aspeed: Add support for the AST2600 SoC
cover aspeed: Add support for the AST2600 SoC
Hello,
The series starts with a watchdog fix and a new model for the SDHCI controller. Follows the code for the AST2600 SoC.
Most of the Aspeed models are reworked with an object class to introduce the AST2600 variant. A model for the AST2600 SoC and a simple AST2600 EVB machine is proposed at the end of the series. It can boot the OpenBMC firmware image which is currently used for HW bringup.
Thanks,
C.
Amithash Prasad (1):
aspeed/wdt: Check correct register for clock source
Cédric Le Goater (13):
aspeed/timer: Introduce an object class per SoC aspeed/timer: Add support for control register 3 aspeed/timer: Add AST2600 support
aspeed/timer: Add support for IRQ status register on the AST2600 aspeed/sdmc: Introduce an object class per SoC watchdog/aspeed: Introduce an object class per SoC aspeed/smc: Add AST2600 support
aspeed/i2c: Introduce an object class per SoC aspeed/i2c: Add AST2600 support
aspeed: Introduce an object class per SoC
aspeed/soc: Add AST2600 support
aspeed: Add an AST2600 eval board
aspeed: add support for the Aspeed MII controller of the AST2600
Eddie James (1):
hw/sd/aspeed_sdhci: New device
Joel Stanley (5):
hw: aspeed_scu: Add AST2600 support
aspeed/sdmc: Add AST2600 support
hw: wdt_aspeed: Add AST2600 support
aspeed: Parameterise number of MACs
aspeed/soc: Add ASPEED Video stub
Rashmica Gupta (1):
hw/gpio: Add in AST2600 specific implementation
include/hw/arm/aspeed_soc.h | 29 +- include/hw/i2c/aspeed_i2c.h | 20 +- include/hw/misc/aspeed_scu.h | 7 +- include/hw/misc/aspeed_sdmc.h | 20 +- include/hw/net/ftgmac100.h | 17 + include/hw/sd/aspeed_sdhci.h | 34 ++ include/hw/ssi/aspeed_smc.h | 2 + include/hw/timer/aspeed_timer.h | 18 ++ include/hw/watchdog/wdt_aspeed.h | 19 +- hw/arm/aspeed.c | 37 ++- hw/arm/aspeed_ast2600.c | 522 +++++++++++++++++++++++++++++++ hw/arm/aspeed_soc.c | 199 +++++++----- hw/gpio/aspeed_gpio.c | 142 ++++++++- hw/i2c/aspeed_i2c.c | 106 ++++++- hw/misc/aspeed_scu.c | 192 +++++++++++- hw/misc/aspeed_sdmc.c | 250 +++++++++++---- hw/net/ftgmac100.c | 162 ++++++++++ hw/sd/aspeed_sdhci.c | 198 ++++++++++++ hw/ssi/aspeed_smc.c | 129 +++++++- hw/timer/aspeed_timer.c | 213 +++++++++++-- hw/watchdog/wdt_aspeed.c | 153 +++++---- hw/arm/Makefile.objs | 2 +- hw/sd/Makefile.objs | 1 + 23 files changed, 2204 insertions(+), 268 deletions(-) create mode 100644 include/hw/sd/aspeed_sdhci.h create mode 100644 hw/arm/aspeed_ast2600.c create mode 100644 hw/sd/aspeed_sdhci.c
Signature · 1 line
2.21.0
PATCH 1/21 aspeed/wdt: Check correct register for clock source
From: Amithash Prasad <[email protected]>
When WDT_RESTART is written, the data is not the contents of the WDT_CTRL register. Hence ensure we are looking at WDT_CTRL to check if bit WDT_CTRL_1MHZ_CLK is set or not.
--- hw/watchdog/wdt_aspeed.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
hw/watchdog/wdt_aspeed.c +1 -1
@@ -140,7 +140,7 @@ static void aspeed_wdt_write(void *opaque, hwaddr offset, uint64_t data,case WDT_RESTART :if ((data &0xFFFF ) ==WDT_RESTART_MAGIC ) { s->regs[WDT_STATUS ] = s->regs[WDT_RELOAD_VALUE ]; -aspeed_wdt_reload (s, !(data &WDT_CTRL_1MHZ_CLK )); +aspeed_wdt_reload (s, !(s->regs[WDT_CTRL ] &WDT_CTRL_1MHZ_CLK )); }break ;case WDT_CTRL :
Signature · 1 line
2.21.0
Re: [PATCH 01/21] aspeed/wdt: Check correct register for clock source
Quoted 10 lines · Cédric Le Goater
From: Amithash Prasad <[email protected]>
When
WDT_RESTARTis written, the data is not the contents of theWDT_CTRLregister. Hence ensure we are looking atWDT_CTRLto check if bitWDT_CTRL_1MHZ_CLKis set or not.Signed-off-by Amithash Prasad <[email protected]>[clg: improved Suject prefix ]Signed-off-by Cédric Le Goater <[email protected]>
A patch from the Open Source Firmware Conference! Thanks Amithash.
Quoted 20 lines
--- hw/watchdog/wdt_aspeed.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)hw/watchdog/wdt_aspeed.c +1
-2@@ -140,7 +140,7 @@ static void aspeed_wdt_write(void *opaque, hwaddr offset, uint64_t data,case WDT_RESTART :if ((data &0xFFFF ) ==WDT_RESTART_MAGIC ) { s->regs[WDT_STATUS ] = s->regs[WDT_RELOAD_VALUE ]; -aspeed_wdt_reload (s, !(data &WDT_CTRL_1MHZ_CLK )); +aspeed_wdt_reload (s, !(s->regs[WDT_CTRL ] &WDT_CTRL_1MHZ_CLK )); }break ;case WDT_CTRL : --2.21.0
PATCH 2/21 hw/sd/aspeed_sdhci: New device
In reply to Cédric Le GoaterFrom: Eddie James <[email protected]>
The Aspeed SOCs have two SD/MMC controllers. Add a device that encapsulates both of these controllers and models the Aspeed-specific registers and behavior.
Tested by reading from mmcblk0 in Linux:
qemu-system-arm -machine romulus-bmc -nographic \
-drive file=flash-romulus,format=raw,if=mtd \
-device sd-card,drive=sd0 -drive file=_tmp/kernel,format=raw,if=sd,id=sd0
[clg: - changed the controller MMIO window size to 0x1000
- moved the MMIO mapping of the SDHCI slots at the SoC level
- merged code to add SD drives on the SD buses at the machine level ]
--- include/hw/arm/aspeed_soc.h | 3 + include/hw/sd/aspeed_sdhci.h | 34 ++++++ hw/arm/aspeed.c | 15 ++- hw/arm/aspeed_soc.c | 23 ++++ hw/sd/aspeed_sdhci.c | 198 +++++++++++++++++++++++++++++++++++ hw/sd/Makefile.objs | 1 + 6 files changed, 273 insertions(+), 1 deletion(-) create mode 100644 include/hw/sd/aspeed_sdhci.h create mode 100644 hw/sd/aspeed_sdhci.c
include/hw/arm/aspeed_soc.h +3 -0
@@ -24,6 +24,7 @@#include "hw/net/ftgmac100.h" #include "target/arm/cpu.h" #include "hw/gpio/aspeed_gpio.h" +#include "hw/sd/aspeed_sdhci.h" #define ASPEED_SPIS_NUM 2#define ASPEED_WDTS_NUM 3
@@ -50,6 +51,7 @@ typedef struct AspeedSoCState {AspeedWDTState wdt[ASPEED_WDTS_NUM ];FTGMAC100State ftgmac100[ASPEED_MACS_NUM ];AspeedGPIOState gpio; +AspeedSDHCIState sdhci; } AspeedSoCState;#define TYPE_ASPEED_SOC "aspeed-soc"
@@ -93,6 +95,7 @@ enum {ASPEED_SCU ,ASPEED_ADC ,ASPEED_SRAM , +ASPEED_SDHCI ,ASPEED_GPIO ,ASPEED_RTC , ASPEED_TIMER1,
include/hw/sd/aspeed_sdhci.h new file +34 -0
@@ -0,0 +1,34 @@ +/* +* Aspeed SD Host Controller +* Eddie James <[email protected]> +* +* Copyright (C) 2019 IBM Corp +* SPDX-License-Identifer: GPL-2.0-or-later +*/ + +#ifndef ASPEED_SDHCI_H +#define ASPEED_SDHCI_H + +#include "hw/sd/sdhci.h" + +#define TYPE_ASPEED_SDHCI "aspeed.sdhci" +#define ASPEED_SDHCI (obj) OBJECT_CHECK(AspeedSDHCIState, (obj), \ + TYPE_ASPEED_SDHCI) + +#define ASPEED_SDHCI_CAPABILITIES 0x01E80080 +#define ASPEED_SDHCI_NUM_SLOTS 2 +#define ASPEED_SDHCI_NUM_REGS (ASPEED_SDHCI_REG_SIZE / sizeof(uint32_t)) +#define ASPEED_SDHCI_REG_SIZE 0x100 + +typedef struct AspeedSDHCIState { +SysBusDevice parent; + +SDHCIState slots[ASPEED_SDHCI_NUM_SLOTS ]; + +MemoryRegion iomem; +qemu_irq irq; + +uint32_t regs[ASPEED_SDHCI_NUM_REGS ]; +}AspeedSDHCIState ; + +#endif /* ASPEED_SDHCI_H */
hw/arm/aspeed.c +14 -1
@@ -170,6 +170,7 @@ static void aspeed_board_init(MachineState *machine,AspeedSoCClass *sc;DriveInfo *drive0 =drive_get (IF_MTD ,0 ,0 );ram_addr_t max_ram_size; +int i; bmc =g_new0 (AspeedBoardState,1 );
@@ -252,6 +253,19 @@ static void aspeed_board_init(MachineState *machine, cfg->i2c_init (bmc); } +for (i =0 ; i <ARRAY_SIZE (bmc->soc.sdhci.slots); i++) { +SDHCIState *sdhci = &bmc->soc.sdhci.slots[i]; +DriveInfo *dinfo =drive_get_next (IF_SD ); +BlockBackend *blk; +DeviceState *card; + + blk = dinfo ?blk_by_legacy_dinfo (dinfo) :NULL ; + card =qdev_create (qdev_get_child_bus (DEVICE (sdhci),"sd-bus" ), +TYPE_SD_CARD ); +qdev_prop_set_drive (card,"drive" , blk, &error_fatal); +object_property_set_bool (OBJECT (card), true,"realized" , &error_fatal); + } +arm_load_kernel (ARM_CPU (first_cpu), machine, &aspeed_board_binfo); }
@@ -373,7 +387,6 @@ static void aspeed_machine_class_init(ObjectClass *oc, void *data) mc->desc = board->desc; mc->init = aspeed_machine_init; mc->max_cpus =ASPEED_CPUS_NUM ; - mc->no_sdcard =1 ; mc->no_floppy =1 ; mc->no_cdrom =1 ; mc->no_parallel =1 ;
hw/arm/aspeed_soc.c +23 -0
@@ -36,6 +36,7 @@ static const hwaddr aspeed_soc_ast2400_memmap[] = { [ASPEED_XDMA] = 0x1E6E7000, [ASPEED_ADC] = 0x1E6E9000, [ASPEED_SRAM] = 0x1E720000, + [ASPEED_SDHCI] = 0x1E740000, [ASPEED_GPIO] = 0x1E780000, [ASPEED_RTC] = 0x1E781000, [ASPEED_TIMER1] = 0x1E782000,
@@ -63,6 +64,7 @@ static const hwaddr aspeed_soc_ast2500_memmap[] = { [ASPEED_XDMA] = 0x1E6E7000, [ASPEED_ADC] = 0x1E6E9000, [ASPEED_SRAM] = 0x1E720000, + [ASPEED_SDHCI] = 0x1E740000, [ASPEED_GPIO] = 0x1E780000, [ASPEED_RTC] = 0x1E781000, [ASPEED_TIMER1] = 0x1E782000,
@@ -108,6 +110,7 @@ static const int aspeed_soc_ast2400_irqmap[] = { [ASPEED_ETH1] = 2, [ASPEED_ETH2] = 3, [ASPEED_XDMA] = 6, + [ASPEED_SDHCI] = 26, };#define aspeed_soc_ast2500_irqmap aspeed_soc_ast2400_irqmap
@@ -230,6 +233,15 @@ static void aspeed_soc_init(Object *obj)snprintf (typename,sizeof (typename),"aspeed.gpio-%s" , socname);sysbus_init_child_obj (obj,"gpio" ,OBJECT (&s->gpio),sizeof (s->gpio), typename); + +sysbus_init_child_obj (obj,"sdc" ,OBJECT (&s->sdhci),sizeof (s->sdhci), +TYPE_ASPEED_SDHCI ); + +/* Init sd card slot class here so that they're under the correct parent */ +for (i =0 ; i <ASPEED_SDHCI_NUM_SLOTS ; ++i) { +sysbus_init_child_obj (obj,"sdhci[*]" ,OBJECT (&s->sdhci.slots[i]), +sizeof (s->sdhci.slots[i]),TYPE_SYSBUS_SDHCI ); + } }static void aspeed_soc_realize (DeviceState *dev,Error **errp)
@@ -419,6 +431,17 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)sysbus_mmio_map (SYS_BUS_DEVICE (&s->gpio),0 , sc->info->memmap[ASPEED_GPIO ]);sysbus_connect_irq (SYS_BUS_DEVICE (&s->gpio),0 ,aspeed_soc_get_irq (s,ASPEED_GPIO )); + +/* SDHCI */ +object_property_set_bool (OBJECT (&s->sdhci), true,"realized" , &err); +if (err) { +error_propagate (errp, err); +return ; + } +sysbus_mmio_map (SYS_BUS_DEVICE (&s->sdhci),0 , + sc->info->memmap[ASPEED_SDHCI ]); +sysbus_connect_irq (SYS_BUS_DEVICE (&s->sdhci),0 , +aspeed_soc_get_irq (s,ASPEED_SDHCI )); }static Property aspeed_soc_properties[] = {DEFINE_PROP_UINT32 ("num-cpus" , AspeedSoCState, num_cpus,0 ),
hw/sd/aspeed_sdhci.c new file +198 -0
198 lines not on this page · raw message
hw/sd/Makefile.objs +1 -0
@@ -8,3 +8,4 @@ obj-$(CONFIG_MILKYMIST) += milkymist-memcard.o obj-$(CONFIG_OMAP) += omap_mmc.o obj-$(CONFIG_PXA2XX) += pxa2xx_mmci.o obj-$(CONFIG_RASPI) += bcm2835_sdhost.o +obj-$(CONFIG_ASPEED_SOC) += aspeed_sdhci.o
Signature · 1 line
2.21.0
PATCH 3/21 hw: aspeed_scu: Add AST2600 support
In reply to Cédric Le GoaterFrom: Joel Stanley <[email protected]>
The SCU controller on the AST2600 SoC has extra registers. Increase the number of regs of the model and introduce a new field in the class to customize the MemoryRegion operations depending on the SoC model.
[clg: - improved commit log
- changed vmstate version
- reworked model integration into new objet class ]
--- include/hw/misc/aspeed_scu.h | 7 +- hw/misc/aspeed_scu.c | 190 +++++++++++++++++++++++++++++++++-- 2 files changed, 189 insertions(+), 8 deletions(-)
include/hw/misc/aspeed_scu.h +6 -1
@@ -17,8 +17,10 @@#define ASPEED_SCU (obj) OBJECT_CHECK(AspeedSCUState, (obj), TYPE_ASPEED_SCU)#define TYPE_ASPEED_2400_SCU TYPE_ASPEED_SCU "-ast2400"#define TYPE_ASPEED_2500_SCU TYPE_ASPEED_SCU "-ast2500" +#define TYPE_ASPEED_2600_SCU TYPE_ASPEED_SCU "-ast2600"#define ASPEED_SCU_NR_REGS (0x1A8 >> 2) +#define ASPEED_AST2600_SCU_NR_REGS (0xE20 >> 2) typedef struct AspeedSCUState { /*< private >*/
@@ -27,7 +29,7 @@ typedef struct AspeedSCUState {/*< public >*/ MemoryRegion iomem; -uint32_t regs[ASPEED_SCU_NR_REGS ]; +uint32_t regs[ASPEED_AST2600_SCU_NR_REGS ];uint32_t silicon_rev;uint32_t hw_strap1;uint32_t hw_strap2;
@@ -38,6 +40,7 @@ typedef struct AspeedSCUState {#define AST2400_A1_SILICON_REV 0x02010303U#define AST2500_A0_SILICON_REV 0x04000303U#define AST2500_A1_SILICON_REV 0x04010303U +#define AST2600_A0_SILICON_REV 0x05000303U#define ASPEED_IS_AST2500 (si_rev) ((((si_rev) >> 24) & 0xff) == 0x04)
@@ -54,6 +57,8 @@ typedef struct AspeedSCUClass {const uint32_t *resets;uint32_t (*calc_hpll)(AspeedSCUState *s,uint32_t hpll_reg);uint32_t apb_divider; +uint32_t nr_regs; +const MemoryRegionOps *ops; } AspeedSCUClass;#define ASPEED_SCU_PROT_KEY 0x1688A8A8
hw/misc/aspeed_scu.c +183 -7
262 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 03/21] hw: aspeed_scu: Add AST2600 support
Quoted 15 lines · Cédric Le Goater
From: Joel Stanley <[email protected]>
The SCU controller on the AST2600 SoC has extra registers. Increase the number of regs of the model and introduce a new field in the class to customize the MemoryRegion operations depending on the SoC model.
Signed-off-by Joel Stanley <[email protected]>[clg: - improved commit log
- changed vmstate version
- reworked model integration into new objet class ]Signed-off-by Cédric Le Goater <[email protected]>--- include/hw/misc/aspeed_scu.h | 7 +- hw/misc/aspeed_scu.c | 190 +++++++++++++++++++++++++++++++++-- 2 files changed, 189 insertions(+), 8 deletions(-)
...
Quoted 33 lines
+static void aspeed_ast2600_scu_write(void *opaque, hwaddr offset,
uint64_tdata,hw/misc/aspeed_scu.c +8
-0+ unsigned size) +{ +AspeedSCUState *s =ASPEED_SCU (opaque); +int reg =TO_REG (offset); + +if (reg >=ASPEED_AST2600_SCU_NR_REGS ) { +qemu_log_mask (LOG_GUEST_ERROR , + "%s: Out-of-bounds write at offset 0x%"
HWADDR_PRIx"\n",hw/misc/aspeed_scu.c +6
-0+ __func__, offset); +return ; + } + +if (reg >PROT_KEY && !s->regs[PROT_KEY ]) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: SCU is locked!\n",
__func__);hw/misc/aspeed_scu.c +15
-0+ } + +trace_aspeed_scu_write (offset, size, data); + +switch (reg) { +case AST2600_PROT_KEY : + s->regs[reg] = (data ==ASPEED_SCU_PROT_KEY ) ?1 :0 ; +return ; +case AST2600_HW_STRAP1 : +case AST2600_HW_STRAP2 : +if (s->regs[reg +2 ]) { +return ; + } +/* fall through */ +case AST2600_SYS_RST_CTRL :
hw/misc/aspeed_scu.c static const TypeInfo aspeed_2500_scu_info = { · line 572
Quoted 10 lines
hw/misc/aspeed_scu.c +10
-0+case AST2600_SYS_RST_CTRL2 : +/* W1S (Write 1 to set) registers */ + s->regs[reg] |= data; +return ; +case AST2600_SYS_RST_CTRL_CLR : +case AST2600_SYS_RST_CTRL2_CLR : +case AST2600_HW_STRAP1_CLR : +case AST2600_HW_STRAP2_CLR : +/* W1C (Write 1 to clear) registers */ + s->regs[reg] &= ~data;
This clear should respect the protection register for each strap case.
Andrew
Re: [PATCH 03/21] hw: aspeed_scu: Add AST2600 support
Quoted 58 lines · Andrew Jeffery
Quoted 15 lines · Cédric Le Goater
From: Joel Stanley <[email protected]>
The SCU controller on the AST2600 SoC has extra registers. Increase the number of regs of the model and introduce a new field in the class to customize the MemoryRegion operations depending on the SoC model.
Signed-off-by Joel Stanley <[email protected]>[clg: - improved commit log
- changed vmstate version
- reworked model integration into new objet class ]Signed-off-by Cédric Le Goater <[email protected]>--- include/hw/misc/aspeed_scu.h | 7 +- hw/misc/aspeed_scu.c | 190 +++++++++++++++++++++++++++++++++-- 2 files changed, 189 insertions(+), 8 deletions(-)...
Quoted 37 lines
+static void aspeed_ast2600_scu_write(void *opaque, hwaddr offset,
uint64_tdata,hw/misc/aspeed_scu.c +8
-0+ unsigned size) +{ +AspeedSCUState *s =ASPEED_SCU (opaque); +int reg =TO_REG (offset); + +if (reg >=ASPEED_AST2600_SCU_NR_REGS ) { +qemu_log_mask (LOG_GUEST_ERROR , + "%s: Out-of-bounds write at offset 0x%"
HWADDR_PRIx"\n",hw/misc/aspeed_scu.c +6
-0+ __func__, offset); +return ; + } + +if (reg >PROT_KEY && !s->regs[PROT_KEY ]) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: SCU is locked!\n",
__func__);hw/misc/aspeed_scu.c +19
-0+ } + +trace_aspeed_scu_write (offset, size, data); + +switch (reg) { +case AST2600_PROT_KEY : + s->regs[reg] = (data ==ASPEED_SCU_PROT_KEY ) ?1 :0 ; +return ; +case AST2600_HW_STRAP1 : +case AST2600_HW_STRAP2 : +if (s->regs[reg +2 ]) { +return ; + } +/* fall through */ +case AST2600_SYS_RST_CTRL : +case AST2600_SYS_RST_CTRL2 : +/* W1S (Write 1 to set) registers */ + s->regs[reg] |= data; +return ;
Quoted 8 lines
hw/misc/aspeed_scu.c static const TypeInfo aspeed_2500_scu_info = { · line 572
Quoted 6 lines
hw/misc/aspeed_scu.c +6
-0+case AST2600_SYS_RST_CTRL_CLR : +case AST2600_SYS_RST_CTRL2_CLR : +case AST2600_HW_STRAP1_CLR : +case AST2600_HW_STRAP2_CLR : +/* W1C (Write 1 to clear) registers */ + s->regs[reg] &= ~data;This clear should respect the protection register for each strap case.
Joel,
You are the expert ! :)
Thanks,
C.
Joel Stanley
maintainerRe: [PATCH 03/21] hw: aspeed_scu: Add AST2600 support
Quoted 33 lines · Cédric Le Goater
Quoted 31 lines · Andrew Jeffery
Quoted 27 lines · Cédric Le Goater
From: Joel Stanley <[email protected]>
The SCU controller on the AST2600 SoC has extra registers. Increase the number of regs of the model and introduce a new field in the class to customize the MemoryRegion operations depending on the SoC model.
hw/misc/aspeed_scu.c +21
-0+switch (reg) { +case AST2600_PROT_KEY : + s->regs[reg] = (data ==ASPEED_SCU_PROT_KEY ) ?1 :0 ; +return ; +case AST2600_HW_STRAP1 : +case AST2600_HW_STRAP2 : +if (s->regs[reg +2 ]) { +return ; + } +/* fall through */ +case AST2600_SYS_RST_CTRL : +case AST2600_SYS_RST_CTRL2 : +/* W1S (Write 1 to set) registers */ + s->regs[reg] |= data; +return ; +case AST2600_SYS_RST_CTRL_CLR : +case AST2600_SYS_RST_CTRL2_CLR : +case AST2600_HW_STRAP1_CLR : +case AST2600_HW_STRAP2_CLR : +/* W1C (Write 1 to clear) registers */ + s->regs[reg] &= ~data;
Quoted 5 lines
Quoted 1 line
This clear should respect the protection register for each strap case.
Joel,
You are the expert ! :)
Someone could implement this if they wanted to. In the future it might be useful to create a detailed model for the OTP and secure boot behavior, and that can affect the strapping.
However it is not critical for running guests under qemu. I think we should defer it until there is some guest code that needs the detailed behavior.
Cheers,
Joel
Re: [PATCH 03/21] hw: aspeed_scu: Add AST2600 support
Quoted 40 lines · Joel Stanley
Quoted 33 lines · Cédric Le Goater
Quoted 31 lines · Andrew Jeffery
Quoted 27 lines · Cédric Le Goater
From: Joel Stanley <[email protected]>
The SCU controller on the AST2600 SoC has extra registers. Increase the number of regs of the model and introduce a new field in the class to customize the MemoryRegion operations depending on the SoC model.
hw/misc/aspeed_scu.c +21
-0+switch (reg) { +case AST2600_PROT_KEY : + s->regs[reg] = (data ==ASPEED_SCU_PROT_KEY ) ?1 :0 ; +return ; +case AST2600_HW_STRAP1 : +case AST2600_HW_STRAP2 : +if (s->regs[reg +2 ]) { +return ; + } +/* fall through */ +case AST2600_SYS_RST_CTRL : +case AST2600_SYS_RST_CTRL2 : +/* W1S (Write 1 to set) registers */ + s->regs[reg] |= data; +return ; +case AST2600_SYS_RST_CTRL_CLR : +case AST2600_SYS_RST_CTRL2_CLR : +case AST2600_HW_STRAP1_CLR : +case AST2600_HW_STRAP2_CLR : +/* W1C (Write 1 to clear) registers */ + s->regs[reg] &= ~data;Quoted 5 lines
Quoted 1 line
This clear should respect the protection register for each strap case.
Joel,
You are the expert ! :)
Quoted 7 lines
Someone could implement this if they wanted to. In the future it might be useful to create a detailed model for the OTP and secure boot behavior, and that can affect the strapping.
However it is not critical for running guests under qemu. I think we should defer it until there is some guest code that needs the detailed behavior.
ok. It think we could trap the invalid writes with a simple mask array at the beginning of the write op .
Thanks,
C.
PATCH 4/21 aspeed/timer: Introduce an object class per SoC
In reply to Cédric Le GoaterThe most important changes will be on the register range 0x34 - 0x3C memops. Introduce class read/write operations to handle the differences between SoCs.
--- include/hw/timer/aspeed_timer.h | 15 +++++ hw/arm/aspeed_soc.c | 3 +- hw/timer/aspeed_timer.c | 107 ++++++++++++++++++++++++++++---- 3 files changed, 113 insertions(+), 12 deletions(-)
include/hw/timer/aspeed_timer.h +15 -0
@@ -28,6 +28,9 @@#define ASPEED_TIMER (obj) \ OBJECT_CHECK(AspeedTimerCtrlState, (obj), TYPE_ASPEED_TIMER);#define TYPE_ASPEED_TIMER "aspeed.timer" +#define TYPE_ASPEED_2400_TIMER TYPE_ASPEED_TIMER "-ast2400" +#define TYPE_ASPEED_2500_TIMER TYPE_ASPEED_TIMER "-ast2500" +#define ASPEED_TIMER_NR_TIMERS 8 typedef struct AspeedTimer {
@@ -60,4 +63,16 @@ typedef struct AspeedTimerCtrlState { AspeedSCUState *scu; } AspeedTimerCtrlState; +#define ASPEED_TIMER_CLASS (klass) \ + OBJECT_CLASS_CHECK(AspeedTimerClass, (klass), TYPE_ASPEED_TIMER) +#define ASPEED_TIMER_GET_CLASS (obj) \ + OBJECT_GET_CLASS(AspeedTimerClass, (obj), TYPE_ASPEED_TIMER) + +typedef struct AspeedTimerClass { +SysBusDeviceClass parent_class; + +uint64_t (*read)(AspeedTimerCtrlState *s,hwaddr offset); +void (*write)(AspeedTimerCtrlState *s,hwaddr offset,uint64_t value); +}AspeedTimerClass ; +#endif /* ASPEED_TIMER_H */
hw/arm/aspeed_soc.c +2 -1
@@ -182,8 +182,9 @@ static void aspeed_soc_init(Object *obj)sysbus_init_child_obj (obj,"rtc" ,OBJECT (&s->rtc),sizeof (s->rtc),TYPE_ASPEED_RTC ); +snprintf (typename,sizeof (typename),"aspeed.timer-%s" , socname);sysbus_init_child_obj (obj,"timerctrl" ,OBJECT (&s->timerctrl), -sizeof (s->timerctrl),TYPE_ASPEED_TIMER ); +sizeof (s->timerctrl), typename);object_property_add_const_link (OBJECT (&s->timerctrl),"scu" ,OBJECT (&s->scu), &error_abort);
hw/timer/aspeed_timer.c +96 -11
138 lines not on this page · raw message
Signature · 1 line
2.21.0
PATCH 5/21 aspeed/timer: Add support for control register 3
In reply to Cédric Le GoaterThe AST2500 timer has a third control register that is used to implement a set-to-clear feature for the main control register.
This models the behaviour expected by the AST2500 while maintaining the same behaviour for the AST2400.
The vmstate version is not increased yet because the structure is modified again in the following patches.
Based on previous work from Joel Stanley.
--- include/hw/timer/aspeed_timer.h | 1 + hw/timer/aspeed_timer.c | 19 +++++++++++++++++++ 2 files changed, 20 insertions(+)
include/hw/timer/aspeed_timer.h +1 -0
@@ -58,6 +58,7 @@ typedef struct AspeedTimerCtrlState {uint32_t ctrl;uint32_t ctrl2; +uint32_t ctrl3;AspeedTimer timers[ASPEED_TIMER_NR_TIMERS ];AspeedSCUState *scu;
hw/timer/aspeed_timer.c +19 -0
@@ -498,6 +498,8 @@ static uint64_t aspeed_2500_timer_read(AspeedTimerCtrlState *s, hwaddr offset)switch (offset) {case 0x38 : + value = s->ctrl3 &BIT (0 ); +break ;case 0x3C :default : qemu_log_mask(LOG_GUEST_ERROR, "%s: Bad offset 0x%" HWADDR_PRIx "\n",
@@ -511,9 +513,24 @@ static uint64_t aspeed_2500_timer_read(AspeedTimerCtrlState *s, hwaddr offset)static void aspeed_2500_timer_write (AspeedTimerCtrlState *s,hwaddr offset,uint64_t value) { +const uint32_t tv = (uint32_t )(value &0xFFFFFFFF ); +uint8_t command; +switch (offset) {case 0x38 : + command = (value >>1 ) &0xFF ; +if (command ==0xAE ) { + s->ctrl3 =0x1 ; + }else if (command ==0xEA ) { + s->ctrl3 =0x0 ; + } +break ;case 0x3C : +if (s->ctrl3 &BIT (0 )) { +aspeed_timer_set_ctrl (s, s->ctrl & ~tv); + } +break ; +default :qemu_log_mask (LOG_GUEST_ERROR ,"%s: Bad offset 0x%" HWADDR_PRIx"\n" , __func__, offset);
@@ -574,6 +591,7 @@ static void aspeed_timer_reset(DeviceState *dev) } s->ctrl =0 ; s->ctrl2 =0 ; + s->ctrl3 =0 ; }static const VMStateDescription vmstate_aspeed_timer = {
@@ -597,6 +615,7 @@ static const VMStateDescription vmstate_aspeed_timer_state = { .fields = (VMStateField[]) {VMSTATE_UINT32 (ctrl, AspeedTimerCtrlState),VMSTATE_UINT32 (ctrl2, AspeedTimerCtrlState), +VMSTATE_UINT32 (ctrl3, AspeedTimerCtrlState),VMSTATE_STRUCT_ARRAY (timers, AspeedTimerCtrlState,ASPEED_TIMER_NR_TIMERS ,1 , vmstate_aspeed_timer, AspeedTimer),
Signature · 1 line
2.21.0
PATCH 6/21 aspeed/timer: Add AST2600 support
In reply to Cédric Le GoaterThe AST2600 timer has a third control register that is used to implement a set-to-clear feature for the main control register.
On the AST2600, it is not configurable via 0x38 (control register 3) as it is on the AST2500.
Based on previous work from Joel Stanley.
--- include/hw/timer/aspeed_timer.h | 1 + hw/timer/aspeed_timer.c | 51 +++++++++++++++++++++++++++++++++ 2 files changed, 52 insertions(+)
include/hw/timer/aspeed_timer.h +1 -0
@@ -30,6 +30,7 @@#define TYPE_ASPEED_TIMER "aspeed.timer"#define TYPE_ASPEED_2400_TIMER TYPE_ASPEED_TIMER "-ast2400"#define TYPE_ASPEED_2500_TIMER TYPE_ASPEED_TIMER "-ast2500" +#define TYPE_ASPEED_2600_TIMER TYPE_ASPEED_TIMER "-ast2600"#define ASPEED_TIMER_NR_TIMERS 8
hw/timer/aspeed_timer.c +51 -0
68 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 06/21] aspeed/timer: Add AST2600 support
Quoted 10 lines · Cédric Le Goater
The AST2600 timer has a third control register that is used to implement a set-to-clear feature for the main control register.
On the AST2600, it is not configurable via
0x38(control register 3) as it is on the AST2500.Based on previous work from Joel Stanley.
Signed-off-by Cédric Le Goater <[email protected]>
PATCH 7/21 aspeed/timer: Add support for IRQ status register on the AST2600
In reply to Cédric Le GoaterThe AST2600 timer replaces control register 2 with a interrupt status register. It is set by hardware when an IRQ occurs and cleared by software.
Modify the vmstate version to take into account the new fields.
Based on previous work from Joel Stanley.
--- include/hw/timer/aspeed_timer.h | 1 + hw/timer/aspeed_timer.c | 36 +++++++++++++++++++++++++-------- 2 files changed, 29 insertions(+), 8 deletions(-)
include/hw/timer/aspeed_timer.h +1 -0
@@ -60,6 +60,7 @@ typedef struct AspeedTimerCtrlState {uint32_t ctrl;uint32_t ctrl2;uint32_t ctrl3; +uint32_t irq_sts;AspeedTimer timers[ASPEED_TIMER_NR_TIMERS ];AspeedSCUState *scu;
hw/timer/aspeed_timer.c +28 -8
@@ -160,7 +160,9 @@ static uint64_t calculate_next(struct AspeedTimer *t)timer_del (&t->timer);if (timer_overflow_interrupt (t)) { +AspeedTimerCtrlState *s =timer_to_ctrl (t); t->level = !t->level; + s->irq_sts |=BIT (t->id);qemu_set_irq (t->irq, t->level); }
@@ -199,7 +201,9 @@ static void aspeed_timer_expire(void *opaque) }if (interrupt) { +AspeedTimerCtrlState *s =timer_to_ctrl (t); t->level = !t->level; + s->irq_sts |=BIT (t->id);qemu_set_irq (t->irq, t->level); }
@@ -244,9 +248,6 @@ static uint64_t aspeed_timer_read(void *opaque, hwaddr offset, unsigned size)case 0x30 :/* Control Register */ value = s->ctrl;break ; -case 0x34 :/* Control Register 2 */ - value = s->ctrl2; -break ; case 0x00 ... 0x2c: /* Timers 1 - 4 */ value =aspeed_timer_get_value (&s->timers[(offset >>4 )], reg);break ;
@@ -438,9 +439,6 @@ static void aspeed_timer_write(void *opaque, hwaddr offset, uint64_t value,case 0x30 :aspeed_timer_set_ctrl (s, tv);break ; -case 0x34 : -aspeed_timer_set_ctrl2 (s, tv); -break ;/* Timer Registers */ case 0x00 ... 0x2c:aspeed_timer_set_value (s, (offset >>TIMER_NR_REGS ), reg, tv);
@@ -468,6 +466,9 @@ static uint64_t aspeed_2400_timer_read(AspeedTimerCtrlState *s, hwaddr offset)uint64_t value;switch (offset) { +case 0x34 : + value = s->ctrl2; +break ;case 0x38 :case 0x3C :default :
@@ -482,7 +483,12 @@ static uint64_t aspeed_2400_timer_read(AspeedTimerCtrlState *s, hwaddr offset)static void aspeed_2400_timer_write (AspeedTimerCtrlState *s,hwaddr offset,uint64_t value) { +const uint32_t tv = (uint32_t )(value &0xFFFFFFFF ); +switch (offset) { +case 0x34 : +aspeed_timer_set_ctrl2 (s, tv); +break ;case 0x38 :case 0x3C :default :
@@ -497,6 +503,9 @@ static uint64_t aspeed_2500_timer_read(AspeedTimerCtrlState *s, hwaddr offset)uint64_t value;switch (offset) { +case 0x34 : + value = s->ctrl2; +break ;case 0x38 : value = s->ctrl3 &BIT (0 );break ;
@@ -517,6 +526,9 @@ static void aspeed_2500_timer_write(AspeedTimerCtrlState *s, hwaddr offset,uint8_t command;switch (offset) { +case 0x34 : +aspeed_timer_set_ctrl2 (s, tv); +break ;case 0x38 : command = (value >>1 ) &0xFF ;if (command ==0xAE ) {
@@ -543,6 +555,9 @@ static uint64_t aspeed_2600_timer_read(AspeedTimerCtrlState *s, hwaddr offset)uint64_t value;switch (offset) { +case 0x34 : + value = s->irq_sts; +break ;case 0x38 :case 0x3C :default :
@@ -560,6 +575,9 @@ static void aspeed_2600_timer_write(AspeedTimerCtrlState *s, hwaddr offset,const uint32_t tv = (uint32_t )(value &0xFFFFFFFF );switch (offset) { +case 0x34 : + s->irq_sts &= tv; +break ;case 0x3C :aspeed_timer_set_ctrl (s, s->ctrl & ~tv);break ;
@@ -626,6 +644,7 @@ static void aspeed_timer_reset(DeviceState *dev) s->ctrl =0 ; s->ctrl2 =0 ; s->ctrl3 =0 ; + s->irq_sts =0 ; }static const VMStateDescription vmstate_aspeed_timer = {
@@ -644,12 +663,13 @@ static const VMStateDescription vmstate_aspeed_timer = {static const VMStateDescription vmstate_aspeed_timer_state = { .name ="aspeed.timerctrl" , - .version_id =1 , - .minimum_version_id =1 , + .version_id =2 , + .minimum_version_id =2 , .fields = (VMStateField []) {VMSTATE_UINT32 (ctrl, AspeedTimerCtrlState),VMSTATE_UINT32 (ctrl2, AspeedTimerCtrlState),VMSTATE_UINT32 (ctrl3, AspeedTimerCtrlState), +VMSTATE_UINT32 (irq_sts, AspeedTimerCtrlState),VMSTATE_STRUCT_ARRAY (timers, AspeedTimerCtrlState,ASPEED_TIMER_NR_TIMERS ,1 , vmstate_aspeed_timer, AspeedTimer),
Signature · 1 line
2.21.0
Re: [PATCH 07/21] aspeed/timer: Add support for IRQ status register on the AST2600
Quoted 10 lines · Cédric Le Goater
The AST2600 timer replaces control register 2 with a interrupt status register. It is set by hardware when an IRQ occurs and cleared by software.
Modify the vmstate version to take into account the new fields.
Based on previous work from Joel Stanley.
Signed-off-by Cédric Le Goater <[email protected]>
PATCH 8/21 aspeed/sdmc: Introduce an object class per SoC
In reply to Cédric Le GoaterUse class handlers and class constants to differentiate the characteristics of the memory controller and remove the 'silicon_rev' property.
--- include/hw/misc/aspeed_sdmc.h | 19 +++- hw/arm/aspeed_soc.c | 5 +- hw/misc/aspeed_sdmc.c | 168 +++++++++++++++++++++------------- 3 files changed, 122 insertions(+), 70 deletions(-)
include/hw/misc/aspeed_sdmc.h +15 -4
@@ -13,6 +13,8 @@#define TYPE_ASPEED_SDMC "aspeed.sdmc"#define ASPEED_SDMC (obj) OBJECT_CHECK(AspeedSDMCState, (obj), TYPE_ASPEED_SDMC) +#define TYPE_ASPEED_2400_SDMC TYPE_ASPEED_SDMC "-ast2400" +#define TYPE_ASPEED_2500_SDMC TYPE_ASPEED_SDMC "-ast2500"#define ASPEED_SDMC_NR_REGS (0x174 >> 2)
@@ -24,12 +26,21 @@ typedef struct AspeedSDMCState {MemoryRegion iomem;uint32_t regs[ASPEED_SDMC_NR_REGS ]; -uint32_t silicon_rev; -uint32_t ram_bits;uint64_t ram_size; uint64_t max_ram_size; - uint32_t fixed_conf; - } AspeedSDMCState; +#define ASPEED_SDMC_CLASS (klass) \ + OBJECT_CLASS_CHECK(AspeedSDMCClass, (klass), TYPE_ASPEED_SDMC) +#define ASPEED_SDMC_GET_CLASS (obj) \ + OBJECT_GET_CLASS(AspeedSDMCClass, (obj), TYPE_ASPEED_SDMC) + +typedef struct AspeedSDMCClass { +SysBusDeviceClass parent_class; + +uint64_t max_ram_size; +uint32_t (*compute_conf)(AspeedSDMCState *s,uint32_t data); +void (*write)(AspeedSDMCState *s,uint32_t reg,uint32_t data); +}AspeedSDMCClass ; +#endif /* ASPEED_SDMC_H */
hw/arm/aspeed_soc.c +2 -3
@@ -205,10 +205,9 @@ static void aspeed_soc_init(Object *obj) sizeof(s->spi[i]), typename); } +snprintf (typename,sizeof (typename),"aspeed.sdmc-%s" , socname);sysbus_init_child_obj (obj,"sdmc" ,OBJECT (&s->sdmc),sizeof (s->sdmc), -TYPE_ASPEED_SDMC ); -qdev_prop_set_uint32 (DEVICE (&s->sdmc),"silicon-rev" , - sc->info->silicon_rev); + typename);object_property_add_alias (obj,"ram-size" ,OBJECT (&s->sdmc),"ram-size" , &error_abort); object_property_add_alias(obj, "max-ram-size", OBJECT(&s->sdmc),
hw/misc/aspeed_sdmc.c +105 -63
214 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 08/21] aspeed/sdmc: Introduce an object class per SoC
Quoted 6 lines · Cédric Le Goater
Use class handlers and class constants to differentiate the characteristics of the memory controller and remove the '
silicon_rev' property.Signed-off-by Cédric Le Goater <[email protected]>
PATCH 9/21 aspeed/sdmc: Add AST2600 support
In reply to Cédric Le GoaterFrom: Joel Stanley <[email protected]>
The AST2600 SDMC controller is slightly different from its predecessor (DRAM training). Max memory is now 2G on the AST2600.
[clg: - improved commit log
- reworked model integration into new objet class ]
--- include/hw/misc/aspeed_sdmc.h | 1 + hw/misc/aspeed_scu.c | 2 + hw/misc/aspeed_sdmc.c | 82 +++++++++++++++++++++++++++++++++++ 3 files changed, 85 insertions(+)
include/hw/misc/aspeed_sdmc.h +1 -0
@@ -15,6 +15,7 @@#define ASPEED_SDMC (obj) OBJECT_CHECK(AspeedSDMCState, (obj), TYPE_ASPEED_SDMC)#define TYPE_ASPEED_2400_SDMC TYPE_ASPEED_SDMC "-ast2400"#define TYPE_ASPEED_2500_SDMC TYPE_ASPEED_SDMC "-ast2500" +#define TYPE_ASPEED_2600_SDMC TYPE_ASPEED_SDMC "-ast2600"#define ASPEED_SDMC_NR_REGS (0x174 >> 2)
hw/misc/aspeed_scu.c +2 -0
@@ -99,6 +99,7 @@#define AST2600_CLK_STOP_CTRL_CLR TO_REG(0x84)#define AST2600_CLK_STOP_CTRL2 TO_REG(0x90)#define AST2600_CLK_STOP_CTR2L_CLR TO_REG(0x94) +#define AST2600_SDRAM_HANDSHAKE TO_REG(0x100)#define AST2600_HPLL_EXT TO_REG(0x204)#define AST2600_MPLL_EXT TO_REG(0x224)#define AST2600_EPLL_EXT TO_REG(0x244)
@@ -601,6 +602,7 @@ static const uint32_t ast2600_a0_resets[ASPEED_AST2600_SCU_NR_REGS] = { [AST2600_SYS_RST_CTRL2] = 0xFFFFFFFC, [AST2600_CLK_STOP_CTRL] = 0xEFF43E8B, [AST2600_CLK_STOP_CTRL2] = 0xFFF0FFF0, + [AST2600_SDRAM_HANDSHAKE] = 0x00000040, /* SoC completed DRAM init */ };static void aspeed_ast2600_scu_reset (DeviceState *dev)
hw/misc/aspeed_sdmc.c +82 -0
111 lines not on this page · raw message
Signature · 1 line
2.21.0
PATCH 10/21 watchdog/aspeed: Introduce an object class per SoC
In reply to Cédric Le GoaterIt cleanups the current models for the Aspeed AST2400 and AST2500 SoCs and prepares ground for future SoCs. It removes the need of the 'silicon_rev' property.
--- include/hw/watchdog/wdt_aspeed.h | 18 ++++- hw/arm/aspeed_soc.c | 9 ++- hw/watchdog/wdt_aspeed.c | 122 ++++++++++++++++--------------- 3 files changed, 86 insertions(+), 63 deletions(-)
include/hw/watchdog/wdt_aspeed.h +16 -2
@@ -16,6 +16,8 @@#define TYPE_ASPEED_WDT "aspeed.wdt"#define ASPEED_WDT (obj) \ OBJECT_CHECK(AspeedWDTState, (obj), TYPE_ASPEED_WDT) +#define TYPE_ASPEED_2400_WDT TYPE_ASPEED_WDT "-ast2400" +#define TYPE_ASPEED_2500_WDT TYPE_ASPEED_WDT "-ast2500"#define ASPEED_WDT_REGS_MAX (0x20 / 4)
@@ -30,8 +32,20 @@ typedef struct AspeedWDTState {AspeedSCUState *scu; uint32_t pclk_freq; -uint32_t silicon_rev; - uint32_t ext_pulse_width_mask; } AspeedWDTState; +#define ASPEED_WDT_CLASS (klass) \ + OBJECT_CLASS_CHECK(AspeedWDTClass, (klass), TYPE_ASPEED_WDT) +#define ASPEED_WDT_GET_CLASS (obj) \ + OBJECT_GET_CLASS(AspeedWDTClass, (obj), TYPE_ASPEED_WDT) + +typedef struct AspeedWDTClass { +SysBusDeviceClass parent_class; + +uint32_t offset; +uint32_t ext_pulse_width_mask; +uint32_t reset_ctrl_reg; +void (*reset_pulse)(AspeedWDTState *s,uint32_t property); +}AspeedWDTClass ; +#endif /* WDT_ASPEED_H */
hw/arm/aspeed_soc.c +5 -4
@@ -214,10 +214,9 @@ static void aspeed_soc_init(Object *obj) "max-ram-size", &error_abort);for (i =0 ; i < sc->info->wdts_num; i++) { +snprintf (typename,sizeof (typename),"aspeed.wdt-%s" , socname);sysbus_init_child_obj (obj,"wdt[*]" ,OBJECT (&s->wdt[i]), -sizeof (s->wdt[i]),TYPE_ASPEED_WDT ); -qdev_prop_set_uint32 (DEVICE (&s->wdt[i]),"silicon-rev" , - sc->info->silicon_rev); +sizeof (s->wdt[i]), typename);object_property_add_const_link (OBJECT (&s->wdt[i]),"scu" ,OBJECT (&s->scu), &error_abort); }
@@ -384,13 +383,15 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)/* Watch dog */ for (i =0 ; i < sc->info->wdts_num; i++) { +AspeedWDTClass *awc =ASPEED_WDT_GET_CLASS (&s->wdt[i]); +object_property_set_bool (OBJECT (&s->wdt[i]), true,"realized" , &err);if (err) {error_propagate (errp, err);return ; }sysbus_mmio_map (SYS_BUS_DEVICE (&s->wdt[i]),0 , - sc->info->memmap[ASPEED_WDT ] + i *0x20 ); + sc->info->memmap[ASPEED_WDT ] + i * awc->offset); }/* Net */
hw/watchdog/wdt_aspeed.c +65 -57
182 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 10/21] watchdog/aspeed: Introduce an object class per SoC
Quoted 6 lines · Cédric Le Goater
It cleanups the current models for the Aspeed AST2400 and AST2500 SoCs and prepares ground for future SoCs. It removes the need of the '
silicon_rev' property.Signed-off-by Cédric Le Goater <[email protected]>
PATCH 11/21 hw: wdt_aspeed: Add AST2600 support
In reply to Cédric Le GoaterFrom: Joel Stanley <[email protected]>
The AST2600 has four watchdogs, and they each have a 0x40 of registers.
When running as part of an ast2600 system we must check a different offset for the system reset control register in the SCU.
--- include/hw/arm/aspeed_soc.h | 2 +- include/hw/watchdog/wdt_aspeed.h | 1 + hw/watchdog/wdt_aspeed.c | 29 +++++++++++++++++++++++++++++ 3 files changed, 31 insertions(+), 1 deletion(-)
include/hw/arm/aspeed_soc.h +1 -1
@@ -27,7 +27,7 @@#include "hw/sd/aspeed_sdhci.h" #define ASPEED_SPIS_NUM 2 -#define ASPEED_WDTS_NUM 3 +#define ASPEED_WDTS_NUM 4#define ASPEED_CPUS_NUM 2#define ASPEED_MACS_NUM 2
include/hw/watchdog/wdt_aspeed.h +1 -0
@@ -18,6 +18,7 @@ OBJECT_CHECK(AspeedWDTState, (obj), TYPE_ASPEED_WDT)#define TYPE_ASPEED_2400_WDT TYPE_ASPEED_WDT "-ast2400"#define TYPE_ASPEED_2500_WDT TYPE_ASPEED_WDT "-ast2500" +#define TYPE_ASPEED_2600_WDT TYPE_ASPEED_WDT "-ast2600"#define ASPEED_WDT_REGS_MAX (0x20 / 4)
hw/watchdog/wdt_aspeed.c +29 -0
@@ -40,12 +40,14 @@#define WDT_DRIVE_TYPE_MASK (0xFF << 24)#define WDT_PUSH_PULL_MAGIC (0xA8 << 24)#define WDT_OPEN_DRAIN_MAGIC (0x8A << 24) +#define WDT_RESET_MASK1 (0x1c / 4)#define WDT_TIMEOUT_STATUS (0x10 / 4)#define WDT_TIMEOUT_CLEAR (0x14 / 4)#define WDT_RESTART_MAGIC 0x4755 +#define AST2600_SCU_RESET_CONTROL1 (0x40 / 4)#define SCU_RESET_CONTROL1 (0x04 / 4)#define SCU_RESET_SDRAM BIT(0)
@@ -74,6 +76,8 @@ static uint64_t aspeed_wdt_read(void *opaque, hwaddr offset, unsigned size)return s->regs[WDT_CTRL ];case WDT_RESET_WIDTH :return s->regs[WDT_RESET_WIDTH ]; +case WDT_RESET_MASK1 : +return s->regs[WDT_RESET_MASK1 ];case WDT_TIMEOUT_STATUS :case WDT_TIMEOUT_CLEAR : qemu_log_mask(LOG_UNIMP,
@@ -146,6 +150,11 @@ static void aspeed_wdt_write(void *opaque, hwaddr offset, uint64_t data, s->regs[WDT_RESET_WIDTH ] |= data & awc->ext_pulse_width_mask;break ; +case WDT_RESET_MASK1 : +/* TODO: implement */ + s->regs[WDT_RESET_MASK1 ] = data; +break ; +case WDT_TIMEOUT_STATUS :case WDT_TIMEOUT_CLEAR : qemu_log_mask(LOG_UNIMP,
@@ -316,12 +325,32 @@ static const TypeInfo aspeed_2500_wdt_info = { .class_init = aspeed_2500_wdt_class_init, }; +static void aspeed_2600_wdt_class_init (ObjectClass *klass,void *data) +{ +DeviceClass *dc =DEVICE_CLASS (klass); +AspeedWDTClass *awc =ASPEED_WDT_CLASS (klass); + + dc->desc ="ASPEED 2600 Watchdog Controller" ; + awc->offset =0x40 ; + awc->ext_pulse_width_mask =0xfffff ;/* TODO */ + awc->reset_ctrl_reg =AST2600_SCU_RESET_CONTROL1 ; + awc->reset_pulse = aspeed_2500_wdt_reset_pulse; +} + +static const TypeInfo aspeed_2600_wdt_info = { + .name =TYPE_ASPEED_2600_WDT , + .parent =TYPE_ASPEED_WDT , + .instance_size =sizeof (AspeedWDTState), + .class_init = aspeed_2600_wdt_class_init, +}; +static void wdt_aspeed_register_types (void ) {watchdog_add_model (&model);type_register_static (&aspeed_wdt_info);type_register_static (&aspeed_2400_wdt_info);type_register_static (&aspeed_2500_wdt_info); +type_register_static (&aspeed_2600_wdt_info); }type_init (wdt_aspeed_register_types)
Signature · 1 line
2.21.0
PATCH 12/21 aspeed/smc: Add AST2600 support
In reply to Cédric Le GoaterThe AST2600 SoC SMC controller is a SPI only controller now and has a few extensions which we will need to take into account when SW requires it.
- 4BYTE mode
- HCLK divider has changed (SPI Training)
- CE0-2 Read Timing Compensation registers
This is enough to support u-boot.
--- include/hw/ssi/aspeed_smc.h | 2 + hw/ssi/aspeed_smc.c | 129 +++++++++++++++++++++++++++++++++--- 2 files changed, 123 insertions(+), 8 deletions(-)
include/hw/ssi/aspeed_smc.h +2 -0
@@ -49,6 +49,8 @@ typedef struct AspeedSMCController {hwaddr dma_flash_mask;hwaddr dma_dram_mask;uint32_t nregs; +uint32_t (*segment_to_reg)(const AspeedSegments *seg); +void (*reg_to_segment)(uint32_t reg,AspeedSegments *seg); } AspeedSMCController; typedef struct AspeedSMCFlash {
hw/ssi/aspeed_smc.c +121 -8
215 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 12/21] aspeed/smc: Add AST2600 support
Quoted 10 lines · Cédric Le Goater
The AST2600 SoC SMC controller is a SPI only controller now and has a few extensions which we will need to take into account when SW requires it.
- 4BYTE mode
- HCLK divider has changed (SPI Training)
- CE0-2 Read Timing Compensation registersThis is enough to support u-boot.
As easy as that! I hope the Linux driver is as simple :)
Quoted 1 line
Signed-off-by Cédric Le Goater <[email protected]>
Re: [PATCH 12/21] aspeed/smc: Add AST2600 support
Quoted 6 lines · Joel Stanley
Quoted 5 lines · Cédric Le Goater
The AST2600 SoC SMC controller is a SPI only controller now and has a few extensions which we will need to take into account when SW requires it.
Quoted 7 lines
Quoted 5 lines
- 4BYTE mode
- HCLK divider has changed (SPI Training)
- CE0-2 Read Timing Compensation registersThis is enough to support u-boot.
As easy as that! I hope the Linux driver is as simple :)
Only HW will tell ! :)
Quoted 3 lines
Quoted 1 line
Signed-off-by Cédric Le Goater <[email protected]>Reviewed-by Joel Stanley <[email protected]>
I had some issues with the segment routines. I will have to resend this one.
thanks,
C.
PATCH 13/21 hw/gpio: Add in AST2600 specific implementation
In reply to Cédric Le GoaterFrom: Rashmica Gupta <[email protected]>
The AST2600 has the same sets of 3.6v gpios as the AST2400 plus an addtional two sets of 1.8V gpios.
--- hw/gpio/aspeed_gpio.c | 142 ++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 137 insertions(+), 5 deletions(-)
hw/gpio/aspeed_gpio.c +137 -5
@@ -169,6 +169,48 @@#define GPIO_3_6V_MEM_SIZE 0x1F0#define GPIO_3_6V_REG_ARRAY_SIZE (GPIO_3_6V_MEM_SIZE >> 2) +/* AST2600 only - 1.8V gpios */ +/* +* The AST2600 has same 3.6V gpios as the AST2400 (memory offsets 0x0-0x198) +* and addtional 1.8V gpios (memory offsets 0x800-0x9D4). +*/ +#define GPIO_1_8V_REG_OFFSET 0x800 +#define GPIO_1_8V_ABCD_DATA_VALUE ((0x800 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_DIRECTION ((0x804 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_INT_ENABLE ((0x808 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_INT_SENS_0 ((0x80C - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_INT_SENS_1 ((0x810 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_INT_SENS_2 ((0x814 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_INT_STATUS ((0x818 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_RESET_TOLERANT ((0x81C - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_DATA_VALUE ((0x820 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_DIRECTION ((0x824 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_INT_ENABLE ((0x828 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_INT_SENS_0 ((0x82C - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_INT_SENS_1 ((0x830 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_INT_SENS_2 ((0x834 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_INT_STATUS ((0x838 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_RESET_TOLERANT ((0x83C - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_DEBOUNCE_1 ((0x840 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_DEBOUNCE_2 ((0x844 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_DEBOUNCE_1 ((0x848 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_DEBOUNCE_2 ((0x84C - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_DEBOUNCE_TIME_1 ((0x850 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_DEBOUNCE_TIME_2 ((0x854 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_DEBOUNCE_TIME_3 ((0x858 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_COMMAND_SRC_0 ((0x860 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_COMMAND_SRC_1 ((0x864 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_COMMAND_SRC_0 ((0x868 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_COMMAND_SRC_1 ((0x86C - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_DATA_READ ((0x8C0 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_DATA_READ ((0x8C4 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_ABCD_INPUT_MASK ((0x9D0 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_E_INPUT_MASK ((0x9D4 - GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_1_8V_MEM_SIZE 0x9D8 +#define GPIO_1_8V_REG_ARRAY_SIZE ((GPIO_1_8V_MEM_SIZE - \ + GPIO_1_8V_REG_OFFSET) >> 2) +#define GPIO_MAX_MEM_SIZE MAX(GPIO_3_6V_MEM_SIZE, GPIO_1_8V_MEM_SIZE) +static int aspeed_evaluate_irq (GPIOSets *regs,int gpio_prev_high,int gpio) {uint32_t falling_edge =0 , rising_edge =0 ;
@@ -465,6 +507,39 @@ static const AspeedGPIOReg aspeed_3_6v_gpios[GPIO_3_6V_REG_ARRAY_SIZE] = { [GPIO_AC_INPUT_MASK] = { 7, gpio_reg_input_mask }, }; +static const AspeedGPIOReg aspeed_1_8v_gpios[GPIO_1_8V_REG_ARRAY_SIZE ] = { +/* 1.8V Set ABCD */ + [GPIO_1_8V_ABCD_DATA_VALUE ] = {0 , gpio_reg_data_value}, + [GPIO_1_8V_ABCD_DIRECTION ] = {0 , gpio_reg_direction}, + [GPIO_1_8V_ABCD_INT_ENABLE ] = {0 , gpio_reg_int_enable}, + [GPIO_1_8V_ABCD_INT_SENS_0 ] = {0 , gpio_reg_int_sens_0}, + [GPIO_1_8V_ABCD_INT_SENS_1 ] = {0 , gpio_reg_int_sens_1}, + [GPIO_1_8V_ABCD_INT_SENS_2 ] = {0 , gpio_reg_int_sens_2}, + [GPIO_1_8V_ABCD_INT_STATUS ] = {0 , gpio_reg_int_status}, + [GPIO_1_8V_ABCD_RESET_TOLERANT ] = {0 , gpio_reg_reset_tolerant}, + [GPIO_1_8V_ABCD_DEBOUNCE_1 ] = {0 , gpio_reg_debounce_1}, + [GPIO_1_8V_ABCD_DEBOUNCE_2 ] = {0 , gpio_reg_debounce_2}, + [GPIO_1_8V_ABCD_COMMAND_SRC_0 ] = {0 , gpio_reg_cmd_source_0}, + [GPIO_1_8V_ABCD_COMMAND_SRC_1 ] = {0 , gpio_reg_cmd_source_1}, + [GPIO_1_8V_ABCD_DATA_READ ] = {0 , gpio_reg_data_read}, + [GPIO_1_8V_ABCD_INPUT_MASK ] = {0 , gpio_reg_input_mask}, +/* 1.8V Set E */ + [GPIO_1_8V_E_DATA_VALUE ] = {1 , gpio_reg_data_value}, + [GPIO_1_8V_E_DIRECTION ] = {1 , gpio_reg_direction}, + [GPIO_1_8V_E_INT_ENABLE ] = {1 , gpio_reg_int_enable}, + [GPIO_1_8V_E_INT_SENS_0 ] = {1 , gpio_reg_int_sens_0}, + [GPIO_1_8V_E_INT_SENS_1 ] = {1 , gpio_reg_int_sens_1}, + [GPIO_1_8V_E_INT_SENS_2 ] = {1 , gpio_reg_int_sens_2}, + [GPIO_1_8V_E_INT_STATUS ] = {1 , gpio_reg_int_status}, + [GPIO_1_8V_E_RESET_TOLERANT ] = {1 , gpio_reg_reset_tolerant}, + [GPIO_1_8V_E_DEBOUNCE_1 ] = {1 , gpio_reg_debounce_1}, + [GPIO_1_8V_E_DEBOUNCE_2 ] = {1 , gpio_reg_debounce_2}, + [GPIO_1_8V_E_COMMAND_SRC_0 ] = {1 , gpio_reg_cmd_source_0}, + [GPIO_1_8V_E_COMMAND_SRC_1 ] = {1 , gpio_reg_cmd_source_1}, + [GPIO_1_8V_E_DATA_READ ] = {1 , gpio_reg_data_read}, + [GPIO_1_8V_E_INPUT_MASK ] = {1 , gpio_reg_input_mask}, +}; +static uint64_t aspeed_gpio_read (void *opaque,hwaddr offset,uint32_t size) {AspeedGPIOState *s =ASPEED_GPIO (opaque);
@@ -663,8 +738,11 @@ static void aspeed_gpio_get_pin(Object *obj, Visitor *v, const char *name,int set_idx, group_idx =0 ;if (sscanf (name,"gpio%2[A-Z]%1d" , group, &pin) !=2 ) { -error_setg (errp,"%s: error reading %s" , __func__, name); -return ; +/* 1.8V gpio */ +if (sscanf (name,"gpio%3s%1d" , group, &pin) !=2 ) { +error_setg (errp,"%s: error reading %s" , __func__, name); +return ; + } } set_idx =get_set_idx (s, group, &group_idx);if (set_idx ==-1 ) {
@@ -692,8 +770,11 @@ static void aspeed_gpio_set_pin(Object *obj, Visitor *v, const char *name,return ; }if (sscanf (name,"gpio%2[A-Z]%1d" , group, &pin) !=2 ) { -error_setg (errp,"%s: error reading %s" , __func__, name); -return ; +/* 1.8V gpio */ +if (sscanf (name,"gpio%3s%1d" , group, &pin) !=2 ) { +error_setg (errp,"%s: error reading %s" , __func__, name); +return ; + } } set_idx =get_set_idx (s, group, &group_idx);if (set_idx ==-1 ) {
@@ -726,6 +807,21 @@ static const GPIOSetProperties ast2500_set_props[] = { [7] = {0x000000ff, 0x000000ff, {"AC"} }, }; +static GPIOSetProperties ast2600_3_6v_set_props[] = { + [0 ] = {0xffffffff ,0xffffffff , {"A" ,"B" ,"C" ,"D" } }, + [1 ] = {0xffffffff ,0xffffffff , {"E" ,"F" ,"G" ,"H" } }, + [2 ] = {0xffffffff ,0xffffffff , {"I" ,"J" ,"K" ,"L" } }, + [3 ] = {0xffffffff ,0xffffffff , {"M" ,"N" ,"O" ,"P" } }, + [4 ] = {0xffffffff ,0xffffffff , {"Q" ,"R" ,"S" ,"T" } }, + [5 ] = {0xffffffff ,0x0000ffff , {"U" ,"V" ,"W" ,"X" } }, + [6 ] = {0xffff0000 ,0x0fff0000 , {"Y" ,"Z" ,"" ,"" } }, +}; + +static GPIOSetProperties ast2600_1_8v_set_props[] = { + [0 ] = {0xffffffff ,0xffffffff , {"18A" ,"18B" ,"18C" ,"18D" } }, + [1 ] = {0x0000000f ,0x0000000f , {"18E" } }, +}; +static const MemoryRegionOps aspeed_gpio_ops = { .read = aspeed_gpio_read, .write = aspeed_gpio_write,
@@ -758,7 +854,7 @@ static void aspeed_gpio_realize(DeviceState *dev, Error **errp) }memory_region_init_io (&s->iomem,OBJECT (s), &aspeed_gpio_ops, s, -TYPE_ASPEED_GPIO ,GPIO_3_6V_MEM_SIZE ); +TYPE_ASPEED_GPIO ,GPIO_MAX_MEM_SIZE );sysbus_init_mmio (sbd, &s->iomem); }
@@ -851,6 +947,26 @@ static void aspeed_gpio_2500_class_init(ObjectClass *klass, void *data) agc->reg_table = aspeed_3_6v_gpios; } +static void aspeed_gpio_ast2600_3_6v_class_init (ObjectClass *klass,void *data) +{ +AspeedGPIOClass *agc =ASPEED_GPIO_CLASS (klass); + + agc->props = ast2600_3_6v_set_props; + agc->nr_gpio_pins =208 ; + agc->nr_gpio_sets =7 ; + agc->reg_table = aspeed_3_6v_gpios; +} + +static void aspeed_gpio_ast2600_1_8v_class_init (ObjectClass *klass,void *data) +{ +AspeedGPIOClass *agc =ASPEED_GPIO_CLASS (klass); + + agc->props = ast2600_1_8v_set_props; + agc->nr_gpio_pins =36 ; + agc->nr_gpio_sets =2 ; + agc->reg_table = aspeed_1_8v_gpios; +} +static const TypeInfo aspeed_gpio_info = { .name =TYPE_ASPEED_GPIO , .parent =TYPE_SYS_BUS_DEVICE ,
@@ -874,11 +990,27 @@ static const TypeInfo aspeed_gpio_ast2500_info = { .instance_init = aspeed_gpio_init, }; +static const TypeInfo aspeed_gpio_ast2600_3_6v_info = { + .name =TYPE_ASPEED_GPIO "-ast2600" , + .parent =TYPE_ASPEED_GPIO , + .class_init = aspeed_gpio_ast2600_3_6v_class_init, + .instance_init = aspeed_gpio_init, +}; + +static const TypeInfo aspeed_gpio_ast2600_1_8v_info = { + .name =TYPE_ASPEED_GPIO "-ast2600-1_8v" , + .parent =TYPE_ASPEED_GPIO , + .class_init = aspeed_gpio_ast2600_1_8v_class_init, + .instance_init = aspeed_gpio_init, +}; +static void aspeed_gpio_register_types (void ) {type_register_static (&aspeed_gpio_info);type_register_static (&aspeed_gpio_ast2400_info);type_register_static (&aspeed_gpio_ast2500_info); +type_register_static (&aspeed_gpio_ast2600_3_6v_info); +type_register_static (&aspeed_gpio_ast2600_1_8v_info); }type_init (aspeed_gpio_register_types);
Signature · 1 line
2.21.0
PATCH 14/21 aspeed/i2c: Introduce an object class per SoC
In reply to Cédric Le GoaterIt prepares ground for register differences between SoCs.
--- include/hw/i2c/aspeed_i2c.h | 15 ++++++++++ hw/arm/aspeed_soc.c | 3 +- hw/i2c/aspeed_i2c.c | 60 ++++++++++++++++++++++++++++++++----- 3 files changed, 69 insertions(+), 9 deletions(-)
include/hw/i2c/aspeed_i2c.h +15 -0
@@ -25,6 +25,8 @@#include "hw/sysbus.h" #define TYPE_ASPEED_I2C "aspeed.i2c" +#define TYPE_ASPEED_2400_I2C TYPE_ASPEED_I2C "-ast2400" +#define TYPE_ASPEED_2500_I2C TYPE_ASPEED_I2C "-ast2500"#define ASPEED_I2C (obj) \ OBJECT_CHECK(AspeedI2CState, (obj), TYPE_ASPEED_I2C)
@@ -59,6 +61,19 @@ typedef struct AspeedI2CState { AspeedI2CBus busses[ASPEED_I2C_NR_BUSSES]; } AspeedI2CState; +#define ASPEED_I2C_CLASS (klass) \ + OBJECT_CLASS_CHECK(AspeedI2CClass, (klass), TYPE_ASPEED_I2C) +#define ASPEED_I2C_GET_CLASS (obj) \ + OBJECT_GET_CLASS(AspeedI2CClass, (obj), TYPE_ASPEED_I2C) + +typedef struct AspeedI2CClass { +SysBusDeviceClass parent_class; + +uint8_t num_busses; +uint8_t reg_size; +uint8_t gap; +}AspeedI2CClass ; +I2CBus *aspeed_i2c_get_bus (DeviceState *dev,int busnr);#endif /* ASPEED_I2C_H */
hw/arm/aspeed_soc.c +2 -1
@@ -188,8 +188,9 @@ static void aspeed_soc_init(Object *obj)object_property_add_const_link (OBJECT (&s->timerctrl),"scu" ,OBJECT (&s->scu), &error_abort); +snprintf (typename,sizeof (typename),"aspeed.i2c-%s" , socname);sysbus_init_child_obj (obj,"i2c" ,OBJECT (&s->i2c),sizeof (s->i2c), -TYPE_ASPEED_I2C ); + typename);snprintf (typename,sizeof (typename),"aspeed.fmc-%s" , socname); sysbus_init_child_obj(obj, "fmc", OBJECT(&s->fmc), sizeof(s->fmc),
hw/i2c/aspeed_i2c.c +52 -8
111 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 14/21] aspeed/i2c: Introduce an object class per SoC
Quoted 4 lines · Cédric Le Goater
It prepares ground for register differences between SoCs.
Signed-off-by Cédric Le Goater <[email protected]>
PATCH 15/21 aspeed/i2c: Add AST2600 support
In reply to Cédric Le GoaterThe I2C controller of the AST2400 and AST2500 SoCs have one IRQ shared by all I2C busses. The AST2600 SoC I2C controller has one IRQ per bus and 16 busses.
--- include/hw/i2c/aspeed_i2c.h | 5 +++- hw/i2c/aspeed_i2c.c | 46 +++++++++++++++++++++++++++++++++++-- 2 files changed, 48 insertions(+), 3 deletions(-)
include/hw/i2c/aspeed_i2c.h +4 -1
@@ -27,10 +27,11 @@#define TYPE_ASPEED_I2C "aspeed.i2c"#define TYPE_ASPEED_2400_I2C TYPE_ASPEED_I2C "-ast2400"#define TYPE_ASPEED_2500_I2C TYPE_ASPEED_I2C "-ast2500" +#define TYPE_ASPEED_2600_I2C TYPE_ASPEED_I2C "-ast2600"#define ASPEED_I2C (obj) \ OBJECT_CHECK(AspeedI2CState, (obj), TYPE_ASPEED_I2C) -#define ASPEED_I2C_NR_BUSSES 14 +#define ASPEED_I2C_NR_BUSSES 16struct AspeedI2CState ;
@@ -41,6 +42,7 @@ typedef struct AspeedI2CBus {I2CBus *bus;uint8_t id; +qemu_irq irq;uint32_t ctrl;uint32_t timing[2 ];
@@ -72,6 +74,7 @@ typedef struct AspeedI2CClass {uint8_t num_busses;uint8_t reg_size;uint8_t gap; +qemu_irq (*bus_get_irq)(AspeedI2CBus *); } AspeedI2CClass;I2CBus *aspeed_i2c_get_bus (DeviceState *dev,int busnr);
hw/i2c/aspeed_i2c.c +44 -2
108 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 15/21] aspeed/i2c: Add AST2600 support
Quoted 6 lines · Cédric Le Goater
The I2C controller of the AST2400 and AST2500 SoCs have one IRQ shared by all I2C busses. The AST2600 SoC I2C controller has one IRQ per bus and 16 busses.
Signed-off-by Cédric Le Goater <[email protected]>
PATCH 16/21 aspeed: Introduce an object class per SoC
In reply to Cédric Le GoaterIt prepares ground for the AST2600.
--- include/hw/arm/aspeed_soc.h | 9 +-- hw/arm/aspeed.c | 4 +- hw/arm/aspeed_soc.c | 148 +++++++++++++++++++----------------- 3 files changed, 84 insertions(+), 77 deletions(-)
include/hw/arm/aspeed_soc.h +3 -6
@@ -57,7 +57,9 @@ typedef struct AspeedSoCState {#define TYPE_ASPEED_SOC "aspeed-soc"#define ASPEED_SOC (obj) OBJECT_CHECK(AspeedSoCState, (obj), TYPE_ASPEED_SOC) -typedef struct AspeedSoCInfo { +typedef struct AspeedSoCClass { + DeviceClass parent_class; + const char *name; const char *cpu_type; uint32_t silicon_rev;
@@ -67,11 +69,6 @@ typedef struct AspeedSoCInfo {const int *irqmap;const hwaddr *memmap;uint32_t num_cpus; -} AspeedSoCInfo; - -typedef struct AspeedSoCClass { -DeviceClass parent_class; -AspeedSoCInfo *info; } AspeedSoCClass;#define ASPEED_SOC_CLASS (klass) \
hw/arm/aspeed.c +2 -2
@@ -215,7 +215,7 @@ static void aspeed_board_init(MachineState *machine,memory_region_allocate_system_memory (&bmc->ram,NULL ,"ram" , ram_size);memory_region_add_subregion (&bmc->ram_container,0 , &bmc->ram);memory_region_add_subregion (get_system_memory (), - sc->info->memmap[ASPEED_SDRAM ], + sc->memmap[ASPEED_SDRAM ], &bmc->ram_container); max_ram_size = object_property_get_uint(OBJECT(&bmc->soc), "max-ram-size",
@@ -246,7 +246,7 @@ static void aspeed_board_init(MachineState *machine, } aspeed_board_binfo.ram_size = ram_size; - aspeed_board_binfo.loader_start = sc->info->memmap[ASPEED_SDRAM ]; + aspeed_board_binfo.loader_start = sc->memmap[ASPEED_SDRAM ]; aspeed_board_binfo.nb_cpus = bmc->soc.num_cpus;if (cfg->i2c_init) {
hw/arm/aspeed_soc.c +79 -69
322 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 16/21] aspeed: Introduce an object class per SoC
Quoted 4 lines · Cédric Le Goater
It prepares ground for the AST2600.
Signed-off-by Cédric Le Goater <[email protected]>
PATCH 17/21 aspeed/soc: Add AST2600 support
In reply to Cédric Le GoaterInitial definitions for a simple machine using an AST2600 SoC (Cortex CPU).
The Cortex CPU and its interrupt controller are too complex to handle in the common Aspeed SoC framework. We introduce a new Aspeed SoC class with instance_init and realize handlers to handle the differences with the AST2400 and the AST2500 SoCs. This will add extra work to keep in sync both models with future extensions but it makes the code clearer.
--- include/hw/arm/aspeed_soc.h | 4 + hw/arm/aspeed_ast2600.c | 494 ++++++++++++++++++++++++++++++++++++ hw/arm/Makefile.objs | 2 +- 3 files changed, 499 insertions(+), 1 deletion(-) create mode 100644 hw/arm/aspeed_ast2600.c
include/hw/arm/aspeed_soc.h +4 -0
@@ -12,6 +12,7 @@#ifndef ASPEED_SOC_H #define ASPEED_SOC_H +#include "hw/cpu/a15mpcore.h" #include "hw/intc/aspeed_vic.h" #include "hw/misc/aspeed_scu.h" #include "hw/misc/aspeed_sdmc.h"
@@ -38,6 +39,7 @@ typedef struct AspeedSoCState {/*< public >*/ ARMCPU cpu[ASPEED_CPUS_NUM ];uint32_t num_cpus; +A15MPPrivState a7mpcore;MemoryRegion sram;AspeedVICState vic;AspeedRtcState rtc;
@@ -51,6 +53,7 @@ typedef struct AspeedSoCState {AspeedWDTState wdt[ASPEED_WDTS_NUM ];FTGMAC100State ftgmac100[ASPEED_MACS_NUM ];AspeedGPIOState gpio; +AspeedGPIOState gpio_1_8v;AspeedSDHCIState sdhci; } AspeedSoCState;
@@ -94,6 +97,7 @@ enum {ASPEED_SRAM ,ASPEED_SDHCI ,ASPEED_GPIO , +ASPEED_GPIO_1_8V ,ASPEED_RTC ,ASPEED_TIMER1 , ASPEED_TIMER2,
hw/arm/aspeed_ast2600.c new file +494 -0
494 lines not on this page · raw message
hw/arm/Makefile.objs +1 -1
@@ -41,7 +41,7 @@ obj-$(CONFIG_XLNX_VERSAL) += xlnx-versal.o xlnx-versal-virt.o obj-$(CONFIG_FSL_IMX25) += fsl-imx25.o imx25_pdk.o obj-$(CONFIG_FSL_IMX31) += fsl-imx31.o kzm.o obj-$(CONFIG_FSL_IMX6) += fsl-imx6.o -obj-$(CONFIG_ASPEED_SOC) += aspeed_soc.o aspeed.o +obj-$(CONFIG_ASPEED_SOC) += aspeed_soc.o aspeed.o aspeed_ast2600.o obj-$(CONFIG_MPS2) += mps2.o obj-$(CONFIG_MPS2) += mps2-tz.o obj-$(CONFIG_MSF2) += msf2-soc.o
Signature · 1 line
2.21.0
Re: [PATCH 17/21] aspeed/soc: Add AST2600 support
Quoted 4 lines · Cédric Le Goater
Initial definitions for a simple machine using an AST2600 SoC (Cortex CPU).
Quoted 8 lines
The Cortex CPU and its interrupt controller are too complex to handle in the common Aspeed SoC framework. We introduce a new Aspeed SoC class with
instance_initand realize handlers to handle the differences with the AST2400 and the AST2500 SoCs. This will add extra work to keep in sync both models with future extensions but it makes the code clearer.Signed-off-by Cédric Le Goater <[email protected]>
One small addition below. If you don't resend I can do a follow up patch for it:
hw/arm/aspeed_ast2600.c line 90
Quoted 5 lines
hw/arm/aspeed_ast2600.c new file +4
-0+ +static const int aspeed_soc_ast2600_irqmap[] = { + [ASPEED_ETH1 ] =2 , + [ASPEED_ETH2 ] =3 ,
We need to add ETH3 and ETH4 here. They look like this:
[ASPEED_ETH3] = 32,
[ASPEED_ETH4] = 33,Re: [PATCH 17/21] aspeed/soc: Add AST2600 support
Quoted 21 lines · Joel Stanley
Quoted 4 lines · Cédric Le Goater
Initial definitions for a simple machine using an AST2600 SoC (Cortex CPU).
Quoted 8 lines
The Cortex CPU and its interrupt controller are too complex to handle in the common Aspeed SoC framework. We introduce a new Aspeed SoC class with
instance_initand realize handlers to handle the differences with the AST2400 and the AST2500 SoCs. This will add extra work to keep in sync both models with future extensions but it makes the code clearer.Signed-off-by Cédric Le Goater <[email protected]>Reviewed-by Joel Stanley <[email protected]>One small addition below. If you don't resend I can do a follow up patch for it:
Quoted 2 lines
(quoted diff) +2
-0+ +static const int aspeed_soc_ast2600_irqmap[] = {
Quoted 8 lines
hw/arm/aspeed_ast2600.c line 90
Quoted 2 lines
hw/arm/aspeed_ast2600.c new file +2
-0+ [ASPEED_ETH1] = 2, + [ASPEED_ETH2] = 3,We need to add ETH3 and ETH4 here. They look like this:
[ASPEED_ETH3] = 32, [ASPEED_ETH4] = 33,
I will include them in :
[PATCH 19/21] aspeed: Parameterise number of MACs
Thanks,
C.
PATCH 18/21 aspeed: Add an AST2600 eval board
In reply to Cédric Le Goater--- hw/arm/aspeed.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+)
hw/arm/aspeed.c +18 -0
@@ -88,6 +88,9 @@ struct AspeedBoardState {/* Witherspoon hardware value: 0xF10AD216 (but use romulus definition) */ #define WITHERSPOON_BMC_HW_STRAP1 ROMULUS_BMC_HW_STRAP1 +/* AST2600 evb hardware value: (QEMU prototype) */ +#define AST2600_EVB_HW_STRAP1 AST2500_EVB_HW_STRAP1 +/* * The max ram region is for firmwares that scan the address space * with load/store to guess how much RAM the SoC has.
@@ -308,6 +311,12 @@ static void ast2500_evb_i2c_init(AspeedBoardState *bmc)i2c_create_slave (aspeed_i2c_get_bus (DEVICE (&soc->i2c),11 ),"ds1338" ,0x32 ); } +static void ast2600_evb_i2c_init (AspeedBoardState *bmc) +{ +/* Start with some devices on our I2C busses */ +ast2500_evb_i2c_init (bmc); +} +static void romulus_bmc_i2c_init (AspeedBoardState *bmc) {AspeedSoCState *soc = &bmc->soc;
@@ -455,6 +464,15 @@ static const AspeedBoardConfig aspeed_boards[] = { .num_cs = 2, .i2c_init = witherspoon_bmc_i2c_init, .ram = 512 * MiB, + }, { + .name = MACHINE_TYPE_NAME("ast2600-evb"), + .desc = "Aspeed AST2600 EVB (Cortex A7)", + .soc_name = "ast2600-a0", + .hw_strap1 = AST2600_EVB_HW_STRAP1, + .fmc_model = "mx25l25635e", + .spi_model = "mx25l25635e", + .num_cs = 1, + .i2c_init = ast2600_evb_i2c_init, }, };
Signature · 1 line
2.21.0
Re: [PATCH 18/21] aspeed: Add an AST2600 eval board
Quoted 11 lines · Cédric Le Goater
Signed-off-by Cédric Le Goater <[email protected]>--- hw/arm/aspeed.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+)hw/arm/aspeed.c +0
-0
hw/arm/aspeed.c struct AspeedBoardState { · line 92
Quoted 5 lines
hw/arm/aspeed.c +2
-0@@ -88,6 +88,9 @@ struct AspeedBoardState {/* Witherspoon hardware value: 0xF10AD216 (but use romulus definition) */ #define WITHERSPOON_BMC_HW_STRAP1 ROMULUS_BMC_HW_STRAP1 +/* AST2600 evb hardware value: (QEMU prototype) */ +#define AST2600_EVB_HW_STRAP1 AST2500_EVB_HW_STRAP1
We can change this to be the actual values if you want:
STRAP1: 0x000000C0
STRAP2: 0x00000003
hw/arm/aspeed.c static const AspeedBoardConfig aspeed_boards[] = { · line 475
Quoted 8 lines
hw/arm/aspeed.c +8
-0+ .name = MACHINE_TYPE_NAME("ast2600-evb"), + .desc = "Aspeed AST2600 EVB (Cortex A7)", + .soc_name = "ast2600-a0", + .hw_strap1 = AST2600_EVB_HW_STRAP1, + .fmc_model = "mx25l25635e", + .spi_model = "mx25l25635e", + .num_cs = 1, + .i2c_init = ast2600_evb_i2c_init,
Can we add a default ram size?
mc->default_ram_size = 2 * GiB;
With these two fixed:
Cheers,
Joel
PATCH 19/21 aspeed: Parameterise number of MACs
In reply to Cédric Le GoaterFrom: Joel Stanley <[email protected]>
To support the ast2600's four MACs allow SoCs to specify the number they have, and create that many.
--- include/hw/arm/aspeed_soc.h | 5 ++++- hw/arm/aspeed_ast2600.c | 7 +++++-- hw/arm/aspeed_soc.c | 6 ++++-- 3 files changed, 13 insertions(+), 5 deletions(-)
include/hw/arm/aspeed_soc.h +4 -1
@@ -30,7 +30,7 @@#define ASPEED_SPIS_NUM 2#define ASPEED_WDTS_NUM 4#define ASPEED_CPUS_NUM 2 -#define ASPEED_MACS_NUM 2 +#define ASPEED_MACS_NUM 4 typedef struct AspeedSoCState { /*< private >*/
@@ -69,6 +69,7 @@ typedef struct AspeedSoCClass {uint64_t sram_size;int spis_num;int wdts_num; +int macs_num;const int *irqmap;const hwaddr *memmap;uint32_t num_cpus;
@@ -114,6 +115,8 @@ enum {ASPEED_I2C ,ASPEED_ETH1 ,ASPEED_ETH2 , +ASPEED_ETH3 , +ASPEED_ETH4 ,ASPEED_SDRAM , ASPEED_XDMA, };
hw/arm/aspeed_ast2600.c +5 -2
@@ -32,7 +32,9 @@ static const hwaddr aspeed_soc_ast2600_memmap[] = { [ASPEED_SPI1] = 0x1E630000, [ASPEED_SPI2] = 0x1E641000, [ASPEED_ETH1] = 0x1E660000, + [ASPEED_ETH3] = 0x1E670000, [ASPEED_ETH2] = 0x1E680000, + [ASPEED_ETH4] = 0x1E690000, [ASPEED_VIC] = 0x1E6C0000, [ASPEED_SDMC] = 0x1E6E0000, [ASPEED_SCU] = 0x1E6E2000,
@@ -175,7 +177,7 @@ static void aspeed_soc_ast2600_init(Object *obj) OBJECT(&s->scu), &error_abort); } -for (i =0 ; i <ASPEED_MACS_NUM ; i++) { +for (i =0 ; i < sc->macs_num; i++) {sysbus_init_child_obj (obj,"ftgmac100[*]" ,OBJECT (&s->ftgmac100[i]),sizeof (s->ftgmac100[i]),TYPE_FTGMAC100 ); }
@@ -399,7 +401,7 @@ static void aspeed_soc_ast2600_realize(DeviceState *dev, Error **errp) }/* Net */ -for (i =0 ; i < nb_nics; i++) { +for (i =0 ; i < nb_nics && i < sc->macs_num; i++) {qdev_set_nic_properties (DEVICE (&s->ftgmac100[i]), &nd_table[i]);object_property_set_bool (OBJECT (&s->ftgmac100[i]), true,"aspeed" , &err);
@@ -472,6 +474,7 @@ static void aspeed_soc_ast2600_class_init(ObjectClass *oc, void *data) sc->sram_size =0x10000 ; sc->spis_num =2 ; sc->wdts_num =4 ; + sc->macs_num =4 ; sc->irqmap = aspeed_soc_ast2600_irqmap; sc->memmap = aspeed_soc_ast2600_memmap; sc->num_cpus =2 ;
hw/arm/aspeed_soc.c +4 -2
@@ -198,7 +198,7 @@ static void aspeed_soc_init(Object *obj) OBJECT(&s->scu), &error_abort); } -for (i =0 ; i <ASPEED_MACS_NUM ; i++) { +for (i =0 ; i < sc->macs_num; i++) {sysbus_init_child_obj (obj,"ftgmac100[*]" ,OBJECT (&s->ftgmac100[i]),sizeof (s->ftgmac100[i]),TYPE_FTGMAC100 ); }
@@ -372,7 +372,7 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp) }/* Net */ -for (i =0 ; i < nb_nics; i++) { +for (i =0 ; i < nb_nics && i < sc->macs_num; i++) {qdev_set_nic_properties (DEVICE (&s->ftgmac100[i]), &nd_table[i]);object_property_set_bool (OBJECT (&s->ftgmac100[i]), true,"aspeed" , &err);
@@ -455,6 +455,7 @@ static void aspeed_soc_ast2400_class_init(ObjectClass *oc, void *data) sc->sram_size =0x8000 ; sc->spis_num =1 ; sc->wdts_num =2 ; + sc->macs_num =2 ; sc->irqmap = aspeed_soc_ast2400_irqmap; sc->memmap = aspeed_soc_ast2400_memmap; sc->num_cpus =1 ;
@@ -478,6 +479,7 @@ static void aspeed_soc_ast2500_class_init(ObjectClass *oc, void *data) sc->sram_size =0x9000 ; sc->spis_num =2 ; sc->wdts_num =3 ; + sc->macs_num =2 ; sc->irqmap = aspeed_soc_ast2500_irqmap; sc->memmap = aspeed_soc_ast2500_memmap; sc->num_cpus =1 ;
Signature · 1 line
2.21.0
PATCH 20/21 aspeed: add support for the Aspeed MII controller of the AST2600
In reply to Cédric Le GoaterThe AST2600 SoC has an extra controller to set the PHY registers.
--- include/hw/arm/aspeed_soc.h | 5 ++ include/hw/net/ftgmac100.h | 17 ++++ hw/arm/aspeed_ast2600.c | 20 +++++ hw/net/ftgmac100.c | 162 ++++++++++++++++++++++++++++++++++++ 4 files changed, 204 insertions(+)
include/hw/arm/aspeed_soc.h +5 -0
@@ -52,6 +52,7 @@ typedef struct AspeedSoCState {AspeedSDMCState sdmc;AspeedWDTState wdt[ASPEED_WDTS_NUM ];FTGMAC100State ftgmac100[ASPEED_MACS_NUM ]; +AspeedMiiState mii[ASPEED_MACS_NUM ];AspeedGPIOState gpio;AspeedGPIOState gpio_1_8v;AspeedSDHCIState sdhci;
@@ -117,6 +118,10 @@ enum {ASPEED_ETH2 ,ASPEED_ETH3 ,ASPEED_ETH4 , +ASPEED_MII1 , +ASPEED_MII2 , +ASPEED_MII3 , +ASPEED_MII4 ,ASPEED_SDRAM , ASPEED_XDMA, };
include/hw/net/ftgmac100.h +17 -0
@@ -66,4 +66,21 @@ typedef struct FTGMAC100State { uint32_t rxdes0_edorr; } FTGMAC100State; +#define TYPE_ASPEED_MII "aspeed-mmi" +#define ASPEED_MII (obj) OBJECT_CHECK(AspeedMiiState, (obj), TYPE_ASPEED_MII) + +/* +* AST2600 MII controller +*/ +typedef struct AspeedMiiState { +/*< private >*/ +SysBusDevice parent_obj; + +FTGMAC100State *nic; + +MemoryRegion iomem; +uint32_t phycr; +uint32_t phydata; +}AspeedMiiState ; +#endif
hw/arm/aspeed_ast2600.c +20 -0
@@ -31,6 +31,10 @@ static const hwaddr aspeed_soc_ast2600_memmap[] = { [ASPEED_FMC] = 0x1E620000, [ASPEED_SPI1] = 0x1E630000, [ASPEED_SPI2] = 0x1E641000, + [ASPEED_MII1] = 0x1E650000, + [ASPEED_MII2] = 0x1E650008, + [ASPEED_MII3] = 0x1E650010, + [ASPEED_MII4] = 0x1E650018, [ASPEED_ETH1] = 0x1E660000, [ASPEED_ETH3] = 0x1E670000, [ASPEED_ETH2] = 0x1E680000,
@@ -180,6 +184,12 @@ static void aspeed_soc_ast2600_init(Object *obj)for (i =0 ; i < sc->macs_num; i++) {sysbus_init_child_obj (obj,"ftgmac100[*]" ,OBJECT (&s->ftgmac100[i]),sizeof (s->ftgmac100[i]),TYPE_FTGMAC100 ); + +sysbus_init_child_obj (obj,"mii[*]" , &s->mii[i],sizeof (s->mii[i]), +TYPE_ASPEED_MII ); +object_property_add_const_link (OBJECT (&s->mii[i]),"nic" , +OBJECT (&s->ftgmac100[i]), + &error_abort); } sysbus_init_child_obj(obj, "xdma", OBJECT(&s->xdma), sizeof(s->xdma),
@@ -416,6 +426,16 @@ static void aspeed_soc_ast2600_realize(DeviceState *dev, Error **errp) sc->memmap[ASPEED_ETH1 + i]);sysbus_connect_irq (SYS_BUS_DEVICE (&s->ftgmac100[i]),0 ,aspeed_soc_get_irq (s,ASPEED_ETH1 + i)); + +object_property_set_bool (OBJECT (&s->mii[i]), true,"realized" , + &err); +if (err) { +error_propagate (errp, err); +return ; + } + +sysbus_mmio_map (SYS_BUS_DEVICE (&s->mii[i]),0 , + sc->memmap[ASPEED_MII1 + i]); }/* XDMA */
hw/net/ftgmac100.c +162 -0
177 lines not on this page · raw message
Signature · 1 line
2.21.0
Re: [PATCH 20/21] aspeed: add support for the Aspeed MII controller of the AST2600
Quoted 4 lines · Cédric Le Goater
The AST2600 SoC has an extra controller to set the PHY registers.
Signed-off-by Cédric Le Goater <[email protected]>
PATCH 21/21 aspeed/soc: Add ASPEED Video stub
In reply to Cédric Le GoaterFrom: Joel Stanley <[email protected]>
--- include/hw/arm/aspeed_soc.h | 1 + hw/arm/aspeed_ast2600.c | 5 +++++ hw/arm/aspeed_soc.c | 6 ++++++ 3 files changed, 12 insertions(+)
include/hw/arm/aspeed_soc.h +1 -0
@@ -96,6 +96,7 @@ enum {ASPEED_SDMC ,ASPEED_SCU ,ASPEED_ADC , +ASPEED_VIDEO ,ASPEED_SRAM ,ASPEED_SDHCI , ASPEED_GPIO,
hw/arm/aspeed_ast2600.c +5 -0
@@ -44,6 +44,7 @@ static const hwaddr aspeed_soc_ast2600_memmap[] = { [ASPEED_SCU] = 0x1E6E2000, [ASPEED_XDMA] = 0x1E6E7000, [ASPEED_ADC] = 0x1E6E9000, + [ASPEED_VIDEO] = 0x1E700000, [ASPEED_SDHCI] = 0x1E740000, [ASPEED_GPIO] = 0x1E780000, [ASPEED_GPIO_1_8V] = 0x1E780800,
@@ -235,6 +236,10 @@ static void aspeed_soc_ast2600_realize(DeviceState *dev, Error **errp)create_unimplemented_device ("aspeed_soc.io" , sc->memmap[ASPEED_IOMEM ],ASPEED_SOC_IOMEM_SIZE ); +/* Video engine stub */ +create_unimplemented_device ("aspeed.video" , sc->memmap[ASPEED_VIDEO ], +0x1000 ); +if (s->num_cpus > sc->num_cpus) {warn_report ("%s: invalid number of CPUs %d, using default %d" , sc->name, s->num_cpus, sc->num_cpus);
hw/arm/aspeed_soc.c +6 -0
@@ -34,6 +34,7 @@ static const hwaddr aspeed_soc_ast2400_memmap[] = { [ASPEED_SDMC] = 0x1E6E0000, [ASPEED_SCU] = 0x1E6E2000, [ASPEED_XDMA] = 0x1E6E7000, + [ASPEED_VIDEO] = 0x1E700000, [ASPEED_ADC] = 0x1E6E9000, [ASPEED_SRAM] = 0x1E720000, [ASPEED_SDHCI] = 0x1E740000,
@@ -63,6 +64,7 @@ static const hwaddr aspeed_soc_ast2500_memmap[] = { [ASPEED_SCU] = 0x1E6E2000, [ASPEED_XDMA] = 0x1E6E7000, [ASPEED_ADC] = 0x1E6E9000, + [ASPEED_VIDEO] = 0x1E700000, [ASPEED_SRAM] = 0x1E720000, [ASPEED_SDHCI] = 0x1E740000, [ASPEED_GPIO] = 0x1E780000,
@@ -231,6 +233,10 @@ static void aspeed_soc_realize(DeviceState *dev, Error **errp)create_unimplemented_device ("aspeed_soc.io" , sc->memmap[ASPEED_IOMEM ],ASPEED_SOC_IOMEM_SIZE ); +/* Video engine stub */ +create_unimplemented_device ("aspeed.video" , sc->memmap[ASPEED_VIDEO ], +0x1000 ); +if (s->num_cpus > sc->num_cpus) {warn_report ("%s: invalid number of CPUs %d, using default %d" , sc->name, s->num_cpus, sc->num_cpus);
Signature · 1 line
2.21.0
Patchew URL: https://patchew.org/QEMU/[email protected]/
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series failed the docker-quick@centos7 build test. Please find the testing commands and their output below. If you have Docker installed, you can probably reproduce it locally.
=== TEST SCRIPT BEGIN ===
#!/bin/bash
make docker-image-centos7 V=1 NETWORK=1
time make docker-test-quick@centos7 SHOW_ENV=1 J=14 NETWORK=1 === TEST SCRIPT END ===
libudev no
default devices yes
warning: Python 2 support is deprecated
warning: Python 3 will be required for building future versions of QEMU cross containers no
NOTE: guest cross-compilers enabled: cc
---
Property '.cntfrq' not found
Broken pipe/tmp/qemu-test/src/tests/libqtest.c:149: kill_qemu() detected QEMU death from signal 6 (Aborted) (core dumped) ERROR - too few tests run (expected 6, got 5)
make: *** [check-qtest-aarch64] Error 1
make: *** Waiting for unfinished jobs....
TEST check-qtest-x86_64: tests/q35-test
The full log is available at
http://patchew.org/logs/[email protected]/testing.docker-quick@centos7/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to [email protected]
Quoted 21 lines · On Fri, 20 Sep 2019, at 06:26, [email protected] wrote:
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series failed the docker-quick@centos7 build test. Please find the testing commands and
their output below. If you have Docker installed, you can probably reproduce it
locally.=== TEST SCRIPT BEGIN ===
#!/bin/bash
make docker-image-centos7 V=1 NETWORK=1
time make docker-test-quick@centos7 SHOW_ENV=1 J=14 NETWORK=1 === TEST SCRIPT END ===libudev no
default devices yes
Quoted 7 lines
warning: Python 2 support is deprecated
warning: Python 3 will be required for building future versions of QEMU cross containers noNOTE: guest cross-compilers enabled: cc
---
Property '.cntfrq' not found
Ah, we should drop the patch introducing the CNTFRQ configuration for the 2600 as well given the patch adding support to configure it isn't part of the series.
Andrew
Quoted 28 lines · Andrew Jeffery
Quoted 25 lines · On Fri, 20 Sep 2019, at 06:26, [email protected] wrote:
Patchew URL: https://patchew.org/QEMU/[email protected]/
Hi,
This series failed the docker-quick@centos7 build test. Please find the testing commands and
their output below. If you have Docker installed, you can probably reproduce it
locally.=== TEST SCRIPT BEGIN ===
#!/bin/bash
make docker-image-centos7 V=1 NETWORK=1
time make docker-test-quick@centos7 SHOW_ENV=1 J=14 NETWORK=1 === TEST SCRIPT END ===libudev no
default devices yeswarning: Python 2 support is deprecated
warning: Python 3 will be required for building future versions of QEMU cross containers no
Quoted 6 lines
Quoted 3 lines
NOTE: guest cross-compilers enabled: cc
---
Property '.cntfrq' not foundAh, we should drop the patch introducing the CNTFRQ configuration for the 2600 as well given the patch adding support to configure it isn't part of the series.
yes. I will move it behind in my branch.
Thanks,
C.