Skip to contentSkip to conversation

aspeed: add support for a witherspoon-bmc board with 128MB flash chip

PATCH In review Cédric Le Goater 4 participants 9 messages May 16, 2019 – May 23, 2019 Arm/Aspeed machine support

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 aspeed: add support for a witherspoon-bmc board with 128MB flash chip

Some Witherspoon boards are shipped with a mt25ql01g flash chip holding the BMC firmware.

---

It helps in tracking https://github.com/openbmc/u-boot/issues/18

 hw/arm/aspeed.c | 11 +++++++++++
 1 file changed, 11 insertions(+)
hw/arm/aspeed.c +11 -0
index efc97df00de0..f26215ab5524 100644
@@ -441,6 +441,17 @@ static const AspeedBoardConfig aspeed_boards[] = {
         .i2c_init  = witherspoon_bmc_i2c_init,
         .ram       = 512 * MiB,
         .num_cpus  = 1,
+     }, {
+        .name      = MACHINE_TYPE_NAME("witherspoon128-bmc"),
+        .desc      = "OpenPOWER Witherspoon128 BMC (ARM1176)",
+        .soc_name  = "ast2500-a1",
+        .hw_strap1 = WITHERSPOON_BMC_HW_STRAP1,
+        .fmc_model = "mt25ql01g",
+        .spi_model = "mx66l1g45g",
+        .num_cs    = 2,
+        .i2c_init  = witherspoon_bmc_i2c_init,
+        .ram       = 512 * MiB,
+        .num_cpus  = 1,
     }, {
         .name      = MACHINE_TYPE_NAME("ast2600-evb"),
         .desc      = "Aspeed AST2600 EVB (Cortex A7)",
Signature · 1 line
2.20.1
Reviewed-by on v1 1/1
#1 lore raw
Quoted 4 lines · Cédric Le Goater

Some Witherspoon boards are shipped with a mt25ql01g flash chip holding the BMC firmware.

"witherspoon128-bmc" is not very creative, but I can't blame you for that :D

Quoted 33 lines

---

It helps in tracking https://github.com/openbmc/u-boot/issues/18

 hw/arm/aspeed.c | 11 +++++++++++
 1 file changed, 11 insertions(+)
hw/arm/aspeed.c +11 -0
index efc97df00de0..f26215ab5524 100644
@@ -441,6 +441,17 @@ static const AspeedBoardConfig aspeed_boards[] = {
         .i2c_init  = witherspoon_bmc_i2c_init,
         .ram       = 512 * MiB,
         .num_cpus  = 1,
+     }, {
+        .name      = MACHINE_TYPE_NAME("witherspoon128-bmc"),
+        .desc      = "OpenPOWER Witherspoon128 BMC (ARM1176)",
+        .soc_name  = "ast2500-a1",
+        .hw_strap1 = WITHERSPOON_BMC_HW_STRAP1,
+        .fmc_model = "mt25ql01g",
+        .spi_model = "mx66l1g45g",
+        .num_cs    = 2,
+        .i2c_init  = witherspoon_bmc_i2c_init,
+        .ram       = 512 * MiB,
+        .num_cpus  = 1,
     }, {
         .name      = MACHINE_TYPE_NAME("ast2600-evb"),
         .desc      = "Aspeed AST2600 EVB (Cortex A7)",
Signature · 1 line
2.20.1
#2 lore raw
Quoted 9 lines · Andrew Jeffery
Quoted 4 lines · Cédric Le Goater

Some Witherspoon boards are shipped with a mt25ql01g flash chip holding the BMC firmware.

"witherspoon128-bmc" is not very creative, but I can't blame you for that :D

he.

Should we introduce a machine option to set the model name of the FMC flash chip ? I think I tried that in the past but at the block level and mainline did not like it. But we could try at the machine level.

Anyhow, it's a good way to debug flash issues on real HW.

Quoted 35 lines
Quoted 33 lines

---

It helps in tracking https://github.com/openbmc/u-boot/issues/18

 hw/arm/aspeed.c | 11 +++++++++++
 1 file changed, 11 insertions(+)
hw/arm/aspeed.c +11 -0
index efc97df00de0..f26215ab5524 100644
@@ -441,6 +441,17 @@ static const AspeedBoardConfig aspeed_boards[] = {
         .i2c_init  = witherspoon_bmc_i2c_init,
         .ram       = 512 * MiB,
         .num_cpus  = 1,
+     }, {
+        .name      = MACHINE_TYPE_NAME("witherspoon128-bmc"),
+        .desc      = "OpenPOWER Witherspoon128 BMC (ARM1176)",
+        .soc_name  = "ast2500-a1",
+        .hw_strap1 = WITHERSPOON_BMC_HW_STRAP1,
+        .fmc_model = "mt25ql01g",
+        .spi_model = "mx66l1g45g",
+        .num_cs    = 2,
+        .i2c_init  = witherspoon_bmc_i2c_init,
+        .ram       = 512 * MiB,
+        .num_cpus  = 1,
     }, {
         .name      = MACHINE_TYPE_NAME("ast2600-evb"),
         .desc      = "Aspeed AST2600 EVB (Cortex A7)",
Signature · 1 line
2.20.1
#3 lore raw
Quoted 11 lines · Cédric Le Goater
Quoted 9 lines · Andrew Jeffery
Quoted 4 lines · Cédric Le Goater

Some Witherspoon boards are shipped with a mt25ql01g flash chip holding the BMC firmware.

"witherspoon128-bmc" is not very creative, but I can't blame you for that :D

Quoted 8 lines

he.

Should we introduce a machine option to set the model name of the FMC flash chip ? I think I tried that in the past but at the block level and mainline did not like it. But we could try at the machine level.

Anyhow, it's a good way to debug flash issues on real HW.

Eh. I'm not fussed by a new machine. I wouldn't do anything for the moment.

Quoted 39 lines
Quoted 37 lines
Quoted 33 lines

---

It helps in tracking https://github.com/openbmc/u-boot/issues/18

 hw/arm/aspeed.c | 11 +++++++++++
 1 file changed, 11 insertions(+)
hw/arm/aspeed.c +11 -0
index efc97df00de0..f26215ab5524 100644
@@ -441,6 +441,17 @@ static const AspeedBoardConfig aspeed_boards[] = {
         .i2c_init  = witherspoon_bmc_i2c_init,
         .ram       = 512 * MiB,
         .num_cpus  = 1,
+     }, {
+        .name      = MACHINE_TYPE_NAME("witherspoon128-bmc"),
+        .desc      = "OpenPOWER Witherspoon128 BMC (ARM1176)",
+        .soc_name  = "ast2500-a1",
+        .hw_strap1 = WITHERSPOON_BMC_HW_STRAP1,
+        .fmc_model = "mt25ql01g",
+        .spi_model = "mx66l1g45g",
+        .num_cs    = 2,
+        .i2c_init  = witherspoon_bmc_i2c_init,
+        .ram       = 512 * MiB,
+        .num_cpus  = 1,
     }, {
         .name      = MACHINE_TYPE_NAME("ast2600-evb"),
         .desc      = "Aspeed AST2600 EVB (Cortex A7)",
Signature · 1 line
2.20.1
maintainer
#4 lore raw
Quoted 17 lines · Andrew Jeffery
Quoted 13 lines · Cédric Le Goater
Quoted 9 lines · Andrew Jeffery
Quoted 4 lines · Cédric Le Goater

Some Witherspoon boards are shipped with a mt25ql01g flash chip holding the BMC firmware.

"witherspoon128-bmc" is not very creative, but I can't blame you for that :D

he.

Quoted 8 lines
Quoted 6 lines

Should we introduce a machine option to set the model name of the FMC flash chip ? I think I tried that in the past but at the block level and mainline did not like it. But we could try at the machine level.

Anyhow, it's a good way to debug flash issues on real HW.

Eh. I'm not fussed by a new machine. I wouldn't do anything for the moment.

We could instead add a swift machine with the 128MB NOR. When we get a model for the SD/MMC it would use that too.

Cheers,

Joel

#5 lore raw
Quoted 23 lines · Joel Stanley
Quoted 22 lines · Andrew Jeffery
Quoted 18 lines · Cédric Le Goater
Quoted 9 lines · Andrew Jeffery
Quoted 4 lines · Cédric Le Goater

Some Witherspoon boards are shipped with a mt25ql01g flash chip holding the BMC firmware.

"witherspoon128-bmc" is not very creative, but I can't blame you for that :D

he.

Should we introduce a machine option to set the model name of the FMC flash chip ? I think I tried that in the past but at the block level and mainline did not like it. But we could try at the machine level.

Quoted 6 lines
Quoted 3 lines
Quoted 1 line

Anyhow, it's a good way to debug flash issues on real HW.

Eh. I'm not fussed by a new machine. I wouldn't do anything for the moment.

We could instead add a swift machine with the 128MB NOR. When we get a model for the SD/MMC it would use that too.

OK. That's a good direction.

C.

maintainer
#6 lore raw
Quoted 25 lines · Cédric Le Goater
Quoted 23 lines · Joel Stanley
Quoted 22 lines · Andrew Jeffery
Quoted 18 lines · Cédric Le Goater
Quoted 9 lines · Andrew Jeffery
Quoted 4 lines · Cédric Le Goater

Some Witherspoon boards are shipped with a mt25ql01g flash chip holding the BMC firmware.

"witherspoon128-bmc" is not very creative, but I can't blame you for that :D

he.

Should we introduce a machine option to set the model name of the FMC flash chip ? I think I tried that in the past but at the block level and mainline did not like it. But we could try at the machine level.

Quoted 8 lines
Quoted 6 lines
Quoted 3 lines
Quoted 1 line

Anyhow, it's a good way to debug flash issues on real HW.

Eh. I'm not fussed by a new machine. I wouldn't do anything for the moment.

We could instead add a swift machine with the 128MB NOR. When we get a model for the SD/MMC it would use that too.

OK. That's a good direction.

Adriana, could you take this on?

#7 lore raw
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 7 lines

Should we introduce a machine option to set the model name of the FMC flash chip ? I think I tried that in the past but at the block level and mainline did not like it. But we could try at the machine level.

Anyhow, it's a good way to debug flash issues on real HW.

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Quoted 1 line

Eh. I'm not fussed by a new machine. I wouldn't do anything for the moment.

We could instead add a swift machine with the 128MB NOR. When we get a model for the SD/MMC it would use that too.

OK. That's a good direction.

Adriana, could you take this on?

Yeah, was thinking about it when Cedric sent out this patch, but since theoretically the witherspoon128-bmc model could be used with a swift image, I put it to the side. But now that you mention it, yeah we can add the SD/MMC to it later so that'd be valuable to have in a qemu model. Will send a patch based on the 128 for a new swift bmc to start with.

#8 lore raw
Quoted 10 lines · Adriana Kobylak
Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 9 lines
Quoted 7 lines

Should we introduce a machine option to set the model name of the FMC flash chip ? I think I tried that in the past but at the block level and mainline did not like it. But we could try at the machine level.

Anyhow, it's a good way to debug flash issues on real HW.

Eh. I'm not fussed by a new machine. I wouldn't do anything for the moment.

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Quoted 2 lines

We could instead add a swift machine with the 128MB NOR. When we get a model for the SD/MMC it would use that too.

OK. That's a good direction.

Adriana, could you take this on?

Yeah, was thinking about it when Cedric sent out this patch, but since theoretically the witherspoon128-bmc model could be used with a swift image, I put it to the side. But now that you mention it, yeah we can add the SD/MMC to it later so that'd be valuable to have in a qemu model. Will send a patch based on the 128 for a new swift bmc to start with.

Yes. I suppose this machine will have a different layout of I2C devices also, different HW strapping, etc.

Thanks,

C.