LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Yinghai Lu <yinghai@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>, "H. Peter Anvin" <hpa@zytor.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	David Miller <davem@davemloft.net>
Subject: Re: [PATCH] memblock: Don't adjust size in memblock_find_base()
Date: Mon, 10 Jan 2011 16:56:29 +1100	[thread overview]
Message-ID: <1294638989.17779.296.camel@pasglop> (raw)
In-Reply-To: <4D277A8C.5010502@kernel.org>

On Fri, 2011-01-07 at 12:41 -0800, Yinghai Lu wrote:
> While applying patch to use memblock to find aperture for 64bit x86.
> Ingo found system with 1g + force_iommu

Please run that through DaveM. I seem to remember that he relies on the
size alignment to ensure that a lot of tiny allocations get properly
coalesced or he runs out of regions at boot time.

A better option would be to use a fixed alignment, or sized based but
with a max threshold such as PAGE_SIZE.

Cheers,
Ben.

> > No AGP bridge found
> > Node 0: aperture @ 38000000 size 32 MB
> > Aperture pointing to e820 RAM. Ignoring.
> > Your BIOS doesn't leave a aperture memory hole
> > Please enable the IOMMU option in the BIOS setup
> > This costs you 64 MB of RAM
> > Cannot allocate aperture memory hole (0,65536K)
> 
> the corresponding code:
> 	addr = memblock_find_in_range(0, 1ULL<<32, aper_size, 512ULL<<20);
> 	if (addr == MEMBLOCK_ERROR || addr + aper_size > 0xffffffff) {
> 		printk(KERN_ERR
> 			"Cannot allocate aperture memory hole (%lx,%uK)\n",
> 				addr, aper_size>>10);
> 		return 0;
> 	}
> 	memblock_x86_reserve_range(addr, addr + aper_size, "aperture64")
> 
> it fails because memblock core code align the size with 512M. that could make
> size way too big.
> 
> So don't align the size in that case.
> 
> actually __memblock_alloc_base, the another caller already align that before calling that function.
> 
> BTW. x86 does not use __memblock_alloc_base...
> 
> Signed-off-by: Yinghai Lu <yinghai@kernel.org>
> 
> ---
>  mm/memblock.c |    2 --
>  1 file changed, 2 deletions(-)
> 
> Index: linux-2.6/mm/memblock.c
> ===================================================================
> --- linux-2.6.orig/mm/memblock.c
> +++ linux-2.6/mm/memblock.c
> @@ -137,8 +137,6 @@ static phys_addr_t __init_memblock membl
>  
>  	BUG_ON(0 == size);
>  
> -	size = memblock_align_up(size, align);
> -
>  	/* Pump up max_addr */
>  	if (end == MEMBLOCK_ALLOC_ACCESSIBLE)
>  		end = memblock.current_limit;
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/



  reply	other threads:[~2011-01-10  5:56 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-18  0:58 [PATCH 0/3] memblock related top down Yinghai Lu
2010-12-28  0:47 ` [PATCH 0/6] memblock related further cleanup for x86 Yinghai Lu
     [not found] ` <4D19320B.7030007@kernel.org>
2010-12-28  0:47   ` [PATCH 1/6] x86: Change get_max_mapped() to inline Yinghai Lu
2010-12-29 23:05     ` H. Peter Anvin
2010-12-29 23:30       ` Yinghai Lu
2010-12-29 23:37         ` Yinghai Lu
2010-12-29 23:42           ` H. Peter Anvin
2010-12-29 23:45     ` [PATCH -v2 " Yinghai Lu
2011-02-16  8:50     ` [tip:x86/mm] " tip-bot for Yinghai Lu
2010-12-28  0:48   ` [PATCH 2/6] x86, 64bit, numa: Allocate memnodemap under max_pfn_mapped Yinghai Lu
2011-02-16  8:50     ` [tip:x86/mm] x86-64, " tip-bot for Yinghai Lu
2010-12-28  0:48   ` [PATCH 3/6] x86, 64bit, numa: Put pgtable to local node memory Yinghai Lu
2010-12-29 23:46     ` [PATCH -v2 " Yinghai Lu
2010-12-29 23:50       ` H. Peter Anvin
2010-12-30  0:11         ` Yinghai Lu
2010-12-30  0:39           ` David Rientjes
2010-12-30  0:58             ` H. Peter Anvin
2010-12-30  1:07               ` David Rientjes
2010-12-30  1:53                 ` H. Peter Anvin
2010-12-30  9:06                   ` Ingo Molnar
2010-12-30 10:28                     ` [boot crash] " Ingo Molnar
2010-12-30 10:30                       ` Ingo Molnar
2010-12-30 12:01                         ` [build failure] " Ingo Molnar
2010-12-30 18:53                           ` David Rientjes
2010-12-30 18:54                             ` [patch] x86, numa: Fix CONFIG_DEBUG_PER_CPU_MAPS without NUMA emulation David Rientjes
2011-01-06 20:39                               ` David Rientjes
2011-01-06 21:25                                 ` Yinghai Lu
2011-01-06 23:58                                   ` David Rientjes
2011-01-07 15:32                               ` [tip:x86/numa] " tip-bot for David Rientjes
2010-12-30 21:18                         ` [boot crash] Re: [PATCH -v2 3/6] x86, 64bit, numa: Put pgtable to local node memory Yinghai Lu
     [not found]                           ` <AANLkTimpyRXS9dNFATh4GqtA46uynutAvK7Qb5S7WPFO@mail.gmail.com>
     [not found]                             ` <20110105134434.GA22816@elte.hu>
2011-01-05 21:24                               ` Yinghai Lu
2011-01-07 20:41                               ` [PATCH] memblock: Don't adjust size in memblock_find_base() Yinghai Lu
2011-01-10  5:56                                 ` Benjamin Herrenschmidt [this message]
2011-01-10  6:59                                   ` Yinghai Lu
2011-01-12  1:35                                   ` Yinghai Lu
2011-01-29  0:37                               ` [PATCH - resend] " Yinghai Lu
2011-02-03 20:25                                 ` Andrew Morton
2011-02-03 20:37                                   ` Yinghai Lu
2011-02-16  8:51     ` [tip:x86/mm] x86-64, numa: Put pgtable to local node memory tip-bot for Yinghai Lu
2010-12-28  0:48   ` [PATCH 4/6] x86: Use early pre-allocated page table buffer top-down Yinghai Lu
2010-12-28  0:48   ` [PATCH 5/6] x86, 64bit: Move out cleanup higmap [_brk_end, _end) out of init_memory_mapping() Yinghai Lu
2011-02-16  8:51     ` [tip:x86/mm] x86-64: " tip-bot for Yinghai Lu
2010-12-28  0:48   ` [PATCH 6/6] x86: Rename e820_table_* to pgt_buf_* Yinghai Lu
2010-12-30 21:54     ` [PATCH -v2 " Yinghai Lu
2010-12-28 20:21 ` [PATCH 0/3] memblock related top down H. Peter Anvin
2010-12-28 21:36   ` Yinghai Lu
2010-12-28 22:09     ` H. Peter Anvin
2010-12-28 22:25       ` Yinghai Lu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1294638989.17779.296.camel@pasglop \
    --to=benh@kernel.crashing.org \
    --cc=davem@davemloft.net \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=yinghai@kernel.org \
    --subject='Re: [PATCH] memblock: Don'\''t adjust size in memblock_find_base()' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).