Re: [PATCH 08/22] lmb: Add get_free_all_memory_range()

[Benjamin Herrenschmidt <benh@kernel.crashing.org>]

On Sat, 2010-05-08 at 08:17 -0700, Yinghai Lu wrote:
> get_free_all_memory_range is for CONFIG_NO_BOOTMEM=y, and will be called by
> free_all_memory_core_early().
> 
> It will use early_node_map aka active ranges subtract lmb.reserved to
> get all free range, and those ranges will convert to slab pages.
> 
> -v3: use __lmb_find_base() to get range free buffer.

So that is more crap on top of crap that shouldn't have been upstream in
the first place. That whole range management stuff that was taken out of
x86 and made generic should just go away (or go back to x86 where it
belongs).

Seriously Ingo, what the hell did you think when you merged that
stuff ? 

So the code here is extremely convluted, only meaningful in the
nobootmem/early_res case and totally NOT something that should be in the
generic lmb code either.

If you're going to do something half-sensible with nobootmem, which is
to use LMB directly, then I see no point in keeping any of that stuff.

Ben.

> Signed-off-by: Yinghai Lu <yinghai@kernel.org>
> Cc: Jan Beulich <jbeulich@novell.com>
> ---
>  include/linux/lmb.h |    2 +
>  mm/lmb.c            |   86 ++++++++++++++++++++++++++++++++++++++++++++++++++-
>  2 files changed, 87 insertions(+), 1 deletions(-)
> 
> diff --git a/include/linux/lmb.h b/include/linux/lmb.h
> index 2c008b9..aaef2ca 100644
> --- a/include/linux/lmb.h
> +++ b/include/linux/lmb.h
> @@ -93,6 +93,8 @@ u64 __lmb_find_area(u64 ei_start, u64 ei_last, u64 start, u64 end,
>  u64 lmb_find_area(u64 start, u64 end, u64 size, u64 align);
>  
>  void lmb_to_bootmem(u64 start, u64 end);
> +struct range;
> +int get_free_all_memory_range(struct range **rangep, int nodeid);
>  
>  #include <asm/lmb.h>
>  
> diff --git a/mm/lmb.c b/mm/lmb.c
> index ccbc76d..2e19df9 100644
> --- a/mm/lmb.c
> +++ b/mm/lmb.c
> @@ -630,7 +630,91 @@ void __init lmb_free_area(u64 start, u64 end)
>  	__check_and_double_region_array(&lmb.reserved, &lmb_reserved_region[0]);
>  }
>  
> -#ifndef CONFIG_NO_BOOTMEM
> +static __init struct range *find_range_array(int count)
> +{
> +	u64 end, size, mem;
> +	struct range *range;
> +
> +	size = sizeof(struct range) * count;
> +	end = lmb.default_alloc_limit;
> +
> +	mem = __lmb_find_base(size, sizeof(struct range), end);
> +	if (mem == -1ULL)
> +		panic("can not find more space for range array");
> +
> +	/*
> +	 * This range is tempoaray, so don't reserve it, it will not be
> +	 * overlapped because We will not alloccate new buffer before
> +	 * We discard this one
> +	 */
> +	range = __va(mem);
> +	memset(range, 0, size);
> +
> +	return range;
> +}
> +
> +#ifdef CONFIG_NO_BOOTMEM
> +static void __init subtract_lmb_reserved(struct range *range, int az)
> +{
> +	int i, count;
> +	u64 final_start, final_end;
> +
> +	/* Take out region array itself at first*/
> +	if (lmb.reserved.region != lmb_reserved_region)
> +		lmb_free(__pa(lmb.reserved.region), sizeof(struct lmb_property) * lmb.reserved.nr_regions);
> +
> +	count  = lmb.reserved.cnt;
> +
> +	pr_info("Subtract (%d early reservations)\n", count);
> +
> +	for (i = 0; i < count; i++) {
> +		struct lmb_property *r = &lmb.reserved.region[i];
> +		pr_info("  #%d [%010llx - %010llx]\n", i, r->base, r->base + r->size);
> +		final_start = PFN_DOWN(r->base);
> +		final_end = PFN_UP(r->base + r->size);
> +		if (final_start >= final_end)
> +			continue;
> +		subtract_range(range, az, final_start, final_end);
> +	}
> +	/* Put region array back ? */
> +	if (lmb.reserved.region != lmb_reserved_region)
> +		lmb_reserve(__pa(lmb.reserved.region), sizeof(struct lmb_property) * lmb.reserved.nr_regions);
> +}
> +
> +int __init get_free_all_memory_range(struct range **rangep, int nodeid)
> +{
> +	int count;
> +	struct range *range;
> +	int nr_range;
> +
> +	count = lmb.reserved.cnt * 2;
> +
> +	range = find_range_array(count);
> +	nr_range = 0;
> +
> +	/*
> +	 * Use early_node_map[] and lmb.reserved.region to get range array
> +	 * at first
> +	 */
> +	nr_range = add_from_early_node_map(range, count, nr_range, nodeid);
> +#ifdef CONFIG_X86_32
> +	subtract_range(range, count, max_low_pfn, -1ULL);
> +#endif
> +	subtract_lmb_reserved(range, count);
> +	nr_range = clean_sort_range(range, count);
> +
> +	/* Need to clear it ? */
> +	if (nodeid == MAX_NUMNODES) {
> +		memset(&lmb.reserved.region[0], 0, sizeof(struct lmb_property) * lmb.reserved.nr_regions);
> +		lmb.reserved.region = NULL;
> +		lmb.reserved.nr_regions = 0;
> +		lmb.reserved.cnt = 0;
> +	}
> +
> +	*rangep = range;
> +	return nr_range;
> +}
> +#else
>  void __init lmb_to_bootmem(u64 start, u64 end)
>  {
>  	int i, count;