Message-ID: <20090512111623.GG25923@csn.ul.ie>
Date:	Tue, 12 May 2009 12:16:24 +0100
From:	Mel Gorman <mel@....ul.ie>
To:	Yinghai Lu <yinghai@...nel.org>
Cc:	Ingo Molnar <mingo@...e.hu>, Thomas Gleixner <tglx@...utronix.de>,
	"H. Peter Anvin" <hpa@...or.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Suresh Siddha <suresh.b.siddha@...el.com>,
	Christoph Lameter <cl@...ux-foundation.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Al Viro <viro@...iv.linux.org.uk>,
	Rusty Russell <rusty@...tcorp.com.au>
Subject: Re: [PATCH 1/3] x86: remove MEMORY_HOTPLUG_RESERVE related code

On Fri, May 08, 2009 at 11:45:49PM -0700, Yinghai Lu wrote:
> 
> after
> | commit b263295dbffd33b0fbff670720fa178c30e3392a
> | Author: Christoph Lameter <clameter@....com>
> | Date:   Wed Jan 30 13:30:47 2008 +0100
> |
> |    x86: 64-bit, make sparsemem vmemmap the only memory model
> 
> we don't have MEMORY_HOTPLUG_RESERVE anymore.
> 
> remove related dead code.
> 

Good spot, this removes a nice amount of code. The changelog could say
more though, how about?

=====
Historically, x86-64 had an architecture-specific method for memory hotplug
whereby it scanned the SRAT for physical memory ranges that could be
potentially used for memory hot-add later. By reserving those ranges
without physical memory, the memmap would be allocated and left dormant
until needed. This depended on the DISCONTIG memory model which has been
removed so the code implementing HOTPLUG_RESERVE is now dead.

This patch removes the dead code used by MEMORY_HOTPLUG_RESERVE
=====

> Signed-off-by: Yinghai Lu <yinghai@...nel.org>
> 
> ---
>  arch/x86/include/asm/numa_64.h |    3 -
>  arch/x86/mm/numa_64.c          |    5 --
>  arch/x86/mm/srat_64.c          |   63 +++++++------------------------------
>  include/linux/mm.h             |    2 -
>  mm/page_alloc.c                |   69 -----------------------------------------
>  5 files changed, 12 insertions(+), 130 deletions(-)
> 
> Index: linux-2.6/arch/x86/include/asm/numa_64.h
> ===================================================================
> --- linux-2.6.orig/arch/x86/include/asm/numa_64.h
> +++ linux-2.6/arch/x86/include/asm/numa_64.h
> @@ -17,9 +17,6 @@ extern int compute_hash_shift(struct boo
>  extern void numa_init_array(void);
>  extern int numa_off;
>  
> -extern void srat_reserve_add_area(int nodeid);
> -extern int hotadd_percent;
> -
>  extern s16 apicid_to_node[MAX_LOCAL_APIC];
>  
>  extern unsigned long numa_free_all_bootmem(void);
> Index: linux-2.6/arch/x86/mm/numa_64.c
> ===================================================================
> --- linux-2.6.orig/arch/x86/mm/numa_64.c
> +++ linux-2.6/arch/x86/mm/numa_64.c
> @@ -272,9 +272,6 @@ void __init setup_node_bootmem(int nodei
>  		reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start,
>  				 bootmap_pages<<PAGE_SHIFT, BOOTMEM_DEFAULT);
>  
> -#ifdef CONFIG_ACPI_NUMA
> -	srat_reserve_add_area(nodeid);
> -#endif
>  	node_set_online(nodeid);
>  }
>  
> @@ -608,8 +605,6 @@ static __init int numa_setup(char *opt)
>  #ifdef CONFIG_ACPI_NUMA
>  	if (!strncmp(opt, "noacpi", 6))
>  		acpi_numa = -1;
> -	if (!strncmp(opt, "hotadd=", 7))
> -		hotadd_percent = simple_strtoul(opt+7, NULL, 10);

Documentation/x86/x86_64/boot-options.txt now needs to be updated to
remove the documentation on hotadd=.

Instead of ignoring the option, should a warning now be printed saying the
option is deprecated?

>  #endif
>  	return 0;
>  }
> Index: linux-2.6/arch/x86/mm/srat_64.c
> ===================================================================
> --- linux-2.6.orig/arch/x86/mm/srat_64.c
> +++ linux-2.6/arch/x86/mm/srat_64.c
> @@ -31,8 +31,6 @@ static nodemask_t nodes_parsed __initdat
>  static nodemask_t cpu_nodes_parsed __initdata;
>  static struct bootnode nodes[MAX_NUMNODES] __initdata;
>  static struct bootnode nodes_add[MAX_NUMNODES];
> -static int found_add_area __initdata;
> -int hotadd_percent __initdata = 0;
>  
>  static int num_node_memblks __initdata;
>  static struct bootnode node_memblk_range[NR_NODE_MEMBLKS] __initdata;
> @@ -66,9 +64,6 @@ static __init void cutoff_node(int i, un
>  {
>  	struct bootnode *nd = &nodes[i];
>  
> -	if (found_add_area)
> -		return;
> -
>  	if (nd->start < start) {
>  		nd->start = start;
>  		if (nd->end < nd->start)
> @@ -86,7 +81,6 @@ static __init void bad_srat(void)
>  	int i;
>  	printk(KERN_ERR "SRAT: SRAT not used.\n");
>  	acpi_numa = -1;
> -	found_add_area = 0;
>  	for (i = 0; i < MAX_LOCAL_APIC; i++)
>  		apicid_to_node[i] = NUMA_NO_NODE;
>  	for (i = 0; i < MAX_NUMNODES; i++)
> @@ -182,24 +176,21 @@ acpi_numa_processor_affinity_init(struct
>  	       pxm, apic_id, node);
>  }
>  
> -static int update_end_of_memory(unsigned long end) {return -1;}
> -static int hotadd_enough_memory(struct bootnode *nd) {return 1;}
>  #ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
>  static inline int save_add_info(void) {return 1;}
>  #else
>  static inline int save_add_info(void) {return 0;}
>  #endif
>  /*
> - * Update nodes_add and decide if to include add are in the zone.
> - * Both SPARSE and RESERVE need nodes_add information.
> - * This code supports one contiguous hot add area per node.
> + * Update nodes_add[]
> + * This code supports one contiguous hot add area per node
>   */
> -static int __init
> -reserve_hotadd(int node, unsigned long start, unsigned long end)
> +static void __init
> +update_nodes_add(int node, unsigned long start, unsigned long end)
>  {

It's now very unclear what the purpose of this function is. I'm guessing it
should be something like

validate_hotadd_region()
This validates that the region of memory described by SRAT as suitable
for use with memory hot-add is sane

What it was for was to validate that the SRAT looked sane and then push out the
end of the node boundaries so that the memmap would get allocated. However,
because we are no longer pushing out the node boundaries, is this doing
anything useful at all any more? For sparsemem, memory-hotadd allocates
the memmap as it required.

>  	unsigned long s_pfn = start >> PAGE_SHIFT;
>  	unsigned long e_pfn = end >> PAGE_SHIFT;
> -	int ret = 0, changed = 0;
> +	int changed = 0;
>  	struct bootnode *nd = &nodes_add[node];
>  
>  	/* I had some trouble with strange memory hotadd regions breaking
> @@ -210,7 +201,7 @@ reserve_hotadd(int node, unsigned long s
>  	   mistakes */
>  	if ((signed long)(end - start) < NODE_MIN_SIZE) {
>  		printk(KERN_ERR "SRAT: Hotplug area too small\n");
> -		return -1;
> +		return;
>  	}
>  
>  	/* This check might be a bit too strict, but I'm keeping it for now. */
> @@ -218,12 +209,7 @@ reserve_hotadd(int node, unsigned long s
>  		printk(KERN_ERR
>  			"SRAT: Hotplug area %lu -> %lu has existing memory\n",
>  			s_pfn, e_pfn);
> -		return -1;
> -	}
> -
> -	if (!hotadd_enough_memory(&nodes_add[node]))  {
> -		printk(KERN_ERR "SRAT: Hotplug area too large\n");
> -		return -1;
> +		return;
>  	}
>  
>  	/* Looks good */
> @@ -245,11 +231,9 @@ reserve_hotadd(int node, unsigned long s
>  			printk(KERN_ERR "SRAT: Hotplug zone not continuous. Partly ignored\n");
>  	}
>  
> -	ret = update_end_of_memory(nd->end);
> -
>  	if (changed)
> -	 	printk(KERN_INFO "SRAT: hot plug zone found %Lx - %Lx\n", nd->start, nd->end);
> -	return ret;
> +		printk(KERN_INFO "SRAT: hot plug zone found %Lx - %Lx\n",
> +				 nd->start, nd->end);
>  }
>  
>  /* Callback for parsing of the Proximity Domain <-> Memory Area mappings */
> @@ -310,13 +294,10 @@ acpi_numa_memory_affinity_init(struct ac
>  	       start, end);
>  	e820_register_active_regions(node, start >> PAGE_SHIFT,
>  				     end >> PAGE_SHIFT);
> -	push_node_boundaries(node, nd->start >> PAGE_SHIFT,
> -						nd->end >> PAGE_SHIFT);
>  
> -	if ((ma->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) &&
> -	    (reserve_hotadd(node, start, end) < 0)) {
> -		/* Ignore hotadd region. Undo damage */
> -		printk(KERN_NOTICE "SRAT: Hotplug region ignored\n");
> +	if (ma->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) {
> +		update_nodes_add(node, start, end);
> +		/* restore nodes[node] */
>  		*nd = oldnode;
>  		if ((nd->start | nd->end) == 0)
>  			node_clear(node, nodes_parsed);
> @@ -510,26 +491,6 @@ static int null_slit_node_compare(int a,
>  }
>  #endif /* CONFIG_NUMA_EMU */
>  
> -void __init srat_reserve_add_area(int nodeid)
> -{
> -	if (found_add_area && nodes_add[nodeid].end) {
> -		u64 total_mb;
> -
> -		printk(KERN_INFO "SRAT: Reserving hot-add memory space "
> -				"for node %d at %Lx-%Lx\n",
> -			nodeid, nodes_add[nodeid].start, nodes_add[nodeid].end);
> -		total_mb = (nodes_add[nodeid].end - nodes_add[nodeid].start)
> -					>> PAGE_SHIFT;
> -		total_mb *= sizeof(struct page);
> -		total_mb >>= 20;
> -		printk(KERN_INFO "SRAT: This will cost you %Lu MB of "
> -				"pre-allocated memory.\n", (unsigned long long)total_mb);
> -		reserve_bootmem_node(NODE_DATA(nodeid), nodes_add[nodeid].start,
> -			       nodes_add[nodeid].end - nodes_add[nodeid].start,
> -			       BOOTMEM_DEFAULT);
> -	}
> -}
> -
>  int __node_distance(int a, int b)
>  {
>  	int index;
> Index: linux-2.6/include/linux/mm.h
> ===================================================================
> --- linux-2.6.orig/include/linux/mm.h
> +++ linux-2.6/include/linux/mm.h
> @@ -1032,8 +1032,6 @@ extern void add_active_range(unsigned in
>  					unsigned long end_pfn);
>  extern void remove_active_range(unsigned int nid, unsigned long start_pfn,
>  					unsigned long end_pfn);
> -extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn,
> -					unsigned long end_pfn);
>  extern void remove_all_active_ranges(void);
>  extern unsigned long absent_pages_in_range(unsigned long start_pfn,
>  						unsigned long end_pfn);
> Index: linux-2.6/mm/page_alloc.c
> ===================================================================
> --- linux-2.6.orig/mm/page_alloc.c
> +++ linux-2.6/mm/page_alloc.c
> @@ -150,10 +150,6 @@ static unsigned long __meminitdata dma_r
>    static int __meminitdata nr_nodemap_entries;
>    static unsigned long __meminitdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES];
>    static unsigned long __meminitdata arch_zone_highest_possible_pfn[MAX_NR_ZONES];
> -#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
> -  static unsigned long __meminitdata node_boundary_start_pfn[MAX_NUMNODES];
> -  static unsigned long __meminitdata node_boundary_end_pfn[MAX_NUMNODES];
> -#endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */
>    static unsigned long __initdata required_kernelcore;
>    static unsigned long __initdata required_movablecore;
>    static unsigned long __meminitdata zone_movable_pfn[MAX_NUMNODES];
> @@ -3121,64 +3117,6 @@ void __init sparse_memory_present_with_a
>  }
>  
>  /**
> - * push_node_boundaries - Push node boundaries to at least the requested boundary
> - * @nid: The nid of the node to push the boundary for
> - * @start_pfn: The start pfn of the node
> - * @end_pfn: The end pfn of the node
> - *
> - * In reserve-based hot-add, mem_map is allocated that is unused until hotadd
> - * time. Specifically, on x86_64, SRAT will report ranges that can potentially
> - * be hotplugged even though no physical memory exists. This function allows
> - * an arch to push out the node boundaries so mem_map is allocated that can
> - * be used later.
> - */
> -#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
> -void __init push_node_boundaries(unsigned int nid,
> -		unsigned long start_pfn, unsigned long end_pfn)
> -{
> -	mminit_dprintk(MMINIT_TRACE, "zoneboundary",
> -			"Entering push_node_boundaries(%u, %lu, %lu)\n",
> -			nid, start_pfn, end_pfn);
> -
> -	/* Initialise the boundary for this node if necessary */
> -	if (node_boundary_end_pfn[nid] == 0)
> -		node_boundary_start_pfn[nid] = -1UL;
> -
> -	/* Update the boundaries */
> -	if (node_boundary_start_pfn[nid] > start_pfn)
> -		node_boundary_start_pfn[nid] = start_pfn;
> -	if (node_boundary_end_pfn[nid] < end_pfn)
> -		node_boundary_end_pfn[nid] = end_pfn;
> -}
> -
> -/* If necessary, push the node boundary out for reserve hotadd */
> -static void __meminit account_node_boundary(unsigned int nid,
> -		unsigned long *start_pfn, unsigned long *end_pfn)
> -{
> -	mminit_dprintk(MMINIT_TRACE, "zoneboundary",
> -			"Entering account_node_boundary(%u, %lu, %lu)\n",
> -			nid, *start_pfn, *end_pfn);
> -
> -	/* Return if boundary information has not been provided */
> -	if (node_boundary_end_pfn[nid] == 0)
> -		return;
> -
> -	/* Check the boundaries and update if necessary */
> -	if (node_boundary_start_pfn[nid] < *start_pfn)
> -		*start_pfn = node_boundary_start_pfn[nid];
> -	if (node_boundary_end_pfn[nid] > *end_pfn)
> -		*end_pfn = node_boundary_end_pfn[nid];
> -}
> -#else
> -void __init push_node_boundaries(unsigned int nid,
> -		unsigned long start_pfn, unsigned long end_pfn) {}
> -
> -static void __meminit account_node_boundary(unsigned int nid,
> -		unsigned long *start_pfn, unsigned long *end_pfn) {}
> -#endif
> -
> -
> -/**
>   * get_pfn_range_for_nid - Return the start and end page frames for a node
>   * @nid: The nid to return the range for. If MAX_NUMNODES, the min and max PFN are returned.
>   * @start_pfn: Passed by reference. On return, it will have the node start_pfn.
> @@ -3203,9 +3141,6 @@ void __meminit get_pfn_range_for_nid(uns
>  
>  	if (*start_pfn == -1UL)
>  		*start_pfn = 0;
> -
> -	/* Push the node boundaries out if requested */
> -	account_node_boundary(nid, start_pfn, end_pfn);
>  }
>  
>  /*
> @@ -3810,10 +3745,6 @@ void __init remove_all_active_ranges(voi
>  {
>  	memset(early_node_map, 0, sizeof(early_node_map));
>  	nr_nodemap_entries = 0;
> -#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
> -	memset(node_boundary_start_pfn, 0, sizeof(node_boundary_start_pfn));
> -	memset(node_boundary_end_pfn, 0, sizeof(node_boundary_end_pfn));
> -#endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */
>  }
>  
>  /* Compare two active node_active_regions */

-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists