Message-ID: <22faee018a36a49e4a507b69d087432c8cd689ec.camel@mediatek.com>
Subject: Re: [PATCH v3 09/33] iommu/mediatek: Remove for_each_m4u in
 tlb_sync_all
From:   Yong Wu <yong.wu@mediatek.com>
To:     AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@collabora.com>
CC:     Joerg Roedel <joro@8bytes.org>, Rob Herring <robh+dt@kernel.org>,
        "Matthias Brugger" <matthias.bgg@gmail.com>,
        Will Deacon <will@kernel.org>,
        Robin Murphy <robin.murphy@arm.com>,
        Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>,
        Tomasz Figa <tfiga@chromium.org>,
        <linux-mediatek@lists.infradead.org>,
        <srv_heupstream@mediatek.com>, <devicetree@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>,
        <linux-arm-kernel@lists.infradead.org>,
        <iommu@lists.linux-foundation.org>,
        Hsin-Yi Wang <hsinyi@chromium.org>, <youlin.pei@mediatek.com>,
        <anan.sun@mediatek.com>, <chao.hao@mediatek.com>,
        <yen-chang.chen@mediatek.com>
Date:   Sun, 9 Jan 2022 10:48:27 +0800
In-Reply-To: <bfa33e94-c2e5-5dab-c9af-b674e1669daa@collabora.com>
References: <20210923115840.17813-1-yong.wu@mediatek.com>
         <20210923115840.17813-10-yong.wu@mediatek.com>
         <bfa33e94-c2e5-5dab-c9af-b674e1669daa@collabora.com>
Content-Type: text/plain; charset="UTF-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Tue, 2022-01-04 at 16:55 +0100, AngeloGioacchino Del Regno wrote:
> Il 23/09/21 13:58, Yong Wu ha scritto:
> > The tlb_sync_all is called from these three functions:
> > a) flush_iotlb_all: it will be called for each a iommu HW.
> > b) tlb_flush_range_sync: it already has for_each_m4u.
> > c) in irq: When IOMMU HW translation fault, Only need flush itself.
> > 
> > Thus, No need for_each_m4u in this tlb_sync_all. Remove it.
> > 
> > Signed-off-by: Yong Wu <yong.wu@mediatek.com>
> > ---
> >   drivers/iommu/mtk_iommu.c | 18 +++++++-----------
> >   1 file changed, 7 insertions(+), 11 deletions(-)
> > 
> > diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c
> > index 6f4f6624e3ac..0b4c30baa864 100644
> > --- a/drivers/iommu/mtk_iommu.c
> > +++ b/drivers/iommu/mtk_iommu.c
> > @@ -206,19 +206,15 @@ static struct mtk_iommu_domain
> > *to_mtk_domain(struct iommu_domain *dom)
> >   
> >   static void mtk_iommu_tlb_flush_all(struct mtk_iommu_data *data)
> >   {
> > -	struct list_head *head = data->hw_list;
> > -
> > -	for_each_m4u(data, head) {
> > -		if (pm_runtime_get_if_in_use(data->dev) <= 0)
> > -			continue;
> > +	if (pm_runtime_get_if_in_use(data->dev) <= 0)
> > +		return;
> >   
> > -		writel_relaxed(F_INVLD_EN1 | F_INVLD_EN0,
> > -			       data->base + data->plat_data-
> > >inv_sel_reg);
> > -		writel_relaxed(F_ALL_INVLD, data->base +
> > REG_MMU_INVALIDATE);
> > -		wmb(); /* Make sure the tlb flush all done */
> > +	writel_relaxed(F_INVLD_EN1 | F_INVLD_EN0,
> > +		       data->base + data->plat_data->inv_sel_reg);
> > +	writel_relaxed(F_ALL_INVLD, data->base + REG_MMU_INVALIDATE);
> > +	wmb(); /* Make sure the tlb flush all done */
> 
> There aren't a lot of writes here - not anymore, since you are no
> longer doing
> this for_each_m4u()...
> ...so, please change writel_relaxed() to writel() calls, allowing you
> to also
> remove the write barrier at the end (since in the non relaxed
> version, order is already ensured).

In the "writel", the "__iowmb()" runs before "write_relaxed". Then how
to guarantee the last register was wrote into the HW. Here the flush
all don't have sync(waiting it complete)

> 
> >   
> > -		pm_runtime_put(data->dev);
> > -	}
> > +	pm_runtime_put(data->dev);
> >   }
> >   
> >   static void mtk_iommu_tlb_flush_range_sync(unsigned long iova,
> > size_t size,
> >