Message ID | 20190501125322.23791-1-pierre-louis.bossart@linux.intel.com (mailing list archive) |
---|---|
State | Accepted, archived |
Delegated to: | Rafael Wysocki |
Headers | show |
Series | [v2] ACPI / device_sysfs: change _ADR representation to 64 bits | expand |
On 01-05-19, 07:53, Pierre-Louis Bossart wrote: > Standards such as the MIPI DisCo for SoundWire 1.0 specification > assume the _ADR field is 64 bits. > > _ADR is defined as an "Integer" represented as 64 bits since ACPI 2.0 > released in 2002. The low levels already use _ADR as 64 bits, e.g. in > struct acpi_device_info. > > This patch bumps the representation used for sysfs to 64 bits. To > avoid any compatibility/ABI issues, the printf format is only extended > to 16 characters when the actual _ADR value exceeds the 32 bit > maximum. > > Example with a SoundWire device, the results show the complete > vendorID and linkID which were omitted before: > > Before: > $ more /sys/bus/acpi/devices/device\:38/adr > 0x5d070000 > After: > $ more /sys/bus/acpi/devices/device\:38/adr > 0x000010025d070000 > > Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> > --- > v2: only use 64 bits when required to avoid compatibility issues > (feedback from Vinod and Rafael) > > drivers/acpi/device_sysfs.c | 6 ++++-- > include/acpi/acpi_bus.h | 2 +- > 2 files changed, 5 insertions(+), 3 deletions(-) > > diff --git a/drivers/acpi/device_sysfs.c b/drivers/acpi/device_sysfs.c > index 8940054d6250..7dda0ee05cd1 100644 > --- a/drivers/acpi/device_sysfs.c > +++ b/drivers/acpi/device_sysfs.c > @@ -428,8 +428,10 @@ static ssize_t acpi_device_adr_show(struct device *dev, > { > struct acpi_device *acpi_dev = to_acpi_device(dev); > > - return sprintf(buf, "0x%08x\n", > - (unsigned int)(acpi_dev->pnp.bus_address)); > + if (acpi_dev->pnp.bus_address > 0xFFFFFFFF) Would prefer to use U32_MAX instead of 0xFFFFFFFF > + return sprintf(buf, "0x%016llx\n", acpi_dev->pnp.bus_address); > + else > + return sprintf(buf, "0x%08llx\n", acpi_dev->pnp.bus_address); > } > static DEVICE_ATTR(adr, 0444, acpi_device_adr_show, NULL); > > diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h > index f7981751ac77..9075e28ea60a 100644 > --- a/include/acpi/acpi_bus.h > +++ b/include/acpi/acpi_bus.h > @@ -230,7 +230,7 @@ struct acpi_device_dir { > /* Plug and Play */ > > typedef char acpi_bus_id[8]; > -typedef unsigned long acpi_bus_address; > +typedef u64 acpi_bus_address; > typedef char acpi_device_name[40]; > typedef char acpi_device_class[20]; > > -- > 2.17.1
> -----Original Message----- > From: Pierre-Louis Bossart [mailto:pierre-louis.bossart@linux.intel.com] > Sent: Wednesday, May 1, 2019 5:53 AM > To: alsa-devel@alsa-project.org > Cc: linux-kernel@vger.kernel.org; tiwai@suse.de; broonie@kernel.org; > vkoul@kernel.org; gregkh@linuxfoundation.org; > liam.r.girdwood@linux.intel.com; jank@cadence.com; joe@perches.com; > srinivas.kandagatla@linaro.org; Pierre-Louis Bossart <pierre- > louis.bossart@linux.intel.com>; Rafael J. Wysocki <rjw@rjwysocki.net>; > Len Brown <lenb@kernel.org>; Moore, Robert <robert.moore@intel.com>; > Schmauss, Erik <erik.schmauss@intel.com>; open list:ACPI <linux- > acpi@vger.kernel.org>; open list:ACPI COMPONENT ARCHITECTURE (ACPICA) > <devel@acpica.org> > Subject: [PATCH v2] ACPI / device_sysfs: change _ADR representation to > 64 bits > > Standards such as the MIPI DisCo for SoundWire 1.0 specification assume > the _ADR field is 64 bits. > > _ADR is defined as an "Integer" represented as 64 bits since ACPI 2.0 > released in 2002. The low levels already use _ADR as 64 bits, e.g. in > struct acpi_device_info. > [Moore, Robert] Just to be precise: since acpi 2.0 the integer width is either 32 bits or 64 bits, depending on the version number of the DSDT (1-->32, 2 or greater --> 64). > This patch bumps the representation used for sysfs to 64 bits. To avoid > any compatibility/ABI issues, the printf format is only extended to 16 > characters when the actual _ADR value exceeds the 32 bit maximum. > > Example with a SoundWire device, the results show the complete vendorID > and linkID which were omitted before: > > Before: > $ more /sys/bus/acpi/devices/device\:38/adr > 0x5d070000 > After: > $ more /sys/bus/acpi/devices/device\:38/adr > 0x000010025d070000 > > Signed-off-by: Pierre-Louis Bossart <pierre- > louis.bossart@linux.intel.com> > --- > v2: only use 64 bits when required to avoid compatibility issues > (feedback from Vinod and Rafael) > > drivers/acpi/device_sysfs.c | 6 ++++-- > include/acpi/acpi_bus.h | 2 +- > 2 files changed, 5 insertions(+), 3 deletions(-) > > diff --git a/drivers/acpi/device_sysfs.c b/drivers/acpi/device_sysfs.c > index 8940054d6250..7dda0ee05cd1 100644 > --- a/drivers/acpi/device_sysfs.c > +++ b/drivers/acpi/device_sysfs.c > @@ -428,8 +428,10 @@ static ssize_t acpi_device_adr_show(struct device > *dev, { > struct acpi_device *acpi_dev = to_acpi_device(dev); > > - return sprintf(buf, "0x%08x\n", > - (unsigned int)(acpi_dev->pnp.bus_address)); > + if (acpi_dev->pnp.bus_address > 0xFFFFFFFF) > + return sprintf(buf, "0x%016llx\n", acpi_dev- > >pnp.bus_address); > + else > + return sprintf(buf, "0x%08llx\n", acpi_dev->pnp.bus_address); > } > static DEVICE_ATTR(adr, 0444, acpi_device_adr_show, NULL); > > diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index > f7981751ac77..9075e28ea60a 100644 > --- a/include/acpi/acpi_bus.h > +++ b/include/acpi/acpi_bus.h > @@ -230,7 +230,7 @@ struct acpi_device_dir { > /* Plug and Play */ > > typedef char acpi_bus_id[8]; > -typedef unsigned long acpi_bus_address; > +typedef u64 acpi_bus_address; > typedef char acpi_device_name[40]; > typedef char acpi_device_class[20]; > > -- > 2.17.1
On Thu, May 2, 2019 at 6:58 AM Vinod Koul <vkoul@kernel.org> wrote: > > On 01-05-19, 07:53, Pierre-Louis Bossart wrote: > > Standards such as the MIPI DisCo for SoundWire 1.0 specification > > assume the _ADR field is 64 bits. > > > > _ADR is defined as an "Integer" represented as 64 bits since ACPI 2.0 > > released in 2002. The low levels already use _ADR as 64 bits, e.g. in > > struct acpi_device_info. > > > > This patch bumps the representation used for sysfs to 64 bits. To > > avoid any compatibility/ABI issues, the printf format is only extended > > to 16 characters when the actual _ADR value exceeds the 32 bit > > maximum. > > > > Example with a SoundWire device, the results show the complete > > vendorID and linkID which were omitted before: > > > > Before: > > $ more /sys/bus/acpi/devices/device\:38/adr > > 0x5d070000 > > After: > > $ more /sys/bus/acpi/devices/device\:38/adr > > 0x000010025d070000 > > > > Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> > > --- > > v2: only use 64 bits when required to avoid compatibility issues > > (feedback from Vinod and Rafael) > > > > drivers/acpi/device_sysfs.c | 6 ++++-- > > include/acpi/acpi_bus.h | 2 +- > > 2 files changed, 5 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/acpi/device_sysfs.c b/drivers/acpi/device_sysfs.c > > index 8940054d6250..7dda0ee05cd1 100644 > > --- a/drivers/acpi/device_sysfs.c > > +++ b/drivers/acpi/device_sysfs.c > > @@ -428,8 +428,10 @@ static ssize_t acpi_device_adr_show(struct device *dev, > > { > > struct acpi_device *acpi_dev = to_acpi_device(dev); > > > > - return sprintf(buf, "0x%08x\n", > > - (unsigned int)(acpi_dev->pnp.bus_address)); > > + if (acpi_dev->pnp.bus_address > 0xFFFFFFFF) > > Would prefer to use U32_MAX instead of 0xFFFFFFFF I would.
On Monday, May 6, 2019 10:36:22 AM CEST Rafael J. Wysocki wrote: > On Thu, May 2, 2019 at 6:58 AM Vinod Koul <vkoul@kernel.org> wrote: > > > > On 01-05-19, 07:53, Pierre-Louis Bossart wrote: > > > Standards such as the MIPI DisCo for SoundWire 1.0 specification > > > assume the _ADR field is 64 bits. > > > > > > _ADR is defined as an "Integer" represented as 64 bits since ACPI 2.0 > > > released in 2002. The low levels already use _ADR as 64 bits, e.g. in > > > struct acpi_device_info. > > > > > > This patch bumps the representation used for sysfs to 64 bits. To > > > avoid any compatibility/ABI issues, the printf format is only extended > > > to 16 characters when the actual _ADR value exceeds the 32 bit > > > maximum. > > > > > > Example with a SoundWire device, the results show the complete > > > vendorID and linkID which were omitted before: > > > > > > Before: > > > $ more /sys/bus/acpi/devices/device\:38/adr > > > 0x5d070000 > > > After: > > > $ more /sys/bus/acpi/devices/device\:38/adr > > > 0x000010025d070000 > > > > > > Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> > > > --- > > > v2: only use 64 bits when required to avoid compatibility issues > > > (feedback from Vinod and Rafael) > > > > > > drivers/acpi/device_sysfs.c | 6 ++++-- > > > include/acpi/acpi_bus.h | 2 +- > > > 2 files changed, 5 insertions(+), 3 deletions(-) > > > > > > diff --git a/drivers/acpi/device_sysfs.c b/drivers/acpi/device_sysfs.c > > > index 8940054d6250..7dda0ee05cd1 100644 > > > --- a/drivers/acpi/device_sysfs.c > > > +++ b/drivers/acpi/device_sysfs.c > > > @@ -428,8 +428,10 @@ static ssize_t acpi_device_adr_show(struct device *dev, > > > { > > > struct acpi_device *acpi_dev = to_acpi_device(dev); > > > > > > - return sprintf(buf, "0x%08x\n", > > > - (unsigned int)(acpi_dev->pnp.bus_address)); > > > + if (acpi_dev->pnp.bus_address > 0xFFFFFFFF) > > > > Would prefer to use U32_MAX instead of 0xFFFFFFFF > > I would. > I have made that change manually and applied the patch. Thanks!
>>>> diff --git a/drivers/acpi/device_sysfs.c b/drivers/acpi/device_sysfs.c >>>> index 8940054d6250..7dda0ee05cd1 100644 >>>> --- a/drivers/acpi/device_sysfs.c >>>> +++ b/drivers/acpi/device_sysfs.c >>>> @@ -428,8 +428,10 @@ static ssize_t acpi_device_adr_show(struct device *dev, >>>> { >>>> struct acpi_device *acpi_dev = to_acpi_device(dev); >>>> >>>> - return sprintf(buf, "0x%08x\n", >>>> - (unsigned int)(acpi_dev->pnp.bus_address)); >>>> + if (acpi_dev->pnp.bus_address > 0xFFFFFFFF) >>> >>> Would prefer to use U32_MAX instead of 0xFFFFFFFF >> >> I would. >> > > I have made that change manually and applied the patch. > > Thanks! Thanks, much appreciated.
diff --git a/drivers/acpi/device_sysfs.c b/drivers/acpi/device_sysfs.c index 8940054d6250..7dda0ee05cd1 100644 --- a/drivers/acpi/device_sysfs.c +++ b/drivers/acpi/device_sysfs.c @@ -428,8 +428,10 @@ static ssize_t acpi_device_adr_show(struct device *dev, { struct acpi_device *acpi_dev = to_acpi_device(dev); - return sprintf(buf, "0x%08x\n", - (unsigned int)(acpi_dev->pnp.bus_address)); + if (acpi_dev->pnp.bus_address > 0xFFFFFFFF) + return sprintf(buf, "0x%016llx\n", acpi_dev->pnp.bus_address); + else + return sprintf(buf, "0x%08llx\n", acpi_dev->pnp.bus_address); } static DEVICE_ATTR(adr, 0444, acpi_device_adr_show, NULL); diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index f7981751ac77..9075e28ea60a 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -230,7 +230,7 @@ struct acpi_device_dir { /* Plug and Play */ typedef char acpi_bus_id[8]; -typedef unsigned long acpi_bus_address; +typedef u64 acpi_bus_address; typedef char acpi_device_name[40]; typedef char acpi_device_class[20];
Standards such as the MIPI DisCo for SoundWire 1.0 specification assume the _ADR field is 64 bits. _ADR is defined as an "Integer" represented as 64 bits since ACPI 2.0 released in 2002. The low levels already use _ADR as 64 bits, e.g. in struct acpi_device_info. This patch bumps the representation used for sysfs to 64 bits. To avoid any compatibility/ABI issues, the printf format is only extended to 16 characters when the actual _ADR value exceeds the 32 bit maximum. Example with a SoundWire device, the results show the complete vendorID and linkID which were omitted before: Before: $ more /sys/bus/acpi/devices/device\:38/adr 0x5d070000 After: $ more /sys/bus/acpi/devices/device\:38/adr 0x000010025d070000 Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> --- v2: only use 64 bits when required to avoid compatibility issues (feedback from Vinod and Rafael) drivers/acpi/device_sysfs.c | 6 ++++-- include/acpi/acpi_bus.h | 2 +- 2 files changed, 5 insertions(+), 3 deletions(-)