All of lore.kernel.org
 help / color / mirror / Atom feed
From: Yinghai Lu <yinghai@kernel.org>
To: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>,
	Christoph Lameter <cl@linux-foundation.org>,
	linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
	Yinghai Lu <yinghai@kernel.org>
Subject: [PATCH 14/35] x86: make early_node_mem get mem > 4g if possible
Date: Wed, 10 Feb 2010 01:20:18 -0800	[thread overview]
Message-ID: <1265793639-15071-15-git-send-email-yinghai@kernel.org> (raw)
In-Reply-To: <1265793639-15071-1-git-send-email-yinghai@kernel.org>

so we could put pgdata for the node high, and later sparse
vmmap will get the section nr that need.

with this patch will make <4g ram will not use sparse vmmap

before this patch, will get, before swiotlb try get bootmem
[    0.000000] nid=1 start=0 end=2080000 aligned=1
[    0.000000]   free [10 - 96]
[    0.000000]   free [b12 - 1000]
[    0.000000]   free [359f - 38a3]
[    0.000000]   free [38b5 - 3a00]
[    0.000000]   free [41e01 - 42000]
[    0.000000]   free [73dde - 73e00]
[    0.000000]   free [73fdd - 74000]
[    0.000000]   free [741dd - 74200]
[    0.000000]   free [743dd - 74400]
[    0.000000]   free [745dd - 74600]
[    0.000000]   free [747dd - 74800]
[    0.000000]   free [749dd - 74a00]
[    0.000000]   free [74bdd - 74c00]
[    0.000000]   free [74ddd - 74e00]
[    0.000000]   free [74fdd - 75000]
[    0.000000]   free [751dd - 75200]
[    0.000000]   free [753dd - 75400]
[    0.000000]   free [755dd - 75600]
[    0.000000]   free [757dd - 75800]
[    0.000000]   free [759dd - 75a00]
[    0.000000]   free [75bdd - 7bf5f]
[    0.000000]   free [7f730 - 7f750]
[    0.000000]   free [100000 - 2080000]
[    0.000000]   total free 1f87170
[   93.301474] Placing 64MB software IO TLB between ffff880075bdd000 - ffff880079bdd000
[   93.311814] software IO TLB at phys 0x75bdd000 - 0x79bdd000

with this patch will get: before swiotlb try get bootmem
[    0.000000] nid=1 start=0 end=2080000 aligned=1
[    0.000000]   free [a - 96]
[    0.000000]   free [702 - 1000]
[    0.000000]   free [359f - 3600]
[    0.000000]   free [37de - 3800]
[    0.000000]   free [39dd - 3a00]
[    0.000000]   free [3bdd - 3c00]
[    0.000000]   free [3ddd - 3e00]
[    0.000000]   free [3fdd - 4000]
[    0.000000]   free [41dd - 4200]
[    0.000000]   free [43dd - 4400]
[    0.000000]   free [45dd - 4600]
[    0.000000]   free [47dd - 4800]
[    0.000000]   free [49dd - 4a00]
[    0.000000]   free [4bdd - 4c00]
[    0.000000]   free [4ddd - 4e00]
[    0.000000]   free [4fdd - 5000]
[    0.000000]   free [51dd - 5200]
[    0.000000]   free [53dd - 5400]
[    0.000000]   free [55dd - 7bf5f]
[    0.000000]   free [7f730 - 7f750]
[    0.000000]   free [100428 - 100600]
[    0.000000]   free [13ea01 - 13ec00]
[    0.000000]   free [170800 - 2080000]
[    0.000000]   total free 1f87170

[   92.689485] PCI-DMA: Using software bounce buffering for IO (SWIOTLB)
[   92.699799] Placing 64MB software IO TLB between ffff8800055dd000 - ffff8800095dd000
[   92.710916] software IO TLB at phys 0x55dd000 - 0x95dd000

so will get enough space below 4G, aka pfn 0x100000

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
---
 arch/x86/mm/numa_64.c |   23 ++++++++++++++++++-----
 1 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 3232148..02f13cb 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -163,14 +163,27 @@ static void * __init early_node_mem(int nodeid, unsigned long start,
 				    unsigned long end, unsigned long size,
 				    unsigned long align)
 {
-	unsigned long mem = find_e820_area(start, end, size, align);
+	unsigned long mem;
 
+	/*
+	 * put it on high as possible
+	 * something will go with NODE_DATA
+	 */
+	if (start < (MAX_DMA_PFN<<PAGE_SHIFT))
+		start = MAX_DMA_PFN<<PAGE_SHIFT;
+	if (start < (MAX_DMA32_PFN<<PAGE_SHIFT) &&
+	    end > (MAX_DMA32_PFN<<PAGE_SHIFT))
+		start = MAX_DMA32_PFN<<PAGE_SHIFT;
+	mem = find_e820_area(start, end, size, align);
 	if (mem != -1L)
 		return __va(mem);
 
-
-	start = __pa(MAX_DMA_ADDRESS);
-	end = max_low_pfn_mapped << PAGE_SHIFT;
+	/* extend the search scope */
+	end = max_pfn_mapped << PAGE_SHIFT;
+	if (end > (MAX_DMA32_PFN<<PAGE_SHIFT))
+		start = MAX_DMA32_PFN<<PAGE_SHIFT;
+	else
+		start = MAX_DMA_PFN<<PAGE_SHIFT;
 	mem = find_e820_area(start, end, size, align);
 	if (mem != -1L)
 		return __va(mem);
-- 
1.6.4.2


  parent reply	other threads:[~2010-02-10  9:26 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-02-10  9:20 [PATCH -v7 0/35] tip related: not use bootmem for x86 Yinghai Lu
2010-02-10  9:20 ` [PATCH 01/35] x86: fix sci on ioapic 1 Yinghai Lu
2010-02-10 22:48   ` [tip:x86/urgent] x86: Fix SCI on IOAPIC != 0 tip-bot for Yinghai Lu
2010-02-10  9:20 ` [PATCH 02/35] x86: keep chip_data in create_irq_nr and destroy_irq Yinghai Lu
2010-02-10 22:39   ` [tip:x86/irq] x86: Avoid race condition in pci_enable_msix() tip-bot for Brandon Phiilps
2010-02-10  9:20 ` [PATCH 03/35] x86: move range related operation to one file Yinghai Lu
2010-02-10  9:20 ` [PATCH 04/35] x86/pci: use resource_size_t in update_res Yinghai Lu
2010-02-10  9:20 ` [PATCH 05/35] x86/pci: amd one chain system to use pci read out res Yinghai Lu
2010-02-10  9:20 ` [PATCH 06/35] x86/pci: use u64 instead of size_t in amd_bus.c Yinghai Lu
2010-02-10  9:20 ` [PATCH 07/35] x86/pci: add cap_resource Yinghai Lu
2010-02-10  9:20 ` [PATCH 08/35] x86/pci: enable pci root res read out for 32bit too Yinghai Lu
2010-02-10  9:20 ` [PATCH 09/35] x86: change range end to start+size Yinghai Lu
2010-02-10  9:20 ` [PATCH 10/35] x86: print out for RAM buffer Yinghai Lu
2010-02-10  9:20 ` [PATCH 11/35] x86: call early_res_to_bootmem one time Yinghai Lu
2010-02-10  9:20 ` [PATCH 12/35] x86: introduce max_early_res and early_res_count Yinghai Lu
2010-02-10  9:20 ` [PATCH 13/35] x86: dynamic increase early_res array size Yinghai Lu
2010-02-10  9:20 ` Yinghai Lu [this message]
2010-02-10  9:20 ` [PATCH 15/35] x86: only call dma32_reserve_bootmem 64bit !CONFIG_NUMA Yinghai Lu
2010-02-10  9:20 ` [PATCH 16/35] x86: make 64 bit use early_res instead of bootmem before slab Yinghai Lu
2010-02-14 14:08   ` Stephen Rothwell
2010-02-14 20:31     ` Yinghai Lu
2010-02-17  1:16     ` Yinghai Lu
2010-02-24 22:59       ` Peter Zijlstra
2010-02-24 23:29         ` Yinghai Lu
2010-02-24 23:32           ` Yinghai Lu
2010-02-25  2:07             ` Tejun Heo
2010-02-25  2:13               ` Yinghai Lu
2010-02-25  2:33                 ` Tejun Heo
2010-02-25  2:36               ` [PATCH] early_res: add free_early_partial Yinghai Lu
2010-02-25 11:10                 ` Peter Zijlstra
2010-03-02  2:48                 ` [PATCH] early_res: need to save name aside with free_early_partial Yinghai Lu
2010-02-10  9:20 ` [PATCH 17/35] sparsemem: put usemap for one node together Yinghai Lu
2010-02-10  9:20 ` [PATCH 18/35] sparsemem: put mem map " Yinghai Lu
2010-02-10  9:20 ` [PATCH 19/35] x86: move bios page reserve early to head32/64.c Yinghai Lu
2010-02-10  9:20 ` [PATCH 20/35] x86: seperate early_res related code from e820.c Yinghai Lu
2010-02-10  9:20 ` [PATCH 21/35] x86: add find_early_area_size Yinghai Lu
2010-02-10  9:20 ` [PATCH 22/35] x86: move back find_e820_area to e820.c Yinghai Lu
2010-02-10  9:20 ` [PATCH 23/35] early_res: enhance check_and_double_early_res Yinghai Lu
2010-02-10  9:20 ` [PATCH 24/35] x86: make 32bit support NO_BOOTMEM Yinghai Lu
2010-02-10  9:20 ` [PATCH 25/35] move round_up/down to kernel.h Yinghai Lu
2010-02-13 18:49   ` Joe Perches
2010-02-13 19:52     ` H. Peter Anvin
2010-02-13 20:11       ` Andrew Morton
2010-02-13 21:57         ` H. Peter Anvin
2010-02-10  9:20 ` [PATCH 26/35] x86: add find_fw_memmap_area Yinghai Lu
2010-02-10  9:20 ` [PATCH 27/35] core: move early_res Yinghai Lu
2010-02-14 14:16   ` Stephen Rothwell
2010-02-14 17:08     ` Ingo Molnar
2010-02-14 23:43       ` Stephen Rothwell
2010-02-15  4:44         ` Ingo Molnar
2010-02-14 20:46     ` Yinghai Lu
2010-02-16 23:46       ` H. Peter Anvin
2010-02-16 23:53         ` Yinghai Lu
2010-02-17  0:01           ` H. Peter Anvin
2010-02-17  0:41             ` Yinghai Lu
2010-02-17  0:46               ` H. Peter Anvin
2010-02-17  1:10                 ` Yinghai Lu
2010-02-17  2:40                   ` Yinghai Lu
2010-02-10  9:20 ` [PATCH 28/35] irq: remove not need bootmem code Yinghai Lu
2010-02-18  1:57   ` [tip:x86/irq] irq: Remove unnecessary " tip-bot for Yinghai Lu
2010-02-10  9:20 ` [PATCH 29/35] radix: move radix init early Yinghai Lu
2010-02-18  1:57   ` [tip:x86/irq] init: Move radix_tree_init() early tip-bot for Yinghai Lu
2010-02-10  9:20 ` [PATCH 30/35] sparseirq: change irq_desc_ptrs to static Yinghai Lu
2010-02-18  1:58   ` [tip:x86/irq] sparseirq: Change " tip-bot for Yinghai Lu
2010-02-10  9:20 ` [PATCH 31/35] sparseirq: use radix_tree instead of ptrs array Yinghai Lu
2010-02-18  1:58   ` [tip:x86/irq] sparseirq: Use " tip-bot for Yinghai Lu
2010-02-10  9:20 ` [PATCH 32/35] x86: remove arch_probe_nr_irqs Yinghai Lu
2010-02-18  1:58   ` [tip:x86/irq] x86, irq: Remove arch_probe_nr_irqs tip-bot for Yinghai Lu
2010-02-10  9:20 ` [PATCH 33/35] use nr_cpus= to set nr_cpu_ids early Yinghai Lu
2010-02-18  1:59   ` [tip:x86/irq] smp: Use " tip-bot for Yinghai Lu
2010-02-10  9:20 ` [PATCH 34/35] x86: use num_processors for possible cpus Yinghai Lu
2010-02-18  1:32   ` H. Peter Anvin
2010-02-18  2:38     ` Yinghai Lu
2010-02-18 17:26       ` H. Peter Anvin
2010-02-18 19:48         ` Christoph Lameter
2010-02-18 19:53           ` H. Peter Anvin
2010-02-19 15:14             ` Christoph Lameter
2010-02-19 16:14               ` H. Peter Anvin
2010-02-10  9:20 ` [PATCH 35/35] x86: make 32bit apic flat to physflat switch like 64bit Yinghai Lu
2010-02-11 16:14 ` [PATCH -v7 0/35] tip related: not use bootmem for x86 Ingo Molnar
2010-02-11 21:10   ` Yinghai Lu
2010-02-15  2:27 ` Benjamin Herrenschmidt
2010-02-15  4:50   ` 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=1265793639-15071-15-git-send-email-yinghai@kernel.org \
    --to=yinghai@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=cl@linux-foundation.org \
    --cc=hpa@zytor.com \
    --cc=jbarnes@virtuousgeek.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.