From: Milton Miller <miltonm@bga.com>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: linuxppc-dev@ozlabs.org
Subject: powerpc: numa.c: always trim to lmb_end_of_DRAM
Date: Thu, 10 Jul 2008 16:19:06 -0500 (CDT) [thread overview]
Message-ID: <for-27-patch3@bga.com> (raw)
In-Reply-To: <for-27-patch1-0@bga.com>
numa_enforce_memory_limit trys to be smart and only trim the memory it
finds to lmb_end_of_DRAM when a memory limit is set. However, the
early boot code may also limit memory when iommu=off is specified. When
this happens, the page allocator is told of pages not in the linear mapping
and gets a fatal DSI.
Signed-off-by: Milton Miller <miltonm@bga.com>
--
Previously this patch tried to check if iommu=off was specified, but it was
requested that the check just be removed.
http://patchwork.ozlabs.org/linuxppc/patch?id=11774
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index cf4bffb..590406c 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -472,12 +472,10 @@ static unsigned long __init numa_enforce_memory_limit(unsigned long start,
/*
* We use lmb_end_of_DRAM() in here instead of memory_limit because
* we've already adjusted it for the limit and it takes care of
- * having memory holes below the limit.
+ * having memory holes below the limit. Also, in the case of
+ * iommu_is_off, memory_limit is not set but is implicitly enforced.
*/
- if (! memory_limit)
- return size;
-
if (start + size <= lmb_end_of_DRAM())
return size;
next prev parent reply other threads:[~2008-07-10 21:20 UTC|newest]
Thread overview: 89+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-07-10 21:12 [lm-sensors] [RFC] (almost) booting allyesconfig -- please don't Milton Miller
2008-07-10 21:12 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Milton Miller
2008-07-10 21:14 ` mtd: remove __PPC__ hardcoded address from nand/diskonchip and devices/docprobe Milton Miller
2008-07-10 21:14 ` [1/3] powerpc: add _HEAD_GLOBAL to place functions in .text.head Milton Miller
2008-07-10 21:16 ` [2/3] powerpc: head_64.S: put irq vectors " Milton Miller
2008-07-10 21:19 ` Milton Miller [this message]
2008-07-10 21:20 ` powerpc: pseries, cell: use cpu_thread_in_core in smp_init for of_spin_map Milton Miller
2008-07-10 21:22 ` powerpc: find and destroy possible stale kernel added properties Milton Miller
2008-07-10 21:23 ` powerpc: add static and ifdef prom_strtoul and prom_memparse Milton Miller
2008-07-10 21:29 ` [PATCH] spufs: correct kcalloc usage Milton Miller
2008-07-10 23:04 ` Jeremy Kerr
2008-07-10 21:33 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Hans de Goede
2008-07-10 21:33 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Hans de Goede
2008-07-10 21:33 ` Hans de Goede
2008-07-10 21:51 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Milton Miller
2008-07-10 21:51 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Milton Miller
2008-07-10 21:51 ` Milton Miller
2008-07-11 6:52 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Jean Delvare
2008-07-11 6:52 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Jean Delvare
2008-07-11 6:52 ` Jean Delvare
2008-07-11 7:27 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Hans de Goede
2008-07-11 7:27 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Hans de Goede
2008-07-11 7:27 ` Hans de Goede
2008-07-11 7:36 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Jean Delvare
2008-07-11 7:36 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Jean Delvare
2008-07-11 7:36 ` Jean Delvare
2008-07-13 6:31 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Hans de Goede
2008-07-13 6:31 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Hans de Goede
2008-07-13 6:31 ` Hans de Goede
2008-07-13 21:11 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please David Hubbard
2008-07-13 21:11 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region David Hubbard
2008-07-13 21:11 ` David Hubbard
2008-07-13 21:22 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Hans de Goede
2008-07-13 21:22 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Hans de Goede
2008-07-13 21:22 ` Hans de Goede
2008-07-13 21:26 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please David Hubbard
2008-07-13 21:26 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region David Hubbard
2008-07-13 21:26 ` David Hubbard
2008-07-14 7:59 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Jean Delvare
2008-07-14 7:59 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Jean Delvare
2008-07-14 7:59 ` Jean Delvare
2008-07-14 17:09 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Milton Miller
2008-07-14 17:09 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Milton Miller
2008-07-14 17:09 ` Milton Miller
2008-07-14 17:30 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- Hans de Goede
2008-07-14 17:30 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Hans de Goede
2008-07-14 17:30 ` Hans de Goede
2008-07-14 17:55 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please David Hubbard
2008-07-14 17:55 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region David Hubbard
2008-07-14 17:55 ` David Hubbard
2008-07-15 8:36 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Jean Delvare
2008-07-15 8:36 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Jean Delvare
2008-07-15 8:36 ` Jean Delvare
2008-07-15 15:31 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please David Hubbard
2008-07-15 15:31 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region David Hubbard
2008-07-15 15:31 ` David Hubbard
2008-07-16 7:46 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Jean Delvare
2008-07-16 7:46 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Jean Delvare
2008-07-16 7:46 ` Jean Delvare
2008-07-16 8:09 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Rene Herman
2008-07-16 8:09 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Rene Herman
2008-07-16 8:09 ` Rene Herman
2008-07-15 8:28 ` [lm-sensors] [RFC] (almost) booting allyesconfig -- please Jean Delvare
2008-07-15 8:28 ` [RFC] (almost) booting allyesconfig -- please don't poke super-io without request_region Jean Delvare
2008-07-15 8:28 ` Jean Delvare
[not found] ` <for-27-patch9@bga.com>
2008-07-12 20:02 ` [PATCH/RESEND] pci: dynids.use_driver_data considered harmful Milton Miller
2008-07-12 20:17 ` Greg KH
2008-07-12 20:58 ` Jean Delvare
2008-07-12 21:17 ` Milton Miller
2008-07-12 21:29 ` Milton Miller
[not found] ` <20080712041137.GA5933@kroah.com>
2008-07-12 21:08 ` [PATCH/RFC] " Milton Miller
2008-07-12 22:48 ` Milton Miller
2008-07-16 10:18 ` Milton Miller
2008-07-17 7:07 ` Greg KH
2008-07-17 14:36 ` Milton Miller
2008-08-06 7:31 ` Jean Delvare
2008-08-14 22:12 ` Greg KH
2008-08-15 14:50 ` Milton Miller
2008-08-15 15:50 ` Jean Delvare
2008-08-15 17:46 ` Jesse Barnes
2008-08-15 18:55 ` Jean Delvare
2008-08-15 19:15 ` Jesse Barnes
2008-08-16 6:22 ` Greg KH
2008-08-17 19:06 ` Jean Delvare
2008-08-18 3:50 ` Greg KH
2008-08-18 17:13 ` Jesse Barnes
2008-08-18 20:41 ` Jesse Barnes
2008-08-19 18:01 ` Milton Miller
2008-08-06 7:22 ` Jean Delvare
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=for-27-patch3@bga.com \
--to=miltonm@bga.com \
--cc=benh@kernel.crashing.org \
--cc=linuxppc-dev@ozlabs.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.