| Message ID | 20231019071611.98885-2-philmd@linaro.org |
|---|---|
| State | Superseded |
| Headers | show |
| Series | hw: Strengthen SysBus & QBus API | expand |
On Thu, Oct 19, 2023 at 09:15:59AM +0200, Philippe Mathieu-Daudé wrote: > Date: Thu, 19 Oct 2023 09:15:59 +0200 > From: Philippe Mathieu-Daudé <philmd@linaro.org> > Subject: [PATCH v2 01/12] hw/i386/amd_iommu: Do not use SysBus API to map > local MMIO region > X-Mailer: git-send-email 2.41.0 > > There is no point in exposing an internal MMIO region via > SysBus and directly mapping it in the very same device. > > Just map it without using the SysBus API. > > Transformation done using the following coccinelle script: > > @@ > expression sbdev; > expression index; > expression addr; > expression subregion; > @@ > - sysbus_init_mmio(sbdev, subregion); > ... when != sbdev > - sysbus_mmio_map(sbdev, index, addr); > + memory_region_add_subregion(get_system_memory(), addr, subregion); > > Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> > Reviewed-by: Michael S. Tsirkin <mst@redhat.com> > --- > hw/i386/amd_iommu.c | 5 ++--- > 1 file changed, 2 insertions(+), 3 deletions(-) Reviewed-by: Zhao Liu <zhao1.liu@intel.com> -Zhao > > diff --git a/hw/i386/amd_iommu.c b/hw/i386/amd_iommu.c > index 8d0f2f99dd..7965415b47 100644 > --- a/hw/i386/amd_iommu.c > +++ b/hw/i386/amd_iommu.c > @@ -1579,9 +1579,8 @@ static void amdvi_sysbus_realize(DeviceState *dev, Error **errp) > /* set up MMIO */ > memory_region_init_io(&s->mmio, OBJECT(s), &mmio_mem_ops, s, "amdvi-mmio", > AMDVI_MMIO_SIZE); > - > - sysbus_init_mmio(SYS_BUS_DEVICE(s), &s->mmio); > - sysbus_mmio_map(SYS_BUS_DEVICE(s), 0, AMDVI_BASE_ADDR); > + memory_region_add_subregion(get_system_memory(), AMDVI_BASE_ADDR, > + &s->mmio); > pci_setup_iommu(bus, amdvi_host_dma_iommu, s); > amdvi_init(s); > } > -- > 2.41.0 > > >
diff --git a/hw/i386/amd_iommu.c b/hw/i386/amd_iommu.c index 8d0f2f99dd..7965415b47 100644 --- a/hw/i386/amd_iommu.c +++ b/hw/i386/amd_iommu.c @@ -1579,9 +1579,8 @@ static void amdvi_sysbus_realize(DeviceState *dev, Error **errp) /* set up MMIO */ memory_region_init_io(&s->mmio, OBJECT(s), &mmio_mem_ops, s, "amdvi-mmio", AMDVI_MMIO_SIZE); - - sysbus_init_mmio(SYS_BUS_DEVICE(s), &s->mmio); - sysbus_mmio_map(SYS_BUS_DEVICE(s), 0, AMDVI_BASE_ADDR); + memory_region_add_subregion(get_system_memory(), AMDVI_BASE_ADDR, + &s->mmio); pci_setup_iommu(bus, amdvi_host_dma_iommu, s); amdvi_init(s); }
There is no point in exposing an internal MMIO region via SysBus and directly mapping it in the very same device. Just map it without using the SysBus API. Transformation done using the following coccinelle script: @@ expression sbdev; expression index; expression addr; expression subregion; @@ - sysbus_init_mmio(sbdev, subregion); ... when != sbdev - sysbus_mmio_map(sbdev, index, addr); + memory_region_add_subregion(get_system_memory(), addr, subregion); Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> --- hw/i386/amd_iommu.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)