From: Linus Torvalds <torvalds@linux-foundation.org>
To: Yinghai Lu <yinghai@kernel.org>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>,
Ingo Molnar <mingo@elte.hu>,
Andrew Morton <akpm@linux-foundation.org>,
Thomas Gleixner <tglx@linutronix.de>,
"H. Peter Anvin" <hpa@zytor.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
linux-pci@vger.kernel.org, yannick.roehlly@free.fr
Subject: Re: [PATCH] x86/pci: make pci_mem_start to be aligned only
Date: Tue, 14 Apr 2009 14:10:38 -0700 (PDT) [thread overview]
Message-ID: <alpine.LFD.2.00.0904141359020.18124@localhost.localdomain> (raw)
In-Reply-To: <49E4F71F.10107@kernel.org>
On Tue, 14 Apr 2009, Yinghai Lu wrote:
>
> Impact: make more big space below 4g for assigning to unassigned pci devices
>
> don't need to reserved one round after the gapstart.
>
> Reported-and-tested-by: Yannick <yannick.roehlly@free.fr>
> Signed-off-by: Yinghai Lu <yinghai@kernel.org>
>
> diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
> index ef2c356..a0ba9b1 100644
> --- a/arch/x86/kernel/e820.c
> +++ b/arch/x86/kernel/e820.c
> @@ -642,7 +642,7 @@ __init void e820_setup_gap(void)
> while ((gapsize >> 4) > round)
> round += round;
> /* Fun with two's complement */
> - pci_mem_start = (gapstart + round) & -round;
> + pci_mem_start = roundup(gapstart, round);
That thing is called "e820_setup_gap()" for a reason. It's supposed to
create a _gap_. That's why it historically doesn't just round up (no
"+round-1" as in round_up()), it rounds up to the _next_ boundary
("+round") in order to guarantee a gap.
The reason? We've definitely seen ACPI code or integrated graphics stuff
that steals a lot of memory at the end, which means that end-of-RAM might
be not at 2GB, but at 2GB-16MB-1MB, for example (1MB of "ACPI data", and
16MB of "stolen video ram").
Now, the BIOS _hopefully_ marks those areas clearly reserved, and as a
result we don't end up allocating PCI data in there, but the gap was there
literally to make sure we always leave that gap, very much on purpose.
So I'm very nervous about this.
At a minimum, if we do this, I'd like to make sure we round up to a big
boundary (eg 32MB or something - exactly because a missing 16MB can easily
be some integrated stolen video memory).
Sure, we do that whole
while ((gapsize >> 4) > round)
round += round;
thing, so that if the gap is large, then we'll certainly get to 32MB too,
but I think your patch matters the most exactly when the gap is small.
Maybe we could just raise the initial minimum rounding from 1MB to 32MB?
Alternatively, maybe we can make sure that we round up to at least X bytes
from the end of RAM, and to at least Y bytes from the end of some RESERVED
thing.
I dunno. Maybe your patch is fine as-is. But I do get nervous.
Linus
next prev parent reply other threads:[~2009-04-14 21:17 UTC|newest]
Thread overview: 117+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <bug-11103-13546@http.bugzilla.kernel.org/>
[not found] ` <200904101913.n3AJDhMm018684@demeter.kernel.org>
2009-04-10 20:27 ` [Bug 11103] Can't use framebuffer or vesa Xorg with two memory modules Yinghai Lu
2009-04-11 3:29 ` Yinghai Lu
2009-04-14 20:49 ` [PATCH] pci: don't assume pref memio are 64bit -v2 Yinghai Lu
2009-04-14 20:50 ` [PATCH] x86/pci: make pci_mem_start to be aligned only Yinghai Lu
2009-04-14 21:10 ` Linus Torvalds [this message]
2009-04-14 21:27 ` H. Peter Anvin
2009-04-14 21:27 ` Yinghai Lu
2009-04-14 22:35 ` Yannick Roehlly
2009-04-15 0:29 ` [PATCH] x86/pci: make pci_mem_start to be aligned only -v2 Yinghai Lu
2009-04-15 0:41 ` [PATCH] x86/pci: make pci_mem_start to be aligned only -v3 Yinghai Lu
2009-04-15 0:42 ` [PATCH] x86/pci: fix -1 calling to e820_all_mapped with mmconfig Yinghai Lu
2009-04-16 16:31 ` [PATCH] x86/pci: make pci_mem_start to be aligned only -v3 Jesse Barnes
2009-04-16 16:44 ` Linus Torvalds
2009-04-16 16:56 ` Ingo Molnar
2009-04-16 17:18 ` Yinghai Lu
2009-04-16 17:27 ` H. Peter Anvin
2009-04-16 17:38 ` Ingo Molnar
2009-04-16 17:28 ` Ingo Molnar
2009-04-16 20:13 ` [PATCH] x86/pci: make pci_mem_start to be aligned only -v4 Yinghai Lu
2009-04-16 23:18 ` Linus Torvalds
2009-04-16 23:54 ` Ingo Molnar
2009-04-17 0:24 ` Linus Torvalds
2009-04-17 13:16 ` Ingo Molnar
2009-04-17 21:59 ` Yinghai Lu
2009-04-17 22:04 ` H. Peter Anvin
2009-04-18 5:37 ` [PATCH] pci: keep pci device resource name pointer right Yinghai Lu
2009-04-18 7:51 ` Ingo Molnar
2009-04-18 16:05 ` Jesse Barnes
2009-04-18 18:42 ` Linus Torvalds
2009-04-18 19:19 ` Yinghai Lu
2009-04-18 19:23 ` Greg KH
2009-04-18 20:00 ` Kay Sievers
2009-04-18 20:27 ` Kay Sievers
2009-04-18 20:37 ` Ingo Molnar
2009-04-18 22:05 ` [PATCH] driver: dont update dev_name via device_add path Yinghai Lu
2009-04-28 7:36 ` [PATCH] driver: make dev_set_name(, NULL) work Yinghai Lu
2009-04-28 7:42 ` [RFC PATCH] use dev_set_name(,NULL) to prevent leaking Yinghai Lu
2009-04-28 8:25 ` Kay Sievers
2009-04-28 15:21 ` Yinghai Lu
2009-04-28 15:34 ` Yinghai Lu
2009-04-28 15:39 ` Greg KH
2009-04-28 15:51 ` Yinghai Lu
2009-04-28 15:56 ` Kay Sievers
2009-04-28 16:08 ` Yinghai Lu
2009-04-28 16:15 ` Kay Sievers
2009-04-28 19:04 ` Yinghai Lu
2009-04-28 16:36 ` Yinghai Lu
2009-04-28 16:50 ` Kay Sievers
2009-04-28 14:52 ` Greg KH
2009-04-28 14:51 ` [PATCH] driver: make dev_set_name(, NULL) work Greg KH
2009-04-28 15:14 ` Yinghai Lu
2009-04-28 15:39 ` Greg KH
2009-04-18 21:07 ` [PATCH] pci: keep pci device resource name pointer right Yinghai Lu
2009-04-18 22:17 ` Linus Torvalds
2009-04-18 20:00 ` [PATCH] driver: dont update dev_name if it is not changed Yinghai Lu
2009-04-18 20:11 ` Ingo Molnar
2009-04-18 20:20 ` Yinghai Lu
2009-04-18 20:27 ` Ingo Molnar
2009-04-18 8:33 ` [PATCH] x86/pci: make pci_mem_start to be aligned only -v4 Yinghai Lu
2009-04-18 9:22 ` Ingo Molnar
2009-04-18 17:07 ` Yinghai Lu
2009-04-18 18:57 ` Linus Torvalds
2009-04-18 19:14 ` Ingo Molnar
2009-04-18 19:26 ` Yinghai Lu
2009-04-18 22:20 ` Yinghai Lu
2009-04-18 22:31 ` Linus Torvalds
2009-04-18 20:13 ` Ivan Kokshaysky
2009-04-18 18:50 ` Linus Torvalds
2009-04-18 22:44 ` Yinghai Lu
2009-04-18 23:01 ` Yinghai Lu
2009-04-18 23:06 ` Linus Torvalds
2009-04-18 23:26 ` Yinghai Lu
2009-04-18 23:30 ` Yinghai Lu
2009-04-18 23:04 ` Linus Torvalds
2009-04-19 0:32 ` H. Peter Anvin
2009-04-19 4:50 ` Linus Torvalds
2009-04-19 5:26 ` Yinghai Lu
2009-04-19 19:35 ` Yannick Roehlly
2009-04-19 19:59 ` Yinghai Lu
2009-04-19 20:24 ` Yannick Roehlly
2009-04-19 9:02 ` Ingo Molnar
2009-04-19 9:06 ` Ingo Molnar
2009-04-19 17:52 ` Jesse Barnes
2009-04-20 22:33 ` Ivan Kokshaysky
2009-04-20 22:52 ` Yinghai Lu
2009-04-21 10:54 ` Ivan Kokshaysky
2009-04-21 0:09 ` Yinghai Lu
2009-04-21 10:56 ` Ivan Kokshaysky
2009-04-21 15:57 ` Yinghai Lu
2009-04-22 22:37 ` [RFC PATCH 1/2] pci: don't assume pref memio are 64bit -v3 Yinghai Lu
2009-04-22 22:38 ` [RFC PATCH 2/2] pci: try to assign res for device under transparent bridges Yinghai Lu
2009-04-22 22:49 ` [RFC PATCH 1/2] pci: don't assume pref memio are 64bit -v3 Jesse Barnes
2009-04-23 0:49 ` Yinghai Lu
2009-04-23 1:05 ` Jesse Barnes
2009-04-23 2:03 ` Yinghai Lu
2009-04-23 12:58 ` Ivan Kokshaysky
2009-04-23 15:30 ` Yinghai Lu
2009-04-23 2:10 ` Yinghai Lu
2009-04-23 13:22 ` Ivan Kokshaysky
2009-04-23 15:13 ` Yinghai Lu
2009-04-23 22:19 ` Ivan Kokshaysky
2009-04-24 3:48 ` [PATCH 1/4] pci/x86: don't assume pref memio are 64bit -v4 Yinghai Lu
2009-04-24 3:49 ` [PATCH 2/4] pci: try to assign res for device under transparent bridges -v2 Yinghai Lu
2009-04-24 3:50 ` [PATCH 3/4] x86: reserve range near the ram Yinghai Lu
2009-04-24 3:50 ` [PATCH 4/4] x86/pci: make pci_mem_start to be aligned only -v5 Yinghai Lu
2009-04-24 13:16 ` [PATCH 1/4] pci/x86: don't assume pref memio are 64bit -v4 Ivan Kokshaysky
2009-05-05 18:52 ` Jesse Barnes
2009-05-06 12:33 ` Ingo Molnar
2009-05-06 15:06 ` [PATCH 1/2] x86: reserve range near the ram Yinghai Lu
2009-05-11 9:51 ` [tip:x86/mm] x86, e820, pci: reserve extra free space near end of RAM tip-bot for Linus Torvalds
2009-05-06 15:07 ` [PATCH 1/2] x86/pci: make pci_mem_start to be aligned only -v5 Yinghai Lu
2009-05-11 9:51 ` [tip:x86/mm] x86/pci: remove rounding quirk from e820_setup_gap() tip-bot for Yinghai Lu
2009-04-23 12:36 ` [RFC PATCH 1/2] pci: don't assume pref memio are 64bit -v3 Ivan Kokshaysky
2009-04-23 12:41 ` Ingo Molnar
2009-04-23 13:09 ` Ivan Kokshaysky
2009-04-23 15:05 ` Yinghai Lu
2009-04-21 15:41 ` [PATCH] x86/pci: make pci_mem_start to be aligned only -v4 Jesse Barnes
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=alpine.LFD.2.00.0904141359020.18124@localhost.localdomain \
--to=torvalds@linux-foundation.org \
--cc=akpm@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=yannick.roehlly@free.fr \
--cc=yinghai@kernel.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.