All of lore.kernel.org
 help / color / mirror / Atom feed
From: Matt Fleming <matt@codeblueprint.co.uk>
To: Lukas Wunner <lukas@wunner.de>
Cc: Bjorn Helgaas <helgaas@kernel.org>,
	linux-pci@vger.kernel.org, linux-wireless@vger.kernel.org,
	zajec5@gmail.com, b43-dev@lists.infradead.org,
	Matthew Garrett <mjg59@srcf.ucam.org>,
	linux-efi@vger.kernel.org
Subject: Re: [PATCH] PCI: Add Broadcom 4331 reset quirk to prevent IRQ storm
Date: Sat, 9 Apr 2016 13:00:55 +0100	[thread overview]
Message-ID: <20160409120055.GR2701@codeblueprint.co.uk> (raw)
In-Reply-To: <20160406213029.GA12421@wunner.de>

On Wed, 06 Apr, at 11:30:29PM, Lukas Wunner wrote:
> 
> More specifically, arch/x86/platform/efi/quirks.c:efi_free_boot_services()
> makes EFI boot services memory available to the page allocator and this
> is called near the end of init/main.c:start_kernel(). *Afterwards*, the
> initcalls are executed via rest_init() -> kernel_init() ->
> kernel_init_freeable() -> do_basic_setup() -> do_initcalls(). In particular,
> PCI header fixups are executed in the subsys_initcall acpi_init().
> 
> Some of the possible solutions are:
> 
> (1) Delay efi_free_boot_services() further, e.g. until late_initcall.
 
I'd prefer not to do this because boot services regions can account
for a large amount of memory (multiple gigabytes).

> (2) Find out exactly which EFI boot services area is used for DMA and
>     amend arch/x86/platform/efi/quirks.c:efi_apply_memmap_quirks() to
>     assign that area a different type if dmi_match(DMI_SYS_VENDOR,
>     "Apple Inc."). That way it's not freed by efi_free_boot_services().
>     Add a late_initcall which calls free_bootmem() for that area.
 
If it is possible to find out which boot services region is the
problematic one, this would be the best solution. If there's any way
to tie it into the PCI quirk, that would be even better so we don't
need to maintain the blacklist in two places.

WARNING: multiple messages have this Message-ID (diff)
From: Matt Fleming <matt@codeblueprint.co.uk>
To: Lukas Wunner <lukas@wunner.de>
Cc: Bjorn Helgaas <helgaas@kernel.org>,
	linux-pci@vger.kernel.org, linux-wireless@vger.kernel.org,
	zajec5@gmail.com, b43-dev@lists.infradead.org,
	Matthew Garrett <mjg59@srcf.ucam.org>,
	linux-efi@vger.kernel.org
Subject: [PATCH] PCI: Add Broadcom 4331 reset quirk to prevent IRQ storm
Date: Sat, 9 Apr 2016 13:00:55 +0100	[thread overview]
Message-ID: <20160409120055.GR2701@codeblueprint.co.uk> (raw)
In-Reply-To: <20160406213029.GA12421@wunner.de>

On Wed, 06 Apr, at 11:30:29PM, Lukas Wunner wrote:
> 
> More specifically, arch/x86/platform/efi/quirks.c:efi_free_boot_services()
> makes EFI boot services memory available to the page allocator and this
> is called near the end of init/main.c:start_kernel(). *Afterwards*, the
> initcalls are executed via rest_init() -> kernel_init() ->
> kernel_init_freeable() -> do_basic_setup() -> do_initcalls(). In particular,
> PCI header fixups are executed in the subsys_initcall acpi_init().
> 
> Some of the possible solutions are:
> 
> (1) Delay efi_free_boot_services() further, e.g. until late_initcall.
 
I'd prefer not to do this because boot services regions can account
for a large amount of memory (multiple gigabytes).

> (2) Find out exactly which EFI boot services area is used for DMA and
>     amend arch/x86/platform/efi/quirks.c:efi_apply_memmap_quirks() to
>     assign that area a different type if dmi_match(DMI_SYS_VENDOR,
>     "Apple Inc."). That way it's not freed by efi_free_boot_services().
>     Add a late_initcall which calls free_bootmem() for that area.
 
If it is possible to find out which boot services region is the
problematic one, this would be the best solution. If there's any way
to tie it into the PCI quirk, that would be even better so we don't
need to maintain the blacklist in two places.

  parent reply	other threads:[~2016-04-09 12:00 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-29 18:20 [PATCH] PCI: Add Broadcom 4331 reset quirk to prevent IRQ storm Lukas Wunner
2016-03-31 18:51 ` Rafał Miłecki
2016-03-31 18:51   ` Rafał Miłecki
2016-04-05 19:40 ` Bjorn Helgaas
2016-04-05 19:40   ` Bjorn Helgaas
2016-04-05 19:49   ` Michael Büsch
2016-04-05 19:49     ` Michael Büsch
2016-04-06 13:31     ` Bjorn Helgaas
2016-04-06 13:31       ` Bjorn Helgaas
2016-04-06 15:17       ` Michael Büsch
2016-04-06 15:17         ` Michael Büsch
2016-04-06 21:36         ` Lukas Wunner
2016-04-06 21:36           ` Lukas Wunner
2016-04-05 19:59   ` Matthew Garrett
2016-04-05 19:59     ` Matthew Garrett
2016-04-06 11:28     ` Andrew Worsley
2016-04-06 11:28       ` Andrew Worsley
2016-04-06 21:30   ` Lukas Wunner
2016-04-06 21:30     ` Lukas Wunner
2016-04-06 21:30     ` Lukas Wunner
2016-04-06 22:19     ` Matthew Garrett
2016-04-06 22:19       ` Matthew Garrett
2016-04-06 22:19       ` Matthew Garrett
2016-04-09 12:00     ` Matt Fleming [this message]
2016-04-09 12:00       ` Matt Fleming
2016-04-24 16:58       ` Lukas Wunner
2016-04-24 16:58         ` Lukas Wunner
     [not found] <E1akxli-00030z-BC@bombadil.infradead.org>
2016-03-31 23:13 ` Chris Bainbridge
2016-03-31 23:13   ` Chris Bainbridge
2016-04-01  4:59   ` Michael Büsch
2016-04-01  4:59     ` Michael Büsch
2016-04-01 22:46   ` Lukas Wunner
2016-04-01 22:46     ` Lukas Wunner
2016-04-02  7:30     ` Michael Büsch
2016-04-02  7:30       ` Michael Büsch
2016-04-02 11:40 ` Andrew Worsley
2016-04-03 11:49   ` Lukas Wunner
2016-04-03 11:49     ` Lukas Wunner
2016-04-07 12:04     ` Andrew Worsley
2016-04-07 12:04       ` Andrew Worsley
2016-04-10 10:09       ` Andrew Worsley
2016-04-10 10:09         ` Andrew Worsley
2016-04-12 18:32         ` Lukas Wunner
2016-04-12 18:32           ` Lukas Wunner
2016-04-13 20:42           ` Andrew Worsley
2016-04-13 20:42             ` Andrew Worsley
2016-04-24 17:04             ` Lukas Wunner
2016-04-24 17:04               ` Lukas Wunner
2016-05-23 14:42               ` Lukas Wunner
2016-05-23 14:42                 ` Lukas Wunner
2016-05-24 23:38                 ` Chris Bainbridge
2016-05-24 23:38                   ` Chris Bainbridge
     [not found] <E1akxli-00030z-Jz@bombadil.infradead.org>
2016-03-31 19:09 ` Michael Büsch
2016-03-31 19:09   ` Michael Büsch
  -- strict thread matches above, loose matches on Subject: below --
2016-03-29 17:46 Lukas Wunner
2016-03-29 17:46 ` Lukas Wunner
2016-03-29 17:41 Lukas Wunner

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=20160409120055.GR2701@codeblueprint.co.uk \
    --to=matt@codeblueprint.co.uk \
    --cc=b43-dev@lists.infradead.org \
    --cc=helgaas@kernel.org \
    --cc=linux-efi@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-wireless@vger.kernel.org \
    --cc=lukas@wunner.de \
    --cc=mjg59@srcf.ucam.org \
    --cc=zajec5@gmail.com \
    /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.