| Message ID | 1432159758-4486-2-git-send-email-Suravee.Suthikulpanit@amd.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Wednesday, May 20, 2015 05:09:14 PM Suravee Suthikulpanit wrote: > This patch implements support for ACPI _CCA object, which is introduced in > ACPIv5.1, can be used for specifying device DMA coherency attribute. > > The parsing logic traverses device namespace to parse coherency > information, and stores it in acpi_device_flags. Then uses it to call > arch_setup_dma_ops() when creating each device enumerated in DSDT > during ACPI scan. > > This patch also introduces acpi_dma_is_coherent(), which provides > an interface for device drivers to check the coherency information > similarly to the of_dma_is_coherent(). > > Signed-off-by: Mark Salter <msalter@redhat.com> > Signed-off-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> > --- > drivers/acpi/Kconfig | 3 +++ > drivers/acpi/acpi_platform.c | 2 +- > drivers/acpi/glue.c | 5 +++++ > drivers/acpi/scan.c | 35 +++++++++++++++++++++++++++++++++++ > include/acpi/acpi_bus.h | 37 ++++++++++++++++++++++++++++++++++++- > include/linux/acpi.h | 5 +++++ > 6 files changed, 85 insertions(+), 2 deletions(-) > > diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig > index ab2cbb5..212735f 100644 > --- a/drivers/acpi/Kconfig > +++ b/drivers/acpi/Kconfig > @@ -54,6 +54,9 @@ config ACPI_GENERIC_GSI > config ACPI_SYSTEM_POWER_STATES_SUPPORT > bool > > +config ACPI_CCA_REQUIRED > + bool > + > config ACPI_SLEEP > bool > depends on SUSPEND || HIBERNATION > diff --git a/drivers/acpi/acpi_platform.c b/drivers/acpi/acpi_platform.c > index 4bf7559..06a67d5 100644 > --- a/drivers/acpi/acpi_platform.c > +++ b/drivers/acpi/acpi_platform.c > @@ -103,7 +103,7 @@ struct platform_device *acpi_create_platform_device(struct acpi_device *adev) > pdevinfo.res = resources; > pdevinfo.num_res = count; > pdevinfo.fwnode = acpi_fwnode_handle(adev); > - pdevinfo.dma_mask = DMA_BIT_MASK(32); > + pdevinfo.dma_mask = acpi_check_dma(adev, NULL) ? DMA_BIT_MASK(32) : 0; > pdev = platform_device_register_full(&pdevinfo); > if (IS_ERR(pdev)) > dev_err(&adev->dev, "platform device creation failed: %ld\n", > diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c > index 39c485b..b9657af 100644 > --- a/drivers/acpi/glue.c > +++ b/drivers/acpi/glue.c > @@ -13,6 +13,7 @@ > #include <linux/slab.h> > #include <linux/rwsem.h> > #include <linux/acpi.h> > +#include <linux/dma-mapping.h> > > #include "internal.h" > > @@ -167,6 +168,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) > struct list_head *physnode_list; > unsigned int node_id; > int retval = -EINVAL; > + bool coherent; > > if (has_acpi_companion(dev)) { > if (acpi_dev) { > @@ -223,6 +225,9 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) > if (!has_acpi_companion(dev)) > ACPI_COMPANION_SET(dev, acpi_dev); > > + if (acpi_check_dma(acpi_dev, &coherent)) > + arch_setup_dma_ops(dev, 0, 0, NULL, coherent); > + Well, so is this going to work for PCI too after all? > acpi_physnode_link_name(physical_node_name, node_id); > retval = sysfs_create_link(&acpi_dev->dev.kobj, &dev->kobj, > physical_node_name);
Not sure if this went out earlier. So I am resending. On 5/22/15 16:56, Rafael J. Wysocki wrote: >> diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c >> >index 39c485b..b9657af 100644 >> >--- a/drivers/acpi/glue.c >> >+++ b/drivers/acpi/glue.c >> >@@ -13,6 +13,7 @@ >> > #include <linux/slab.h> >> > #include <linux/rwsem.h> >> > #include <linux/acpi.h> >> >+#include <linux/dma-mapping.h> >> > >> > #include "internal.h" >> > >> >@@ -167,6 +168,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) >> > struct list_head *physnode_list; >> > unsigned int node_id; >> > int retval = -EINVAL; >> >+ bool coherent; >> > >> > if (has_acpi_companion(dev)) { >> > if (acpi_dev) { >> >@@ -223,6 +225,9 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) >> > if (!has_acpi_companion(dev)) >> > ACPI_COMPANION_SET(dev, acpi_dev); >> > >> >+ if (acpi_check_dma(acpi_dev, &coherent)) >> >+ arch_setup_dma_ops(dev, 0, 0, NULL, coherent); >> >+ > Well, so is this going to work for PCI too after all? > No, as Bjorn suggested, PCI changes for setting DMA coherent from _CCA (patch 3/6 in V4) will be submitted separately. We are working on cleaning up and up-streaming the PCI ACPI support for ARM64. Thanks, Suravee
On Friday, May 22, 2015 05:24:15 PM Suravee Suthikulanit wrote: > Not sure if this went out earlier. So I am resending. > > On 5/22/15 16:56, Rafael J. Wysocki wrote: > >> diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c > >> >index 39c485b..b9657af 100644 > >> >--- a/drivers/acpi/glue.c > >> >+++ b/drivers/acpi/glue.c > >> >@@ -13,6 +13,7 @@ > >> > #include <linux/slab.h> > >> > #include <linux/rwsem.h> > >> > #include <linux/acpi.h> > >> >+#include <linux/dma-mapping.h> > >> > > >> > #include "internal.h" > >> > > >> >@@ -167,6 +168,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) > >> > struct list_head *physnode_list; > >> > unsigned int node_id; > >> > int retval = -EINVAL; > >> >+ bool coherent; > >> > > >> > if (has_acpi_companion(dev)) { > >> > if (acpi_dev) { > >> >@@ -223,6 +225,9 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) > >> > if (!has_acpi_companion(dev)) > >> > ACPI_COMPANION_SET(dev, acpi_dev); > >> > > >> >+ if (acpi_check_dma(acpi_dev, &coherent)) > >> >+ arch_setup_dma_ops(dev, 0, 0, NULL, coherent); > >> >+ > > Well, so is this going to work for PCI too after all? > > > > No, as Bjorn suggested, PCI changes for setting DMA coherent from _CCA > (patch 3/6 in V4) will be submitted separately. We are working on > cleaning up and up-streaming the PCI ACPI support for ARM64. OK, but acpi_bind_one() is called for PCI devices too. Won't that be a problem?
On 5/22/2015 6:05 PM, Rafael J. Wysocki wrote: > On Friday, May 22, 2015 05:24:15 PM Suravee Suthikulanit wrote: >> Not sure if this went out earlier. So I am resending. >> >> On 5/22/15 16:56, Rafael J. Wysocki wrote: >>>> diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c >>>>> index 39c485b..b9657af 100644 >>>>> --- a/drivers/acpi/glue.c >>>>> +++ b/drivers/acpi/glue.c >>>>> @@ -13,6 +13,7 @@ >>>>> #include <linux/slab.h> >>>>> #include <linux/rwsem.h> >>>>> #include <linux/acpi.h> >>>>> +#include <linux/dma-mapping.h> >>>>> >>>>> #include "internal.h" >>>>> >>>>> @@ -167,6 +168,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) >>>>> struct list_head *physnode_list; >>>>> unsigned int node_id; >>>>> int retval = -EINVAL; >>>>> + bool coherent; >>>>> >>>>> if (has_acpi_companion(dev)) { >>>>> if (acpi_dev) { >>>>> @@ -223,6 +225,9 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) >>>>> if (!has_acpi_companion(dev)) >>>>> ACPI_COMPANION_SET(dev, acpi_dev); >>>>> >>>>> + if (acpi_check_dma(acpi_dev, &coherent)) >>>>> + arch_setup_dma_ops(dev, 0, 0, NULL, coherent); >>>>> + >>> Well, so is this going to work for PCI too after all? >>> >> >> No, as Bjorn suggested, PCI changes for setting DMA coherent from _CCA >> (patch 3/6 in V4) will be submitted separately. We are working on >> cleaning up and up-streaming the PCI ACPI support for ARM64. > > OK, but acpi_bind_one() is called for PCI devices too. Won't that be a problem? > > > In this case, we would be going through the following call path: --> pci_device_add() |--> pci_dma_configure() ** 1 ** |--> device_add() |--> platform_notify() |--> acpi_platform_notify() |--> acpi_bind_one() ** 2 ** At (1), we would be calling arch_setup_dma_ops() with the PCI host bridge _CCA information. So, it should have already taken care of setting up device coherency here. At (2), if there is no acpi_dev for endpoint devices (which I believe this is normally the case), it would return early and skip arch_setup_dma_ops(). At (2), if there is an acpi_dev, the coherent_dma flag should have already been setup by the acpi_init_device_object during ACPI scan. However, I am not certain about this case since I don't have the DSDT containing PCI endpoint devices to test with. Thanks, Suravee
On Friday, May 22, 2015 07:15:17 PM Suravee Suthikulanit wrote: > On 5/22/2015 6:05 PM, Rafael J. Wysocki wrote: > > On Friday, May 22, 2015 05:24:15 PM Suravee Suthikulanit wrote: > >> Not sure if this went out earlier. So I am resending. > >> > >> On 5/22/15 16:56, Rafael J. Wysocki wrote: > >>>> diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c > >>>>> index 39c485b..b9657af 100644 > >>>>> --- a/drivers/acpi/glue.c > >>>>> +++ b/drivers/acpi/glue.c > >>>>> @@ -13,6 +13,7 @@ > >>>>> #include <linux/slab.h> > >>>>> #include <linux/rwsem.h> > >>>>> #include <linux/acpi.h> > >>>>> +#include <linux/dma-mapping.h> > >>>>> > >>>>> #include "internal.h" > >>>>> > >>>>> @@ -167,6 +168,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) > >>>>> struct list_head *physnode_list; > >>>>> unsigned int node_id; > >>>>> int retval = -EINVAL; > >>>>> + bool coherent; > >>>>> > >>>>> if (has_acpi_companion(dev)) { > >>>>> if (acpi_dev) { > >>>>> @@ -223,6 +225,9 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) > >>>>> if (!has_acpi_companion(dev)) > >>>>> ACPI_COMPANION_SET(dev, acpi_dev); > >>>>> > >>>>> + if (acpi_check_dma(acpi_dev, &coherent)) > >>>>> + arch_setup_dma_ops(dev, 0, 0, NULL, coherent); > >>>>> + > >>> Well, so is this going to work for PCI too after all? > >>> > >> > >> No, as Bjorn suggested, PCI changes for setting DMA coherent from _CCA > >> (patch 3/6 in V4) will be submitted separately. We are working on > >> cleaning up and up-streaming the PCI ACPI support for ARM64. > > > > OK, but acpi_bind_one() is called for PCI devices too. Won't that be a problem? > > > > > > > In this case, we would be going through the following call path: > > --> pci_device_add() > |--> pci_dma_configure() ** 1 ** > |--> device_add() > |--> platform_notify() > |--> acpi_platform_notify() > |--> acpi_bind_one() ** 2 ** > > At (1), we would be calling arch_setup_dma_ops() with the PCI host > bridge _CCA information. So, it should have already taken care of > setting up device coherency here. > > At (2), if there is no acpi_dev for endpoint devices (which I believe > this is normally the case), it would return early and skip > arch_setup_dma_ops(). That's not correct. There may be ACPI companions for endpoint devices too. > At (2), if there is an acpi_dev, the coherent_dma flag should have > already been setup by the acpi_init_device_object during ACPI scan. That one sets the flag for the *ACPI* *companion* of the device, which I'm still thinking is pointless, isn't it? > However, I am not certain about this case since I don't have the DSDT > containing PCI endpoint devices to test with. Every x86 PC has them (as far as I can say), but in that case there's no _CCA and they are all coherent.
On 5/22/2015 8:25 PM, Rafael J. Wysocki wrote: > On Friday, May 22, 2015 07:15:17 PM Suravee Suthikulanit wrote: >> On 5/22/2015 6:05 PM, Rafael J. Wysocki wrote: >>> On Friday, May 22, 2015 05:24:15 PM Suravee Suthikulanit wrote: >>>> Not sure if this went out earlier. So I am resending. >>>> >>>> On 5/22/15 16:56, Rafael J. Wysocki wrote: >>>>>> diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c >>>>>>> index 39c485b..b9657af 100644 >>>>>>> --- a/drivers/acpi/glue.c >>>>>>> +++ b/drivers/acpi/glue.c >>>>>>> @@ -13,6 +13,7 @@ >>>>>>> #include <linux/slab.h> >>>>>>> #include <linux/rwsem.h> >>>>>>> #include <linux/acpi.h> >>>>>>> +#include <linux/dma-mapping.h> >>>>>>> >>>>>>> #include "internal.h" >>>>>>> >>>>>>> @@ -167,6 +168,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) >>>>>>> struct list_head *physnode_list; >>>>>>> unsigned int node_id; >>>>>>> int retval = -EINVAL; >>>>>>> + bool coherent; >>>>>>> >>>>>>> if (has_acpi_companion(dev)) { >>>>>>> if (acpi_dev) { >>>>>>> @@ -223,6 +225,9 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) >>>>>>> if (!has_acpi_companion(dev)) >>>>>>> ACPI_COMPANION_SET(dev, acpi_dev); >>>>>>> >>>>>>> + if (acpi_check_dma(acpi_dev, &coherent)) >>>>>>> + arch_setup_dma_ops(dev, 0, 0, NULL, coherent); >>>>>>> + >>>>> Well, so is this going to work for PCI too after all? >>>>> >>>> >>>> No, as Bjorn suggested, PCI changes for setting DMA coherent from _CCA >>>> (patch 3/6 in V4) will be submitted separately. We are working on >>>> cleaning up and up-streaming the PCI ACPI support for ARM64. >>> >>> OK, but acpi_bind_one() is called for PCI devices too. Won't that be a problem? >>> >>> >> > >> In this case, we would be going through the following call path: >> >> --> pci_device_add() >> |--> pci_dma_configure() ** 1 ** >> |--> device_add() >> |--> platform_notify() >> |--> acpi_platform_notify() >> |--> acpi_bind_one() ** 2 ** >> >> At (1), we would be calling arch_setup_dma_ops() with the PCI host >> bridge _CCA information. So, it should have already taken care of >> setting up device coherency here. >> >> At (2), if there is no acpi_dev for endpoint devices (which I believe >> this is normally the case), it would return early and skip >> arch_setup_dma_ops(). > > That's not correct. There may be ACPI companions for endpoint devices too. Ok. Duly noted :) >> At (2), if there is an acpi_dev, the coherent_dma flag should have >> already been setup by the acpi_init_device_object during ACPI scan. > > That one sets the flag for the *ACPI* *companion* of the device, which > I'm still thinking is pointless, isn't it? When you say pointless, are you referring to the part where we are end up calling arch_setup_dma_coherent() twice in this case? I am not quite following your point. >> However, I am not certain about this case since I don't have the DSDT >> containing PCI endpoint devices to test with. > > Every x86 PC has them (as far as I can say), but in that case there's no > _CCA and they are all coherent. Ok. Thanks, Suravee
diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig index ab2cbb5..212735f 100644 --- a/drivers/acpi/Kconfig +++ b/drivers/acpi/Kconfig @@ -54,6 +54,9 @@ config ACPI_GENERIC_GSI config ACPI_SYSTEM_POWER_STATES_SUPPORT bool +config ACPI_CCA_REQUIRED + bool + config ACPI_SLEEP bool depends on SUSPEND || HIBERNATION diff --git a/drivers/acpi/acpi_platform.c b/drivers/acpi/acpi_platform.c index 4bf7559..06a67d5 100644 --- a/drivers/acpi/acpi_platform.c +++ b/drivers/acpi/acpi_platform.c @@ -103,7 +103,7 @@ struct platform_device *acpi_create_platform_device(struct acpi_device *adev) pdevinfo.res = resources; pdevinfo.num_res = count; pdevinfo.fwnode = acpi_fwnode_handle(adev); - pdevinfo.dma_mask = DMA_BIT_MASK(32); + pdevinfo.dma_mask = acpi_check_dma(adev, NULL) ? DMA_BIT_MASK(32) : 0; pdev = platform_device_register_full(&pdevinfo); if (IS_ERR(pdev)) dev_err(&adev->dev, "platform device creation failed: %ld\n", diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c index 39c485b..b9657af 100644 --- a/drivers/acpi/glue.c +++ b/drivers/acpi/glue.c @@ -13,6 +13,7 @@ #include <linux/slab.h> #include <linux/rwsem.h> #include <linux/acpi.h> +#include <linux/dma-mapping.h> #include "internal.h" @@ -167,6 +168,7 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) struct list_head *physnode_list; unsigned int node_id; int retval = -EINVAL; + bool coherent; if (has_acpi_companion(dev)) { if (acpi_dev) { @@ -223,6 +225,9 @@ int acpi_bind_one(struct device *dev, struct acpi_device *acpi_dev) if (!has_acpi_companion(dev)) ACPI_COMPANION_SET(dev, acpi_dev); + if (acpi_check_dma(acpi_dev, &coherent)) + arch_setup_dma_ops(dev, 0, 0, NULL, coherent); + acpi_physnode_link_name(physical_node_name, node_id); retval = sysfs_create_link(&acpi_dev->dev.kobj, &dev->kobj, physical_node_name); diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index 849b699..fdedb63 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -11,6 +11,7 @@ #include <linux/kthread.h> #include <linux/dmi.h> #include <linux/nls.h> +#include <linux/dma-mapping.h> #include <asm/pgtable.h> @@ -2137,6 +2138,39 @@ void acpi_free_pnp_ids(struct acpi_device_pnp *pnp) kfree(pnp->unique_id); } +static void acpi_init_coherency(struct acpi_device *adev) +{ + unsigned long long cca = 0; + acpi_status status; + struct acpi_device *parent = adev->parent; + + if (parent && parent->flags.cca_seen) { + /* + * From ACPI spec, OSPM will ignore _CCA if an ancestor + * already saw one. + */ + adev->flags.cca_seen = 1; + cca = parent->flags.coherent_dma; + } else { + status = acpi_evaluate_integer(adev->handle, "_CCA", + NULL, &cca); + if (ACPI_SUCCESS(status)) + adev->flags.cca_seen = 1; + else if (!IS_ENABLED(CONFIG_ACPI_CCA_REQUIRED)) + /* + * If architecture does not specify that _CCA is + * required for DMA-able devices (e.g. x86), + * we default to _CCA=1. + */ + cca = 1; + else + acpi_handle_debug(adev->handle, + "ACPI device is missing _CCA.\n"); + } + + adev->flags.coherent_dma = cca; +} + void acpi_init_device_object(struct acpi_device *device, acpi_handle handle, int type, unsigned long long sta) { @@ -2155,6 +2189,7 @@ void acpi_init_device_object(struct acpi_device *device, acpi_handle handle, device->flags.visited = false; device_initialize(&device->dev); dev_set_uevent_suppress(&device->dev, true); + acpi_init_coherency(device); } void acpi_device_add_finalize(struct acpi_device *device) diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index 8de4fa9..cbd1b57 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -208,7 +208,9 @@ struct acpi_device_flags { u32 visited:1; u32 hotplug_notify:1; u32 is_dock_station:1; - u32 reserved:23; + u32 coherent_dma:1; + u32 cca_seen:1; + u32 reserved:21; }; /* File System */ @@ -380,6 +382,39 @@ struct acpi_device { void (*remove)(struct acpi_device *); }; +static inline bool acpi_check_dma(struct acpi_device *adev, bool *coherent) +{ + bool ret = false; + + if (!adev) + return ret; + + /** + * Currently, we only support _CCA=1 (i.e. coherent_dma=1) + * This should be equivalent to specifyig dma-coherent for + * a device in OF. + * + * For the case when _CCA=0 (i.e. coherent_dma=0 && cca_seen=1), + * There are two cases: + * case 1. Do not support and disable DMA. + * case 2. Support but rely on arch-specific cache maintenance for + * non-coherence DMA operations. + * Currently, we implement case 1 above. + * + * For the case when _CCA is missing (i.e. cca_seen=0) and + * platform specifies ACPI_CCA_REQUIRED, we do not support DMA, + * and fallback to arch-specific default handling. + * + * See acpi_init_coherency() for more info. + */ + if (adev->flags.coherent_dma) { + ret = true; + if (coherent) + *coherent = adev->flags.coherent_dma; + } + return ret; +} + static inline bool is_acpi_node(struct fwnode_handle *fwnode) { return fwnode && fwnode->type == FWNODE_ACPI; diff --git a/include/linux/acpi.h b/include/linux/acpi.h index b10c4a6..30c81ac 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -583,6 +583,11 @@ static inline int acpi_device_modalias(struct device *dev, return -ENODEV; } +static inline bool acpi_check_dma(struct acpi_device *adev, bool *coherent) +{ + return false; +} + #define ACPI_PTR(_ptr) (NULL) #endif /* !CONFIG_ACPI */