Hi Liviu, Please ignore last email and please find the latest feedback inline as below.
On Wed, Jun 05, 2019 at 07:19:37PM +0800, Liviu Dudau wrote: > Hi Lowry, > > On Tue, Apr 30, 2019 at 07:19:29AM +0100, Lowry Li (Arm Technology China) > wrote: > > Adds iommu_connect and disconnect for SMMU support, and configures > > TBU translation once SMMU has been attached to the display device. > > > > Signed-off-by: Lowry Li (Arm Technology China) <lowry...@arm.com> > > --- > > .../gpu/drm/arm/display/komeda/d71/d71_component.c | 5 +++ > > drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c | 49 > > ++++++++++++++++++++++ > > drivers/gpu/drm/arm/display/komeda/komeda_dev.c | 17 ++++++++ > > drivers/gpu/drm/arm/display/komeda/komeda_dev.h | 7 ++++ > > .../drm/arm/display/komeda/komeda_framebuffer.c | 2 + > > .../drm/arm/display/komeda/komeda_framebuffer.h | 2 + > > 6 files changed, 82 insertions(+) > > > > diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c > > b/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c > > index 33ca171..9065040 100644 > > --- a/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c > > +++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_component.c > > @@ -215,6 +215,8 @@ static void d71_layer_update(struct komeda_component *c, > > malidp_write32(reg, LAYER_FMT, kfb->format_caps->hw_id); > > malidp_write32(reg, BLK_IN_SIZE, HV_SIZE(st->hsize, st->vsize)); > > > > + if (kfb->is_va) > > + ctrl |= L_TBU_EN; > > malidp_write32_mask(reg, BLK_CONTROL, ctrl_mask, ctrl); > > } > > > > @@ -348,6 +350,9 @@ static void d71_wb_layer_update(struct komeda_component > > *c, > > fb->pitches[i] & 0xFFFF); > > } > > > > + if (kfb->is_va) > > + ctrl |= LW_TBU_EN; > > + > > malidp_write32(reg, LAYER_FMT, kfb->format_caps->hw_id); > > malidp_write32(reg, BLK_IN_SIZE, HV_SIZE(st->hsize, st->vsize)); > > malidp_write32(reg, BLK_INPUT_ID0, to_d71_input_id(&state->inputs[0])); > > diff --git a/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c > > b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c > > index 9603de9..45c98a7 100644 > > --- a/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c > > +++ b/drivers/gpu/drm/arm/display/komeda/d71/d71_dev.c > > @@ -517,6 +517,53 @@ static void d71_init_fmt_tbl(struct komeda_dev *mdev) > > table->n_formats = ARRAY_SIZE(d71_format_caps_table); > > } > > > > +static int d71_connect_iommu(struct komeda_dev *mdev) > > +{ > > + struct d71_dev *d71 = mdev->chip_data; > > + u32 __iomem *reg = d71->gcu_addr; > > + u32 check_bits = (d71->num_pipelines == 2) ? > > + GCU_STATUS_TCS0 | GCU_STATUS_TCS1 : GCU_STATUS_TCS0; > > + int i, ret; > > + > > + if (!d71->integrates_tbu) > > + return -1; > > + > > + malidp_write32_mask(reg, BLK_CONTROL, 0x7, TBU_CONNECT_MODE); > > + > > + ret = dp_wait_cond(has_bits(check_bits, malidp_read32(reg, BLK_STATUS)), > > + 100, 1000, 1000); > > + if (ret <= 0) { > > + DRM_ERROR("connect to TCU timeout!\n"); > > + malidp_write32_mask(reg, BLK_CONTROL, 0x7, INACTIVE_MODE); > > + return -ETIMEDOUT; > > + } > > + > > + for (i = 0; i < d71->num_pipelines; i++) > > + malidp_write32_mask(d71->pipes[i]->lpu_addr, LPU_TBU_CONTROL, > > + LPU_TBU_CTRL_TLBPEN, LPU_TBU_CTRL_TLBPEN); > > + return 0; > > +} > > + > > +static int d71_disconnect_iommu(struct komeda_dev *mdev) > > +{ > > + struct d71_dev *d71 = mdev->chip_data; > > + u32 __iomem *reg = d71->gcu_addr; > > + u32 check_bits = (d71->num_pipelines == 2) ? > > + GCU_STATUS_TCS0 | GCU_STATUS_TCS1 : GCU_STATUS_TCS0; > > + int ret; > > + > > + malidp_write32_mask(reg, BLK_CONTROL, 0x7, TBU_DISCONNECT_MODE); > > + > > + ret = dp_wait_cond(((malidp_read32(reg, BLK_STATUS) & check_bits) == 0), > > + 100, 1000, 1000); > > + if (ret <= 0) { > > + DRM_ERROR("disconnect from TCU timeout!\n"); > > + malidp_write32_mask(reg, BLK_CONTROL, 0x7, INACTIVE_MODE); > > + } > > + > > + return ret > 0 ? 0 : -1; > > For consistency with d71_connect_iommu() you should return -ETIMEDOUT as well. > I'm sending a patch to convert dp_wait_cond() to return that when it times > out, > so we can be consistent in the future. > Yes, will change to use -ETIMEDOUT. > > +} > > + > > static struct komeda_dev_funcs d71_chip_funcs = { > > .init_format_table = d71_init_fmt_tbl, > > .enum_resources = d71_enum_resources, > > @@ -527,6 +574,8 @@ static void d71_init_fmt_tbl(struct komeda_dev *mdev) > > .on_off_vblank = d71_on_off_vblank, > > .change_opmode = d71_change_opmode, > > .flush = d71_flush, > > + .connect_iommu = d71_connect_iommu, > > + .disconnect_iommu = d71_disconnect_iommu, > > }; > > > > struct komeda_dev_funcs * > > diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c > > b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c > > index e4e5b58..2d97c82 100644 > > --- a/drivers/gpu/drm/arm/display/komeda/komeda_dev.c > > +++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.c > > @@ -251,6 +251,18 @@ struct komeda_dev *komeda_dev_create(struct device > > *dev) > > dev->dma_parms = &mdev->dma_parms; > > dma_set_max_seg_size(dev, DMA_BIT_MASK(32)); > > > > + mdev->iommu = iommu_get_domain_for_dev(mdev->dev); > > + if (!mdev->iommu) > > + DRM_INFO("continue without IOMMU support!\n"); > > + > > + if (mdev->iommu && mdev->funcs->connect_iommu) { > > + err = mdev->funcs->connect_iommu(mdev); > > + if (err) { > > + DRM_ERROR("connect iommu failed.\n"); > > connect_iommu() has already printed a DRM_ERROR(), do we need another one > here? > This log printing is necessary at core layer for the case that some other chip layer code doesn't have error messages. So suggest keeping this code here. > > + goto err_cleanup; > > err_cleanup doesn't set the new mdev->iommu to NULL, forgot to add it? > > Sorry, I forgot adding this. Will set to NULL here once connected failed. Thanks again. > > + } > > + } > > + > > err = sysfs_create_group(&dev->kobj, &komeda_sysfs_attr_group); > > if (err) { > > DRM_ERROR("create sysfs group failed.\n"); > > @@ -280,6 +292,11 @@ void komeda_dev_destroy(struct komeda_dev *mdev) > > debugfs_remove_recursive(mdev->debugfs_root); > > #endif > > > > + if (mdev->iommu && mdev->funcs->disconnect_iommu) > > + if (mdev->funcs->disconnect_iommu(mdev)) > > + DRM_ERROR("disconnect iommu failed.\n"); > > Same comment an in komeda_dev_create(), disconnect_iommu() already printed a > DRM_ERROR() > > > + mdev->iommu = NULL; > > + > > for (i = 0; i < mdev->n_pipelines; i++) { > > komeda_pipeline_destroy(mdev, mdev->pipelines[i]); > > mdev->pipelines[i] = NULL; > > diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_dev.h > > b/drivers/gpu/drm/arm/display/komeda/komeda_dev.h > > index 83ace71..dac1eda 100644 > > --- a/drivers/gpu/drm/arm/display/komeda/komeda_dev.h > > +++ b/drivers/gpu/drm/arm/display/komeda/komeda_dev.h > > @@ -92,6 +92,10 @@ struct komeda_dev_funcs { > > int (*enum_resources)(struct komeda_dev *mdev); > > /** @cleanup: call to chip to cleanup komeda_dev->chip data */ > > void (*cleanup)(struct komeda_dev *mdev); > > + /** @connect_iommu: Optional, connect to external iommu */ > > + int (*connect_iommu)(struct komeda_dev *mdev); > > + /** @disconnect_iommu: Optional, disconnect to external iommu */ > > + int (*disconnect_iommu)(struct komeda_dev *mdev); > > /** > > * @irq_handler: > > * > > @@ -184,6 +188,9 @@ struct komeda_dev { > > */ > > void *chip_data; > > > > + /** @iommu: iommu domain */ > > + struct iommu_domain *iommu; > > + > > /** @debugfs_root: root directory of komeda debugfs */ > > struct dentry *debugfs_root; > > }; > > diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.c > > b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.c > > index d5822a3..360ab70 100644 > > --- a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.c > > +++ b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.c > > @@ -201,6 +201,8 @@ struct drm_framebuffer * > > goto err_cleanup; > > } > > > > + kfb->is_va = mdev->iommu ? true : false; > > + > > return &kfb->base; > > > > err_cleanup: > > diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h > > b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h > > index 6cbb2f6..f4046e2 100644 > > --- a/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h > > +++ b/drivers/gpu/drm/arm/display/komeda/komeda_framebuffer.h > > @@ -21,6 +21,8 @@ struct komeda_fb { > > * extends drm_format_info for komeda specific information > > */ > > const struct komeda_format_caps *format_caps; > > + /** @is_va: if smmu is enabled, it will be true */ > > + bool is_va; > > /** @aligned_w: aligned frame buffer width */ > > u32 aligned_w; > > /** @aligned_h: aligned frame buffer height */ > > -- > > 1.9.1 > > > > Best regards, > Liviu > > > -- > ==================== > | I would like to | > | fix the world, | > | but they're not | > | giving me the | > \ source code! / > --------------- > ¯\_(ツ)_/¯ -- Regards, Lowry _______________________________________________ dri-devel mailing list dri-devel@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/dri-devel