[U-Boot] [PATCH v1 0/7] x86: acpi: Few enhancements and improvements

[U-Boot] [PATCH v1 0/7] x86: acpi: Few enhancements and improvements

[Andy Shevchenko]

Patch 1 is a build improvement to have better understanding why DSDT
is not compiled.

Patches 2 and 3 add date of U-Boot build as OEM revision to
the generated tables.

Patch 4 marks fields in FADT in accordance with latest ACPI spec.

Patches 5 and 6 provides a generic "fill" functions that can be replaced
in specific board / architecture code.

Patch 7 allows to define HW reduced platforms by avoiding ACPI hardware
communication as required by spec.

The series has been tested on Intel Edison board.

Andy Shevchenko (7):
  Makefile: Don't shadow actual error when compile ASL
  Makefile: Export build date as integer
  x86: acpi: Fill OEM revision
  x86: acpi: Name fields in FADT in accordance with specification
  x86: acpi: Deduplicate acpi_fill_madt() implementation
  x86: acpi: Export acpi_fill_mcfg() with __weak attribute
  x86: acpi: Don't touch hardware on HW reduced platforms

 Makefile                          |  2 ++
 arch/x86/cpu/baytrail/acpi.c      | 29 ---------------------------
 arch/x86/cpu/quark/acpi.c         | 29 ---------------------------
 arch/x86/include/asm/acpi_table.h |  8 +++++---
 arch/x86/lib/acpi_table.c         | 41 ++++++++++++++++++++++++++++++++++++---
 scripts/Makefile.lib              |  2 +-
 6 files changed, 46 insertions(+), 65 deletions(-)

-- 
2.13.2

[U-Boot] [PATCH v1 1/7] Makefile: Don't shadow actual error when compile ASL

[Andy Shevchenko]

If ASL compiler failed by any reason do not produce output C file.
Otherwise sequential run of make will shadow the actual error in ASL,
i.e.

  CC      board/intel/edison/dsdt.o
board/intel/edison/dsdt.c:1:1: error: unterminated comment
 /*

and user has to remove dsdt.c and run make in order to see the error again.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 scripts/Makefile.lib | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index 80ddb08474..9ce47b4d22 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -386,7 +386,7 @@ $(obj)/helloworld.so: $(obj)/helloworld.o arch/$(ARCH)/lib/$(EFI_CRT0) \
 quiet_cmd_acpi_c_asl= ASL     $<
 cmd_acpi_c_asl=         \
 	$(CPP) -x assembler-with-cpp -D__ASSEMBLY__ -P $(UBOOTINCLUDE) -o $<.tmp $<; \
-	iasl -p $< -tc $<.tmp $(if $(KBUILD_VERBOSE:1=), >/dev/null); \
+	iasl -p $< -tc $<.tmp $(if $(KBUILD_VERBOSE:1=), >/dev/null) && \
 	mv $(patsubst %.asl,%.hex,$<) $@
 
 $(obj)/dsdt.c:    $(src)/dsdt.asl
-- 
2.13.2

[U-Boot] [PATCH v1 2/7] Makefile: Export build date as integer

[Andy Shevchenko]

In some cases we would need build date as integer value.
Export U_BOOT_BUILD_DATE as %Y%m%d integer value.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 Makefile | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Makefile b/Makefile
index 452596485d..71de5c656c 100644
--- a/Makefile
+++ b/Makefile
@@ -1352,6 +1352,7 @@ define filechk_timestamp.h
 			LC_ALL=C $${DATE} -u -d "$${SOURCE_DATE}" +'#define U_BOOT_TIME "%T"'; \
 			LC_ALL=C $${DATE} -u -d "$${SOURCE_DATE}" +'#define U_BOOT_TZ "%z"'; \
 			LC_ALL=C $${DATE} -u -d "$${SOURCE_DATE}" +'#define U_BOOT_DMI_DATE "%m/%d/%Y"'; \
+			LC_ALL=C $${DATE} -u -d "$${SOURCE_DATE}" +'#define U_BOOT_BUILD_DATE 0x%Y%m%d'; \
 		else \
 			return 42; \
 		fi; \
@@ -1360,6 +1361,7 @@ define filechk_timestamp.h
 		LC_ALL=C date +'#define U_BOOT_TIME "%T"'; \
 		LC_ALL=C date +'#define U_BOOT_TZ "%z"'; \
 		LC_ALL=C date +'#define U_BOOT_DMI_DATE "%m/%d/%Y"'; \
+		LC_ALL=C date +'#define U_BOOT_BUILD_DATE 0x%Y%m%d'; \
 	fi)
 endef
 
-- 
2.13.2

[U-Boot] [PATCH v1 3/7] x86: acpi: Fill OEM revision

[Andy Shevchenko]

Fill OEM revision field in the tables by U-Boot build date.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 arch/x86/lib/acpi_table.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
index 01d5b6fff0..c6f4d8941c 100644
--- a/arch/x86/lib/acpi_table.c
+++ b/arch/x86/lib/acpi_table.c
@@ -11,6 +11,7 @@
 #include <cpu.h>
 #include <dm.h>
 #include <dm/uclass-internal.h>
+#include <version.h>
 #include <asm/acpi/global_nvs.h>
 #include <asm/acpi_table.h>
 #include <asm/io.h>
@@ -60,6 +61,7 @@ void acpi_fill_header(struct acpi_table_header *header, char *signature)
 	memcpy(header->signature, signature, 4);
 	memcpy(header->oem_id, OEM_ID, 6);
 	memcpy(header->oem_table_id, OEM_TABLE_ID, 8);
+	header->oem_revision = U_BOOT_BUILD_DATE;
 	memcpy(header->aslc_id, ASLC_ID, 4);
 }
 
-- 
2.13.2

[U-Boot] [PATCH v1 4/7] x86: acpi: Name fields in FADT in accordance with specification

[Andy Shevchenko]

ACPI specification defines FADT fields marked as reserved in U-Boot.

Name these fields in accordance with ACPI specification.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 arch/x86/include/asm/acpi_table.h | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/arch/x86/include/asm/acpi_table.h b/arch/x86/include/asm/acpi_table.h
index dd7a946b6c..e177c441e7 100644
--- a/arch/x86/include/asm/acpi_table.h
+++ b/arch/x86/include/asm/acpi_table.h
@@ -178,9 +178,8 @@ struct __packed acpi_fadt {
 	u32 flags;
 	struct acpi_gen_regaddr reset_reg;
 	u8 reset_value;
-	u8 res3;
-	u8 res4;
-	u8 res5;
+	u16 arm_boot_arch;
+	u8 minor_revision;
 	u32 x_firmware_ctl_l;
 	u32 x_firmware_ctl_h;
 	u32 x_dsdt_l;
-- 
2.13.2

[U-Boot] [PATCH v1 5/7] x86: acpi: Deduplicate acpi_fill_madt() implementation

[Andy Shevchenko]

In Baytrail and Quark support code acpi_fill_madt() is identical.

Deduplicate its implementation by moving to lib/acpi_tables.c.

At the same time mark acpi_fill_madt() with __weak attribute to keep a
possibility to override it in platform code

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 arch/x86/cpu/baytrail/acpi.c | 29 -----------------------------
 arch/x86/cpu/quark/acpi.c    | 29 -----------------------------
 arch/x86/lib/acpi_table.c    | 29 +++++++++++++++++++++++++++++
 3 files changed, 29 insertions(+), 58 deletions(-)

diff --git a/arch/x86/cpu/baytrail/acpi.c b/arch/x86/cpu/baytrail/acpi.c
index 55ed7de781..cbefdf871d 100644
--- a/arch/x86/cpu/baytrail/acpi.c
+++ b/arch/x86/cpu/baytrail/acpi.c
@@ -11,8 +11,6 @@
 #include <asm/acpi_s3.h>
 #include <asm/acpi_table.h>
 #include <asm/io.h>
-#include <asm/ioapic.h>
-#include <asm/mpspec.h>
 #include <asm/tables.h>
 #include <asm/arch/global_nvs.h>
 #include <asm/arch/iomap.h>
@@ -141,33 +139,6 @@ void acpi_create_fadt(struct acpi_fadt *fadt, struct acpi_facs *facs,
 	header->checksum = table_compute_checksum(fadt, header->length);
 }
 
-static int acpi_create_madt_irq_overrides(u32 current)
-{
-	struct acpi_madt_irqoverride *irqovr;
-	u16 sci_flags = MP_IRQ_TRIGGER_LEVEL | MP_IRQ_POLARITY_HIGH;
-	int length = 0;
-
-	irqovr = (void *)current;
-	length += acpi_create_madt_irqoverride(irqovr, 0, 0, 2, 0);
-
-	irqovr = (void *)(current + length);
-	length += acpi_create_madt_irqoverride(irqovr, 0, 9, 9, sci_flags);
-
-	return length;
-}
-
-u32 acpi_fill_madt(u32 current)
-{
-	current += acpi_create_madt_lapics(current);
-
-	current += acpi_create_madt_ioapic((struct acpi_madt_ioapic *)current,
-			io_apic_read(IO_APIC_ID) >> 24, IO_APIC_ADDR, 0);
-
-	current += acpi_create_madt_irq_overrides(current);
-
-	return current;
-}
-
 void acpi_create_gnvs(struct acpi_global_nvs *gnvs)
 {
 	struct udevice *dev;
diff --git a/arch/x86/cpu/quark/acpi.c b/arch/x86/cpu/quark/acpi.c
index 3968f7a8bf..5717a620b5 100644
--- a/arch/x86/cpu/quark/acpi.c
+++ b/arch/x86/cpu/quark/acpi.c
@@ -6,8 +6,6 @@
 
 #include <common.h>
 #include <asm/acpi_table.h>
-#include <asm/ioapic.h>
-#include <asm/mpspec.h>
 #include <asm/tables.h>
 #include <asm/arch/global_nvs.h>
 #include <asm/arch/iomap.h>
@@ -136,33 +134,6 @@ void acpi_create_fadt(struct acpi_fadt *fadt, struct acpi_facs *facs,
 	header->checksum = table_compute_checksum(fadt, header->length);
 }
 
-static int acpi_create_madt_irq_overrides(u32 current)
-{
-	struct acpi_madt_irqoverride *irqovr;
-	u16 sci_flags = MP_IRQ_TRIGGER_LEVEL | MP_IRQ_POLARITY_HIGH;
-	int length = 0;
-
-	irqovr = (void *)current;
-	length += acpi_create_madt_irqoverride(irqovr, 0, 0, 2, 0);
-
-	irqovr = (void *)(current + length);
-	length += acpi_create_madt_irqoverride(irqovr, 0, 9, 9, sci_flags);
-
-	return length;
-}
-
-u32 acpi_fill_madt(u32 current)
-{
-	current += acpi_create_madt_lapics(current);
-
-	current += acpi_create_madt_ioapic((struct acpi_madt_ioapic *)current,
-			io_apic_read(IO_APIC_ID) >> 24, IO_APIC_ADDR, 0);
-
-	current += acpi_create_madt_irq_overrides(current);
-
-	return current;
-}
-
 void acpi_create_gnvs(struct acpi_global_nvs *gnvs)
 {
 	/* quark is a uni-processor */
diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
index c6f4d8941c..14d10c34cd 100644
--- a/arch/x86/lib/acpi_table.c
+++ b/arch/x86/lib/acpi_table.c
@@ -15,7 +15,9 @@
 #include <asm/acpi/global_nvs.h>
 #include <asm/acpi_table.h>
 #include <asm/io.h>
+#include <asm/ioapic.h>
 #include <asm/lapic.h>
+#include <asm/mpspec.h>
 #include <asm/tables.h>
 #include <asm/arch/global_nvs.h>
 
@@ -241,6 +243,33 @@ int acpi_create_madt_lapic_nmi(struct acpi_madt_lapic_nmi *lapic_nmi,
 	return lapic_nmi->length;
 }
 
+static int acpi_create_madt_irq_overrides(u32 current)
+{
+	struct acpi_madt_irqoverride *irqovr;
+	u16 sci_flags = MP_IRQ_TRIGGER_LEVEL | MP_IRQ_POLARITY_HIGH;
+	int length = 0;
+
+	irqovr = (void *)current;
+	length += acpi_create_madt_irqoverride(irqovr, 0, 0, 2, 0);
+
+	irqovr = (void *)(current + length);
+	length += acpi_create_madt_irqoverride(irqovr, 0, 9, 9, sci_flags);
+
+	return length;
+}
+
+__weak u32 acpi_fill_madt(u32 current)
+{
+	current += acpi_create_madt_lapics(current);
+
+	current += acpi_create_madt_ioapic((struct acpi_madt_ioapic *)current,
+			io_apic_read(IO_APIC_ID) >> 24, IO_APIC_ADDR, 0);
+
+	current += acpi_create_madt_irq_overrides(current);
+
+	return current;
+}
+
 static void acpi_create_madt(struct acpi_madt *madt)
 {
 	struct acpi_table_header *header = &(madt->header);
-- 
2.13.2

[U-Boot] [PATCH v1 6/7] x86: acpi: Export acpi_fill_mcfg() with __weak attribute

[Andy Shevchenko]

Some platforms might require different approach when filling memory
mappings configuration table.

Allow them to override the common method.

At the same time export acpi_create_mcfg_mmconfig().

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 arch/x86/include/asm/acpi_table.h | 3 +++
 arch/x86/lib/acpi_table.c         | 6 +++---
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/arch/x86/include/asm/acpi_table.h b/arch/x86/include/asm/acpi_table.h
index e177c441e7..80038504dd 100644
--- a/arch/x86/include/asm/acpi_table.h
+++ b/arch/x86/include/asm/acpi_table.h
@@ -314,6 +314,9 @@ int acpi_create_madt_irqoverride(struct acpi_madt_irqoverride *irqoverride,
 int acpi_create_madt_lapic_nmi(struct acpi_madt_lapic_nmi *lapic_nmi,
 			       u8 cpu, u16 flags, u8 lint);
 u32 acpi_fill_madt(u32 current);
+int acpi_create_mcfg_mmconfig(struct acpi_mcfg_mmconfig *mmconfig, u32 base,
+			      u16 seg_nr, u8 start, u8 end);
+u32 acpi_fill_mcfg(u32 current);
 void acpi_create_gnvs(struct acpi_global_nvs *gnvs);
 /**
  * enter_acpi_mode() - enter into ACPI mode
diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
index 14d10c34cd..da6014fb41 100644
--- a/arch/x86/lib/acpi_table.c
+++ b/arch/x86/lib/acpi_table.c
@@ -293,8 +293,8 @@ static void acpi_create_madt(struct acpi_madt *madt)
 	header->checksum = table_compute_checksum((void *)madt, header->length);
 }
 
-static int acpi_create_mcfg_mmconfig(struct acpi_mcfg_mmconfig *mmconfig,
-				     u32 base, u16 seg_nr, u8 start, u8 end)
+int acpi_create_mcfg_mmconfig(struct acpi_mcfg_mmconfig *mmconfig, u32 base,
+			      u16 seg_nr, u8 start, u8 end)
 {
 	memset(mmconfig, 0, sizeof(*mmconfig));
 	mmconfig->base_address_l = base;
@@ -306,7 +306,7 @@ static int acpi_create_mcfg_mmconfig(struct acpi_mcfg_mmconfig *mmconfig,
 	return sizeof(struct acpi_mcfg_mmconfig);
 }
 
-static u32 acpi_fill_mcfg(u32 current)
+__weak u32 acpi_fill_mcfg(u32 current)
 {
 	current += acpi_create_mcfg_mmconfig
 		((struct acpi_mcfg_mmconfig *)current,
-- 
2.13.2

[U-Boot] [PATCH v1 7/7] x86: acpi: Don't touch hardware on HW reduced platforms

[Andy Shevchenko]

If ACPI HW reduced bit in FADT is set we should ignore any ACPI hardware
communications.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 arch/x86/lib/acpi_table.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
index da6014fb41..3eb101105b 100644
--- a/arch/x86/lib/acpi_table.c
+++ b/arch/x86/lib/acpi_table.c
@@ -463,6 +463,10 @@ ulong write_acpi_tables(ulong start)
 
 	debug("ACPI: done\n");
 
+	/* Don't touch ACPI hardware on HW reduced platforms */
+	if (fadt->flags & ACPI_FADT_HW_REDUCED_ACPI)
+		return current;
+
 	/*
 	 * Other than waiting for OSPM to request us to switch to ACPI mode,
 	 * do it by ourselves, since SMI will not be triggered.
-- 
2.13.2

[U-Boot] [PATCH v1 0/7] x86: acpi: Few enhancements and improvements

[Andy Shevchenko]

On Fri, 2017-07-21 at 22:31 +0300, Andy Shevchenko wrote:
> Patch 1 is a build improvement to have better understanding why DSDT
> is not compiled.
> 
> Patches 2 and 3 add date of U-Boot build as OEM revision to
> the generated tables.
> 
> Patch 4 marks fields in FADT in accordance with latest ACPI spec.
> 
> Patches 5 and 6 provides a generic "fill" functions that can be
> replaced
> in specific board / architecture code.
> 
> Patch 7 allows to define HW reduced platforms by avoiding ACPI
> hardware
> communication as required by spec.
> 
> The series has been tested on Intel Edison board.

# dmesg | grep -i acpi
[    0.000000] Command line: console=tty1 console=ttyS2,115200n8
rootfstype=ramfs rw ignore_loglevel apic=debug
acpi.debug_layer=0xffff056d acpi.debug_level=0x0000200f
[    0.000000] ACPI: Early table checksum verification disabled
[    0.000000] ACPI: RSDP 0x00000000000E4500 000024 (v02 U-BOOT)
[    0.000000] ACPI: XSDT 0x00000000000E45E0 00003C (v01 U-BOOT U-BOOTBL 
20170721 INTL 00000000)
[    0.000000] ACPI: FACP 0x00000000000E4BA0 0000F4 (v06 U-BOOT U-BOOTBL 
20170721 INTL 00000000)
[    0.000000] ACPI: DSDT 0x00000000000E4780 000317 (v02 U-BOOT U-BOOTBL 
00010000 INTL 20170303)
[    0.000000] ACPI: APIC 0x00000000000E4CA0 000048 (v04 U-BOOT U-BOOTBL 
20170721 INTL 00000000)
[    0.000000] ACPI: MCFG 0x00000000000E4CF0 00003C (v01 U-BOOT U-BOOTBL 
20170721 INTL 00000000)
[    0.000000] ACPI: Local APIC address 0xfee00000
[    0.000000] ACPI: no legacy devices present
[    0.000000] ACPI: Local APIC address 0xfee00000
[    0.000000] ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled)
[    0.000000] ACPI: LAPIC (acpi_id[0x02] lapic_id[0x02] enabled)
[    0.000000] ACPI: IOAPIC (id[0x02] address[0xfec00000] gsi_base[0])
[    0.000000] Using ACPI (MADT) for SMP configuration information
[    0.000000] Kernel command line: console=tty1 console=ttyS2,115200n8
rootfstype=ramfs rw ignore_loglevel apic=debug
acpi.debug_layer=0xffff056d acpi.debug_level=0x0000200f
[    0.000492] ACPI: Core revision 20170531
[    0.001000] ACPI: 1 ACPI AML tables successfully acquired and loaded
[    0.077715] ACPI FADT declares the system doesn't support PCIe ASPM,
so disable it
[    0.078017] ACPI: bus type PCI registered
[    0.128219] ACPI: Added _OSI(Module Device)
[    0.128361] ACPI: Added _OSI(Processor Device)
[    0.128496] ACPI: Added _OSI(3.0 _SCP Extensions)
[    0.128635] ACPI: Added _OSI(Processor Aggregator Device)
[    0.130366] ACPI: Interpreter enabled
[    0.130541] ACPI: (supports S0)
[    0.130647] ACPI: Using IOAPIC for interrupt routing
[    0.130897] PCI: Using host bridge windows from ACPI; if necessary,
use "pci=nocrs" and report a bug
[    0.155018]       bus-0135 bus_get_status        : Device [ACPI]
status [0000000f]
[    0.158317] ACPI: PCI Root Bridge [PCI0] (domain 0000 [bus 00-ff])
[    0.158776] acpi PNP0A08:00: _OSC: OS supports [ExtendedConfig ASPM
ClockPM Segments MSI]
[    0.159628] acpi PNP0A08:00: _OSC: OS now controls [PCIeHotplug PME
AER PCIeCapability]
[    0.159869] acpi PNP0A08:00: FADT indicates ASPM is unsupported,
using BIOS configuration
[    0.161535] acpi PNP0A08:00: [Firmware Info]: MMCONFIG for domain
0000 [bus 00-00] only partially covers this bridge
[    0.221280] ACPI: bus type USB registered
[    0.304179] pnp: PnP ACPI init
[    0.304830] pnp: PnP ACPI: found 0 devices


> 
> Andy Shevchenko (7):
>   Makefile: Don't shadow actual error when compile ASL
>   Makefile: Export build date as integer
>   x86: acpi: Fill OEM revision
>   x86: acpi: Name fields in FADT in accordance with specification
>   x86: acpi: Deduplicate acpi_fill_madt() implementation
>   x86: acpi: Export acpi_fill_mcfg() with __weak attribute
>   x86: acpi: Don't touch hardware on HW reduced platforms
> 
>  Makefile                          |  2 ++
>  arch/x86/cpu/baytrail/acpi.c      | 29 ---------------------------
>  arch/x86/cpu/quark/acpi.c         | 29 ---------------------------
>  arch/x86/include/asm/acpi_table.h |  8 +++++---
>  arch/x86/lib/acpi_table.c         | 41
> ++++++++++++++++++++++++++++++++++++---
>  scripts/Makefile.lib              |  2 +-
>  6 files changed, 46 insertions(+), 65 deletions(-)
> 

-- 
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Intel Finland Oy

[U-Boot] [PATCH v1 1/7] Makefile: Don't shadow actual error when compile ASL

[Bin Meng]

Hi Andy,

On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> If ASL compiler failed by any reason do not produce output C file.
> Otherwise sequential run of make will shadow the actual error in ASL,
> i.e.
>
>   CC      board/intel/edison/dsdt.o
> board/intel/edison/dsdt.c:1:1: error: unterminated comment
>  /*
>
> and user has to remove dsdt.c and run make in order to see the error again.
>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  scripts/Makefile.lib | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>

Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

Although I am not sure how to reproduce "ASL compiler failed by any
reason do not produce output C file"..

Regards,
Bin

[U-Boot] [PATCH v1 2/7] Makefile: Export build date as integer

[Bin Meng]

On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> In some cases we would need build date as integer value.
> Export U_BOOT_BUILD_DATE as %Y%m%d integer value.
>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  Makefile | 2 ++
>  1 file changed, 2 insertions(+)
>

Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

[U-Boot] [PATCH v1 3/7] x86: acpi: Fill OEM revision

[Bin Meng]

On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> Fill OEM revision field in the tables by U-Boot build date.
>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  arch/x86/lib/acpi_table.c | 2 ++
>  1 file changed, 2 insertions(+)
>

Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

[U-Boot] [PATCH v1 4/7] x86: acpi: Name fields in FADT in accordance with specification

[Bin Meng]

On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> ACPI specification defines FADT fields marked as reserved in U-Boot.
>
> Name these fields in accordance with ACPI specification.
>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  arch/x86/include/asm/acpi_table.h | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
>

Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

[U-Boot] [PATCH v1 5/7] x86: acpi: Deduplicate acpi_fill_madt() implementation

[Bin Meng]

On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> In Baytrail and Quark support code acpi_fill_madt() is identical.
>
> Deduplicate its implementation by moving to lib/acpi_tables.c.
>
> At the same time mark acpi_fill_madt() with __weak attribute to keep a
> possibility to override it in platform code
>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  arch/x86/cpu/baytrail/acpi.c | 29 -----------------------------
>  arch/x86/cpu/quark/acpi.c    | 29 -----------------------------
>  arch/x86/lib/acpi_table.c    | 29 +++++++++++++++++++++++++++++
>  3 files changed, 29 insertions(+), 58 deletions(-)
>

Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

[U-Boot] [PATCH v1 6/7] x86: acpi: Export acpi_fill_mcfg() with __weak attribute

[Bin Meng]

On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> Some platforms might require different approach when filling memory
> mappings configuration table.
>
> Allow them to override the common method.
>
> At the same time export acpi_create_mcfg_mmconfig().
>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  arch/x86/include/asm/acpi_table.h | 3 +++
>  arch/x86/lib/acpi_table.c         | 6 +++---
>  2 files changed, 6 insertions(+), 3 deletions(-)
>

Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

[U-Boot] [PATCH v1 7/7] x86: acpi: Don't touch hardware on HW reduced platforms

[Bin Meng]

Hi Andy,

On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
> If ACPI HW reduced bit in FADT is set we should ignore any ACPI hardware
> communications.
>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  arch/x86/lib/acpi_table.c | 4 ++++
>  1 file changed, 4 insertions(+)
>
> diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
> index da6014fb41..3eb101105b 100644
> --- a/arch/x86/lib/acpi_table.c
> +++ b/arch/x86/lib/acpi_table.c
> @@ -463,6 +463,10 @@ ulong write_acpi_tables(ulong start)
>
>         debug("ACPI: done\n");
>
> +       /* Don't touch ACPI hardware on HW reduced platforms */
> +       if (fadt->flags & ACPI_FADT_HW_REDUCED_ACPI)
> +               return current;
> +

Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

I assume this is needed on Intel Edison. If that's the case, will you
support S3 resume on Edison? And if yes, I believe we will need update
acpi_resume() to add the same check there?

>         /*
>          * Other than waiting for OSPM to request us to switch to ACPI mode,
>          * do it by ourselves, since SMI will not be triggered.
> --

Regards,
Bin

[U-Boot] [PATCH v1 1/7] Makefile: Don't shadow actual error when compile ASL

[Andy Shevchenko]

On Sat, Jul 22, 2017 at 3:59 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
> Hi Andy,
>
> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
>> If ASL compiler failed by any reason do not produce output C file.
>> Otherwise sequential run of make will shadow the actual error in ASL,
>> i.e.
>>
>>   CC      board/intel/edison/dsdt.o
>> board/intel/edison/dsdt.c:1:1: error: unterminated comment
>>  /*
>>
>> and user has to remove dsdt.c and run make in order to see the error again.
>>
>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>> ---
>>  scripts/Makefile.lib | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>
> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
>

Thanks!

> Although I am not sure how to reproduce "ASL compiler failed by any
> reason do not produce output C file"..

Add

Name (XXX) ({

line to dsdt.asl for Minnowboard MAX and you will see

  ASL     board/intel/minnowmax/dsdt.asl
Compiler aborting due to parser-detected syntax error(s)
board/intel/minnowmax/dsdt.asl.tmp    739:  Name(XXX) ({
Error    6126 -                                    ^ syntax error,
unexpected PARSEOP_CLOSE_PAREN, expecting ','

Next run will get you above error (see commit message) which has been
quite confusing.

P.S. Of course it's artificial, but better to handle iasl (potential) failures.

-- 
With Best Regards,
Andy Shevchenko

[U-Boot] [PATCH v1 7/7] x86: acpi: Don't touch hardware on HW reduced platforms

[Andy Shevchenko]

On Sat, Jul 22, 2017 at 4:45 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
> Hi Andy,
>
> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
>> If ACPI HW reduced bit in FADT is set we should ignore any ACPI hardware
>> communications.
>>
>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>> ---
>>  arch/x86/lib/acpi_table.c | 4 ++++
>>  1 file changed, 4 insertions(+)
>>
>> diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
>> index da6014fb41..3eb101105b 100644
>> --- a/arch/x86/lib/acpi_table.c
>> +++ b/arch/x86/lib/acpi_table.c
>> @@ -463,6 +463,10 @@ ulong write_acpi_tables(ulong start)
>>
>>         debug("ACPI: done\n");
>>
>> +       /* Don't touch ACPI hardware on HW reduced platforms */
>> +       if (fadt->flags & ACPI_FADT_HW_REDUCED_ACPI)
>> +               return current;
>> +
>
> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

Thanks!

>
> I assume this is needed on Intel Edison.

Actually it has very little with Edison, though I had stumbled over
the bug on that platform.

ARM64 support, if any will come, or I even can suggest to try ASUS
T100 transformer (Baytrail based HW reduced).

> If that's the case, will you
> support S3 resume on Edison? And if yes, I believe we will need update
> acpi_resume() to add the same check there?

It might affect any of HW reduced one, needs to be checked.

I fixed only one issue I found, I didn't look at every places where
U-Boot goes to ACPI HW,

>
>>         /*
>>          * Other than waiting for OSPM to request us to switch to ACPI mode,
>>          * do it by ourselves, since SMI will not be triggered.
>> --
>
> Regards,
> Bin
> _______________________________________________
> U-Boot mailing list
> U-Boot at lists.denx.de
> https://lists.denx.de/listinfo/u-boot



-- 
With Best Regards,
Andy Shevchenko

[U-Boot] [PATCH v1 1/7] Makefile: Don't shadow actual error when compile ASL

[Bin Meng]

Hi Andy,

On Sun, Jul 23, 2017 at 12:38 AM, Andy Shevchenko
<andy.shevchenko@gmail.com> wrote:
> On Sat, Jul 22, 2017 at 3:59 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
>> Hi Andy,
>>
>> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
>> <andriy.shevchenko@linux.intel.com> wrote:
>>> If ASL compiler failed by any reason do not produce output C file.
>>> Otherwise sequential run of make will shadow the actual error in ASL,
>>> i.e.
>>>
>>>   CC      board/intel/edison/dsdt.o
>>> board/intel/edison/dsdt.c:1:1: error: unterminated comment
>>>  /*
>>>
>>> and user has to remove dsdt.c and run make in order to see the error again.
>>>
>>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>>> ---
>>>  scripts/Makefile.lib | 2 +-
>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>
>> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
>>
>
> Thanks!
>
>> Although I am not sure how to reproduce "ASL compiler failed by any
>> reason do not produce output C file"..
>
> Add
>
> Name (XXX) ({

Ah, I tried some pattern with no success but yours indeed triggers the issue.

>
> line to dsdt.asl for Minnowboard MAX and you will see
>
>   ASL     board/intel/minnowmax/dsdt.asl
> Compiler aborting due to parser-detected syntax error(s)
> board/intel/minnowmax/dsdt.asl.tmp    739:  Name(XXX) ({
> Error    6126 -                                    ^ syntax error,
> unexpected PARSEOP_CLOSE_PAREN, expecting ','
>
> Next run will get you above error (see commit message) which has been
> quite confusing.
>
> P.S. Of course it's artificial, but better to handle iasl (potential) failures.
>

Thanks for the clarification!
Tested-by: Bin Meng <bmeng.cn@gmail.com>

Regards,
Bin

[U-Boot] [PATCH v1 7/7] x86: acpi: Don't touch hardware on HW reduced platforms

[Bin Meng]

Hi Andy,

On Sun, Jul 23, 2017 at 12:43 AM, Andy Shevchenko
<andy.shevchenko@gmail.com> wrote:
> On Sat, Jul 22, 2017 at 4:45 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
>> Hi Andy,
>>
>> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
>> <andriy.shevchenko@linux.intel.com> wrote:
>>> If ACPI HW reduced bit in FADT is set we should ignore any ACPI hardware
>>> communications.
>>>
>>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>>> ---
>>>  arch/x86/lib/acpi_table.c | 4 ++++
>>>  1 file changed, 4 insertions(+)
>>>
>>> diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
>>> index da6014fb41..3eb101105b 100644
>>> --- a/arch/x86/lib/acpi_table.c
>>> +++ b/arch/x86/lib/acpi_table.c
>>> @@ -463,6 +463,10 @@ ulong write_acpi_tables(ulong start)
>>>
>>>         debug("ACPI: done\n");
>>>
>>> +       /* Don't touch ACPI hardware on HW reduced platforms */
>>> +       if (fadt->flags & ACPI_FADT_HW_REDUCED_ACPI)
>>> +               return current;
>>> +
>>
>> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
>
> Thanks!
>
>>
>> I assume this is needed on Intel Edison.
>
> Actually it has very little with Edison, though I had stumbled over
> the bug on that platform.
>
> ARM64 support, if any will come, or I even can suggest to try ASUS
> T100 transformer (Baytrail based HW reduced).
>
>> If that's the case, will you
>> support S3 resume on Edison? And if yes, I believe we will need update
>> acpi_resume() to add the same check there?
>
> It might affect any of HW reduced one, needs to be checked.
>
> I fixed only one issue I found, I didn't look at every places where
> U-Boot goes to ACPI HW,
>

OK, we will revisit that then once a HW reduced platform wants to support S3.

Regards,
Bin

[U-Boot] [PATCH v1 7/7] x86: acpi: Don't touch hardware on HW reduced platforms

[Bin Meng]

Hi Andy,

On Sun, Jul 23, 2017 at 12:43 AM, Andy Shevchenko
<andy.shevchenko@gmail.com> wrote:
> On Sat, Jul 22, 2017 at 4:45 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
>> Hi Andy,
>>
>> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
>> <andriy.shevchenko@linux.intel.com> wrote:
>>> If ACPI HW reduced bit in FADT is set we should ignore any ACPI hardware
>>> communications.
>>>
>>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>>> ---
>>>  arch/x86/lib/acpi_table.c | 4 ++++
>>>  1 file changed, 4 insertions(+)
>>>
>>> diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
>>> index da6014fb41..3eb101105b 100644
>>> --- a/arch/x86/lib/acpi_table.c
>>> +++ b/arch/x86/lib/acpi_table.c
>>> @@ -463,6 +463,10 @@ ulong write_acpi_tables(ulong start)
>>>
>>>         debug("ACPI: done\n");
>>>
>>> +       /* Don't touch ACPI hardware on HW reduced platforms */
>>> +       if (fadt->flags & ACPI_FADT_HW_REDUCED_ACPI)
>>> +               return current;
>>> +
>>
>> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
>
> Thanks!
>
>>
>> I assume this is needed on Intel Edison.
>
> Actually it has very little with Edison, though I had stumbled over
> the bug on that platform.
>
> ARM64 support, if any will come, or I even can suggest to try ASUS
> T100 transformer (Baytrail based HW reduced).
>

Just out of curiosity, if the ASUS T100 transformer is BayTrail based,
why is that HW reduced? My understanding is that BayTrail has these
ACPI registers.

>> If that's the case, will you
>> support S3 resume on Edison? And if yes, I believe we will need update
>> acpi_resume() to add the same check there?
>
> It might affect any of HW reduced one, needs to be checked.
>
> I fixed only one issue I found, I didn't look at every places where
> U-Boot goes to ACPI HW,
>

Regards,
Bin

[U-Boot] [PATCH v1 7/7] x86: acpi: Don't touch hardware on HW reduced platforms

[Andy Shevchenko]

On Sun, 2017-07-23 at 08:48 +0800, Bin Meng wrote:
> Hi Andy,
> 
> On Sun, Jul 23, 2017 at 12:43 AM, Andy Shevchenko
> <andy.shevchenko@gmail.com> wrote:
> > On Sat, Jul 22, 2017 at 4:45 PM, Bin Meng <bmeng.cn@gmail.com>
> > wrote:

> > > I assume this is needed on Intel Edison.
> > 
> > Actually it has very little with Edison, though I had stumbled over
> > the bug on that platform.
> > 
> > ARM64 support, if any will come, or I even can suggest to try ASUS
> > T100 transformer (Baytrail based HW reduced).
> > 
> 
> Just out of curiosity, if the ASUS T100 transformer is BayTrail based,
> why is that HW reduced? My understanding is that BayTrail has these
> ACPI registers.

Yeah, I don't know the reason why, but the decision had been made by
department without our team involvement. So, they make it HW reduced and
BIOS also had been adapted for this.

-- 
Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Intel Finland Oy

[U-Boot] [PATCH v1 1/7] Makefile: Don't shadow actual error when compile ASL

[Bin Meng]

On Sun, Jul 23, 2017 at 12:47 AM, Bin Meng <bmeng.cn@gmail.com> wrote:
> Hi Andy,
>
> On Sun, Jul 23, 2017 at 12:38 AM, Andy Shevchenko
> <andy.shevchenko@gmail.com> wrote:
>> On Sat, Jul 22, 2017 at 3:59 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
>>> Hi Andy,
>>>
>>> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
>>> <andriy.shevchenko@linux.intel.com> wrote:
>>>> If ASL compiler failed by any reason do not produce output C file.
>>>> Otherwise sequential run of make will shadow the actual error in ASL,
>>>> i.e.
>>>>
>>>>   CC      board/intel/edison/dsdt.o
>>>> board/intel/edison/dsdt.c:1:1: error: unterminated comment
>>>>  /*
>>>>
>>>> and user has to remove dsdt.c and run make in order to see the error again.
>>>>
>>>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>>>> ---
>>>>  scripts/Makefile.lib | 2 +-
>>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>>
>>>
>>> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
>>>
>>
>> Thanks!
>>
>>> Although I am not sure how to reproduce "ASL compiler failed by any
>>> reason do not produce output C file"..
>>
>> Add
>>
>> Name (XXX) ({
>
> Ah, I tried some pattern with no success but yours indeed triggers the issue.
>
>>
>> line to dsdt.asl for Minnowboard MAX and you will see
>>
>>   ASL     board/intel/minnowmax/dsdt.asl
>> Compiler aborting due to parser-detected syntax error(s)
>> board/intel/minnowmax/dsdt.asl.tmp    739:  Name(XXX) ({
>> Error    6126 -                                    ^ syntax error,
>> unexpected PARSEOP_CLOSE_PAREN, expecting ','
>>
>> Next run will get you above error (see commit message) which has been
>> quite confusing.
>>
>> P.S. Of course it's artificial, but better to handle iasl (potential) failures.
>>
>
> Thanks for the clarification!
> Tested-by: Bin Meng <bmeng.cn@gmail.com>

applied to u-boot-x86, thanks!

[U-Boot] [PATCH v1 2/7] Makefile: Export build date as integer

[Bin Meng]

On Sat, Jul 22, 2017 at 9:11 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
>> In some cases we would need build date as integer value.
>> Export U_BOOT_BUILD_DATE as %Y%m%d integer value.
>>
>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>> ---
>>  Makefile | 2 ++
>>  1 file changed, 2 insertions(+)
>>
>
> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

applied to u-boot-x86, thanks!

[U-Boot] [PATCH v1 3/7] x86: acpi: Fill OEM revision

[Bin Meng]

On Sat, Jul 22, 2017 at 9:12 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
>> Fill OEM revision field in the tables by U-Boot build date.
>>
>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>> ---
>>  arch/x86/lib/acpi_table.c | 2 ++
>>  1 file changed, 2 insertions(+)
>>
>
> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

applied to u-boot-x86, thanks!

[U-Boot] [PATCH v1 4/7] x86: acpi: Name fields in FADT in accordance with specification

[Bin Meng]

On Sat, Jul 22, 2017 at 9:14 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
>> ACPI specification defines FADT fields marked as reserved in U-Boot.
>>
>> Name these fields in accordance with ACPI specification.
>>
>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>> ---
>>  arch/x86/include/asm/acpi_table.h | 5 ++---
>>  1 file changed, 2 insertions(+), 3 deletions(-)
>>
>
> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

applied to u-boot-x86, thanks!

[U-Boot] [PATCH v1 5/7] x86: acpi: Deduplicate acpi_fill_madt() implementation

[Bin Meng]

On Sat, Jul 22, 2017 at 9:33 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
>> In Baytrail and Quark support code acpi_fill_madt() is identical.
>>
>> Deduplicate its implementation by moving to lib/acpi_tables.c.
>>
>> At the same time mark acpi_fill_madt() with __weak attribute to keep a
>> possibility to override it in platform code
>>
>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>> ---
>>  arch/x86/cpu/baytrail/acpi.c | 29 -----------------------------
>>  arch/x86/cpu/quark/acpi.c    | 29 -----------------------------
>>  arch/x86/lib/acpi_table.c    | 29 +++++++++++++++++++++++++++++
>>  3 files changed, 29 insertions(+), 58 deletions(-)
>>
>
> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

applied to u-boot-x86, thanks!

[U-Boot] [PATCH v1 6/7] x86: acpi: Export acpi_fill_mcfg() with __weak attribute

[Bin Meng]

On Sat, Jul 22, 2017 at 9:34 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
>> Some platforms might require different approach when filling memory
>> mappings configuration table.
>>
>> Allow them to override the common method.
>>
>> At the same time export acpi_create_mcfg_mmconfig().
>>
>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>> ---
>>  arch/x86/include/asm/acpi_table.h | 3 +++
>>  arch/x86/lib/acpi_table.c         | 6 +++---
>>  2 files changed, 6 insertions(+), 3 deletions(-)
>>
>
> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>

applied to u-boot-x86, thanks!

[U-Boot] [PATCH v1 7/7] x86: acpi: Don't touch hardware on HW reduced platforms

[Bin Meng]

On Sat, Jul 22, 2017 at 9:45 PM, Bin Meng <bmeng.cn@gmail.com> wrote:
> Hi Andy,
>
> On Sat, Jul 22, 2017 at 3:32 AM, Andy Shevchenko
> <andriy.shevchenko@linux.intel.com> wrote:
>> If ACPI HW reduced bit in FADT is set we should ignore any ACPI hardware
>> communications.
>>
>> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
>> ---
>>  arch/x86/lib/acpi_table.c | 4 ++++
>>  1 file changed, 4 insertions(+)
>>
>> diff --git a/arch/x86/lib/acpi_table.c b/arch/x86/lib/acpi_table.c
>> index da6014fb41..3eb101105b 100644
>> --- a/arch/x86/lib/acpi_table.c
>> +++ b/arch/x86/lib/acpi_table.c
>> @@ -463,6 +463,10 @@ ulong write_acpi_tables(ulong start)
>>
>>         debug("ACPI: done\n");
>>
>> +       /* Don't touch ACPI hardware on HW reduced platforms */
>> +       if (fadt->flags & ACPI_FADT_HW_REDUCED_ACPI)
>> +               return current;
>> +
>
> Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
>

applied to u-boot-x86, thanks!