| Message ID | 1544258371-4600-9-git-send-email-yong.wu@mediatek.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | MT8183 IOMMU SUPPORT | expand |
On Sat, Dec 8, 2018 at 4:42 PM Yong Wu <yong.wu@mediatek.com> wrote: > > The larb-id may be remapped in the smi-common, this means the > larb-id reported in the mtk_iommu_isr isn't the real larb-id, > > Take mt8183 as a example: > M4U > | > --------------------------------------------- > | SMI common | > -0-----7-----5-----6-----1-----2------3-----4- <- Id remapped > | | | | | | | | > larb0 larb1 IPU0 IPU1 larb4 larb5 larb6 CCU > disp vdec img cam venc img cam > As above, larb0 connects with the id 0 in smi-common. > larb1 connects with the id 7 in smi-common. > ... > If the larb-id reported in the isr is 7, actually it's larb1(vdec). > In order to output the right larb-id in the isr, we add a larb-id > remapping relationship in this patch. > > This also is a preparing patch for mt8183. > > Signed-off-by: Yong Wu <yong.wu@mediatek.com> > --- > drivers/iommu/mtk_iommu.c | 3 +++ > drivers/iommu/mtk_iommu.h | 4 ++++ > 2 files changed, 7 insertions(+) > > diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c > index eda062a..8ab3b69 100644 > --- a/drivers/iommu/mtk_iommu.c > +++ b/drivers/iommu/mtk_iommu.c > @@ -220,6 +220,9 @@ static irqreturn_t mtk_iommu_isr(int irq, void *dev_id) > fault_larb = F_MMU0_INT_ID_LARB_ID(regval); > fault_port = F_MMU0_INT_ID_PORT_ID(regval); > > + if (data->plat_data->larbid_remap_enable) > + fault_larb = data->plat_data->larbid_remapped[fault_larb]; > + > if (report_iommu_fault(&dom->domain, data->dev, fault_iova, > write ? IOMMU_FAULT_WRITE : IOMMU_FAULT_READ)) { > dev_err_ratelimited( > diff --git a/drivers/iommu/mtk_iommu.h b/drivers/iommu/mtk_iommu.h > index b8749ac..3877050 100644 > --- a/drivers/iommu/mtk_iommu.h > +++ b/drivers/iommu/mtk_iommu.h > @@ -47,6 +47,10 @@ struct mtk_iommu_plat_data { > > /* HW will use the EMI clock if there isn't the "bclk". */ > bool has_bclk; > + > + /* The larb-id may be remapped in the smi-common. */ > + bool larbid_remap_enable; > + unsigned int larbid_remapped[MTK_LARB_NR_MAX]; Wouldn't it be a little simpler if you just had larbid_remap[MTK_LARB_NR_MAX] (no larbid_remap_enable), and just set it to {0, 1, 2, 3, 4, 5, 6, 7} in platforms that don't need complicated remapping? Also, unsigned char/u8 array would be enough. > }; > > struct mtk_iommu_domain; > -- > 1.9.1 >
Hi Nicolas, Thanks for the review of this patchset. On Fri, 2018-12-21 at 11:35 +0800, Nicolas Boichat wrote: > On Sat, Dec 8, 2018 at 4:42 PM Yong Wu <yong.wu@mediatek.com> wrote: > > > > The larb-id may be remapped in the smi-common, this means the > > larb-id reported in the mtk_iommu_isr isn't the real larb-id, > > > > Take mt8183 as a example: > > M4U > > | > > --------------------------------------------- > > | SMI common | > > -0-----7-----5-----6-----1-----2------3-----4- <- Id remapped > > | | | | | | | | > > larb0 larb1 IPU0 IPU1 larb4 larb5 larb6 CCU > > disp vdec img cam venc img cam > > As above, larb0 connects with the id 0 in smi-common. > > larb1 connects with the id 7 in smi-common. > > ... > > If the larb-id reported in the isr is 7, actually it's larb1(vdec). > > In order to output the right larb-id in the isr, we add a larb-id > > remapping relationship in this patch. > > > > This also is a preparing patch for mt8183. > > > > Signed-off-by: Yong Wu <yong.wu@mediatek.com> > > --- > > drivers/iommu/mtk_iommu.c | 3 +++ > > drivers/iommu/mtk_iommu.h | 4 ++++ > > 2 files changed, 7 insertions(+) > > > > diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c > > index eda062a..8ab3b69 100644 > > --- a/drivers/iommu/mtk_iommu.c > > +++ b/drivers/iommu/mtk_iommu.c > > @@ -220,6 +220,9 @@ static irqreturn_t mtk_iommu_isr(int irq, void *dev_id) > > fault_larb = F_MMU0_INT_ID_LARB_ID(regval); > > fault_port = F_MMU0_INT_ID_PORT_ID(regval); > > > > + if (data->plat_data->larbid_remap_enable) > > + fault_larb = data->plat_data->larbid_remapped[fault_larb]; > > + > > if (report_iommu_fault(&dom->domain, data->dev, fault_iova, > > write ? IOMMU_FAULT_WRITE : IOMMU_FAULT_READ)) { > > dev_err_ratelimited( > > diff --git a/drivers/iommu/mtk_iommu.h b/drivers/iommu/mtk_iommu.h > > index b8749ac..3877050 100644 > > --- a/drivers/iommu/mtk_iommu.h > > +++ b/drivers/iommu/mtk_iommu.h > > @@ -47,6 +47,10 @@ struct mtk_iommu_plat_data { > > > > /* HW will use the EMI clock if there isn't the "bclk". */ > > bool has_bclk; > > + > > + /* The larb-id may be remapped in the smi-common. */ > > + bool larbid_remap_enable; > > + unsigned int larbid_remapped[MTK_LARB_NR_MAX]; > > Wouldn't it be a little simpler if you just had > larbid_remap[MTK_LARB_NR_MAX] (no larbid_remap_enable), and just set > it to {0, 1, 2, 3, 4, 5, 6, 7} in platforms that don't need > complicated remapping? Actually I'd like the original way(Print the larb-id from the register directly if larb-id is not remapped). But this solution is also ok for me. > > Also, unsigned char/u8 array would be enough. OK. "unsigned char" is enough. Originally I think "int" or "long" may be better to access in ARM/ARM64. > > > }; > > > > struct mtk_iommu_domain; > > -- > > 1.9.1 > >
diff --git a/drivers/iommu/mtk_iommu.c b/drivers/iommu/mtk_iommu.c index eda062a..8ab3b69 100644 --- a/drivers/iommu/mtk_iommu.c +++ b/drivers/iommu/mtk_iommu.c @@ -220,6 +220,9 @@ static irqreturn_t mtk_iommu_isr(int irq, void *dev_id) fault_larb = F_MMU0_INT_ID_LARB_ID(regval); fault_port = F_MMU0_INT_ID_PORT_ID(regval); + if (data->plat_data->larbid_remap_enable) + fault_larb = data->plat_data->larbid_remapped[fault_larb]; + if (report_iommu_fault(&dom->domain, data->dev, fault_iova, write ? IOMMU_FAULT_WRITE : IOMMU_FAULT_READ)) { dev_err_ratelimited( diff --git a/drivers/iommu/mtk_iommu.h b/drivers/iommu/mtk_iommu.h index b8749ac..3877050 100644 --- a/drivers/iommu/mtk_iommu.h +++ b/drivers/iommu/mtk_iommu.h @@ -47,6 +47,10 @@ struct mtk_iommu_plat_data { /* HW will use the EMI clock if there isn't the "bclk". */ bool has_bclk; + + /* The larb-id may be remapped in the smi-common. */ + bool larbid_remap_enable; + unsigned int larbid_remapped[MTK_LARB_NR_MAX]; }; struct mtk_iommu_domain;
The larb-id may be remapped in the smi-common, this means the larb-id reported in the mtk_iommu_isr isn't the real larb-id, Take mt8183 as a example: M4U | --------------------------------------------- | SMI common | -0-----7-----5-----6-----1-----2------3-----4- <- Id remapped | | | | | | | | larb0 larb1 IPU0 IPU1 larb4 larb5 larb6 CCU disp vdec img cam venc img cam As above, larb0 connects with the id 0 in smi-common. larb1 connects with the id 7 in smi-common. ... If the larb-id reported in the isr is 7, actually it's larb1(vdec). In order to output the right larb-id in the isr, we add a larb-id remapping relationship in this patch. This also is a preparing patch for mt8183. Signed-off-by: Yong Wu <yong.wu@mediatek.com> --- drivers/iommu/mtk_iommu.c | 3 +++ drivers/iommu/mtk_iommu.h | 4 ++++ 2 files changed, 7 insertions(+)