From mboxrd@z Thu Jan 1 00:00:00 1970 From: mel@skynet.ie (Mel Gorman) Date: Thu, 16 Nov 2006 16:05:41 +0000 Subject: Re: 05e0caad3b7bd0d0fbeff980bca22f186241a501 breaks ia64 kdump Message-Id: <20061116160540.GA6747@skynet.ie> List-Id: References: <20061026075951.GA30910@verge.net.au> In-Reply-To: <20061026075951.GA30910@verge.net.au> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-ia64@vger.kernel.org > > As you say, it's not clear why the normal discontig kernel boots because > > the regions should have been skipped by add_active_range(). > > > I think that depends on the init value of memmap, if they are all zero, > free_pages_check will be happy and not report any thing. So I guess we may > see this bug in normal kernel with a warm reboot, or with a machine which > PROM does not clear memory to all zero. It could have been just luck. See more below > > > Try your patch and see does it work for kdump. It should work fine in the > > normal case because at very worst, slightly more memmap is allocated than > > is strictly required. > > > It works for kdump. > Can you try out the patch below please as a possibly simplier alternative? It's a patch from Bob Picco that addresses a similar problem with initmem on a kernel booting as normal. The only difference between Bob's original patch and this one is the addition of the CONFIG_KEXEC stuff from your patch. It survived a simply boot test. Thanks >>> Begin Bob's patch While pursuing and unrelated issue with 64Mb granules I noticed a problem related to inconsistent use of add_active_range. There doesn't appear any reason to me why FLATMEM versus DISCONTIG_MEM should register memory to add_active_range with different code. So I've changed the code into a common implementation. The other subtle issue fixed by this patch was calling add_active_range in count_node_pages before granule aligning is performed. We were lucky with 16MB granules but not so with 64MB granules. count_node_pages has reserved regions filtered out and as a consequence linked kernel text and data aren't covered by calls to count_node_pages. So linked kernel regions wasn't reported to add_active_regions. This resulted in free_initmem causing numerous bad_page reports. This won't occur with this patch because now all known memory regions are reported by register_active_ranges. Acked-by: Mel Gorman Signed-off-by: Bob Picco arch/ia64/mm/contig.c | 3 +-- arch/ia64/mm/discontig.c | 4 +++- arch/ia64/mm/init.c | 24 +++++++++++++++++------- include/asm-ia64/meminit.h | 3 ++- 4 files changed, 23 insertions(+), 11 deletions(-) diff -rup -X /usr/src/patchset-0.6/bin//dontdiff linux-2.6.19-rc5-clean/arch/ia64/mm/contig.c linux-2.6.19-rc5-register_all_memory/arch/ia64/mm/contig.c --- linux-2.6.19-rc5-clean/arch/ia64/mm/contig.c 2006-11-08 02:24:20.000000000 +0000 +++ linux-2.6.19-rc5-register_all_memory/arch/ia64/mm/contig.c 2006-11-16 15:29:31.000000000 +0000 @@ -226,7 +226,6 @@ void __init paging_init (void) { unsigned long max_dma; - unsigned long nid = 0; unsigned long max_zone_pfns[MAX_NR_ZONES]; num_physpages = 0; @@ -238,7 +237,7 @@ paging_init (void) max_zone_pfns[ZONE_NORMAL] = max_low_pfn; #ifdef CONFIG_VIRTUAL_MEM_MAP - efi_memmap_walk(register_active_ranges, &nid); + efi_memmap_walk(register_active_ranges, NULL); efi_memmap_walk(find_largest_hole, (u64 *)&max_gap); if (max_gap < LARGE_GAP) { vmem_map = (struct page *) 0; diff -rup -X /usr/src/patchset-0.6/bin//dontdiff linux-2.6.19-rc5-clean/arch/ia64/mm/discontig.c linux-2.6.19-rc5-register_all_memory/arch/ia64/mm/discontig.c --- linux-2.6.19-rc5-clean/arch/ia64/mm/discontig.c 2006-11-08 02:24:20.000000000 +0000 +++ linux-2.6.19-rc5-register_all_memory/arch/ia64/mm/discontig.c 2006-11-16 15:29:31.000000000 +0000 @@ -473,6 +473,9 @@ void __init find_memory(void) node_clear(node, memory_less_mask); mem_data[node].min_pfn = ~0UL; } + + efi_memmap_walk(register_active_ranges, NULL); + /* * Initialize the boot memory maps in reverse order since that's * what the bootmem allocator expects @@ -654,7 +657,6 @@ static __init int count_node_pages(unsig { unsigned long end = start + len; - add_active_range(node, start >> PAGE_SHIFT, end >> PAGE_SHIFT); mem_data[node].num_physpages += len >> PAGE_SHIFT; if (start <= __pa(MAX_DMA_ADDRESS)) mem_data[node].num_dma_physpages +diff -rup -X /usr/src/patchset-0.6/bin//dontdiff linux-2.6.19-rc5-clean/arch/ia64/mm/init.c linux-2.6.19-rc5-register_all_memory/arch/ia64/mm/init.c --- linux-2.6.19-rc5-clean/arch/ia64/mm/init.c 2006-11-08 02:24:20.000000000 +0000 +++ linux-2.6.19-rc5-register_all_memory/arch/ia64/mm/init.c 2006-11-16 15:33:25.000000000 +0000 @@ -594,18 +594,28 @@ find_largest_hole (u64 start, u64 end, v return 0; } +#endif /* CONFIG_VIRTUAL_MEM_MAP */ + int __init -register_active_ranges(u64 start, u64 end, void *nid) +register_active_ranges(u64 start, u64 end, void *arg) { - BUG_ON(nid = NULL); - BUG_ON(*(unsigned long *)nid >= MAX_NUMNODES); + int nid = paddr_to_nid(__pa(start)); + + if (nid < 0) + nid = 0; - add_active_range(*(unsigned long *)nid, - __pa(start) >> PAGE_SHIFT, - __pa(end) >> PAGE_SHIFT); +#ifdef CONFIG_KEXEC + if (start > crashk_res.start && start < crashk_res.end) + start = max(start, crashk_res.end); + if (end > crashk_res.start && end < crashk_res.end) + end = min(end, crashk_res.start); +#endif + + if (start < end) + add_active_range(nid, __pa(start) >> PAGE_SHIFT, + __pa(end) >> PAGE_SHIFT); return 0; } -#endif /* CONFIG_VIRTUAL_MEM_MAP */ static int __init count_reserved_pages (u64 start, u64 end, void *arg) diff -rup -X /usr/src/patchset-0.6/bin//dontdiff linux-2.6.19-rc5-clean/include/asm-ia64/meminit.h linux-2.6.19-rc5-register_all_memory/include/asm-ia64/meminit.h --- linux-2.6.19-rc5-clean/include/asm-ia64/meminit.h 2006-11-08 02:24:20.000000000 +0000 +++ linux-2.6.19-rc5-register_all_memory/include/asm-ia64/meminit.h 2006-11-16 15:29:31.000000000 +0000 @@ -50,12 +50,13 @@ extern void efi_memmap_init(unsigned lon #define IGNORE_PFN0 1 /* XXX fix me: ignore pfn 0 until TLB miss handler is updated... */ +extern int register_active_ranges (u64 start, u64 end, void *arg); + #ifdef CONFIG_VIRTUAL_MEM_MAP # define LARGE_GAP 0x40000000 /* Use virtual mem map if hole is > than this */ extern unsigned long vmalloc_end; extern struct page *vmem_map; extern int find_largest_hole (u64 start, u64 end, void *arg); - extern int register_active_ranges (u64 start, u64 end, void *arg); extern int create_mem_map_page_table (u64 start, u64 end, void *arg); extern int vmemmap_find_next_valid_pfn(int, int); #else -- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab