Skip to contentSkip to conversation

ARM: OMAP5: Add AVS support for ES2.0

PATCH New Nishanth Menon 3 participants 7 messages Aug 5, 2017 – Aug 14, 2017

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

cover ARM: OMAP5: Add AVS support for ES2.0

Hi,

OMAP5 went into production with AVS Class0 Vs pre-production configuration of AVS 1.5. Most of the current "in-the-wild" OMAP5 uEVMs are based on pre-production samples, however, the support for "production configuration" helps platforms such as pyra-handheld[1].

This original issue was reported by zmatt on #beagle freenode IRC channel, However, I donot have a email address to attribute to him/her :( (zmatt, if you are on the list, feel free to add your reported and/or tested)

Series verified (with scope) on:
a) 720-2644-001 OMAP5UEVM with production sample.
b) 750-2628-222(A) UEVM5432G-02 with pre-production sample.

build tested with:
buildman -b omap5-evm omap5

[1] https://www.pyra-handheld.com/wiki/index.php?title=OMAP_5

Nishanth Menon (2):
ARM: OMAP5: Remove OPP_LOW Definitions for ES2.0
ARM: OMAP5: Enable support for AVS0 for OMAP5 production devices

 arch/arm/include/asm/arch-omap5/clock.h | 17 +++++++++++++++--
 arch/arm/include/asm/arch-omap5/omap.h  |  4 ++--
 arch/arm/mach-omap2/omap5/abb.c         |  4 ++--
 arch/arm/mach-omap2/omap5/hw_data.c     |  9 +++++++++
 4 files changed, 28 insertions(+), 6 deletions(-)
Signature · 1 line
2.12.2
#1 lore raw

PATCH 1/2 ARM: OMAP5: Remove OPP_LOW Definitions for ES2.0

ES2.0 descopes OPP_LOW definition. So remove it from macros defined.

---
 arch/arm/include/asm/arch-omap5/clock.h | 3 ---
 1 file changed, 3 deletions(-)
arch/arm/include/asm/arch-omap5/clock.h +0 -3
index 0c99bbdc9320..dd05e8ba8317 100644
@@ -236,9 +236,6 @@
 #define VDD_MPU_ES2_HIGH 1250
 #define VDD_MM_ES2_OD  1120
 
-#define VDD_MPU_ES2_LOW 880
-#define VDD_MM_ES2_LOW 880
-
 /* DRA74x/75x/72x voltage settings in mv for OPP_NOM per DM */
 #define VDD_MPU_DRA7_NOM	1150
 #define VDD_CORE_DRA7_NOM	1150
Signature · 1 line
2.12.2
#4 lore raw

PATCH 2/2 ARM: OMAP5: Enable support for AVS0 for OMAP5 production devices

In reply to Nishanth Menon

OMAP5432 did go into production with AVS class0 registers which were mutually exclusive from AVS Class 1.5 registers.

Most OMAP5-uEVM boards use the pre-production Class1.5 which has production efuse registers set to 0. However on production devices, these are set to valid data.

scale_vcore logic is already smart enough to detect this and use the "Nominal voltage" on devices that do not have efuse registers populated.

On a test production device populated as follows:
MPU OPP_NOM:
=> md.l 0x04A0021C4 1
4a0021c4: 03a003e9 ....
(0x3e9 = 1.01v) vs nom voltage of 1.06v
MPU OPP_HIGH:
=> md.l 0x04A0021C8 1
4a0021c8: 03400485 .. at .

MM OPP_NOM:
=> md.l 0x04A0021A4 1
4a0021a4: 038003d4 ....
(0x3d4 = 980mV) vs nom voltage of 1.025v
MM OPP_OD:
=> md.l 0x04A0021A8 1
4a0021a8: 03600403 ..`.

CORE OPP_NOM:
=> md.l 0x04A0021D8 1
4a0021d8: 000003cf ....
(0x3cf = 975mV) vs nom voltage of 1.040v

Since the efuse values are'nt currently used, we do not regress on existing pre-production samples (they continue to use nominal voltage).

But on boards that do have production samples populated, we can leverage the optimal voltages necessary for proper operation.

Tested on:
a) 720-2644-001 OMAP5UEVM with production sample.
b) 750-2628-222(A) UEVM5432G-02 with pre-production sample.

Data based on OMAP5432 Technical reference Manual SWPU282AF (May 2012-Revised Aug 2016)

NOTE: All collaterals on OMAP5432 silicon itself seems to have been removed from ti.com, though EVM details are still available:
http://www.ti.com/tool/OMAP5432-EVM

---
 arch/arm/include/asm/arch-omap5/clock.h | 16 ++++++++++++++++
 arch/arm/include/asm/arch-omap5/omap.h  |  4 ++--
 arch/arm/mach-omap2/omap5/abb.c         |  4 ++--
 arch/arm/mach-omap2/omap5/hw_data.c     |  9 +++++++++
 4 files changed, 29 insertions(+), 4 deletions(-)
arch/arm/include/asm/arch-omap5/clock.h +16 -0
index dd05e8ba8317..dbe340d23e4f 100644
@@ -236,6 +236,22 @@
 #define VDD_MPU_ES2_HIGH 1250
 #define VDD_MM_ES2_OD  1120
 
+/* Efuse register offsets for OMAP5 platform */
+#define OMAP5_ES2_EFUSE_BASE	0x4A002000
+#define OMAP5_ES2_PROD_REGBITS	16
+
+/* CONTROL_STD_FUSE_OPP_VDD_CORE_3 */
+#define OMAP5_ES2_PROD_CORE_OPNO_VMIN	(OMAP5_ES2_EFUSE_BASE + 0x1D8)
+
+/* CONTROL_STD_FUSE_OPP_VDD_MM_4 */
+#define OMAP5_ES2_PROD_MM_OPNO_VMIN	(OMAP5_ES2_EFUSE_BASE + 0x1A4)
+/* CONTROL_STD_FUSE_OPP_VDD_MM_5 */
+#define OMAP5_ES2_PROD_MM_OPOD_VMIN	(OMAP5_ES2_EFUSE_BASE + 0x1A8)
+/* CONTROL_STD_FUSE_OPP_VDD_MPU_6 */
+#define OMAP5_ES2_PROD_MPU_OPNO_VMIN	(OMAP5_ES2_EFUSE_BASE + 0x1C4)
+/* CONTROL_STD_FUSE_OPP_VDD_MPU_7 */
+#define OMAP5_ES2_PROD_MPU_OPHI_VMIN	(OMAP5_ES2_EFUSE_BASE + 0x1C8)
+
 /* DRA74x/75x/72x voltage settings in mv for OPP_NOM per DM */
 #define VDD_MPU_DRA7_NOM	1150
 #define VDD_CORE_DRA7_NOM	1150
arch/arm/include/asm/arch-omap5/omap.h +2 -2
index 2f005dd3ad92..b047f0d65081 100644
@@ -224,8 +224,8 @@ struct s32ktimer {
 #define OMAP_ABB_GPU_TXDONE_MASK		(0x1 << 28)
 
 /* ABB efuse masks */
-#define OMAP5_ABB_FUSE_VSET_MASK		(0x1F << 24)
-#define OMAP5_ABB_FUSE_ENABLE_MASK		(0x1 << 29)
+#define OMAP5_PROD_ABB_FUSE_VSET_MASK		(0x1F << 20)
+#define OMAP5_PROD_ABB_FUSE_ENABLE_MASK		(0x1 << 25)
 #define DRA7_ABB_FUSE_VSET_MASK			(0x1F << 20)
 #define DRA7_ABB_FUSE_ENABLE_MASK		(0x1 << 25)
 #define OMAP5_ABB_LDOVBBMPU_MUX_CTRL_MASK	(0x1 << 10)
arch/arm/mach-omap2/omap5/abb.c +2 -2
index 3bf88979e5d6..1882c49e7df6 100644
@@ -28,8 +28,8 @@
 s8 abb_setup_ldovbb(u32 fuse, u32 ldovbb)
 {
 	u32 vset;
-	u32 fuse_enable_mask = OMAP5_ABB_FUSE_ENABLE_MASK;
-	u32 fuse_vset_mask = OMAP5_ABB_FUSE_VSET_MASK;
+	u32 fuse_enable_mask = OMAP5_PROD_ABB_FUSE_ENABLE_MASK;
+	u32 fuse_vset_mask = OMAP5_PROD_ABB_FUSE_VSET_MASK;
 
 	if (!is_omap54xx()) {
 		/* DRA7 */
arch/arm/mach-omap2/omap5/hw_data.c +9 -0
index a8a6b8a869e5..4ad6b530d29c 100644
@@ -329,6 +329,15 @@ struct vcores_data omap5430_volts_es2 = {
 	.mm.addr = SMPS_REG_ADDR_45_IVA,
 	.mm.pmic = &palmas,
 	.mm.abb_tx_done_mask = OMAP_ABB_MM_TXDONE_MASK,
+
+	.mpu.efuse.reg[OPP_NOM]	= OMAP5_ES2_PROD_MPU_OPNO_VMIN,
+	.mpu.efuse.reg_bits	= OMAP5_ES2_PROD_REGBITS,
+
+	.core.efuse.reg[OPP_NOM] = OMAP5_ES2_PROD_CORE_OPNO_VMIN,
+	.core.efuse.reg_bits	= OMAP5_ES2_PROD_REGBITS,
+
+	.mm.efuse.reg[OPP_NOM]	= OMAP5_ES2_PROD_MM_OPNO_VMIN,
+	.mm.efuse.reg_bits	= OMAP5_ES2_PROD_REGBITS,
 };
 
 /*
Signature · 1 line
2.12.2
#5 lore raw

PATCH 2/2 ARM: OMAP5: Enable support for AVS0 for OMAP5 production devices

In reply to Nishanth Menon
Quoted 42 lines · Nishanth Menon

OMAP5432 did go into production with AVS class0 registers which were mutually exclusive from AVS Class 1.5 registers.

Most OMAP5-uEVM boards use the pre-production Class1.5 which has production efuse registers set to 0. However on production devices, these are set to valid data.

scale_vcore logic is already smart enough to detect this and use the "Nominal voltage" on devices that do not have efuse registers populated.

On a test production device populated as follows:
MPU OPP_NOM:
=> md.l 0x04A0021C4 1
4a0021c4: 03a003e9 ....
(0x3e9 = 1.01v) vs nom voltage of 1.06v
MPU OPP_HIGH:
=> md.l 0x04A0021C8 1
4a0021c8: 03400485 .. at .

MM OPP_NOM:
=> md.l 0x04A0021A4 1
4a0021a4: 038003d4 ....
(0x3d4 = 980mV) vs nom voltage of 1.025v
MM OPP_OD:
=> md.l 0x04A0021A8 1
4a0021a8: 03600403 ..`.

CORE OPP_NOM:
=> md.l 0x04A0021D8 1
4a0021d8: 000003cf ....
(0x3cf = 975mV) vs nom voltage of 1.040v

Since the efuse values are'nt currently used, we do not regress on existing pre-production samples (they continue to use nominal voltage).

But on boards that do have production samples populated, we can leverage the optimal voltages necessary for proper operation.

Tested on:
a) 720-2644-001 OMAP5UEVM with production sample.
b) 750-2628-222(A) UEVM5432G-02 with pre-production sample.

Quoted 7 lines

Data based on OMAP5432 Technical reference Manual SWPU282AF (May 2012-Revised Aug 2016)

NOTE: All collaterals on OMAP5432 silicon itself seems to have been removed from ti.com, though EVM details are still available:
http://www.ti.com/tool/OMAP5432-EVM

Thanks and regards,
Lokesh

Quoted 1 line
#3 lore raw

PATCH 1/2 ARM: OMAP5: Remove OPP_LOW Definitions for ES2.0

In reply to Nishanth Menon
Quoted 4 lines · Nishanth Menon

ES2.0 descopes OPP_LOW definition. So remove it from macros defined.

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20170813/39284935/attachment.sig>
#6 lore raw

PATCH 2/2 ARM: OMAP5: Enable support for AVS0 for OMAP5 production devices

In reply to Nishanth Menon
Quoted 45 lines · Nishanth Menon

OMAP5432 did go into production with AVS class0 registers which were mutually exclusive from AVS Class 1.5 registers.

Most OMAP5-uEVM boards use the pre-production Class1.5 which has production efuse registers set to 0. However on production devices, these are set to valid data.

scale_vcore logic is already smart enough to detect this and use the "Nominal voltage" on devices that do not have efuse registers populated.

On a test production device populated as follows:
MPU OPP_NOM:
=> md.l 0x04A0021C4 1
4a0021c4: 03a003e9 ....
(0x3e9 = 1.01v) vs nom voltage of 1.06v
MPU OPP_HIGH:
=> md.l 0x04A0021C8 1
4a0021c8: 03400485 .. at .

MM OPP_NOM:
=> md.l 0x04A0021A4 1
4a0021a4: 038003d4 ....
(0x3d4 = 980mV) vs nom voltage of 1.025v
MM OPP_OD:
=> md.l 0x04A0021A8 1
4a0021a8: 03600403 ..`.

CORE OPP_NOM:
=> md.l 0x04A0021D8 1
4a0021d8: 000003cf ....
(0x3cf = 975mV) vs nom voltage of 1.040v

Since the efuse values are'nt currently used, we do not regress on existing pre-production samples (they continue to use nominal voltage).

But on boards that do have production samples populated, we can leverage the optimal voltages necessary for proper operation.

Tested on:
a) 720-2644-001 OMAP5UEVM with production sample.
b) 750-2628-222(A) UEVM5432G-02 with pre-production sample.

Data based on OMAP5432 Technical reference Manual SWPU282AF (May 2012-Revised Aug 2016)

Quoted 6 lines

NOTE: All collaterals on OMAP5432 silicon itself seems to have been removed from ti.com, though EVM details are still available:
http://www.ti.com/tool/OMAP5432-EVM

Applied to u-boot/master, thanks!

Signature · 8 lines
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20170813/e62c8f57/attachment.sig>