least 1G of RAM at boot + */ + if (mmu_psize_defs[MMU_PAGE_16M].shift && + lmb_phys_mem_size() >= 0x40000000) + mmu_vmemmap_psize = MMU_PAGE_16M; + else if (mmu_psize_defs[MMU_PAGE_64K].shift) + mmu_vmemmap_psize = MMU_PAGE_64K; + else + mmu_vmemmap_psize = MMU_PAGE_4K; +#endif /* CONFIG_SPARSEMEM_VMEMMAP */ + printk(KERN_DEBUG "Page orders: linear mapping = %d, " - "virtual = %d, io = %d\n", + "virtual = %d, io = %d" +#ifdef CONFIG_SPARSEMEM_VMEMMAP + ", vmemmap = %d" +#endif + "\n", mmu_psize_defs[mmu_linear_psize].shift, mmu_psize_defs[mmu_virtual_psize].shift, - mmu_psize_defs[mmu_io_psize].shift); + mmu_psize_defs[mmu_io_psize].shift +#ifdef CONFIG_SPARSEMEM_VMEMMAP + ,mmu_psize_defs[mmu_vmemmap_psize].shift +#endif + ); #ifdef CONFIG_HUGETLB_PAGE /* Init large page size. Currently, we pick 16M or 1M depending -- cgit v1.2.3