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:   Wed, 16 Aug 2017 21:53:48 +0200
From:   Robert Richter <robert.richter@...ium.com>
To:     Marc Zyngier <marc.zyngier@....com>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Jason Cooper <jason@...edaemon.net>,
        Shanker Donthineni <shankerd@...eaurora.org>,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3 0/8] irqchip/gic-v3-its: Implement ITS nodes as kernel
 devices and use CMA

Marc,

On 08.08.17 14:22:50, Robert Richter wrote:
> This patch series implements ITS nodes as kernel devices. The
> advantage of that is that CMA can be used to allocate large ITS
> tables, where standard memory and page allocation fails (above
> MAX_ORDER - 1). This approach was suggested first here:
> 
>  https://marc.info/?i=56D44812.6000309%40arm.com
> 
> Another advantage is that all memory resources of the device are now
> device managed, thus the code for releasing the device becomes much
> easier.
> 
> The following is required to implement this:
> 
>  * each ITS node must be assigned to a struct device each,
> 
>  * the ITS initialization must be moved to a later point during boot,
> 
>  * the ITS enablement must be separated from the GIC code (the GIC
>    still needs to be enabled early),
> 
>  * the early ITS node probing by DT or ACPI must be separated from the
>    ITS initialization,
> 
>  * ITS table allocation must be changed to use dma_alloc_coherent()
>    which uses CMA for allocation of large memory ranges.
> 
> This is an update of my last patch submission here:
> 
>  https://lkml.org/lkml/2017/3/6/570
> 
> V3:
>  * rebased onto v4.13-rc1,
>  * fixed use after free in error path in its_init_one(),
>  * added comment for arch_setup_dma_ops() to describe coherency,

what is your opinion on this series?

There might be conflicts with your gicv4 series you sent around. Any
advice how we can move forward with this one?

Thanks,

-Robert

> V2:
>  * rebased onto v4.11-rc1,
>  * fixed syntax error in its_init() function (split probing patch),
>  * added comment in its_create_device(),
>  * fixed GITS_BASER_PAGE_SIZE_MASK usage in its_setup_baser()
> 
> Robert Richter (8):
>   irqchip/gic-v3-its: Initialize its nodes in probe order
>   irqchip/gic-v3-its: Initialize MSIs with subsys_initcalls
>   irqchip/gic-v3-its: Split probing from its node initialization
>   irqchip/gic-v3-its: Decouple its initialization from gic
>   irqchip/gic-v3-its: Prevent its init ordering dependencies
>   irqchip/gic-v3-its: Initialize its nodes later
>   irqchip/gic-v3-its: Handle its nodes as kernel devices
>   irqchip, gicv3-its, cma: Use CMA for allocation of large device tables
> 
>  drivers/irqchip/irq-gic-v3-its-pci-msi.c      |   6 +-
>  drivers/irqchip/irq-gic-v3-its-platform-msi.c |   6 +-
>  drivers/irqchip/irq-gic-v3-its.c              | 285 ++++++++++++++++----------
>  drivers/irqchip/irq-gic-v3.c                  |   8 +-
>  include/linux/cpuhotplug.h                    |   1 +
>  include/linux/irqchip/arm-gic-v3.h            |   5 +-
>  6 files changed, 195 insertions(+), 116 deletions(-)
> 
> -- 
> 2.11.0
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ