[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20080712055444.GA13420@elte.hu>
Date: Sat, 12 Jul 2008 07:54:44 +0200
From: Ingo Molnar <mingo@...e.hu>
To: Matthew Wilcox <matthew@....cx>
Cc: linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
grundler@...isc-linux.org, tglx@...utronix.de, jgarzik@...ox.com,
linux-ide@...r.kernel.org, suresh.b.siddha@...el.com,
benh@...nel.crashing.org, jbarnes@...tuousgeek.org,
rdunlap@...otime.net, mtk.manpages@...il.com,
ebiederm@...ssion.com, Matthew Wilcox <willy@...ux.intel.com>,
Yinghai Lu <yhlu.kernel@...il.com>
Subject: Re: [PATCH 6/6] x86-64: Support for multiple MSIs
* Matthew Wilcox <matthew@....cx> wrote:
> Add support for allocating an aligned block of interrupt vectors.
> Allow interrupts to have up to 32 subchannels. Implement the
> arch_setup_msi_irqs() and arch_teardown_msi_irqs() interfaces.
>
> Signed-off-by: Matthew Wilcox <willy@...ux.intel.com>
> ---
> arch/x86/kernel/io_apic_64.c | 221 +++++++++++++++++++++++++++++++++++------
> arch/x86/kernel/irq_64.c | 2 +-
> include/asm-x86/irq_64.h | 2 +
> 3 files changed, 191 insertions(+), 34 deletions(-)
hm, please implement this symmetrically on 64-bit and 32-bit as well.
That will need more IRQ infrastructure changes but that should be OK as
it will reduce the gap between the 32-bit and 64-bit side instead of
widening it. A good starting point would be to unify/generalize the IRQ
vector allocators. (and that would get most of what you need for
multi-vector MSI)
Ingo
--
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