lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Thu, 8 Mar 2012 17:43:36 -0700
From:	Bjorn Helgaas <bhelgaas@...gle.com>
To:	Yinghai Lu <yinghai@...nel.org>
Cc:	Bjorn Helgaas <bjorn.helgaas@...il.com>,
	Jesse Barnes <jbarnes@...tuousgeek.org>, x86 <x86@...nel.org>,
	linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 00/23] PCI, x86: pci root bus hotplug support

On Tue, Mar 6, 2012 at 11:58 PM, Yinghai Lu <yinghai@...nel.org> wrote:
> On Tue, Mar 6, 2012 at 8:44 PM, Bjorn Helgaas <bjorn.helgaas@...il.com> wrote:
>> On Tue, Mar 6, 2012 at 12:13 AM, Yinghai Lu <yinghai@...nel.org> wrote:
>> Does this series support hot-add, too, or just removal?
>>
>> If it is intended to support hot-add, I didn't see any support for
>> _CBA, which I *think* is the way hotplug-capable host bridges are
>> required to report MMCONFIG areas.
>
> to make the whole hotplug working with intel new cpus with IIO,
> in addition to physical cpu and memory hotplug, will still need
> a. ioapic controller hotplug
> b. intr-remapping hotplug
> c. intel-iommu hotplug
> d. mmconf support.
> e. root bus hotplug
>
> this patch set is only involved with partitally root bus removal/add back.
> and assume ioapic, intr-remap, intel-iommu, and mmconf related are
> working from early initialization.
>
> that will help us to help to fix bugs like devices refcount leaking.

OK, so this series only supports host bridge removal.  There are some
situations where it allows a bridge to be added, but only if the other
things you mentioned (ioapic, intr-remapping, etc.) are already
present, probably because we didn't remove them when removing the host
bridge.

In the case of MMCONFIG in particular, I think this should be moved
into the host bridge add path, e.g., in acpi_pci_root_add().  The fact
that it's totally separate today is a mistake.  But that can be fixed
later.

Overall nits, because kernel development is obsessive by nature:
Please format your changelogs with textwidth=75, so that when you read
them with "git log" in an 80-column terminal, the normal text doesn't
wrap across a line ending.

Please make your subject lines lines consistent in style and
capitalization.  Sometimes you have "PCI", sometimes "pci".  Sometimes
the first word after "PCI: " is capitalized, sometimes not.  One of
your intel-iommu.c patches is "IOMMU:"; the other is "pci, dmar:" (and
these patches are not adjacent in the series, even though they could
be).  This is just sloppy and distracting.

In English, the first word of a sentence is capitalized.  Words in the
middle are not, unless they are acronyms or proper nouns.  It would be
helpful if you could follow these conventions in your changelogs,
because I expect to be reading them for years to come, and I'm easily
distracted.

The notes about "-v2, -v3, -v4, etc." are not really useful in the
commit changelogs.  They're helpful in the [00/] message, but not in
the changelogs.

I appreciate the effort you're making to write changelogs.  They're
definitely starting to contain some useful information.

Bjorn
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ