From mboxrd@z Thu Jan  1 00:00:00 1970
From: andre.przywara@arm.com (Andre Przywara)
Date: Fri, 2 Oct 2015 13:49:17 +0100
Subject: [PATCH v2 01/15] KVM: arm/arm64: VGIC: don't track used LRs in
 the distributor
In-Reply-To: <00f301d0fd0f$6b21ac10$41650430$@samsung.com>
References: <1436538111-4294-1-git-send-email-andre.przywara@arm.com>
 <1436538111-4294-2-git-send-email-andre.przywara@arm.com>
 <00cb01d0fcf8$6a906000$3fb12000$@samsung.com> <560E5D2F.1000401@arm.com>
 <00f301d0fd0f$6b21ac10$41650430$@samsung.com>
Message-ID: <560E7D4D.1090009@arm.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Hi Pavel,

On 02/10/15 13:39, Pavel Fedin wrote:
>  Hello!
> 
>> Can't you use the ELRSR bitmap instead? The idea of lr_used sounds like
>> a moot optimization to me.
> 
>  This perfectly works on 4.2, but will break HW interrupt forwarding on 4.3. If you look at 4.3
> __kvm_vgic_sync_hwstate(), you'll notice that for HW interrupts lr_used and elrsr_ptr will diverge
> at this point, and this function actually brings them into sync. And it relies on lr_used for the
> loop to operate correctly (no idea why we use "for" loop here with extra check instead of
> "for_each_set_bit(lr, vgic_cpu->lr_used, vgic->nr_lr)", looks stupid to me).

I know, because I have reworked my patch lately to work on top of 4.3-rc
and Christoffer's timer rework series. I have something which "works
now"(TM), but wanted to wait for Christoffer's respin to send out.
I will send you this new version this as a sneak preview in private,
maybe that helps.

Cheers,
Andre.