subject:"eMMC support"


On 5/31/22 21:07, Philippe Mathieu-Daudé wrote:

On 31/5/22 11:19, Cédric Le Goater wrote:

On 5/30/22 21:37, Philippe Mathieu-Daudé wrote:

From: Philippe Mathieu-Daudé 

Rebase/respin of Cédric RFC:
https://lore.kernel.org/qemu-devel/20220318132824.1134400-1-...@kaod.org/
(sorry it took me so long guys...)

Pushed at https://gitlab.com/philmd/qemu/-/commits/emmc-v2

I plan to queue patches 1-12 via sdmmc-next later this week.

Cédric, if you are happy with this series, it should be easy to rebase
your other patches on top and address the comments I left on the RFC :)


I pushed an update on :

   https://github.com/legoater/qemu/commits/aspeed-7.1

Here is an image :

   https://www.kaod.org/qemu/aspeed/mmc-p10bmc.qcow2

run with :

  qemu-system-arm -M rainier-bmc -net nic -net user -drive 
file=./mmc-p10bmc.qcow2,format=qcow2,if=sd,id=sd0,index=2 -nographic 
-nodefaults -snapshot -serial mon:stdio


Useful, thanks.

I see in hw/arm/aspeed_ast2600.c:

     /* Init sd card slot class here so that they're under the correct parent */
     for (i = 0; i < ASPEED_SDHCI_NUM_SLOTS; ++i) {
     object_initialize_child(obj, "sd-controller.sdhci[*]",
     >sdhci.slots[i], TYPE_SYSBUS_SDHCI);
     }

     object_initialize_child(obj, "emmc-controller.sdhci", >emmc.slots[0],
     TYPE_SYSBUS_SDHCI);

     /* eMMC Boot Controller stub */
     create_unimplemented_device("aspeed.emmc-boot-controller",
     sc->memmap[ASPEED_DEV_EMMC_BC],
     0x1000);

     /* eMMC */
     if (!sysbus_realize(SYS_BUS_DEVICE(>emmc), errp)) {
     return;
     }
     sysbus_mmio_map(SYS_BUS_DEVICE(>emmc), 0, sc->memmap[ASPEED_DEV_EMMC]);
     sysbus_connect_irq(SYS_BUS_DEVICE(>emmc), 0,
    aspeed_soc_get_irq(s, ASPEED_DEV_EMMC));

Where is 'emmc-controller.sdhci' realized?


the slots are realized in aspeed_sdhci_realize(). It's not very
symmetric and the names are confusing.

I think that one of the problems is that the instance_init routine
of TYPE_ASPEED_SDHCI object doesn't know on how much slots
object_initialize_child() should be called since it depends on
its flavor : SD/eMMC.


In aspeed_sdhci_realize() you set sd-spec-version" = 2, is that OK
with eMMC?


ah yes. it boots anyhow.


What expects the real hw?

  ast2400 ast2500 ast2600
 
SDHC cardv2.0/v3.0   v2.0/v3.0   v2.0/v3.0	

SDIO Host  v2.0v2.0v2.0 
SD slots2   2   2   
eMMCx v4.51   v5.1
eMMC slots  x   1   1

on the ast2500, the SDIO and eMMC logics are combined in one controller
but since it is not used, QEMU does not model the eMMC part.

Thanks,

C.

Re: [PATCH v2 00/25] hw/sd: Rework models for eMMC support


On 31/5/22 11:19, Cédric Le Goater wrote:

On 5/30/22 21:37, Philippe Mathieu-Daudé wrote:

From: Philippe Mathieu-Daudé 

Rebase/respin of Cédric RFC:
https://lore.kernel.org/qemu-devel/20220318132824.1134400-1-...@kaod.org/
(sorry it took me so long guys...)

Pushed at https://gitlab.com/philmd/qemu/-/commits/emmc-v2

I plan to queue patches 1-12 via sdmmc-next later this week.

Cédric, if you are happy with this series, it should be easy to rebase
your other patches on top and address the comments I left on the RFC :)


I pushed an update on :

   https://github.com/legoater/qemu/commits/aspeed-7.1

Here is an image :

   https://www.kaod.org/qemu/aspeed/mmc-p10bmc.qcow2

run with :

  qemu-system-arm -M rainier-bmc -net nic -net user -drive 
file=./mmc-p10bmc.qcow2,format=qcow2,if=sd,id=sd0,index=2 -nographic 
-nodefaults -snapshot -serial mon:stdio


Useful, thanks.

I see in hw/arm/aspeed_ast2600.c:

/* Init sd card slot class here so that they're under the correct 
parent */

for (i = 0; i < ASPEED_SDHCI_NUM_SLOTS; ++i) {
object_initialize_child(obj, "sd-controller.sdhci[*]",
>sdhci.slots[i], TYPE_SYSBUS_SDHCI);
}

object_initialize_child(obj, "emmc-controller.sdhci", 
>emmc.slots[0],

TYPE_SYSBUS_SDHCI);

/* eMMC Boot Controller stub */
create_unimplemented_device("aspeed.emmc-boot-controller",
sc->memmap[ASPEED_DEV_EMMC_BC],
0x1000);

/* eMMC */
if (!sysbus_realize(SYS_BUS_DEVICE(>emmc), errp)) {
return;
}
sysbus_mmio_map(SYS_BUS_DEVICE(>emmc), 0, 
sc->memmap[ASPEED_DEV_EMMC]);

sysbus_connect_irq(SYS_BUS_DEVICE(>emmc), 0,
   aspeed_soc_get_irq(s, ASPEED_DEV_EMMC));

Where is 'emmc-controller.sdhci' realized?

In aspeed_sdhci_realize() you set sd-spec-version" = 2, is that OK
with eMMC?

What expects the real hw?

Thanks,

Phil.

Re: [PATCH v2 00/25] hw/sd: Rework models for eMMC support


On 5/30/22 21:37, Philippe Mathieu-Daudé wrote:

From: Philippe Mathieu-Daudé 

Rebase/respin of Cédric RFC:
https://lore.kernel.org/qemu-devel/20220318132824.1134400-1-...@kaod.org/
(sorry it took me so long guys...)

Pushed at https://gitlab.com/philmd/qemu/-/commits/emmc-v2

I plan to queue patches 1-12 via sdmmc-next later this week.

Cédric, if you are happy with this series, it should be easy to rebase
your other patches on top and address the comments I left on the RFC :)


I pushed an update on :

  https://github.com/legoater/qemu/commits/aspeed-7.1

Here is an image :

  https://www.kaod.org/qemu/aspeed/mmc-p10bmc.qcow2

run with :

 qemu-system-arm -M rainier-bmc -net nic -net user -drive 
file=./mmc-p10bmc.qcow2,format=qcow2,if=sd,id=sd0,index=2 -nographic 
-nodefaults -snapshot -serial mon:stdio

Thanks,

C.


Regards,

Phil.

Cédric Le Goater (6):
   hw/sd: Add sd_emmc_cmd_SEND_OP_CMD() handler
   hw/sd: Add sd_emmc_cmd_ALL_SEND_CID() handler
   hw/sd: Add sd_emmc_cmd_SEND_RELATIVE_ADDR() handler
   hw/sd: Add sd_emmc_cmd_APP_CMD() handler
   hw/sd: add sd_emmc_cmd_SEND_TUNING_BLOCK() handler
   hw/sd: Add sd_emmc_cmd_SEND_EXT_CSD() handler

Joel Stanley (4):
   hw/sd: Add sd_cmd_SEND_TUNING_BLOCK() handler
   hw/sd: Support boot area in emmc image
   hw/sd: Subtract bootarea size from blk
   hw/sd: Add boot config support

Philippe Mathieu-Daudé (13):
   hw/sd/sdcard: Return ILLEGAL for CMD19/CMD23 prior SD spec v3.01
   hw/sd: When card is in wrong state, log which state it is
   hw/sd: When card is in wrong state, log which spec version is used
   hw/sd: Move proto_name to SDProto structure
   hw/sd: Introduce sd_cmd_handler type
   hw/sd: Add sd_cmd_illegal() handler
   hw/sd: Add sd_cmd_unimplemented() handler
   hw/sd: Add sd_cmd_GO_IDLE_STATE() handler
   hw/sd: Add sd_cmd_SEND_OP_CMD() handler
   hw/sd: Add sd_cmd_ALL_SEND_CID() handler
   hw/sd: Add sd_cmd_SEND_RELATIVE_ADDR() handler
   hw/sd: Add sd_cmd_SET_BLOCK_COUNT() handler
   hw/sd: Basis for eMMC support

Sai Pavan Boddu (2):
   hw/sd: Add CMD21 tuning sequence
   hw/sd: Add mmc switch function support

  hw/sd/sd.c | 645 +
  hw/sd/sdmmc-internal.c |   2 +-
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   7 +
  4 files changed, 627 insertions(+), 124 deletions(-)

Re: [PATCH v2 14/25] hw/sd: Basis for eMMC support


On 5/30/22 21:38, Philippe Mathieu-Daudé wrote:

From: Philippe Mathieu-Daudé 

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework.

Signed-off-by: Cédric Le Goater 
Signed-off-by: Philippe Mathieu-Daudé 
---
TODO: Do not inherit TYPE_SD_CARD, duplicate sd_class_init()
---
  hw/sd/sd.c | 42 ++
  include/hw/sd/sd.h |  3 +++
  2 files changed, 45 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index b2f16dbb73..8b178aa261 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -2166,6 +2166,19 @@ static const SDProto sd_proto_sd = {
  },
  };
  
+static const SDProto sd_proto_emmc = {

+.name = "eMMC",
+.cmd = {
+[0] = sd_cmd_GO_IDLE_STATE,
+[5] = sd_cmd_illegal,



And this needed an extra :

+[23]= sd_cmd_SET_BLOCK_COUNT,


Thanks,

C.



+[19]= sd_cmd_SEND_TUNING_BLOCK,
+[41]= sd_cmd_illegal,
+[52 ... 54] = sd_cmd_illegal,
+[58]= sd_cmd_illegal,
+[59]= sd_cmd_illegal,
+},
+};
+
  static void sd_instance_init(Object *obj)
  {
  SDState *sd = SD_CARD(obj);
@@ -2284,9 +2297,38 @@ static const TypeInfo sd_info = {
  .instance_finalize = sd_instance_finalize,
  };
  
+static void emmc_realize(DeviceState *dev, Error **errp)

+{
+SDState *sd = SD_CARD(dev);
+
+if (sd->spec_version < SD_PHY_SPECv3_01_VERS) {
+error_setg(errp, "Minimum spec for eMMC is v3.01");
+return;
+}
+
+sd_realize(dev, errp);
+}
+
+static void emmc_class_init(ObjectClass *klass, void *data)
+{
+DeviceClass *dc = DEVICE_CLASS(klass);
+SDCardClass *sc = SD_CARD_CLASS(klass);
+
+dc->desc = "eMMC";
+dc->realize = emmc_realize;
+sc->proto = _proto_emmc;
+}
+
+static const TypeInfo emmc_info = {
+.name = TYPE_EMMC,
+.parent = TYPE_SD_CARD,
+.class_init = emmc_class_init,
+ };
+
  static void sd_register_types(void)
  {
  type_register_static(_info);
+type_register_static(_info);
  }
  
  type_init(sd_register_types)

diff --git a/include/hw/sd/sd.h b/include/hw/sd/sd.h
index 0d94e1f346..e52436b7a5 100644
--- a/include/hw/sd/sd.h
+++ b/include/hw/sd/sd.h
@@ -93,6 +93,9 @@ typedef struct {
  #define TYPE_SD_CARD "sd-card"
  OBJECT_DECLARE_TYPE(SDState, SDCardClass, SD_CARD)
  
+#define TYPE_EMMC "emmc"

+DECLARE_INSTANCE_CHECKER(SDState, EMMC, TYPE_EMMC)
+
  struct SDCardClass {
  /*< private >*/
  DeviceClass parent_class;

Re: [PATCH v2 14/25] hw/sd: Basis for eMMC support


On 5/30/22 21:38, Philippe Mathieu-Daudé wrote:

From: Philippe Mathieu-Daudé 

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework.

Signed-off-by: Cédric Le Goater 
Signed-off-by: Philippe Mathieu-Daudé 
---
TODO: Do not inherit TYPE_SD_CARD, duplicate sd_class_init()
---
  hw/sd/sd.c | 42 ++
  include/hw/sd/sd.h |  3 +++
  2 files changed, 45 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index b2f16dbb73..8b178aa261 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -2166,6 +2166,19 @@ static const SDProto sd_proto_sd = {
  },
  };
  
+static const SDProto sd_proto_emmc = {

+.name = "eMMC",
+.cmd = {
+[0] = sd_cmd_GO_IDLE_STATE,
+[5] = sd_cmd_illegal,
+[19]= sd_cmd_SEND_TUNING_BLOCK,
+[41]= sd_cmd_illegal,
+[52 ... 54] = sd_cmd_illegal,
+[58]= sd_cmd_illegal,
+[59]= sd_cmd_illegal,
+},
+};
+
  static void sd_instance_init(Object *obj)
  {
  SDState *sd = SD_CARD(obj);
@@ -2284,9 +2297,38 @@ static const TypeInfo sd_info = {
  .instance_finalize = sd_instance_finalize,
  };
  
+static void emmc_realize(DeviceState *dev, Error **errp)

+{
+SDState *sd = SD_CARD(dev);
+
+if (sd->spec_version < SD_PHY_SPECv3_01_VERS) {
+error_setg(errp, "Minimum spec for eMMC is v3.01");
+return;
+}
+
+sd_realize(dev, errp);


sd_realize overwrites sc->proto. We should not write to the class anyway.

C.


+}
+
+static void emmc_class_init(ObjectClass *klass, void *data)
+{
+DeviceClass *dc = DEVICE_CLASS(klass);
+SDCardClass *sc = SD_CARD_CLASS(klass);
+
+dc->desc = "eMMC";
+dc->realize = emmc_realize;
+sc->proto = _proto_emmc;
+}
+
+static const TypeInfo emmc_info = {
+.name = TYPE_EMMC,
+.parent = TYPE_SD_CARD,
+.class_init = emmc_class_init,
+ };
+
  static void sd_register_types(void)
  {
  type_register_static(_info);
+type_register_static(_info);
  }
  
  type_init(sd_register_types)

diff --git a/include/hw/sd/sd.h b/include/hw/sd/sd.h
index 0d94e1f346..e52436b7a5 100644
--- a/include/hw/sd/sd.h
+++ b/include/hw/sd/sd.h
@@ -93,6 +93,9 @@ typedef struct {
  #define TYPE_SD_CARD "sd-card"
  OBJECT_DECLARE_TYPE(SDState, SDCardClass, SD_CARD)
  
+#define TYPE_EMMC "emmc"

+DECLARE_INSTANCE_CHECKER(SDState, EMMC, TYPE_EMMC)
+
  struct SDCardClass {
  /*< private >*/
  DeviceClass parent_class;

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support


On 5/31/22 10:03, Philippe Mathieu-Daudé wrote:

On 31/5/22 07:58, Cédric Le Goater wrote:

On 5/30/22 19:40, Philippe Mathieu-Daudé wrote:

On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
    - HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)




+static void emmc_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    SDCardClass *sc = SD_CARD_CLASS(klass);
+
+    dc->desc = "eMMC";
+    sc->proto = _proto_emmc;
+    sc->spec_version = SD_PHY_SPECv3_01_VERS; /* eMMC requirement */
+    sc->set_csd = sd_emmc_set_csd;
+}
+
+static const TypeInfo emmc_info = {
+    .name = TYPE_EMMC,
+    .parent = TYPE_SD_CARD,


Hmm this is odd to have the model inheriting features from SD_CARD but then 
behaving differently (one could enumerate QDEV objects implementing
TYPE_SD_CARD then use them expecting they match the SD card protocol).

Why do you need to have TYPE_SD_CARD as parent?


Simply for the initialization.

Could we simply duplicate sd_class_init() assignations instead? That
would likely make it easier to modify eMMC handlers.


May be we lack a base abstract class ?


I've been thinking about it but maybe not enough. I'll revisit.


It would clean up this section in the realize routine :

    sd->proto = sd->spi ? _proto_spi : _proto_sd;

 if (sc->proto) {
 sd->proto = sc->proto;
 }


In v2 I moved the 'proto' field from instance to class, so we don't need
this hack anymore.


Indeed :

   static void sd_realize(DeviceState *dev, Error **errp)
   {
   SDState *sd = SD_CARD(dev);
   SDCardClass *sc = SD_CARD_GET_CLASS(sd);
   int ret;
   
   sc->proto = sd->spi ? _proto_spi : _proto_sd;

   ...

but this is assigning a class attribute from an instance :/

C.

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support


On 31/5/22 07:58, Cédric Le Goater wrote:

On 5/30/22 19:40, Philippe Mathieu-Daudé wrote:

On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org 


[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
    - HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)




+static void emmc_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    SDCardClass *sc = SD_CARD_CLASS(klass);
+
+    dc->desc = "eMMC";
+    sc->proto = _proto_emmc;
+    sc->spec_version = SD_PHY_SPECv3_01_VERS; /* eMMC requirement */
+    sc->set_csd = sd_emmc_set_csd;
+}
+
+static const TypeInfo emmc_info = {
+    .name = TYPE_EMMC,
+    .parent = TYPE_SD_CARD,


Hmm this is odd to have the model inheriting features from SD_CARD but 
then behaving differently (one could enumerate QDEV objects implementing

TYPE_SD_CARD then use them expecting they match the SD card protocol).

Why do you need to have TYPE_SD_CARD as parent?


Simply for the initialization.

Could we simply duplicate sd_class_init() assignations instead? That
would likely make it easier to modify eMMC handlers.


May be we lack a base abstract class ?


I've been thinking about it but maybe not enough. I'll revisit.


It would clean up this section in the realize routine :

    sd->proto = sd->spi ? _proto_spi : _proto_sd;

     if (sc->proto) {
     sd->proto = sc->proto;
     }


In v2 I moved the 'proto' field from instance to class, so we don't need
this hack anymore.

Re: [PATCH v2 00/25] hw/sd: Rework models for eMMC support

On Tue, May 31, 2022 at 9:56 AM Philippe Mathieu-Daudé  wrote:
> On 31/5/22 08:31, Cédric Le Goater wrote:
> > On 5/30/22 21:37, Philippe Mathieu-Daudé wrote:

> >> I plan to queue patches 1-12 via sdmmc-next later this week.
> >>
> >> Cédric, if you are happy with this series, it should be easy to rebase
> >> your other patches on top and address the comments I left on the RFC :)
> >
> > Sure. I will for the first patches to be merged and I might introduce
> > a base class.
>
> Then consider patches 1-13 queued.

Oops too fast, new patches 3 & 13 are not reviewed.

Re: [PATCH v2 00/25] hw/sd: Rework models for eMMC support


On 31/5/22 08:31, Cédric Le Goater wrote:

On 5/30/22 21:37, Philippe Mathieu-Daudé wrote:

From: Philippe Mathieu-Daudé 

Rebase/respin of Cédric RFC:
https://lore.kernel.org/qemu-devel/20220318132824.1134400-1-...@kaod.org/
(sorry it took me so long guys...)

Pushed at https://gitlab.com/philmd/qemu/-/commits/emmc-v2

I plan to queue patches 1-12 via sdmmc-next later this week.

Cédric, if you are happy with this series, it should be easy to rebase
your other patches on top and address the comments I left on the RFC :)


Sure. I will for the first patches to be merged and I might introduce
a base class.


Then consider patches 1-13 queued.


Thanks,

C.



Regards,

Phil.

Re: [PATCH v2 14/25] hw/sd: Basis for eMMC support


On 30/5/22 21:38, Philippe Mathieu-Daudé wrote:

From: Philippe Mathieu-Daudé 


I missed something during the cherry-pick, this should be:

From: Cédric Le Goater 


The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework.

Signed-off-by: Cédric Le Goater 
Signed-off-by: Philippe Mathieu-Daudé 
---
TODO: Do not inherit TYPE_SD_CARD, duplicate sd_class_init()
---
  hw/sd/sd.c | 42 ++
  include/hw/sd/sd.h |  3 +++
  2 files changed, 45 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index b2f16dbb73..8b178aa261 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -2166,6 +2166,19 @@ static const SDProto sd_proto_sd = {
  },
  };
  
+static const SDProto sd_proto_emmc = {

+.name = "eMMC",
+.cmd = {
+[0] = sd_cmd_GO_IDLE_STATE,
+[5] = sd_cmd_illegal,
+[19]= sd_cmd_SEND_TUNING_BLOCK,
+[41]= sd_cmd_illegal,
+[52 ... 54] = sd_cmd_illegal,
+[58]= sd_cmd_illegal,
+[59]= sd_cmd_illegal,
+},
+};
+
  static void sd_instance_init(Object *obj)
  {
  SDState *sd = SD_CARD(obj);
@@ -2284,9 +2297,38 @@ static const TypeInfo sd_info = {
  .instance_finalize = sd_instance_finalize,
  };
  
+static void emmc_realize(DeviceState *dev, Error **errp)

+{
+SDState *sd = SD_CARD(dev);
+
+if (sd->spec_version < SD_PHY_SPECv3_01_VERS) {
+error_setg(errp, "Minimum spec for eMMC is v3.01");
+return;
+}
+
+sd_realize(dev, errp);
+}
+
+static void emmc_class_init(ObjectClass *klass, void *data)
+{
+DeviceClass *dc = DEVICE_CLASS(klass);
+SDCardClass *sc = SD_CARD_CLASS(klass);
+
+dc->desc = "eMMC";
+dc->realize = emmc_realize;
+sc->proto = _proto_emmc;
+}
+
+static const TypeInfo emmc_info = {
+.name = TYPE_EMMC,
+.parent = TYPE_SD_CARD,
+.class_init = emmc_class_init,
+ };
+
  static void sd_register_types(void)
  {
  type_register_static(_info);
+type_register_static(_info);
  }
  
  type_init(sd_register_types)

diff --git a/include/hw/sd/sd.h b/include/hw/sd/sd.h
index 0d94e1f346..e52436b7a5 100644
--- a/include/hw/sd/sd.h
+++ b/include/hw/sd/sd.h
@@ -93,6 +93,9 @@ typedef struct {
  #define TYPE_SD_CARD "sd-card"
  OBJECT_DECLARE_TYPE(SDState, SDCardClass, SD_CARD)
  
+#define TYPE_EMMC "emmc"

+DECLARE_INSTANCE_CHECKER(SDState, EMMC, TYPE_EMMC)
+
  struct SDCardClass {
  /*< private >*/
  DeviceClass parent_class;

Re: [PATCH v2 00/25] hw/sd: Rework models for eMMC support


On 5/30/22 21:37, Philippe Mathieu-Daudé wrote:

From: Philippe Mathieu-Daudé 

Rebase/respin of Cédric RFC:
https://lore.kernel.org/qemu-devel/20220318132824.1134400-1-...@kaod.org/
(sorry it took me so long guys...)

Pushed at https://gitlab.com/philmd/qemu/-/commits/emmc-v2

I plan to queue patches 1-12 via sdmmc-next later this week.

Cédric, if you are happy with this series, it should be easy to rebase
your other patches on top and address the comments I left on the RFC :)


Sure. I will for the first patches to be merged and I might introduce
a base class.

Thanks,

C.



Regards,

Phil.

Cédric Le Goater (6):
   hw/sd: Add sd_emmc_cmd_SEND_OP_CMD() handler
   hw/sd: Add sd_emmc_cmd_ALL_SEND_CID() handler
   hw/sd: Add sd_emmc_cmd_SEND_RELATIVE_ADDR() handler
   hw/sd: Add sd_emmc_cmd_APP_CMD() handler
   hw/sd: add sd_emmc_cmd_SEND_TUNING_BLOCK() handler
   hw/sd: Add sd_emmc_cmd_SEND_EXT_CSD() handler

Joel Stanley (4):
   hw/sd: Add sd_cmd_SEND_TUNING_BLOCK() handler
   hw/sd: Support boot area in emmc image
   hw/sd: Subtract bootarea size from blk
   hw/sd: Add boot config support

Philippe Mathieu-Daudé (13):
   hw/sd/sdcard: Return ILLEGAL for CMD19/CMD23 prior SD spec v3.01
   hw/sd: When card is in wrong state, log which state it is
   hw/sd: When card is in wrong state, log which spec version is used
   hw/sd: Move proto_name to SDProto structure
   hw/sd: Introduce sd_cmd_handler type
   hw/sd: Add sd_cmd_illegal() handler
   hw/sd: Add sd_cmd_unimplemented() handler
   hw/sd: Add sd_cmd_GO_IDLE_STATE() handler
   hw/sd: Add sd_cmd_SEND_OP_CMD() handler
   hw/sd: Add sd_cmd_ALL_SEND_CID() handler
   hw/sd: Add sd_cmd_SEND_RELATIVE_ADDR() handler
   hw/sd: Add sd_cmd_SET_BLOCK_COUNT() handler
   hw/sd: Basis for eMMC support

Sai Pavan Boddu (2):
   hw/sd: Add CMD21 tuning sequence
   hw/sd: Add mmc switch function support

  hw/sd/sd.c | 645 +
  hw/sd/sdmmc-internal.c |   2 +-
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   7 +
  4 files changed, 627 insertions(+), 124 deletions(-)

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support


On 5/30/22 20:29, Philippe Mathieu-Daudé wrote:

On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
    - HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)



  static void sd_instance_init(Object *obj)
  {
  SDState *sd = SD_CARD(obj);
@@ -2162,10 +2338,19 @@ static void sd_instance_finalize(Object *obj)
  static void sd_realize(DeviceState *dev, Error **errp)
  {
  SDState *sd = SD_CARD(dev);
+    SDCardClass *sc = SD_CARD_GET_CLASS(sd);
  int ret;
  sd->proto = sd->spi ? _proto_spi : _proto_sd;
+    if (sc->proto) {
+    sd->proto = sc->proto;
+    }
+
+    if (sc->spec_version) {
+    sd->spec_version = sc->spec_version;
+    }
+
  switch (sd->spec_version) {
  case SD_PHY_SPECv1_10_VERS
   ... SD_PHY_SPECv3_01_VERS:



Instead I'd use:



Yes. This is better.

Thanks,

C.


-- >8 --
@@ -2301,14 +2297,26 @@ static const TypeInfo sd_info = {
  .instance_finalize = sd_instance_finalize,
  };

+static void emmc_realize(DeviceState *dev, Error **errp)
+{
+    SDState *sd = SD_CARD(dev);
+
+    if (sd->spec_version < SD_PHY_SPECv3_01_VERS) {
+    error_setg(errp, "Minimum spec for eMMC is v3.01");
+    return;
+    }
+
+    sd_realize(dev, errp);
+}
+
  static void emmc_class_init(ObjectClass *klass, void *data)
  {
  DeviceClass *dc = DEVICE_CLASS(klass);
  SDCardClass *sc = SD_CARD_CLASS(klass);

  dc->desc = "eMMC";
+    dc->realize = emmc_realize;
  sc->proto = _proto_emmc;
  }

---

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-05-30 Thread Cédric Le Goater


On 5/30/22 19:40, Philippe Mathieu-Daudé wrote:

On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
    - HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)




+static void emmc_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    SDCardClass *sc = SD_CARD_CLASS(klass);
+
+    dc->desc = "eMMC";
+    sc->proto = _proto_emmc;
+    sc->spec_version = SD_PHY_SPECv3_01_VERS; /* eMMC requirement */
+    sc->set_csd = sd_emmc_set_csd;
+}
+
+static const TypeInfo emmc_info = {
+    .name = TYPE_EMMC,
+    .parent = TYPE_SD_CARD,


Hmm this is odd to have the model inheriting features from SD_CARD but then 
behaving differently (one could enumerate QDEV objects implementing
TYPE_SD_CARD then use them expecting they match the SD card protocol).

Why do you need to have TYPE_SD_CARD as parent?


Simply for the initialization.

Could we simply duplicate sd_class_init() assignations instead? That
would likely make it easier to modify eMMC handlers.


May be we lack a base abstract class ?

It would clean up this section in the realize routine :

   sd->proto = sd->spi ? _proto_spi : _proto_sd;

if (sc->proto) {
sd->proto = sc->proto;
}

Thanks,

C.


+    .class_init = emmc_class_init,
+ };

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-05-30 Thread Cédric Le Goater


On 5/30/22 19:02, Philippe Mathieu-Daudé wrote:

Hi Cédric,

On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
    - HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)



+static const SDProto sd_proto_emmc = {


What about renaming as:

     ... emmc_proto = {


yes. These are internal functions. Fine with me.

Thanks,

C.




+    .name = "eMMC",
+    .cmd = {
+    [0] = sd_cmd_GO_IDLE_STATE,
+    [1] = sd_emmc_cmd_SEND_OP_CMD,


    = emmc_cmd_SEND_OP_CMD,


+    [2] = sd_emmc_cmd_ALL_SEND_CID,


  ...

?


+    [3] = sd_emmc_cmd_SEND_RELATIVE_ADDR,
+    [5] = sd_cmd_illegal,
+    [8] = sd_emmc_cmd_SEND_EXT_CSD,
+    [19]    = sd_cmd_SEND_TUNING_BLOCK,
+    [21]    = sd_emmc_cmd_SEND_TUNING_BLOCK,
+    [41]    = sd_cmd_illegal,
+    [52 ... 54] = sd_cmd_illegal,
+    [55]    = sd_emmc_cmd_APP_CMD,
+    [58]    = sd_cmd_illegal,
+    [59]    = sd_cmd_illegal,
+    },
+};

[PATCH v2 14/25] hw/sd: Basis for eMMC support

2022-05-30 Thread Philippe Mathieu-Daudé

From: Philippe Mathieu-Daudé 

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework.

Signed-off-by: Cédric Le Goater 
Signed-off-by: Philippe Mathieu-Daudé 
---
TODO: Do not inherit TYPE_SD_CARD, duplicate sd_class_init()
---
 hw/sd/sd.c | 42 ++
 include/hw/sd/sd.h |  3 +++
 2 files changed, 45 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index b2f16dbb73..8b178aa261 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -2166,6 +2166,19 @@ static const SDProto sd_proto_sd = {
 },
 };
 
+static const SDProto sd_proto_emmc = {
+.name = "eMMC",
+.cmd = {
+[0] = sd_cmd_GO_IDLE_STATE,
+[5] = sd_cmd_illegal,
+[19]= sd_cmd_SEND_TUNING_BLOCK,
+[41]= sd_cmd_illegal,
+[52 ... 54] = sd_cmd_illegal,
+[58]= sd_cmd_illegal,
+[59]= sd_cmd_illegal,
+},
+};
+
 static void sd_instance_init(Object *obj)
 {
 SDState *sd = SD_CARD(obj);
@@ -2284,9 +2297,38 @@ static const TypeInfo sd_info = {
 .instance_finalize = sd_instance_finalize,
 };
 
+static void emmc_realize(DeviceState *dev, Error **errp)
+{
+SDState *sd = SD_CARD(dev);
+
+if (sd->spec_version < SD_PHY_SPECv3_01_VERS) {
+error_setg(errp, "Minimum spec for eMMC is v3.01");
+return;
+}
+
+sd_realize(dev, errp);
+}
+
+static void emmc_class_init(ObjectClass *klass, void *data)
+{
+DeviceClass *dc = DEVICE_CLASS(klass);
+SDCardClass *sc = SD_CARD_CLASS(klass);
+
+dc->desc = "eMMC";
+dc->realize = emmc_realize;
+sc->proto = _proto_emmc;
+}
+
+static const TypeInfo emmc_info = {
+.name = TYPE_EMMC,
+.parent = TYPE_SD_CARD,
+.class_init = emmc_class_init,
+ };
+
 static void sd_register_types(void)
 {
 type_register_static(_info);
+type_register_static(_info);
 }
 
 type_init(sd_register_types)
diff --git a/include/hw/sd/sd.h b/include/hw/sd/sd.h
index 0d94e1f346..e52436b7a5 100644
--- a/include/hw/sd/sd.h
+++ b/include/hw/sd/sd.h
@@ -93,6 +93,9 @@ typedef struct {
 #define TYPE_SD_CARD "sd-card"
 OBJECT_DECLARE_TYPE(SDState, SDCardClass, SD_CARD)
 
+#define TYPE_EMMC "emmc"
+DECLARE_INSTANCE_CHECKER(SDState, EMMC, TYPE_EMMC)
+
 struct SDCardClass {
 /*< private >*/
 DeviceClass parent_class;
-- 
2.36.1

[PATCH v2 00/25] hw/sd: Rework models for eMMC support

2022-05-30 Thread Philippe Mathieu-Daudé

From: Philippe Mathieu-Daudé 

Rebase/respin of Cédric RFC:
https://lore.kernel.org/qemu-devel/20220318132824.1134400-1-...@kaod.org/
(sorry it took me so long guys...)

Pushed at https://gitlab.com/philmd/qemu/-/commits/emmc-v2

I plan to queue patches 1-12 via sdmmc-next later this week.

Cédric, if you are happy with this series, it should be easy to rebase
your other patches on top and address the comments I left on the RFC :)

Regards,

Phil.

Cédric Le Goater (6):
  hw/sd: Add sd_emmc_cmd_SEND_OP_CMD() handler
  hw/sd: Add sd_emmc_cmd_ALL_SEND_CID() handler
  hw/sd: Add sd_emmc_cmd_SEND_RELATIVE_ADDR() handler
  hw/sd: Add sd_emmc_cmd_APP_CMD() handler
  hw/sd: add sd_emmc_cmd_SEND_TUNING_BLOCK() handler
  hw/sd: Add sd_emmc_cmd_SEND_EXT_CSD() handler

Joel Stanley (4):
  hw/sd: Add sd_cmd_SEND_TUNING_BLOCK() handler
  hw/sd: Support boot area in emmc image
  hw/sd: Subtract bootarea size from blk
  hw/sd: Add boot config support

Philippe Mathieu-Daudé (13):
  hw/sd/sdcard: Return ILLEGAL for CMD19/CMD23 prior SD spec v3.01
  hw/sd: When card is in wrong state, log which state it is
  hw/sd: When card is in wrong state, log which spec version is used
  hw/sd: Move proto_name to SDProto structure
  hw/sd: Introduce sd_cmd_handler type
  hw/sd: Add sd_cmd_illegal() handler
  hw/sd: Add sd_cmd_unimplemented() handler
  hw/sd: Add sd_cmd_GO_IDLE_STATE() handler
  hw/sd: Add sd_cmd_SEND_OP_CMD() handler
  hw/sd: Add sd_cmd_ALL_SEND_CID() handler
  hw/sd: Add sd_cmd_SEND_RELATIVE_ADDR() handler
  hw/sd: Add sd_cmd_SET_BLOCK_COUNT() handler
  hw/sd: Basis for eMMC support

Sai Pavan Boddu (2):
  hw/sd: Add CMD21 tuning sequence
  hw/sd: Add mmc switch function support

 hw/sd/sd.c | 645 +
 hw/sd/sdmmc-internal.c |   2 +-
 hw/sd/sdmmc-internal.h |  97 +++
 include/hw/sd/sd.h |   7 +
 4 files changed, 627 insertions(+), 124 deletions(-)

-- 
2.36.1

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-05-30 Thread Philippe Mathieu-Daudé via


On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
- HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)



  static void sd_instance_init(Object *obj)
  {
  SDState *sd = SD_CARD(obj);
@@ -2162,10 +2338,19 @@ static void sd_instance_finalize(Object *obj)
  static void sd_realize(DeviceState *dev, Error **errp)
  {
  SDState *sd = SD_CARD(dev);
+SDCardClass *sc = SD_CARD_GET_CLASS(sd);
  int ret;
  
  sd->proto = sd->spi ? _proto_spi : _proto_sd;
  
+if (sc->proto) {

+sd->proto = sc->proto;
+}
+
+if (sc->spec_version) {
+sd->spec_version = sc->spec_version;
+}
+
  switch (sd->spec_version) {
  case SD_PHY_SPECv1_10_VERS
   ... SD_PHY_SPECv3_01_VERS:



Instead I'd use:

-- >8 --
@@ -2301,14 +2297,26 @@ static const TypeInfo sd_info = {
 .instance_finalize = sd_instance_finalize,
 };

+static void emmc_realize(DeviceState *dev, Error **errp)
+{
+SDState *sd = SD_CARD(dev);
+
+if (sd->spec_version < SD_PHY_SPECv3_01_VERS) {
+error_setg(errp, "Minimum spec for eMMC is v3.01");
+return;
+}
+
+sd_realize(dev, errp);
+}
+
 static void emmc_class_init(ObjectClass *klass, void *data)
 {
 DeviceClass *dc = DEVICE_CLASS(klass);
 SDCardClass *sc = SD_CARD_CLASS(klass);

 dc->desc = "eMMC";
+dc->realize = emmc_realize;
 sc->proto = _proto_emmc;
 }

---

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-05-30 Thread Philippe Mathieu-Daudé via


On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
- HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)




+static void emmc_class_init(ObjectClass *klass, void *data)
+{
+DeviceClass *dc = DEVICE_CLASS(klass);
+SDCardClass *sc = SD_CARD_CLASS(klass);
+
+dc->desc = "eMMC";
+sc->proto = _proto_emmc;
+sc->spec_version = SD_PHY_SPECv3_01_VERS; /* eMMC requirement */
+sc->set_csd = sd_emmc_set_csd;
+}
+
+static const TypeInfo emmc_info = {
+.name = TYPE_EMMC,
+.parent = TYPE_SD_CARD,


Hmm this is odd to have the model inheriting features from SD_CARD but 
then behaving differently (one could enumerate QDEV objects implementing

TYPE_SD_CARD then use them expecting they match the SD card protocol).

Why do you need to have TYPE_SD_CARD as parent?

Could we simply duplicate sd_class_init() assignations instead? That
would likely make it easier to modify eMMC handlers.


+.class_init = emmc_class_init,
+ };

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-05-30 Thread Philippe Mathieu-Daudé via


Hi Cédric,

On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
- HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)



+static const SDProto sd_proto_emmc = {


What about renaming as:

... emmc_proto = {


+.name = "eMMC",
+.cmd = {
+[0] = sd_cmd_GO_IDLE_STATE,
+[1] = sd_emmc_cmd_SEND_OP_CMD,


   = emmc_cmd_SEND_OP_CMD,


+[2] = sd_emmc_cmd_ALL_SEND_CID,


 ...

?


+[3] = sd_emmc_cmd_SEND_RELATIVE_ADDR,
+[5] = sd_cmd_illegal,
+[8] = sd_emmc_cmd_SEND_EXT_CSD,
+[19]= sd_cmd_SEND_TUNING_BLOCK,
+[21]= sd_emmc_cmd_SEND_TUNING_BLOCK,
+[41]= sd_cmd_illegal,
+[52 ... 54] = sd_cmd_illegal,
+[55]= sd_emmc_cmd_APP_CMD,
+[58]= sd_cmd_illegal,
+[59]= sd_cmd_illegal,
+},
+};

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-05-10 Thread Cédric Le Goater


On 5/10/22 09:15, Cédric Le Goater wrote:

[ ... ]



Do you mind splitting as:

- Add TYPE_EMMC, emmc_class_init and sd_proto_emmc[] with
   already existing handlers (1 patch)

- Add new handlers, from smaller to sd_emmc_set_csd(),
   and finally mmc_set_ext_csd() with the EXT_CSD definitions
   (various patches).


Sure, I will reorganize the patchset. It will ease the review.


I just did in https://github.com/legoater/qemu/commits/aspeed-7.1

The commit logs need more information, at least a reference to the specs.




Otherwise LGTM!

What is your test suite?


Not simple.

Get the latest FW images for the rainier platform from :

   
https://jenkins.openbmc.org/view/latest/job/latest-master/label=docker-builder,target=p10bmc/lastSuccessfulBuild/artifact/openbmc/build/tmp/deploy/images/p10bmc/

and build a bootable eMMC with such a sequence :

   dd of=$mmc if=/dev/zero bs=1M count=128
   dd of=$mmc if=${fw_dir}/u-boot-spl.bin conv=notrunc
   dd of=$mmc if=${fw_dir}/u-boot.bin conv=notrunc bs=1K seek=64
   dd of=$mmc if=${fw_dir}/u-boot-env.bin conv=notrunc bs=1K seek=$((896 + 64))
   xzdec $wicxz | dd status=progress of=$mmc conv=notrunc bs=1M seek=2
   truncate --size 16G $mmc

We could have a smaller image with a buildroot rootfs.

Then, boot with:

   qemu-system-arm -M rainier-bmc -net nic,netdev=net0 -netdev user \
 -drive file=$mmc.qcow2,format=qcow2,if=sd,id=sd0,index=2


and you will need my branch for that.

Thanks,

C.




The Aspeed machines (AST2600) can boot from flash or eMMC and the above
command line does the trick. It is not optimal.


Thanks,

C.

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-05-10 Thread Cédric Le Goater


[ ... ]



Do you mind splitting as:

- Add TYPE_EMMC, emmc_class_init and sd_proto_emmc[] with
   already existing handlers (1 patch)

- Add new handlers, from smaller to sd_emmc_set_csd(),
   and finally mmc_set_ext_csd() with the EXT_CSD definitions
   (various patches).


Sure, I will reorganize the patchset. It will ease the review.


Otherwise LGTM!

What is your test suite?


Not simple.

Get the latest FW images for the rainier platform from :

  
https://jenkins.openbmc.org/view/latest/job/latest-master/label=docker-builder,target=p10bmc/lastSuccessfulBuild/artifact/openbmc/build/tmp/deploy/images/p10bmc/

and build a bootable eMMC with such a sequence :

  dd of=$mmc if=/dev/zero bs=1M count=128
  dd of=$mmc if=${fw_dir}/u-boot-spl.bin conv=notrunc
  dd of=$mmc if=${fw_dir}/u-boot.bin conv=notrunc bs=1K seek=64
  dd of=$mmc if=${fw_dir}/u-boot-env.bin conv=notrunc bs=1K seek=$((896 + 64))

  xzdec $wicxz | dd status=progress of=$mmc conv=notrunc bs=1M seek=2

  truncate --size 16G $mmc

We could have a smaller image with a buildroot rootfs.

Then, boot with:

  qemu-system-arm -M rainier-bmc -net nic,netdev=net0 -netdev user \
-drive file=$mmc.qcow2,format=qcow2,if=sd,id=sd0,index=2

The Aspeed machines (AST2600) can boot from flash or eMMC and the above
command line does the trick. It is not optimal.


Thanks,

C.

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-05-09 Thread Philippe Mathieu-Daudé via


Hi Cédric,

On 18/3/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
- HPI activation ]
Signed-off-by: Cédric Le Goater 
---
  hw/sd/sdmmc-internal.h |  97 +++
  include/hw/sd/sd.h |   9 ++
  hw/sd/sd.c | 205 -
  hw/sd/sdmmc-internal.c |   2 +-
  4 files changed, 311 insertions(+), 2 deletions(-)


Do you mind splitting as:

- Add TYPE_EMMC, emmc_class_init and sd_proto_emmc[] with
  already existing handlers (1 patch)

- Add new handlers, from smaller to sd_emmc_set_csd(),
  and finally mmc_set_ext_csd() with the EXT_CSD definitions
  (various patches).

Otherwise LGTM!

What is your test suite?

Thanks,

Phil.

Re: [RFC PATCH 00/17] hw/sd: Rework models for eMMC support

2022-04-21 Thread Cédric Le Goater


Hello,

On 3/18/22 14:28, Cédric Le Goater wrote:

Hello Philippe,

I am restarting the discussion we started in :

   http://patchwork.ozlabs.org/project/qemu-devel/list/?series=250563

This series adds an extension for a new eMMC device using the
framework you put in place. It's not perfect but we are getting close.
The SPI variant would need its own class I suppose and there are more
cleanups to be done in the set of commands. Please comment !



Since patch 01-09 are nearly all reviewed, may be we could start by
merging those ?

Thanks,

C.

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-03-28 Thread Cédric Le Goater

Hello Jerome,

On 3/28/22 14:10, Jerome Forissier wrote:

Hi Cédric,

On 3/18/22 14:28, Cédric Le Goater wrote:

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

That is interesting.

This series is really a pre-pre-alpha-0. It's a mixed of changes from
Joel and I on top of Vincent Palatin patches sent ~10 years ago. I
reworked them recently to take into account a proposal of Philippe to
improve the sd models and possibly add eMMC support. They are good
enough to boot a rainier-bmc board.

Is RPMB emulation implemented? I doesn't look like so, although...

no. it's not.

+sd->ext_csd[EXT_CSD_RPMB_MULT] = 0x1; /* RPMB size */

...this indicates that the device has one block (128 KB) of RPMB if I'm
not mistaken.

I would be quite interested in testing and possibly helping out
implement RPMB, although I must admit don't know much about QEMU
internals. I have written some quick & dirty emulation code for RPMB at
the Linux ioctl() level, see [1]. This code is useful for CI testing of
the OP-TEE OS project [2], but having a lower level emulation in QEMU
would be much better since the Linux kernel would "see" the device and
report it in sysfs etc.

[1]
https://github.com/OP-TEE/optee_client/blob/3.16.0/tee-supplicant/src/rpmb.c#L494-L571
[2] https://github.com/OP-TEE/optee_os/

You can give these patches a try using this branch:

https://github.com/legoater/qemu/commits/aspeed-7.0

Using the above QEMU, I would run a rainier-bmc machine which requires eMMC
support :

qemu-system-arm -M rainier-bmc \
-nic user,mac=C0:FF:EE:00:00:02,hostfwd=::-:22 \
-drive file=/path/to/mmc-p10bmc.qcow2,format=qcow2,if=sd,id=sd0,index=2
\
-nographic -nodefaults -snapshot -serial mon:stdio

The bootable qcow2 image :

https://www.kaod.org/qemu/aspeed/mmc-p10bmc.qcow2

was created from :

https://jenkins.openbmc.org/view/latest/job/latest-master/label=docker-builder,target=p10bmc/lastSuccessfulBuild/artifact/openbmc/build/tmp/deploy/images/p10bmc/

Booting from the eMMC is a topic by itself but instead of booting
from the eMMC, I would use a custom kernel and userspace.

Compile an AST2600 kernel or grab :

https://www.kaod.org/qemu/aspeed/zImage
https://www.kaod.org/qemu/aspeed/aspeed-bmc-ibm-rainier.dtb

Compile a buildroot image or grab :

https://www.kaod.org/qemu/aspeed/rootfs.cpio.xz

and run :

qemu-system-arm -M rainier-bmc \
-kernel /path/to/linux/build_ast2600/arch/arm/boot/zImage \
-initrd /path/to/rootfs.cpio.xz \
-dtb
/path/to/linux/build_ast2600/arch/arm/boot/dts/aspeed-bmc-ibm-rainier.dtb \
-append 'console=ttyS4,115200n8 root=/dev/ram ro' \
-nic user,mac=C0:FF:EE:00:00:02,hostfwd=::-:22 \
-drive file=/path/to/mmc-p10bmc.qcow2,format=qcow2,if=sd,id=sd0,index=2
\
-nographic -nodefaults -serial mon:stdio

root/0penBmc to login.

From there, modify the kernel and the buildroot image, you can scp
some binary, do your testing and possibly send patches to improve QEMU
support !

Re: [RFC PATCH 11/17] hw/sd: Add eMMC support

2022-03-28 Thread Jerome Forissier

Hi Cédric,

On 3/18/22 14:28, Cédric Le Goater wrote:
> The initial eMMC support from Vincent Palatin was largely reworked to
> match the current SD framework. The parameters mimick a real 4GB eMMC,
> but it can be set to various sizes.
> 
> This adds a new QOM object class for EMMC devices.

That is interesting. Is RPMB emulation implemented? I doesn't look like
so, although...

> +sd->ext_csd[EXT_CSD_RPMB_MULT] = 0x1; /* RPMB size */

...this indicates that the device has one block (128 KB) of RPMB if I'm
not mistaken.

I would be quite interested in testing and possibly helping out
implement RPMB, although I must admit don't know much about QEMU
internals. I have written some quick & dirty emulation code for RPMB at
the Linux ioctl() level, see [1]. This code is useful for CI testing of
the OP-TEE OS project [2], but having a lower level emulation in QEMU
would be much better since the Linux kernel would "see" the device and
report it in sysfs etc.

[1] 
https://github.com/OP-TEE/optee_client/blob/3.16.0/tee-supplicant/src/rpmb.c#L494-L571
[2] https://github.com/OP-TEE/optee_os/

Thanks,
-- 
Jerome

[RFC PATCH 00/17] hw/sd: Rework models for eMMC support

2022-03-18 Thread Cédric Le Goater

Hello Philippe,

I am restarting the discussion we started in : 

  http://patchwork.ozlabs.org/project/qemu-devel/list/?series=250563

This series adds an extension for a new eMMC device using the
framework you put in place. It's not perfect but we are getting close.
The SPI variant would need its own class I suppose and there are more
cleanups to be done in the set of commands. Please comment !

Thanks,

C.

Cédric Le Goater (2):
  hw/sd: Add eMMC support
  hw/sd: Fix SET_BLOCK_COUNT command argument

Joel Stanley (3):
  hw/sd: Add sd_cmd_SEND_TUNING_BLOCK() handler
  hw/sd: Support boot area in emmc image
  hw/sd: Subtract bootarea size from blk

Philippe Mathieu-Daudé (9):
  hw/sd: When card is in wrong state, log which state it is
  hw/sd: Move proto_name to SDProto structure
  hw/sd: Introduce sd_cmd_handler type
  hw/sd: Add sd_cmd_illegal() handler
  hw/sd: Add sd_cmd_unimplemented() handler
  hw/sd: Add sd_cmd_GO_IDLE_STATE() handler
  hw/sd: Add sd_cmd_SEND_OP_CMD() handler
  hw/sd: Add sd_cmd_ALL_SEND_CID() handler
  hw/sd: Add sd_cmd_SEND_RELATIVE_ADDR() handler

Sai Pavan Boddu (3):
  hw/sd: Update CMD1 definition for MMC
  hw/sd: Add CMD21 tuning sequence
  hw/sd: Add mmc switch function support

 hw/sd/sdmmc-internal.h |  97 +++
 include/hw/sd/sd.h |  10 +
 hw/sd/sd.c | 585 +
 hw/sd/sdmmc-internal.c |   2 +-
 4 files changed, 583 insertions(+), 111 deletions(-)

-- 
2.34.1

[RFC PATCH 11/17] hw/sd: Add eMMC support

2022-03-18 Thread Cédric Le Goater

The initial eMMC support from Vincent Palatin was largely reworked to
match the current SD framework. The parameters mimick a real 4GB eMMC,
but it can be set to various sizes.

This adds a new QOM object class for EMMC devices.

Signed-off-by: Vincent Palatin 
Link: 
https://lore.kernel.org/r/1311635951-11047-5-git-send-email-vpala...@chromium.org
[ jms: - Forward ported to QEMU 5.2 ]
Signed-off-by: Joel Stanley 
[ clg: - ported on aspeed-7.0 patchset
   - HPI activation ]
Signed-off-by: Cédric Le Goater 
---
 hw/sd/sdmmc-internal.h |  97 +++
 include/hw/sd/sd.h |   9 ++
 hw/sd/sd.c | 205 -
 hw/sd/sdmmc-internal.c |   2 +-
 4 files changed, 311 insertions(+), 2 deletions(-)

diff --git a/hw/sd/sdmmc-internal.h b/hw/sd/sdmmc-internal.h
index d8bf17d204fc..2b98f117cd8f 100644
--- a/hw/sd/sdmmc-internal.h
+++ b/hw/sd/sdmmc-internal.h
@@ -37,4 +37,101 @@ const char *sd_cmd_name(uint8_t cmd);
  */
 const char *sd_acmd_name(uint8_t cmd);
 
+/*
+ * EXT_CSD fields
+ */
+
+#define EXT_CSD_CMDQ_MODE_EN15  /* R/W */
+#define EXT_CSD_FLUSH_CACHE 32  /* W */
+#define EXT_CSD_CACHE_CTRL  33  /* R/W */
+#define EXT_CSD_POWER_OFF_NOTIFICATION  34  /* R/W */
+#define EXT_CSD_PACKED_FAILURE_INDEX35  /* RO */
+#define EXT_CSD_PACKED_CMD_STATUS   36  /* RO */
+#define EXT_CSD_EXP_EVENTS_STATUS   54  /* RO, 2 bytes */
+#define EXT_CSD_EXP_EVENTS_CTRL 56  /* R/W, 2 bytes */
+#define EXT_CSD_DATA_SECTOR_SIZE61  /* R */
+#define EXT_CSD_GP_SIZE_MULT143 /* R/W */
+#define EXT_CSD_PARTITION_SETTING_COMPLETED 155 /* R/W */
+#define EXT_CSD_PARTITION_ATTRIBUTE 156 /* R/W */
+#define EXT_CSD_PARTITION_SUPPORT   160 /* RO */
+#define EXT_CSD_HPI_MGMT161 /* R/W */
+#define EXT_CSD_RST_N_FUNCTION  162 /* R/W */
+#define EXT_CSD_BKOPS_EN163 /* R/W */
+#define EXT_CSD_BKOPS_START 164 /* W */
+#define EXT_CSD_SANITIZE_START  165 /* W */
+#define EXT_CSD_WR_REL_PARAM166 /* RO */
+#define EXT_CSD_RPMB_MULT   168 /* RO */
+#define EXT_CSD_FW_CONFIG   169 /* R/W */
+#define EXT_CSD_BOOT_WP 173 /* R/W */
+#define EXT_CSD_ERASE_GROUP_DEF 175 /* R/W */
+#define EXT_CSD_PART_CONFIG 179 /* R/W */
+#define EXT_CSD_ERASED_MEM_CONT 181 /* RO */
+#define EXT_CSD_BUS_WIDTH   183 /* R/W */
+#define EXT_CSD_STROBE_SUPPORT  184 /* RO */
+#define EXT_CSD_HS_TIMING   185 /* R/W */
+#define EXT_CSD_POWER_CLASS 187 /* R/W */
+#define EXT_CSD_REV 192 /* RO */
+#define EXT_CSD_STRUCTURE   194 /* RO */
+#define EXT_CSD_CARD_TYPE   196 /* RO */
+#define EXT_CSD_DRIVER_STRENGTH 197 /* RO */
+#define EXT_CSD_OUT_OF_INTERRUPT_TIME   198 /* RO */
+#define EXT_CSD_PART_SWITCH_TIME199 /* RO */
+#define EXT_CSD_PWR_CL_52_195   200 /* RO */
+#define EXT_CSD_PWR_CL_26_195   201 /* RO */
+#define EXT_CSD_PWR_CL_52_360   202 /* RO */
+#define EXT_CSD_PWR_CL_26_360   203 /* RO */
+#define EXT_CSD_SEC_CNT 212 /* RO, 4 bytes */
+#define EXT_CSD_S_A_TIMEOUT 217 /* RO */
+#define EXT_CSD_S_C_VCCQ219 /* RO */
+#define EXT_CSD_S_C_VCC 220 /* RO */
+#define EXT_CSD_REL_WR_SEC_C222 /* RO */
+#define EXT_CSD_HC_WP_GRP_SIZE  221 /* RO */
+#define EXT_CSD_ERASE_TIMEOUT_MULT  223 /* RO */
+#define EXT_CSD_HC_ERASE_GRP_SIZE   224 /* RO */
+#define EXT_CSD_ACC_SIZE225 /* RO */
+#define EXT_CSD_BOOT_MULT   226 /* RO */
+#define EXT_CSD_BOOT_INFO   228 /* RO */
+#define EXT_CSD_SEC_TRIM_MULT   229 /* RO */
+#define EXT_CSD_SEC_ERASE_MULT  230 /* RO */
+#define EXT_CSD_SEC_FEATURE_SUPPORT 231 /* RO */
+#define EXT_CSD_TRIM_MULT   232 /* RO */
+#define EXT_CSD_PWR_CL_200_195  236 /* RO */
+#define EXT_CSD_PWR_CL_200_360  237 /* RO */
+#define EXT_CSD_PWR_CL_DDR_52_195   238 /* RO */
+#define EXT_CSD_PWR_CL_DDR_52_360   239 /* RO */
+#define EXT_CSD_BKOPS_STATUS246 /* RO */
+#define EXT_CSD_POWER_OFF_LONG_TIME 247 /* RO */
+#define EXT_CSD_GENERIC_CMD6_TIME   248 /* RO */
+#define EXT_CSD_CACHE_SIZE  249 /* RO, 4 bytes */
+#define EXT_CSD_PWR_CL_DDR_200_360  253 /* RO */
+#define EXT_CSD_FIRMWARE_VERSION254 /* RO, 8 bytes */
+#define EXT_CSD_PRE_EOL_INFO267 /* RO */
+#define EXT_CSD_DEVICE_LIFE_TIME_EST_TYP_A  268 /* RO */
+#define EXT_CSD_DEVICE_LIFE_TIME_EST_TYP_B

Re: [PATCH v3 00/21] eMMC support

2021-03-02 Thread Cédric Le Goater

Hello,

Adding the SD maintainers for more feedback. 

Thanks,

C.

On 2/28/21 8:33 PM, Sai Pavan Boddu wrote:
> Hi,
> 
> This patch series add support for eMMC cards. This work was previosly
> submitted by Vincent, rebased few changes on top.
> 
> Cedric & Joel has helped to added boot partition access support. I
> expect them to make a follow-up series to use it with aspeed machines.> 
> Present series adds eMMC support to Versal SOC.
> 
> Initial patch series version is RFC
> Changes for V2:
>   Split Patch 1
>   Add comments for eMMC Erase commands
>   Added documentation about eMMC and Versal-virt board.
>   Make eMMC optional for xlnx-versal-virt machines
> Changes for V3:
>   Revome addition of EMMC drive flag
>   Add TYPE_EMMC device
>   Add id strings for shci instances
>   Update versal doc with eMMC example
>   Fix signed-off-by lines for few patches
> 
> Cédric Le Goater (1):
>   sd: sdmmc-internal: Add command string for SEND_OP_CMD
> 
> Joel Stanley (2):
>   sd: emmc: Support boot area in emmc image
>   sd: emmc: Subtract bootarea size from blk
> 
> Sai Pavan Boddu (14):
>   sd: sd: Remove usage of tabs in the file
>   sd: emmc: Add support for eMMC cards
>   sd: emmc: Dont not update CARD_CAPACITY for eMMC cards
>   sd: emmc: Update CMD1 definition for eMMC
>   sd: emmc: support idle state in CMD2
>   sd: emmc: Add mmc switch function support
>   sd: emmc: add CMD21 tuning sequence
>   sd: emmc: Make ACMD41 illegal for mmc
>   sd: emmc: Add support for emmc erase
>   sd: emmc: Update CID structure for eMMC
>   sd: sdhci: Support eMMC devices
>   arm: xlnx-versal: Add emmc to versal
>   docs: devel: emmc: Add a doc for emmc card emulation
>   docs: arm: xlnx-versal-virt: Add eMMC support documentation
> 
> Vincent Palatin (4):
>   sd: emmc: Update SET_RELATIVE_ADDR command
>   sd: emmc: update OCR fields for eMMC
>   sd: emmc: Add support for EXT_CSD & CSD for eMMC
>   sd: emmc: Update CMD8 to send EXT_CSD register
> 
>  docs/devel/emmc.txt  |  16 +
>  docs/system/arm/xlnx-versal-virt.rst |  14 +
>  hw/sd/sdmmc-internal.h   |  97 ++
>  include/hw/arm/xlnx-versal.h |   1 +
>  include/hw/sd/sd.h   |   2 +
>  hw/arm/xlnx-versal-virt.c|  29 +-
>  hw/arm/xlnx-versal.c |  14 +-
>  hw/sd/sd.c   | 563 
> ++-
>  hw/sd/sdhci.c|   4 -
>  hw/sd/sdmmc-internal.c   |   2 +-
>  10 files changed, 594 insertions(+), 148 deletions(-)
>  create mode 100644 docs/devel/emmc.txt
>

Re: [PATCH v3 16/21] sd: emmc: Support boot area in emmc image

2021-03-01 Thread Cédric Le Goater

On 2/28/21 8:33 PM, Sai Pavan Boddu wrote:
> From: Joel Stanley 
> 
> This assumes a specially constructued image:
> 
>   dd if=/dev/zero of=mmc-bootarea.img count=2 bs=1M
>   dd if=u-boot-spl.bin of=mmc-bootarea.img conv=notrunc
>   dd if=u-boot.bin of=mmc-bootarea.img conv=notrunc count=64 bs=1K
>   cat mmc-bootarea.img obmc-phosphor-image.wic > mmc.img
>   truncate --size 16GB mmc.img
>   truncate --size 128MB mmc-bootarea.img
> 
> Signed-off-by: Joel Stanley 
> [clg: - changes on the definition names ]
> Signed-off-by: Cédric Le Goater 
> [spb: use data_start property to access right emmc partition,
>   Clean up PARTITION_ENABLE support as incomplete,
>   Fix commit message to be generic.]
> Signed-off-by: Sai Pavan Boddu 
> ---
>  hw/sd/sd.c | 40 
>  1 file changed, 40 insertions(+)
> 
> diff --git a/hw/sd/sd.c b/hw/sd/sd.c
> index 08b77ad..d311477 100644
> --- a/hw/sd/sd.c
> +++ b/hw/sd/sd.c
> @@ -1044,6 +1044,34 @@ static void sd_lock_command(SDState *sd)
>  sd->card_status &= ~CARD_IS_LOCKED;
>  }
>  
> +/*
> + * This requires a disk image that has two boot partitions inserted at the
> + * beginning of it. The size of the boot partitions are configured in the
> + * ext_csd structure, which is hardcoded in qemu. They are currently set to
> + * 1MB each.
> + */
> +static uint32_t sd_bootpart_offset(SDState *sd)
> +{
> +unsigned int access = sd->ext_csd[EXT_CSD_PART_CONFIG] &
> +EXT_CSD_PART_CONFIG_ACC_MASK;
> +unsigned int boot_capacity = sd->ext_csd[EXT_CSD_BOOT_MULT] << 17;
> +
> +if (!sd->emmc) {
> +return 0;
> +}
> +
> +switch (access) {
> +case EXT_CSD_PART_CONFIG_ACC_DEFAULT:
> +return boot_capacity * 2;
> +case EXT_CSD_PART_CONFIG_ACC_BOOT0:
> +return 0;
> +case EXT_CSD_PART_CONFIG_ACC_BOOT0 + 1:
> +return boot_capacity * 1;
> +default:
> + g_assert_not_reached();
> +}
> +}
> +
>  static sd_rsp_type_t sd_normal_command(SDState *sd, SDRequest req)
>  {
>  uint32_t rca = 0x;
> @@ -1359,6 +1387,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  return sd_r1;
>  }
>  
> +if (sd->emmc) {
> +addr += sd_bootpart_offset(sd);


This could be a class handler. The default behavior would be to return 0.

C.

> +}
>  sd->state = sd_sendingdata_state;
>  sd->data_start = addr;
>  sd->data_offset = 0;
> @@ -1378,6 +1409,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  return sd_r1;
>  }
>  
> +if (sd->emmc) {
> +addr += sd_bootpart_offset(sd);
> +}
>  sd->state = sd_sendingdata_state;
>  sd->data_start = addr;
>  sd->data_offset = 0;
> @@ -1434,6 +1468,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  return sd_r1;
>  }
>  
> +if (sd->emmc) {
> +addr += sd_bootpart_offset(sd);
> +}
>  sd->state = sd_receivingdata_state;
>  sd->data_start = addr;
>  sd->data_offset = 0;
> @@ -1464,6 +1501,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  return sd_r1;
>  }
>  
> +if (sd->emmc) {
> +addr += sd_bootpart_offset(sd);
> +}
>  sd->state = sd_receivingdata_state;
>  sd->data_start = addr;
>  sd->data_offset = 0;
>

[PATCH v3 16/21] sd: emmc: Support boot area in emmc image

From: Joel Stanley 

This assumes a specially constructued image:

  dd if=/dev/zero of=mmc-bootarea.img count=2 bs=1M
  dd if=u-boot-spl.bin of=mmc-bootarea.img conv=notrunc
  dd if=u-boot.bin of=mmc-bootarea.img conv=notrunc count=64 bs=1K
  cat mmc-bootarea.img obmc-phosphor-image.wic > mmc.img
  truncate --size 16GB mmc.img
  truncate --size 128MB mmc-bootarea.img

Signed-off-by: Joel Stanley 
[clg: - changes on the definition names ]
Signed-off-by: Cédric Le Goater 
[spb: use data_start property to access right emmc partition,
  Clean up PARTITION_ENABLE support as incomplete,
  Fix commit message to be generic.]
Signed-off-by: Sai Pavan Boddu 
---
 hw/sd/sd.c | 40 
 1 file changed, 40 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index 08b77ad..d311477 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -1044,6 +1044,34 @@ static void sd_lock_command(SDState *sd)
 sd->card_status &= ~CARD_IS_LOCKED;
 }
 
+/*
+ * This requires a disk image that has two boot partitions inserted at the
+ * beginning of it. The size of the boot partitions are configured in the
+ * ext_csd structure, which is hardcoded in qemu. They are currently set to
+ * 1MB each.
+ */
+static uint32_t sd_bootpart_offset(SDState *sd)
+{
+unsigned int access = sd->ext_csd[EXT_CSD_PART_CONFIG] &
+EXT_CSD_PART_CONFIG_ACC_MASK;
+unsigned int boot_capacity = sd->ext_csd[EXT_CSD_BOOT_MULT] << 17;
+
+if (!sd->emmc) {
+return 0;
+}
+
+switch (access) {
+case EXT_CSD_PART_CONFIG_ACC_DEFAULT:
+return boot_capacity * 2;
+case EXT_CSD_PART_CONFIG_ACC_BOOT0:
+return 0;
+case EXT_CSD_PART_CONFIG_ACC_BOOT0 + 1:
+return boot_capacity * 1;
+default:
+ g_assert_not_reached();
+}
+}
+
 static sd_rsp_type_t sd_normal_command(SDState *sd, SDRequest req)
 {
 uint32_t rca = 0x;
@@ -1359,6 +1387,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_sendingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1378,6 +1409,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_sendingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1434,6 +1468,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_receivingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1464,6 +1501,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_receivingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
-- 
2.7.4

[PATCH v3 21/21] docs: arm: xlnx-versal-virt: Add eMMC support documentation

Add details of eMMC specific machine property and example for passing
eMMC device.

Signed-off-by: Sai Pavan Boddu 
---
 docs/system/arm/xlnx-versal-virt.rst | 14 ++
 1 file changed, 14 insertions(+)

diff --git a/docs/system/arm/xlnx-versal-virt.rst 
b/docs/system/arm/xlnx-versal-virt.rst
index 2602d0f..d1099fa 100644
--- a/docs/system/arm/xlnx-versal-virt.rst
+++ b/docs/system/arm/xlnx-versal-virt.rst
@@ -29,6 +29,7 @@ Implemented devices:
 - 2 GEMs (Cadence MACB Ethernet MACs)
 - 8 ADMA (Xilinx zDMA) channels
 - 2 SD Controllers
+* SDHCI0 can be configured as eMMC
 - OCM (256KB of On Chip Memory)
 - DDR memory
 
@@ -43,6 +44,19 @@ Other differences between the hardware and the QEMU model:
 - QEMU provides 8 virtio-mmio virtio transports; these start at
   address ``0xa000`` and have IRQs from 111 and upwards.
 
+Enabling eMMC
+"
+In order to enable eMMC pass the following machine property "emmc=on".
+ex:
+"-M xlnx-versal-virt,emmc=on"
+
+Above switch would configure SDHCI0 to accept eMMC. More details on eMMC
+emulation can be found in docs/devel/emmc.txt.
+
+Below is the command to pass eMMC device.
+"-drive file=emmc.img,if=none,id=emmc,format=raw
+ -device emmc,drive=emmc,id=emmc0,bus=/sdhci0/sd-bus"
+
 Running
 """
 If the user provides an Operating System to be loaded, we expect users
-- 
2.7.4

[PATCH v3 00/21] eMMC support

Hi,

This patch series add support for eMMC cards. This work was previosly
submitted by Vincent, rebased few changes on top.

Cedric & Joel has helped to added boot partition access support. I
expect them to make a follow-up series to use it with aspeed machines.

Present series adds eMMC support to Versal SOC.

Initial patch series version is RFC
Changes for V2:
Split Patch 1
Add comments for eMMC Erase commands
Added documentation about eMMC and Versal-virt board.
Make eMMC optional for xlnx-versal-virt machines
Changes for V3:
Revome addition of EMMC drive flag
Add TYPE_EMMC device
Add id strings for shci instances
Update versal doc with eMMC example
Fix signed-off-by lines for few patches

Cédric Le Goater (1):
  sd: sdmmc-internal: Add command string for SEND_OP_CMD

Joel Stanley (2):
  sd: emmc: Support boot area in emmc image
  sd: emmc: Subtract bootarea size from blk

Sai Pavan Boddu (14):
  sd: sd: Remove usage of tabs in the file
  sd: emmc: Add support for eMMC cards
  sd: emmc: Dont not update CARD_CAPACITY for eMMC cards
  sd: emmc: Update CMD1 definition for eMMC
  sd: emmc: support idle state in CMD2
  sd: emmc: Add mmc switch function support
  sd: emmc: add CMD21 tuning sequence
  sd: emmc: Make ACMD41 illegal for mmc
  sd: emmc: Add support for emmc erase
  sd: emmc: Update CID structure for eMMC
  sd: sdhci: Support eMMC devices
  arm: xlnx-versal: Add emmc to versal
  docs: devel: emmc: Add a doc for emmc card emulation
  docs: arm: xlnx-versal-virt: Add eMMC support documentation

Vincent Palatin (4):
  sd: emmc: Update SET_RELATIVE_ADDR command
  sd: emmc: update OCR fields for eMMC
  sd: emmc: Add support for EXT_CSD & CSD for eMMC
  sd: emmc: Update CMD8 to send EXT_CSD register

 docs/devel/emmc.txt  |  16 +
 docs/system/arm/xlnx-versal-virt.rst |  14 +
 hw/sd/sdmmc-internal.h   |  97 ++
 include/hw/arm/xlnx-versal.h |   1 +
 include/hw/sd/sd.h   |   2 +
 hw/arm/xlnx-versal-virt.c|  29 +-
 hw/arm/xlnx-versal.c |  14 +-
 hw/sd/sd.c   | 563 ++-
 hw/sd/sdhci.c|   4 -
 hw/sd/sdmmc-internal.c   |   2 +-
 10 files changed, 594 insertions(+), 148 deletions(-)
 create mode 100644 docs/devel/emmc.txt

-- 
2.7.4

[PATCH v3 10/21] sd: emmc: support idle state in CMD2

eMMC is expected to be in idle-state post CMD1. Ready state is an
intermediate stage which we don't come across in Device identification
mode.

Signed-off-by: Sai Pavan Boddu 
Signed-off-by: Edgar E. Iglesias 
Reviewed-by: Alistair Francis 
---
 hw/sd/sd.c | 4 
 1 file changed, 4 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index 054ad1e..8a7d0de 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -1059,6 +1059,10 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 if (sd->spi)
 goto bad_cmd;
 switch (sd->state) {
+case sd_idle_state:
+if (!sd->emmc) {
+break;
+}
 case sd_ready_state:
 sd->state = sd_identification_state;
 return sd_r2_i;
-- 
2.7.4

[PATCH v2 10/22] sd: emmc: support idle state in CMD2

eMMC is expected to be in idle-state post CMD1. Ready state is an
intermediate stage which we don't come across in Device identification
mode.

Signed-off-by: Sai Pavan Boddu 
Signed-off-by: Edgar E. Iglesias 
Reviewed-by: Alistair Francis 
---
 hw/sd/sd.c | 4 
 1 file changed, 4 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index 8bc8d5d..ae5c5e8 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -1051,6 +1051,10 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 if (sd->spi)
 goto bad_cmd;
 switch (sd->state) {
+case sd_idle_state:
+if (!sd->emmc) {
+break;
+}
 case sd_ready_state:
 sd->state = sd_identification_state;
 return sd_r2_i;
-- 
2.7.4

[PATCH v2 17/22] sd: emmc: Support boot area in emmc image

From: Joel Stanley 

This assumes a specially constructued image:

  dd if=/dev/zero of=mmc-bootarea.img count=2 bs=1M
  dd if=u-boot-spl.bin of=mmc-bootarea.img conv=notrunc
  dd if=u-boot.bin of=mmc-bootarea.img conv=notrunc count=64 bs=1K
  cat mmc-bootarea.img obmc-phosphor-image.wic > mmc.img
  truncate --size 16GB mmc.img
  truncate --size 128MB mmc-bootarea.img

Signed-off-by: Joel Stanley 
[clg: - changes on the definition names ]
Signed-off-by: Cédric Le Goater 
[spb: use data_start property to access right emmc partition,
  Clean up PARTITION_ENABLE support as incomplete,
  Fix commit message to be generic.]
Signed-off-by: Sai Pavan Boddu 
---
 hw/sd/sd.c | 40 
 1 file changed, 40 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index 60799aa..ab29e54 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -1040,6 +1040,34 @@ static void sd_lock_command(SDState *sd)
 sd->card_status &= ~CARD_IS_LOCKED;
 }
 
+/*
+ * This requires a disk image that has two boot partitions inserted at the
+ * beginning of it. The size of the boot partitions are configured in the
+ * ext_csd structure, which is hardcoded in qemu. They are currently set to
+ * 1MB each.
+ */
+static uint32_t sd_bootpart_offset(SDState *sd)
+{
+unsigned int access = sd->ext_csd[EXT_CSD_PART_CONFIG] &
+EXT_CSD_PART_CONFIG_ACC_MASK;
+unsigned int boot_capacity = sd->ext_csd[EXT_CSD_BOOT_MULT] << 17;
+
+if (!sd->emmc) {
+return 0;
+}
+
+switch (access) {
+case EXT_CSD_PART_CONFIG_ACC_DEFAULT:
+return boot_capacity * 2;
+case EXT_CSD_PART_CONFIG_ACC_BOOT0:
+return 0;
+case EXT_CSD_PART_CONFIG_ACC_BOOT0 + 1:
+return boot_capacity * 1;
+default:
+ g_assert_not_reached();
+}
+}
+
 static sd_rsp_type_t sd_normal_command(SDState *sd, SDRequest req)
 {
 uint32_t rca = 0x;
@@ -1355,6 +1383,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_sendingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1374,6 +1405,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_sendingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1430,6 +1464,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_receivingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1460,6 +1497,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_receivingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
-- 
2.7.4

[PATCH v2 22/22] docs: arm: xlnx-versal-virt: Add eMMC support documentation

Add details of eMMC specific machine property.

Signed-off-by: Sai Pavan Boddu 
---
 docs/system/arm/xlnx-versal-virt.rst | 10 ++
 1 file changed, 10 insertions(+)

diff --git a/docs/system/arm/xlnx-versal-virt.rst 
b/docs/system/arm/xlnx-versal-virt.rst
index 2602d0f..a48a88d 100644
--- a/docs/system/arm/xlnx-versal-virt.rst
+++ b/docs/system/arm/xlnx-versal-virt.rst
@@ -29,6 +29,7 @@ Implemented devices:
 - 2 GEMs (Cadence MACB Ethernet MACs)
 - 8 ADMA (Xilinx zDMA) channels
 - 2 SD Controllers
+* SDHCI0 can be configured as eMMC
 - OCM (256KB of On Chip Memory)
 - DDR memory
 
@@ -43,6 +44,15 @@ Other differences between the hardware and the QEMU model:
 - QEMU provides 8 virtio-mmio virtio transports; these start at
   address ``0xa000`` and have IRQs from 111 and upwards.
 
+Enabling eMMC
+"
+In order to enable eMMC pass the following machine property "emmc=on".
+ex:
+"-M xlnx-versal-virt,emmc=on"
+
+Above switch would configure SDHCI0 to accept eMMC card. More details on eMMC
+emulation can be found in docs/devel/emmc.txt.
+
 Running
 """
 If the user provides an Operating System to be loaded, we expect users
-- 
2.7.4

[PATCH v2 00/22] eMMC support

Hi,

This patch series add support for eMMC cards. This work was previosly
submitted by Vincent, rebased few changes on top.

Cedric & Joel has helped to added boot partition access support. I
expect them to make a follow-up series to use it with aspeed machines.

Present series adds eMMC support to Versal SOC.

Initial patch series version is RFC
Changes for V2:
Split Patch 1
Add comments for eMMC Erase commands
Added documentation about eMMC and Versal-virt board.
Make eMMC optional for xlnx-versal-virt machines

Regards,
Sai Pavan

Cédric Le Goater (2):
  sd: sdmmc-internal: Add command string for SEND_OP_CMD
  sd: emmc: Add Extended CSD register definitions

Joel Stanley (2):
  sd: emmc: Support boot area in emmc image
  sd: emmc: Subtract bootarea size from blk

Sai Pavan Boddu (13):
  sd: sd: Remove usage of tabs in the file
  sd: emmc: Dont not update CARD_CAPACITY for eMMC cards
  sd: emmc: Update CMD1 definition for eMMC
  sd: emmc: support idle state in CMD2
  sd: emmc: Add mmc switch function support
  sd: emmc: add CMD21 tuning sequence
  sd: emmc: Make ACMD41 illegal for mmc
  sd: emmc: Add support for emmc erase
  sd: emmc: Update CID structure for eMMC
  sd: sdhci: Support eMMC devices
  arm: xlnx-versal: Add emmc to versal
  docs: devel: emmc: Add a doc for emmc card emulation
  docs: arm: xlnx-versal-virt: Add eMMC support documentation

Vincent Palatin (5):
  block: add eMMC block device type
  sd: emmc: Add support for eMMC cards
  sd: emmc: update OCR fields for eMMC
  sd: emmc: Add support for EXT_CSD & CSD for eMMC
  sd: emmc: Update CMD8 to send EXT_CSD register

 docs/devel/emmc.txt  |  16 +
 docs/system/arm/xlnx-versal-virt.rst |  10 +
 hw/sd/sdmmc-internal.h   |  97 +++
 include/hw/arm/xlnx-versal.h |   1 +
 include/sysemu/blockdev.h|   1 +
 blockdev.c   |   1 +
 hw/arm/xlnx-versal-virt.c|  30 +-
 hw/arm/xlnx-versal.c |  13 +-
 hw/sd/sd.c   | 545 ++-
 hw/sd/sdhci.c|   4 -
 hw/sd/sdmmc-internal.c   |   2 +-
 11 files changed, 571 insertions(+), 149 deletions(-)
 create mode 100644 docs/devel/emmc.txt

-- 
2.7.4

Re: [RFC PATCH 02/15] sd: add eMMC support

2021-02-13 Thread Luc Michel


On 2/11/21 9:17 AM, Sai Pavan Boddu wrote:

From: Vincent Palatin 

The parameters mimick a real 4GB eMMC, but it can be set to various
sizes.

Signed-off-by: Vincent Palatin 
[SPB: Rebased the patch over qemu 5.1,
   Mark eMMC to support all timing modes]
Signed-off-by: Sai Pavan Boddu 
Signed-off-by: Joel Stanley 
Signed-off-by: Cédric Le Goater 


I find this commit hard to review in its current state. Maybe by 
splitting it into smaller commits with a small description in the commit 
message would make things easier. Eg. one patch to set the ext_csd 
fields, one for the csd, and one per command you modify.


I also noted one small thing bellow.


---
  hw/sd/sd.c | 143 +
  hw/sd/sdmmc-internal.c |   2 +-
  2 files changed, 122 insertions(+), 23 deletions(-)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index 8517dbc..a75fa1c 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c


[snip]


@@ -371,6 +374,51 @@ static const uint8_t sd_csd_rw_mask[16] = {
  0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xfc, 0xfe,
  };
  
+static void mmc_set_ext_csd(SDState *sd, uint64_t size)

+{
+uint32_t sectcount = size >> HWBLOCK_SHIFT;
+
+memset(sd->ext_csd, 0, 512);

memset(sd->ext_csd, 0, sizeof(sd->ext_csd));

--
Luc


+sd->ext_csd[504] = 0x1; /* supported command sets */
+sd->ext_csd[503] = 0x1; /* HPI features  */
+sd->ext_csd[502] = 0x1; /* Background operations support */
+sd->ext_csd[241] = 0xA; /* 1st initialization time after partitioning */
+sd->ext_csd[232] = 0x1; /* Trim multiplier */
+sd->ext_csd[231] = 0x15; /* Secure feature support */
+sd->ext_csd[230] = 0x96; /* Secure erase support */
+sd->ext_csd[229] = 0x96; /* Secure TRIM multiplier */
+sd->ext_csd[228] = 0x7; /* Boot information */
+sd->ext_csd[226] = 0x8; /* Boot partition size */
+sd->ext_csd[225] = 0x6; /* Access size */
+sd->ext_csd[224] = 0x4; /* HC Erase unit size */
+sd->ext_csd[223] = 0x1; /* HC erase timeout */
+sd->ext_csd[222] = 0x1; /* Reliable write sector count */
+sd->ext_csd[221] = 0x4; /* HC write protect group size */
+sd->ext_csd[220] = 0x8; /* Sleep current VCC  */
+sd->ext_csd[219] = 0x7; /* Sleep current VCCQ */
+sd->ext_csd[217] = 0x11; /* Sleep/Awake timeout */
+sd->ext_csd[215] = (sectcount >> 24) & 0xff; /* Sector count */
+sd->ext_csd[214] = (sectcount >> 16) & 0xff; /* ... */
+sd->ext_csd[213] = (sectcount >> 8) & 0xff;  /* ... */
+sd->ext_csd[212] = (sectcount & 0xff);   /* ... */
+sd->ext_csd[210] = 0xa; /* Min write perf for 8bit@52Mhz */
+sd->ext_csd[209] = 0xa; /* Min read perf for 8bit@52Mhz  */
+sd->ext_csd[208] = 0xa; /* Min write perf for 4bit@52Mhz */
+sd->ext_csd[207] = 0xa; /* Min read perf for 4bit@52Mhz */
+sd->ext_csd[206] = 0xa; /* Min write perf for 4bit@26Mhz */
+sd->ext_csd[205] = 0xa; /* Min read perf for 4bit@26Mhz */
+sd->ext_csd[199] = 0x1; /* Partition switching timing */
+sd->ext_csd[198] = 0x1; /* Out-of-interrupt busy timing */
+sd->ext_csd[196] = 0xFF; /* Card type */
+sd->ext_csd[194] = 0x2; /* CSD Structure version */
+sd->ext_csd[192] = 0x5; /* Extended CSD revision */
+sd->ext_csd[168] = 0x1; /* RPMB size */
+sd->ext_csd[160] = 0x3; /* Partinioning support */
+sd->ext_csd[159] = 0x00; /* Max enhanced area size */
+sd->ext_csd[158] = 0x00; /* ... */
+sd->ext_csd[157] = 0xEC; /* ... */
+}
+
  static void sd_set_csd(SDState *sd, uint64_t size)
  {
  int hwblock_shift = HWBLOCK_SHIFT;
@@ -384,7 +432,34 @@ static void sd_set_csd(SDState *sd, uint64_t size)
  }
  csize = (size >> (CMULT_SHIFT + hwblock_shift)) - 1;
  
-if (size <= SDSC_MAX_CAPACITY) { /* Standard Capacity SD */

+if (sd->emmc) { /* eMMC */
+sd->csd[0] = 0xd0;
+sd->csd[1] = 0x0f;
+sd->csd[2] = 0x00;
+sd->csd[3] = 0x32;
+sd->csd[4] = 0x0f;
+if (size <= 0x8000ULL) {
+/* use 1k blocks */
+uint32_t csize1k = (size >> (CMULT_SHIFT + 10)) - 1;
+sd->csd[5] = 0x5a;
+sd->csd[6] = 0x80 | ((csize1k >> 10) & 0xf);
+sd->csd[7] = (csize1k >> 2) & 0xff;
+} else { /* >= 2GB : size stored in ext CSD, block addressing */
+sd->csd[5] = 0x59;
+sd->csd[6] = 0x8f;
+sd->csd[7] = 0xff;
+sd->ocr |= 1 << 30;
+}
+sd->csd[8] = 0xff;
+sd->csd[9] = 0xff;
+sd->csd[10] = 0xf7;
+sd->csd[11] = 0xfe;
+sd->csd[12] = 0x49;
+sd->csd[13] = 0x10;
+sd->csd[14] = 0x00;
+sd->csd[15] = (sd_crc7(sd->csd, 15) << 1) | 1;
+mmc_set_ext_csd(sd, size);
+} else if (size <= SDSC_MAX_CAPACITY) { /* Standard Capacity SD */
  sd->csd[0] = 0x00; /* CSD structure */
  sd->csd[1] = 0x26; /* Data read access-time-1 */
  sd->csd[2] = 0x00; /* Data read access-time-2 */
@@ -431,9 +506,13 @@

Re: [RFC PATCH 12/15] sd: emmc: Support boot area in emmc image

2021-02-12 Thread Alistair Francis

On Thu, Feb 11, 2021 at 12:15 AM Sai Pavan Boddu
 wrote:
>
> From: Joel Stanley 
>
> This assumes a specially constructued image:
>
>   dd if=/dev/zero of=mmc-bootarea.img count=2 bs=1M
>   dd if=u-boot-spl.bin of=mmc-bootarea.img conv=notrunc
>   dd if=u-boot.bin of=mmc-bootarea.img conv=notrunc count=64 bs=1K
>   cat mmc-bootarea.img obmc-phosphor-image.wic > mmc.img
>   truncate --size 16GB mmc.img
>   truncate --size 128MB mmc-bootarea.img

Could we document this somewhere user accessible?

Alistair

>
> Signed-off-by: Joel Stanley 
> [clg: - changes on the definition names ]
> Signed-off-by: Cédric Le Goater 
> [spb: use data_start property to access right emmc partition,
>   Clean up PARTITION_ENABLE support as incomplete,
>   Fix commit message to be generic.]
> Signed-off-by: Sai Pavan Boddu 
> ---
>  hw/sd/sd.c | 40 
>  1 file changed, 40 insertions(+)
>
> diff --git a/hw/sd/sd.c b/hw/sd/sd.c
> index 54fba7b..55c1104 100644
> --- a/hw/sd/sd.c
> +++ b/hw/sd/sd.c
> @@ -1045,6 +1045,34 @@ static void sd_lock_command(SDState *sd)
>  sd->card_status &= ~CARD_IS_LOCKED;
>  }
>
> +/*
> + * This requires a disk image that has two boot partitions inserted at the
> + * beginning of it. The size of the boot partitions are configured in the
> + * ext_csd structure, which is hardcoded in qemu. They are currently set to
> + * 1MB each.
> + */
> +static uint32_t sd_bootpart_offset(SDState *sd)
> +{
> +unsigned int access = sd->ext_csd[EXT_CSD_PART_CONFIG] &
> +EXT_CSD_PART_CONFIG_ACC_MASK;
> +unsigned int boot_capacity = sd->ext_csd[EXT_CSD_BOOT_MULT] << 17;
> +
> +if (!sd->emmc) {
> +return 0;
> +}
> +
> +switch (access) {
> +case EXT_CSD_PART_CONFIG_ACC_DEFAULT:
> +return boot_capacity * 2;
> +case EXT_CSD_PART_CONFIG_ACC_BOOT0:
> +return 0;
> +case EXT_CSD_PART_CONFIG_ACC_BOOT0 + 1:
> +return boot_capacity * 1;
> +default:
> + g_assert_not_reached();
> +}
> +}
> +
>  static sd_rsp_type_t sd_normal_command(SDState *sd, SDRequest req)
>  {
>  uint32_t rca = 0x;
> @@ -1360,6 +1388,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  return sd_r1;
>  }
>
> +if (sd->emmc) {
> +addr += sd_bootpart_offset(sd);
> +}
>  sd->state = sd_sendingdata_state;
>  sd->data_start = addr;
>  sd->data_offset = 0;
> @@ -1379,6 +1410,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  return sd_r1;
>  }
>
> +if (sd->emmc) {
> +addr += sd_bootpart_offset(sd);
> +}
>  sd->state = sd_sendingdata_state;
>  sd->data_start = addr;
>  sd->data_offset = 0;
> @@ -1435,6 +1469,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  return sd_r1;
>  }
>
> +if (sd->emmc) {
> +addr += sd_bootpart_offset(sd);
> +}
>  sd->state = sd_receivingdata_state;
>  sd->data_start = addr;
>  sd->data_offset = 0;
> @@ -1465,6 +1502,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  return sd_r1;
>  }
>
> +if (sd->emmc) {
> +addr += sd_bootpart_offset(sd);
> +}
>  sd->state = sd_receivingdata_state;
>  sd->data_start = addr;
>  sd->data_offset = 0;
> --
> 2.7.4
>
>

Re: [RFC PATCH 05/15] sd: emmc: support idle state in CMD2

2021-02-12 Thread Alistair Francis

On Thu, Feb 11, 2021 at 12:18 AM Sai Pavan Boddu
 wrote:
>
> eMMC is expected to be in idle-state post CMD1. Ready state is an
> intermediate stage which we don't come across in Device identification
> mode.
>
> Signed-off-by: Sai Pavan Boddu 
> Signed-off-by: Edgar E. Iglesias 

Acked-by: Alistair Francis 

Alistair

> ---
>  hw/sd/sd.c | 4 
>  1 file changed, 4 insertions(+)
>
> diff --git a/hw/sd/sd.c b/hw/sd/sd.c
> index e3738b2..69289e0 100644
> --- a/hw/sd/sd.c
> +++ b/hw/sd/sd.c
> @@ -1051,6 +1051,10 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
> SDRequest req)
>  if (sd->spi)
>  goto bad_cmd;
>  switch (sd->state) {
> +case sd_idle_state:
> +if (!sd->emmc) {
> +break;
> +}
>  case sd_ready_state:
>  sd->state = sd_identification_state;
>  return sd_r2_i;
> --
> 2.7.4
>
>

[RFC PATCH 02/15] sd: add eMMC support

From: Vincent Palatin 

The parameters mimick a real 4GB eMMC, but it can be set to various
sizes.

Signed-off-by: Vincent Palatin 
[SPB: Rebased the patch over qemu 5.1,
  Mark eMMC to support all timing modes]
Signed-off-by: Sai Pavan Boddu 
Signed-off-by: Joel Stanley 
Signed-off-by: Cédric Le Goater 
---
 hw/sd/sd.c | 143 +
 hw/sd/sdmmc-internal.c |   2 +-
 2 files changed, 122 insertions(+), 23 deletions(-)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index 8517dbc..a75fa1c 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -108,6 +108,7 @@ struct SDState {
 uint8_t spec_version;
 BlockBackend *blk;
 bool spi;
+bool emmc;
 
 /* Runtime changeables */
 
@@ -134,6 +135,7 @@ struct SDState {
 uint64_t data_start;
 uint32_t data_offset;
 uint8_t data[512];
+uint8_t ext_csd[512];
 qemu_irq readonly_cb;
 qemu_irq inserted_cb;
 QEMUTimer *ocr_power_timer;
@@ -287,7 +289,8 @@ FIELD(OCR, CARD_POWER_UP,  31,  1)
| R_OCR_ACCEPT_SWITCH_1V8_MASK \
| R_OCR_UHS_II_CARD_MASK \
| R_OCR_CARD_CAPACITY_MASK \
-   | R_OCR_CARD_POWER_UP_MASK)
+   | R_OCR_CARD_POWER_UP_MASK \
+   | R_OCR_DUAL_VOLTAGE_CARD_MASK)
 
 static void sd_set_ocr(SDState *sd)
 {
@@ -371,6 +374,51 @@ static const uint8_t sd_csd_rw_mask[16] = {
 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0xfc, 0xfe,
 };
 
+static void mmc_set_ext_csd(SDState *sd, uint64_t size)
+{
+uint32_t sectcount = size >> HWBLOCK_SHIFT;
+
+memset(sd->ext_csd, 0, 512);
+sd->ext_csd[504] = 0x1; /* supported command sets */
+sd->ext_csd[503] = 0x1; /* HPI features  */
+sd->ext_csd[502] = 0x1; /* Background operations support */
+sd->ext_csd[241] = 0xA; /* 1st initialization time after partitioning */
+sd->ext_csd[232] = 0x1; /* Trim multiplier */
+sd->ext_csd[231] = 0x15; /* Secure feature support */
+sd->ext_csd[230] = 0x96; /* Secure erase support */
+sd->ext_csd[229] = 0x96; /* Secure TRIM multiplier */
+sd->ext_csd[228] = 0x7; /* Boot information */
+sd->ext_csd[226] = 0x8; /* Boot partition size */
+sd->ext_csd[225] = 0x6; /* Access size */
+sd->ext_csd[224] = 0x4; /* HC Erase unit size */
+sd->ext_csd[223] = 0x1; /* HC erase timeout */
+sd->ext_csd[222] = 0x1; /* Reliable write sector count */
+sd->ext_csd[221] = 0x4; /* HC write protect group size */
+sd->ext_csd[220] = 0x8; /* Sleep current VCC  */
+sd->ext_csd[219] = 0x7; /* Sleep current VCCQ */
+sd->ext_csd[217] = 0x11; /* Sleep/Awake timeout */
+sd->ext_csd[215] = (sectcount >> 24) & 0xff; /* Sector count */
+sd->ext_csd[214] = (sectcount >> 16) & 0xff; /* ... */
+sd->ext_csd[213] = (sectcount >> 8) & 0xff;  /* ... */
+sd->ext_csd[212] = (sectcount & 0xff);   /* ... */
+sd->ext_csd[210] = 0xa; /* Min write perf for 8bit@52Mhz */
+sd->ext_csd[209] = 0xa; /* Min read perf for 8bit@52Mhz  */
+sd->ext_csd[208] = 0xa; /* Min write perf for 4bit@52Mhz */
+sd->ext_csd[207] = 0xa; /* Min read perf for 4bit@52Mhz */
+sd->ext_csd[206] = 0xa; /* Min write perf for 4bit@26Mhz */
+sd->ext_csd[205] = 0xa; /* Min read perf for 4bit@26Mhz */
+sd->ext_csd[199] = 0x1; /* Partition switching timing */
+sd->ext_csd[198] = 0x1; /* Out-of-interrupt busy timing */
+sd->ext_csd[196] = 0xFF; /* Card type */
+sd->ext_csd[194] = 0x2; /* CSD Structure version */
+sd->ext_csd[192] = 0x5; /* Extended CSD revision */
+sd->ext_csd[168] = 0x1; /* RPMB size */
+sd->ext_csd[160] = 0x3; /* Partinioning support */
+sd->ext_csd[159] = 0x00; /* Max enhanced area size */
+sd->ext_csd[158] = 0x00; /* ... */
+sd->ext_csd[157] = 0xEC; /* ... */
+}
+
 static void sd_set_csd(SDState *sd, uint64_t size)
 {
 int hwblock_shift = HWBLOCK_SHIFT;
@@ -384,7 +432,34 @@ static void sd_set_csd(SDState *sd, uint64_t size)
 }
 csize = (size >> (CMULT_SHIFT + hwblock_shift)) - 1;
 
-if (size <= SDSC_MAX_CAPACITY) { /* Standard Capacity SD */
+if (sd->emmc) { /* eMMC */
+sd->csd[0] = 0xd0;
+sd->csd[1] = 0x0f;
+sd->csd[2] = 0x00;
+sd->csd[3] = 0x32;
+sd->csd[4] = 0x0f;
+if (size <= 0x8000ULL) {
+/* use 1k blocks */
+uint32_t csize1k = (size >> (CMULT_SHIFT + 10)) - 1;
+sd->csd[5] = 0x5a;
+sd->csd[6] = 0x80 | ((csize1k >> 10) & 0xf);
+sd->csd[7] = (csize1k >> 2) & 0xff;
+} else { /* >= 2GB : size stored in ext CSD, block addressing */
+sd->csd[5] = 0x59;
+sd->csd[6] = 0x8f;
+sd->csd[7] = 0xff;
+sd->ocr |= 1 << 30;
+}
+sd->csd[8] = 0xff;
+sd->csd[9] = 0xff;
+sd->csd[10] = 0xf7;
+sd->csd[11] = 0xfe;
+

[RFC PATCH 05/15] sd: emmc: support idle state in CMD2

eMMC is expected to be in idle-state post CMD1. Ready state is an
intermediate stage which we don't come across in Device identification
mode.

Signed-off-by: Sai Pavan Boddu 
Signed-off-by: Edgar E. Iglesias 
---
 hw/sd/sd.c | 4 
 1 file changed, 4 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index e3738b2..69289e0 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -1051,6 +1051,10 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 if (sd->spi)
 goto bad_cmd;
 switch (sd->state) {
+case sd_idle_state:
+if (!sd->emmc) {
+break;
+}
 case sd_ready_state:
 sd->state = sd_identification_state;
 return sd_r2_i;
-- 
2.7.4

[RFC PATCH 12/15] sd: emmc: Support boot area in emmc image

From: Joel Stanley 

This assumes a specially constructued image:

  dd if=/dev/zero of=mmc-bootarea.img count=2 bs=1M
  dd if=u-boot-spl.bin of=mmc-bootarea.img conv=notrunc
  dd if=u-boot.bin of=mmc-bootarea.img conv=notrunc count=64 bs=1K
  cat mmc-bootarea.img obmc-phosphor-image.wic > mmc.img
  truncate --size 16GB mmc.img
  truncate --size 128MB mmc-bootarea.img

Signed-off-by: Joel Stanley 
[clg: - changes on the definition names ]
Signed-off-by: Cédric Le Goater 
[spb: use data_start property to access right emmc partition,
  Clean up PARTITION_ENABLE support as incomplete,
  Fix commit message to be generic.]
Signed-off-by: Sai Pavan Boddu 
---
 hw/sd/sd.c | 40 
 1 file changed, 40 insertions(+)

diff --git a/hw/sd/sd.c b/hw/sd/sd.c
index 54fba7b..55c1104 100644
--- a/hw/sd/sd.c
+++ b/hw/sd/sd.c
@@ -1045,6 +1045,34 @@ static void sd_lock_command(SDState *sd)
 sd->card_status &= ~CARD_IS_LOCKED;
 }
 
+/*
+ * This requires a disk image that has two boot partitions inserted at the
+ * beginning of it. The size of the boot partitions are configured in the
+ * ext_csd structure, which is hardcoded in qemu. They are currently set to
+ * 1MB each.
+ */
+static uint32_t sd_bootpart_offset(SDState *sd)
+{
+unsigned int access = sd->ext_csd[EXT_CSD_PART_CONFIG] &
+EXT_CSD_PART_CONFIG_ACC_MASK;
+unsigned int boot_capacity = sd->ext_csd[EXT_CSD_BOOT_MULT] << 17;
+
+if (!sd->emmc) {
+return 0;
+}
+
+switch (access) {
+case EXT_CSD_PART_CONFIG_ACC_DEFAULT:
+return boot_capacity * 2;
+case EXT_CSD_PART_CONFIG_ACC_BOOT0:
+return 0;
+case EXT_CSD_PART_CONFIG_ACC_BOOT0 + 1:
+return boot_capacity * 1;
+default:
+ g_assert_not_reached();
+}
+}
+
 static sd_rsp_type_t sd_normal_command(SDState *sd, SDRequest req)
 {
 uint32_t rca = 0x;
@@ -1360,6 +1388,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_sendingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1379,6 +1410,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_sendingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1435,6 +1469,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_receivingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
@@ -1465,6 +1502,9 @@ static sd_rsp_type_t sd_normal_command(SDState *sd, 
SDRequest req)
 return sd_r1;
 }
 
+if (sd->emmc) {
+addr += sd_bootpart_offset(sd);
+}
 sd->state = sd_receivingdata_state;
 sd->data_start = addr;
 sd->data_offset = 0;
-- 
2.7.4

[RFC PATCH 00/15] eMMC support