[Qemu-devel] [PATCH 0/4] Misc MIPS fulong2e improvements

[Qemu-devel] [PATCH 0/4] Misc MIPS fulong2e improvements

[BALATON Zoltan]

Hello,

These are some misc patches to get the fulong2e emulation in a better
shape. With my recent via-ide changes and the separately submitted
hw/display: Add basic ATI VGA emulation
patch and after this series the pmon_2e.bin firmware from
https://mirrors.cloud.tencent.com/loongson/pmon/
can actually run and appears to work. I could not find an image to
boot so I could not test that further but this appears to be an
improvement that worth submitting now.

Regards,
BALATON Zoltan

BALATON Zoltan (4):
  hw/pci-host/bonito.c: Add PCI mem region mapped at the correct address
  mips_fulong2e: Fix bios flash size
  mips_fulong2e: Dynamically generate SPD EEPROM data
  mips_fulong2e: Add on-board graphics chip

 hw/mips/mips_fulong2e.c | 46 +++++++++++++++++++++++-----------------------
 hw/pci-host/bonito.c    |  5 ++++-
 2 files changed, 27 insertions(+), 24 deletions(-)

-- 
2.13.7

[Qemu-devel] [PATCH 4/4] mips_fulong2e: Add on-board graphics chip

[BALATON Zoltan]

Add (partial) emulation of the on-board GPU of the machine. This
allows the PMON2000 firmware to run and should also work with Linux
console but probably not with X yet.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
---
Depends on hw/display: Add basic ATI VGA emulation

 hw/mips/mips_fulong2e.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
index eec6fd02c8..68bd030fc1 100644
--- a/hw/mips/mips_fulong2e.c
+++ b/hw/mips/mips_fulong2e.c
@@ -287,6 +287,7 @@ static void mips_fulong2e_init(MachineState *machine)
     I2CBus *smbus;
     MIPSCPU *cpu;
     CPUMIPSState *env;
+    DeviceState *dev;
 
     /* init CPUs */
     cpu = MIPS_CPU(cpu_create(machine->cpu_type));
@@ -347,6 +348,11 @@ static void mips_fulong2e_init(MachineState *machine)
     vt82c686b_southbridge_init(pci_bus, FULONG2E_VIA_SLOT, env->irq[5],
                                &smbus, &isa_bus);
 
+    /* GPU */
+    dev = DEVICE(pci_create(pci_bus, -1, "ati-vga"));
+    qdev_prop_set_uint16(dev, "device_id", 0x5159);
+    qdev_init_nofail(dev);
+
     /* Populate SPD eeprom data */
     spd_data = spd_data_generate(DDR, ram_size, &err);
     if (err) {
-- 
2.13.7

Re: [Qemu-devel] [PATCH 4/4] mips_fulong2e: Add on-board graphics chip

[Philippe Mathieu-Daudé]

On 2/11/19 5:01 AM, BALATON Zoltan wrote:
> Add (partial) emulation of the on-board GPU of the machine. This
> allows the PMON2000 firmware to run and should also work with Linux
> console but probably not with X yet.
> 
> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
> ---
> Depends on hw/display: Add basic ATI VGA emulation
> 
>  hw/mips/mips_fulong2e.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
> index eec6fd02c8..68bd030fc1 100644
> --- a/hw/mips/mips_fulong2e.c
> +++ b/hw/mips/mips_fulong2e.c
> @@ -287,6 +287,7 @@ static void mips_fulong2e_init(MachineState *machine)
>      I2CBus *smbus;
>      MIPSCPU *cpu;
>      CPUMIPSState *env;
> +    DeviceState *dev;
>  
>      /* init CPUs */
>      cpu = MIPS_CPU(cpu_create(machine->cpu_type));
> @@ -347,6 +348,11 @@ static void mips_fulong2e_init(MachineState *machine)
>      vt82c686b_southbridge_init(pci_bus, FULONG2E_VIA_SLOT, env->irq[5],
>                                 &smbus, &isa_bus);
>  
> +    /* GPU */
> +    dev = DEVICE(pci_create(pci_bus, -1, "ati-vga"));

You missed in your cover:
Based-on: 20190211040434.1C98674569F@zero.eik.bme.hu

Else testing we get:
qemu-system-mips64el: Unknown device 'ati-vga' for bus 'PCI'
Aborted (core dumped)

> +    qdev_prop_set_uint16(dev, "device_id", 0x5159);
> +    qdev_init_nofail(dev);
> +
>      /* Populate SPD eeprom data */
>      spd_data = spd_data_generate(DDR, ram_size, &err);
>      if (err) {
> 

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Re: [Qemu-devel] [PATCH 4/4] mips_fulong2e: Add on-board graphics chip

[BALATON Zoltan]

On Tue, 12 Feb 2019, Philippe Mathieu-Daudé wrote:
> On 2/11/19 5:01 AM, BALATON Zoltan wrote:
>> Add (partial) emulation of the on-board GPU of the machine. This
>> allows the PMON2000 firmware to run and should also work with Linux
>> console but probably not with X yet.
>>
>> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
>> ---
>> Depends on hw/display: Add basic ATI VGA emulation
>>
>>  hw/mips/mips_fulong2e.c | 6 ++++++
>>  1 file changed, 6 insertions(+)
>>
>> diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
>> index eec6fd02c8..68bd030fc1 100644
>> --- a/hw/mips/mips_fulong2e.c
>> +++ b/hw/mips/mips_fulong2e.c
>> @@ -287,6 +287,7 @@ static void mips_fulong2e_init(MachineState *machine)
>>      I2CBus *smbus;
>>      MIPSCPU *cpu;
>>      CPUMIPSState *env;
>> +    DeviceState *dev;
>>
>>      /* init CPUs */
>>      cpu = MIPS_CPU(cpu_create(machine->cpu_type));
>> @@ -347,6 +348,11 @@ static void mips_fulong2e_init(MachineState *machine)
>>      vt82c686b_southbridge_init(pci_bus, FULONG2E_VIA_SLOT, env->irq[5],
>>                                 &smbus, &isa_bus);
>>
>> +    /* GPU */
>> +    dev = DEVICE(pci_create(pci_bus, -1, "ati-vga"));
>
> You missed in your cover:
> Based-on: 20190211040434.1C98674569F@zero.eik.bme.hu

I forgot what this was called, thanks for reminding. But patchew seems to 
be down anyway.

Regards,
BALATON Zoltan

>
> Else testing we get:
> qemu-system-mips64el: Unknown device 'ati-vga' for bus 'PCI'
> Aborted (core dumped)
>
>> +    qdev_prop_set_uint16(dev, "device_id", 0x5159);
>> +    qdev_init_nofail(dev);
>> +
>>      /* Populate SPD eeprom data */
>>      spd_data = spd_data_generate(DDR, ram_size, &err);
>>      if (err) {
>>
>
> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
>
>

[Qemu-devel] [PATCH 2/4] mips_fulong2e: Fix bios flash size

[BALATON Zoltan]

According to both the specifications on linux-mips.org referenced in a
comment at the beginning of the file and the flash chip part number
the bios size should be 512k not 1M.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
---
 hw/mips/mips_fulong2e.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
index 02549d5c7e..10e6ed585a 100644
--- a/hw/mips/mips_fulong2e.c
+++ b/hw/mips/mips_fulong2e.c
@@ -21,6 +21,7 @@
 #include "qemu/osdep.h"
 #include "qemu/units.h"
 #include "qapi/error.h"
+#include "cpu.h"
 #include "hw/hw.h"
 #include "hw/i386/pc.h"
 #include "hw/dma/i8257.h"
@@ -35,7 +36,6 @@
 #include "audio/audio.h"
 #include "qemu/log.h"
 #include "hw/loader.h"
-#include "hw/mips/bios.h"
 #include "hw/ide.h"
 #include "elf.h"
 #include "hw/isa/vt82c686.h"
@@ -51,6 +51,8 @@
 #define ENVP_NB_ENTRIES	 	16
 #define ENVP_ENTRY_SIZE	 	256
 
+/* fulong 2e has a 512k flash: Winbond W39L040AP70Z */
+#define BIOS_SIZE (512 * KiB)
 #define MAX_IDE_BUS 2
 
 /*
@@ -307,12 +309,9 @@ static void mips_fulong2e_init(MachineState *machine)
     /* fulong 2e has 256M ram. */
     ram_size = 256 * MiB;
 
-    /* fulong 2e has a 1M flash.Winbond W39L040AP70Z */
-    bios_size = 1 * MiB;
-
     /* allocate RAM */
     memory_region_allocate_system_memory(ram, NULL, "fulong2e.ram", ram_size);
-    memory_region_init_ram(bios, NULL, "fulong2e.bios", bios_size,
+    memory_region_init_ram(bios, NULL, "fulong2e.bios", BIOS_SIZE,
                            &error_fatal);
     memory_region_set_readonly(bios, true);
 
-- 
2.13.7

Re: [Qemu-devel] [PATCH 2/4] mips_fulong2e: Fix bios flash size

[Philippe Mathieu-Daudé]

On 2/11/19 5:01 AM, BALATON Zoltan wrote:
> According to both the specifications on linux-mips.org referenced in a

This link: https://www.linux-mips.org/wiki/Fulong#Front:

> comment at the beginning of the file and the flash chip part number
> the bios size should be 512k not 1M.
> 
> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>

> ---
>  hw/mips/mips_fulong2e.c | 9 ++++-----
>  1 file changed, 4 insertions(+), 5 deletions(-)
> 
> diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
> index 02549d5c7e..10e6ed585a 100644
> --- a/hw/mips/mips_fulong2e.c
> +++ b/hw/mips/mips_fulong2e.c
> @@ -21,6 +21,7 @@
>  #include "qemu/osdep.h"
>  #include "qemu/units.h"
>  #include "qapi/error.h"
> +#include "cpu.h"
>  #include "hw/hw.h"
>  #include "hw/i386/pc.h"
>  #include "hw/dma/i8257.h"
> @@ -35,7 +36,6 @@
>  #include "audio/audio.h"
>  #include "qemu/log.h"
>  #include "hw/loader.h"
> -#include "hw/mips/bios.h"
>  #include "hw/ide.h"
>  #include "elf.h"
>  #include "hw/isa/vt82c686.h"
> @@ -51,6 +51,8 @@
>  #define ENVP_NB_ENTRIES	 	16
>  #define ENVP_ENTRY_SIZE	 	256
>  
> +/* fulong 2e has a 512k flash: Winbond W39L040AP70Z */
> +#define BIOS_SIZE (512 * KiB)
>  #define MAX_IDE_BUS 2
>  
>  /*
> @@ -307,12 +309,9 @@ static void mips_fulong2e_init(MachineState *machine)
>      /* fulong 2e has 256M ram. */
>      ram_size = 256 * MiB;
>  
> -    /* fulong 2e has a 1M flash.Winbond W39L040AP70Z */
> -    bios_size = 1 * MiB;
> -
>      /* allocate RAM */
>      memory_region_allocate_system_memory(ram, NULL, "fulong2e.ram", ram_size);
> -    memory_region_init_ram(bios, NULL, "fulong2e.bios", bios_size,
> +    memory_region_init_ram(bios, NULL, "fulong2e.bios", BIOS_SIZE,
>                             &error_fatal);
>      memory_region_set_readonly(bios, true);
>  
> 

[Qemu-devel] [PATCH 3/4] mips_fulong2e: Dynamically generate SPD EEPROM data

[BALATON Zoltan]

The machine comes with 256M memory module by default but it's
upgradable so it could have different memory size. There was a TODO
comment to replace static SPD EEPROM data with dynamically generated
one to support this. Now that we have a function for that, it's easy
to do. Although this would allow larger RAM sizes, the peculiar memory
map of the machine may need some special handling to map it as low and
high memory. Because I don't know what the correct place would be for
highmem, I've left memory size fixed at 256M for now and TODO is moved
there instead.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
---
 hw/mips/mips_fulong2e.c | 31 +++++++++++++------------------
 1 file changed, 13 insertions(+), 18 deletions(-)

diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
index 10e6ed585a..eec6fd02c8 100644
--- a/hw/mips/mips_fulong2e.c
+++ b/hw/mips/mips_fulong2e.c
@@ -214,20 +214,6 @@ static void main_cpu_reset(void *opaque)
     }
 }
 
-static const uint8_t eeprom_spd[0x80] = {
-    0x80,0x08,0x07,0x0d,0x09,0x02,0x40,0x00,0x04,0x70,
-    0x70,0x00,0x82,0x10,0x00,0x01,0x0e,0x04,0x0c,0x01,
-    0x02,0x20,0x80,0x75,0x70,0x00,0x00,0x50,0x3c,0x50,
-    0x2d,0x20,0xb0,0xb0,0x50,0x50,0x00,0x00,0x00,0x00,
-    0x00,0x41,0x48,0x3c,0x32,0x75,0x00,0x00,0x00,0x00,
-    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
-    0x00,0x00,0x00,0x9c,0x7b,0x07,0x00,0x00,0x00,0x00,
-    0x00,0x00,0x00,0x00,0x48,0x42,0x35,0x34,0x41,0x32,
-    0x35,0x36,0x38,0x4b,0x4e,0x2d,0x41,0x37,0x35,0x42,
-    0x20,0x30,0x20
-};
-
 static void vt82c686b_southbridge_init(PCIBus *pci_bus, int slot, qemu_irq intc,
                                        I2CBus **i2c_bus, ISABus **p_isa_bus)
 {
@@ -284,7 +270,6 @@ static void network_init (PCIBus *pci_bus)
 
 static void mips_fulong2e_init(MachineState *machine)
 {
-    ram_addr_t ram_size = machine->ram_size;
     const char *kernel_filename = machine->kernel_filename;
     const char *kernel_cmdline = machine->kernel_cmdline;
     const char *initrd_filename = machine->initrd_filename;
@@ -292,7 +277,10 @@ static void mips_fulong2e_init(MachineState *machine)
     MemoryRegion *address_space_mem = get_system_memory();
     MemoryRegion *ram = g_new(MemoryRegion, 1);
     MemoryRegion *bios = g_new(MemoryRegion, 1);
+    ram_addr_t ram_size = machine->ram_size;
     long bios_size;
+    uint8_t *spd_data;
+    Error *err = NULL;
     int64_t kernel_entry;
     PCIBus *pci_bus;
     ISABus *isa_bus;
@@ -306,7 +294,7 @@ static void mips_fulong2e_init(MachineState *machine)
 
     qemu_register_reset(main_cpu_reset, cpu);
 
-    /* fulong 2e has 256M ram. */
+    /* TODO: support more than 256M RAM as highmem */
     ram_size = 256 * MiB;
 
     /* allocate RAM */
@@ -359,8 +347,14 @@ static void mips_fulong2e_init(MachineState *machine)
     vt82c686b_southbridge_init(pci_bus, FULONG2E_VIA_SLOT, env->irq[5],
                                &smbus, &isa_bus);
 
-    /* TODO: Populate SPD eeprom data.  */
-    smbus_eeprom_init(smbus, 1, eeprom_spd, sizeof(eeprom_spd));
+    /* Populate SPD eeprom data */
+    spd_data = spd_data_generate(DDR, ram_size, &err);
+    if (err) {
+        warn_report_err(err);
+    }
+    if (spd_data) {
+        smbus_eeprom_init_one(smbus, 0x50, spd_data);
+    }
 
     mc146818_rtc_init(isa_bus, 2000, NULL);
 
@@ -374,6 +368,7 @@ static void mips_fulong2e_machine_init(MachineClass *mc)
     mc->init = mips_fulong2e_init;
     mc->block_default_type = IF_IDE;
     mc->default_cpu_type = MIPS_CPU_TYPE_NAME("Loongson-2E");
+    mc->default_ram_size = 256 * MiB;
 }
 
 DEFINE_MACHINE("fulong2e", mips_fulong2e_machine_init)
-- 
2.13.7

Re: [Qemu-devel] [PATCH 3/4] mips_fulong2e: Dynamically generate SPD EEPROM data

[Philippe Mathieu-Daudé]

On 2/11/19 5:01 AM, BALATON Zoltan wrote:
> The machine comes with 256M memory module by default but it's
> upgradable so it could have different memory size. There was a TODO
> comment to replace static SPD EEPROM data with dynamically generated
> one to support this. Now that we have a function for that, it's easy
> to do. Although this would allow larger RAM sizes, the peculiar memory
> map of the machine may need some special handling to map it as low and
> high memory. Because I don't know what the correct place would be for
> highmem, I've left memory size fixed at 256M for now and TODO is moved
> there instead.
> 
> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>

> ---
>  hw/mips/mips_fulong2e.c | 31 +++++++++++++------------------
>  1 file changed, 13 insertions(+), 18 deletions(-)
> 
> diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
> index 10e6ed585a..eec6fd02c8 100644
> --- a/hw/mips/mips_fulong2e.c
> +++ b/hw/mips/mips_fulong2e.c
> @@ -214,20 +214,6 @@ static void main_cpu_reset(void *opaque)
>      }
>  }
>  
> -static const uint8_t eeprom_spd[0x80] = {
> -    0x80,0x08,0x07,0x0d,0x09,0x02,0x40,0x00,0x04,0x70,
> -    0x70,0x00,0x82,0x10,0x00,0x01,0x0e,0x04,0x0c,0x01,
> -    0x02,0x20,0x80,0x75,0x70,0x00,0x00,0x50,0x3c,0x50,
> -    0x2d,0x20,0xb0,0xb0,0x50,0x50,0x00,0x00,0x00,0x00,
> -    0x00,0x41,0x48,0x3c,0x32,0x75,0x00,0x00,0x00,0x00,
> -    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
> -    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
> -    0x00,0x00,0x00,0x9c,0x7b,0x07,0x00,0x00,0x00,0x00,
> -    0x00,0x00,0x00,0x00,0x48,0x42,0x35,0x34,0x41,0x32,
> -    0x35,0x36,0x38,0x4b,0x4e,0x2d,0x41,0x37,0x35,0x42,
> -    0x20,0x30,0x20
> -};
> -
>  static void vt82c686b_southbridge_init(PCIBus *pci_bus, int slot, qemu_irq intc,
>                                         I2CBus **i2c_bus, ISABus **p_isa_bus)
>  {
> @@ -284,7 +270,6 @@ static void network_init (PCIBus *pci_bus)
>  
>  static void mips_fulong2e_init(MachineState *machine)
>  {
> -    ram_addr_t ram_size = machine->ram_size;
>      const char *kernel_filename = machine->kernel_filename;
>      const char *kernel_cmdline = machine->kernel_cmdline;
>      const char *initrd_filename = machine->initrd_filename;
> @@ -292,7 +277,10 @@ static void mips_fulong2e_init(MachineState *machine)
>      MemoryRegion *address_space_mem = get_system_memory();
>      MemoryRegion *ram = g_new(MemoryRegion, 1);
>      MemoryRegion *bios = g_new(MemoryRegion, 1);
> +    ram_addr_t ram_size = machine->ram_size;
>      long bios_size;
> +    uint8_t *spd_data;
> +    Error *err = NULL;
>      int64_t kernel_entry;
>      PCIBus *pci_bus;
>      ISABus *isa_bus;
> @@ -306,7 +294,7 @@ static void mips_fulong2e_init(MachineState *machine)
>  
>      qemu_register_reset(main_cpu_reset, cpu);
>  
> -    /* fulong 2e has 256M ram. */
> +    /* TODO: support more than 256M RAM as highmem */
>      ram_size = 256 * MiB;
>  
>      /* allocate RAM */
> @@ -359,8 +347,14 @@ static void mips_fulong2e_init(MachineState *machine)
>      vt82c686b_southbridge_init(pci_bus, FULONG2E_VIA_SLOT, env->irq[5],
>                                 &smbus, &isa_bus);
>  
> -    /* TODO: Populate SPD eeprom data.  */
> -    smbus_eeprom_init(smbus, 1, eeprom_spd, sizeof(eeprom_spd));
> +    /* Populate SPD eeprom data */
> +    spd_data = spd_data_generate(DDR, ram_size, &err);
> +    if (err) {
> +        warn_report_err(err);
> +    }
> +    if (spd_data) {
> +        smbus_eeprom_init_one(smbus, 0x50, spd_data);
> +    }
>  
>      mc146818_rtc_init(isa_bus, 2000, NULL);
>  
> @@ -374,6 +368,7 @@ static void mips_fulong2e_machine_init(MachineClass *mc)
>      mc->init = mips_fulong2e_init;
>      mc->block_default_type = IF_IDE;
>      mc->default_cpu_type = MIPS_CPU_TYPE_NAME("Loongson-2E");
> +    mc->default_ram_size = 256 * MiB;
>  }
>  
>  DEFINE_MACHINE("fulong2e", mips_fulong2e_machine_init)
> 

Re: [Qemu-devel] [PATCH 3/4] mips_fulong2e: Dynamically generate SPD EEPROM data

[Aleksandar Markovic]

> From: BALATON Zoltan <balaton@eik.bme.hu>
> Sent: Monday, February 11, 2019 5:01 AM
> To: qemu-devel@nongnu.org
> Cc: Aurelien Jarno; Aleksandar Markovic; Huacai Chen
> Subject: [PATCH 3/4] mips_fulong2e: Dynamically generate SPD EEPROM data
> 
> The machine comes with 256M memory module by default but it's
> upgradable so it could have different memory size. There was a TODO
> comment to replace static SPD EEPROM data with dynamically generated
> one to support this. Now that we have a function for that, it's easy
> to do. Although this would allow larger RAM sizes, the peculiar memory
> map of the machine may need some special handling to map it as low and
> high memory. Because I don't know what the correct place would be for
> highmem, I've left memory size fixed at 256M for now and TODO is moved
> there instead.
> 
> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
> ---
>  hw/mips/mips_fulong2e.c | 31 +++++++++++++------------------
>  1 file changed, 13 insertions(+), 18 deletions(-)
> 

Hello, Zoltan.

Thank you for your work in this area. I genarally support this series.
(When can we expect v2? I would like to integrate this series before
4.0 soft freeze planned for March 12th.)

However, I have just a couple of questions:

1. Is this series dependent on the patches outside of this series?
(meaning, the functionality won't work unless those other patches
are in the tree?)

2. Can you spell out the test procedure that you used, after this
series (and possibly other needed patches) is applied?

3. What exactly this series fixes, or improves?

Thanks again!

Aleksandar

> diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
> index 10e6ed585a..eec6fd02c8 100644
> --- a/hw/mips/mips_fulong2e.c
> +++ b/hw/mips/mips_fulong2e.c
> @@ -214,20 +214,6 @@ static void main_cpu_reset(void *opaque)
>      }
>  }
> 
> -static const uint8_t eeprom_spd[0x80] = {
> -    0x80,0x08,0x07,0x0d,0x09,0x02,0x40,0x00,0x04,0x70,
> -    0x70,0x00,0x82,0x10,0x00,0x01,0x0e,0x04,0x0c,0x01,
> -    0x02,0x20,0x80,0x75,0x70,0x00,0x00,0x50,0x3c,0x50,
> -    0x2d,0x20,0xb0,0xb0,0x50,0x50,0x00,0x00,0x00,0x00,
> -    0x00,0x41,0x48,0x3c,0x32,0x75,0x00,0x00,0x00,0x00,
> -    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
> -    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
> -    0x00,0x00,0x00,0x9c,0x7b,0x07,0x00,0x00,0x00,0x00,
> -    0x00,0x00,0x00,0x00,0x48,0x42,0x35,0x34,0x41,0x32,
> -    0x35,0x36,0x38,0x4b,0x4e,0x2d,0x41,0x37,0x35,0x42,
> -    0x20,0x30,0x20
> -};
> -
>  static void vt82c686b_southbridge_init(PCIBus *pci_bus, int slot, qemu_irq intc,
>                                         I2CBus **i2c_bus, ISABus **p_isa_bus)
>  {
> @@ -284,7 +270,6 @@ static void network_init (PCIBus *pci_bus)
> 
>  static void mips_fulong2e_init(MachineState *machine)
>  {
> -    ram_addr_t ram_size = machine->ram_size;
>      const char *kernel_filename = machine->kernel_filename;
>      const char *kernel_cmdline = machine->kernel_cmdline;
>      const char *initrd_filename = machine->initrd_filename;
> @@ -292,7 +277,10 @@ static void mips_fulong2e_init(MachineState *machine)
>      MemoryRegion *address_space_mem = get_system_memory();
>      MemoryRegion *ram = g_new(MemoryRegion, 1);
>      MemoryRegion *bios = g_new(MemoryRegion, 1);
> +    ram_addr_t ram_size = machine->ram_size;
>      long bios_size;
> +    uint8_t *spd_data;
> +    Error *err = NULL;
>      int64_t kernel_entry;
>      PCIBus *pci_bus;
>      ISABus *isa_bus;
> @@ -306,7 +294,7 @@ static void mips_fulong2e_init(MachineState *machine)
> 
>      qemu_register_reset(main_cpu_reset, cpu);
> 
> -    /* fulong 2e has 256M ram. */
> +    /* TODO: support more than 256M RAM as highmem */
>      ram_size = 256 * MiB;
> 
>      /* allocate RAM */
> @@ -359,8 +347,14 @@ static void mips_fulong2e_init(MachineState *machine)
>      vt82c686b_southbridge_init(pci_bus, FULONG2E_VIA_SLOT, env->irq[5],
>                                 &smbus, &isa_bus);
> 
> -    /* TODO: Populate SPD eeprom data.  */
> -    smbus_eeprom_init(smbus, 1, eeprom_spd, sizeof(eeprom_spd));
> +    /* Populate SPD eeprom data */
> +    spd_data = spd_data_generate(DDR, ram_size, &err);
> +    if (err) {
> +        warn_report_err(err);
> +    }
> +    if (spd_data) {
> +        smbus_eeprom_init_one(smbus, 0x50, spd_data);
> +    }
> 
>      mc146818_rtc_init(isa_bus, 2000, NULL);
> 
> @@ -374,6 +368,7 @@ static void mips_fulong2e_machine_init(MachineClass *mc)
>      mc->init = mips_fulong2e_init;
>      mc->block_default_type = IF_IDE;
>      mc->default_cpu_type = MIPS_CPU_TYPE_NAME("Loongson-2E");
> +    mc->default_ram_size = 256 * MiB;
>  }
> 
>  DEFINE_MACHINE("fulong2e", mips_fulong2e_machine_init)
> --
> 2.13.7
> 

Re: [Qemu-devel] [PATCH 3/4] mips_fulong2e: Dynamically generate SPD EEPROM data

[BALATON Zoltan]

Hello,

On Thu, 14 Feb 2019, Aleksandar Markovic wrote:
>> The machine comes with 256M memory module by default but it's
>> upgradable so it could have different memory size. There was a TODO
>> comment to replace static SPD EEPROM data with dynamically generated
>> one to support this. Now that we have a function for that, it's easy
>> to do. Although this would allow larger RAM sizes, the peculiar memory
>> map of the machine may need some special handling to map it as low and
>> high memory. Because I don't know what the correct place would be for
>> highmem, I've left memory size fixed at 256M for now and TODO is moved
>> there instead.
>>
>> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
>> ---
>>  hw/mips/mips_fulong2e.c | 31 +++++++++++++------------------
>>  1 file changed, 13 insertions(+), 18 deletions(-)
>>
>
> Hello, Zoltan.
>
> Thank you for your work in this area. I genarally support this series.
> (When can we expect v2? I would like to integrate this series before
> 4.0 soft freeze planned for March 12th.)

I plan to submit v2 definitely before freeze (maybe this week end or next 
week) I was just waiting for review comments and to see what happens to 
the ati-vga patch this depends on.

> However, I have just a couple of questions:
>
> 1. Is this series dependent on the patches outside of this series?
> (meaning, the functionality won't work unless those other patches
> are in the tree?)

To quote my original cover letter:

"my recent via-ide changes and the separately submitted
hw/display: Add basic ATI VGA emulation
patch and after this series the pmon_2e.bin firmware from
https://mirrors.cloud.tencent.com/loongson/pmon/
can actually run and appears to work. I could not find an image to
boot so I could not test that further but this appears to be an
improvement that worth submitting now."

Maybe this wasn't clear but only the last patch in this series 
(mips_fulong2e: Add on-board graphics chip) depends on the separate 
ati-vga patch mentioned above, the rest are not dependent on anything 
else. My via changes are already merged through the ide tree so those are 
already in master.

> 2. Can you spell out the test procedure that you used, after this
> series (and possibly other needed patches) is applied?

I've tried:

qemu-system-mips64el -M fulong2e -bios pmon_2e.bin

with the pmon ROM image from the above link. I've also tried booting some 
Linux kernel but I could not find a suitable iso so if anyone is aware of 
a boot CD or image that should work on real hardware I could try that. I 
think Debian 8.11.0 should boot but needs external kernels which I don't 
have. Maybe Philippe can add to this what he tested.

> 3. What exactly this series fixes, or improves?

The series improves the fulong2e machine model to work with its original 
firmware. Previously it needed some modified firmware which was not 
available so this actually makes this board a bit more useful.

>
> Thanks again!
>
> Aleksandar
>
>> diff --git a/hw/mips/mips_fulong2e.c b/hw/mips/mips_fulong2e.c
>> index 10e6ed585a..eec6fd02c8 100644
>> --- a/hw/mips/mips_fulong2e.c
>> +++ b/hw/mips/mips_fulong2e.c
>> @@ -214,20 +214,6 @@ static void main_cpu_reset(void *opaque)
>>      }
>>  }
>>
>> -static const uint8_t eeprom_spd[0x80] = {
>> -    0x80,0x08,0x07,0x0d,0x09,0x02,0x40,0x00,0x04,0x70,
>> -    0x70,0x00,0x82,0x10,0x00,0x01,0x0e,0x04,0x0c,0x01,
>> -    0x02,0x20,0x80,0x75,0x70,0x00,0x00,0x50,0x3c,0x50,
>> -    0x2d,0x20,0xb0,0xb0,0x50,0x50,0x00,0x00,0x00,0x00,
>> -    0x00,0x41,0x48,0x3c,0x32,0x75,0x00,0x00,0x00,0x00,
>> -    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
>> -    0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,
>> -    0x00,0x00,0x00,0x9c,0x7b,0x07,0x00,0x00,0x00,0x00,
>> -    0x00,0x00,0x00,0x00,0x48,0x42,0x35,0x34,0x41,0x32,
>> -    0x35,0x36,0x38,0x4b,0x4e,0x2d,0x41,0x37,0x35,0x42,
>> -    0x20,0x30,0x20
>> -};
>> -
>>  static void vt82c686b_southbridge_init(PCIBus *pci_bus, int slot, qemu_irq intc,
>>                                         I2CBus **i2c_bus, ISABus **p_isa_bus)
>>  {
>> @@ -284,7 +270,6 @@ static void network_init (PCIBus *pci_bus)
>>
>>  static void mips_fulong2e_init(MachineState *machine)
>>  {
>> -    ram_addr_t ram_size = machine->ram_size;
>>      const char *kernel_filename = machine->kernel_filename;
>>      const char *kernel_cmdline = machine->kernel_cmdline;
>>      const char *initrd_filename = machine->initrd_filename;
>> @@ -292,7 +277,10 @@ static void mips_fulong2e_init(MachineState *machine)
>>      MemoryRegion *address_space_mem = get_system_memory();
>>      MemoryRegion *ram = g_new(MemoryRegion, 1);
>>      MemoryRegion *bios = g_new(MemoryRegion, 1);
>> +    ram_addr_t ram_size = machine->ram_size;
>>      long bios_size;
>> +    uint8_t *spd_data;
>> +    Error *err = NULL;
>>      int64_t kernel_entry;
>>      PCIBus *pci_bus;
>>      ISABus *isa_bus;
>> @@ -306,7 +294,7 @@ static void mips_fulong2e_init(MachineState *machine)
>>
>>      qemu_register_reset(main_cpu_reset, cpu);
>>
>> -    /* fulong 2e has 256M ram. */
>> +    /* TODO: support more than 256M RAM as highmem */
>>      ram_size = 256 * MiB;

This particular patch about SPD EEPROM just removes the static EEPROM data 
in favour of the spd_data_generate() function I've recently added which 
also allows to use different RAM size than hard coded 256 that was 
previously also limited by the static SPD data. RAM size is still fixed to 
256M here because I don't know where to map the highmem beyond the first 
256M on MIPS in general and on this board in particular but PMON now 
recognises bigger RAM sizes as highmem when I remove this fixed setting 
after this patch. I've added a TODO comment here for this, when this is 
resolved these lines could be removed as well but I don't intend to do 
that as I don't know how it should work.

Similar cleanup could be done for the malta board which I could try to do 
but I couldn't find its YAMON firmware to test with (a comment in the code 
mentions that it's the way it is due to a bug in that firmware so that 
should be tested to make sure it works, that's why I haven't touched 
that.)

Regards,
BALATON Zoltan

>>      /* allocate RAM */
>> @@ -359,8 +347,14 @@ static void mips_fulong2e_init(MachineState *machine)
>>      vt82c686b_southbridge_init(pci_bus, FULONG2E_VIA_SLOT, env->irq[5],
>>                                 &smbus, &isa_bus);
>>
>> -    /* TODO: Populate SPD eeprom data.  */
>> -    smbus_eeprom_init(smbus, 1, eeprom_spd, sizeof(eeprom_spd));
>> +    /* Populate SPD eeprom data */
>> +    spd_data = spd_data_generate(DDR, ram_size, &err);
>> +    if (err) {
>> +        warn_report_err(err);
>> +    }
>> +    if (spd_data) {
>> +        smbus_eeprom_init_one(smbus, 0x50, spd_data);
>> +    }
>>
>>      mc146818_rtc_init(isa_bus, 2000, NULL);
>>
>> @@ -374,6 +368,7 @@ static void mips_fulong2e_machine_init(MachineClass *mc)
>>      mc->init = mips_fulong2e_init;
>>      mc->block_default_type = IF_IDE;
>>      mc->default_cpu_type = MIPS_CPU_TYPE_NAME("Loongson-2E");
>> +    mc->default_ram_size = 256 * MiB;
>>  }
>>
>>  DEFINE_MACHINE("fulong2e", mips_fulong2e_machine_init)
>> --
>> 2.13.7
>>
>
>

[Qemu-devel] [PATCH 1/4] hw/pci-host/bonito.c: Add PCI mem region mapped at the correct address

[BALATON Zoltan]

Stop using system memory as PCI memory otherwise devices such as VGA
that have regions mapped to PCI memory clash with RAM. Use a separate
memory region for PCI memory and map it to the correct address in
system memory which allows PCI mem regions to show at the correct
address where clients expect them.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
---
 hw/pci-host/bonito.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/hw/pci-host/bonito.c b/hw/pci-host/bonito.c
index 9f33582706..c940ec6e48 100644
--- a/hw/pci-host/bonito.c
+++ b/hw/pci-host/bonito.c
@@ -598,11 +598,14 @@ static const VMStateDescription vmstate_bonito = {
 static void bonito_pcihost_realize(DeviceState *dev, Error **errp)
 {
     PCIHostState *phb = PCI_HOST_BRIDGE(dev);
+    MemoryRegion *mr = g_new0(MemoryRegion, 1);
 
+    memory_region_init(mr, OBJECT(dev), "pci.mem", BONITO_PCILO_SIZE);
     phb->bus = pci_register_root_bus(DEVICE(dev), "pci",
                                      pci_bonito_set_irq, pci_bonito_map_irq,
-                                     dev, get_system_memory(), get_system_io(),
+                                     dev, mr, get_system_io(),
                                      0x28, 32, TYPE_PCI_BUS);
+    memory_region_add_subregion(get_system_memory(), BONITO_PCILO_BASE, mr);
 }
 
 static void bonito_realize(PCIDevice *dev, Error **errp)
-- 
2.13.7

Re: [Qemu-devel] [PATCH 1/4] hw/pci-host/bonito.c: Add PCI mem region mapped at the correct address

[Philippe Mathieu-Daudé]

On 2/11/19 5:01 AM, BALATON Zoltan wrote:
> Stop using system memory as PCI memory otherwise devices such as VGA
> that have regions mapped to PCI memory clash with RAM. Use a separate
> memory region for PCI memory and map it to the correct address in
> system memory which allows PCI mem regions to show at the correct
> address where clients expect them.
> 
> Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
> ---
>  hw/pci-host/bonito.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/hw/pci-host/bonito.c b/hw/pci-host/bonito.c
> index 9f33582706..c940ec6e48 100644
> --- a/hw/pci-host/bonito.c
> +++ b/hw/pci-host/bonito.c
> @@ -598,11 +598,14 @@ static const VMStateDescription vmstate_bonito = {
>  static void bonito_pcihost_realize(DeviceState *dev, Error **errp)
>  {
>      PCIHostState *phb = PCI_HOST_BRIDGE(dev);
> +    MemoryRegion *mr = g_new0(MemoryRegion, 1);

No need to alloc, move that to BonitoState, see [*].

>  
> +    memory_region_init(mr, OBJECT(dev), "pci.mem", BONITO_PCILO_SIZE);
>      phb->bus = pci_register_root_bus(DEVICE(dev), "pci",
>                                       pci_bonito_set_irq, pci_bonito_map_irq,
> -                                     dev, get_system_memory(), get_system_io(),
> +                                     dev, mr, get_system_io(),
>                                       0x28, 32, TYPE_PCI_BUS);
> +    memory_region_add_subregion(get_system_memory(), BONITO_PCILO_BASE, mr);
>  }
>  
>  static void bonito_realize(PCIDevice *dev, Error **errp)
> 

[*]:
-- >8 --
@@ -217,6 +217,7 @@ struct BonitoState {
     PCIHostState parent_obj;
     qemu_irq *pic;
     PCIBonitoState *pci_dev;
+    MemoryRegion pci_mmio;
 };

 #define TYPE_BONITO_PCI_HOST_BRIDGE "Bonito-pcihost"
@@ -598,14 +599,15 @@ static const VMStateDescription vmstate_bonito = {
 static void bonito_pcihost_realize(DeviceState *dev, Error **errp)
 {
     PCIHostState *phb = PCI_HOST_BRIDGE(dev);
-    MemoryRegion *mr = g_new0(MemoryRegion, 1);
+    BonitoState *s = BONITO_PCI_HOST_BRIDGE(dev);

-    memory_region_init(mr, OBJECT(dev), "pci.mem", BONITO_PCILO_SIZE);
+    memory_region_init(&s->pci_mmio, OBJECT(dev), "pci.mem",
BONITO_PCILO_SIZE);
     phb->bus = pci_register_root_bus(DEVICE(dev), "pci",
                                      pci_bonito_set_irq,
pci_bonito_map_irq,
-                                     dev, mr, get_system_io(),
+                                     dev, &s->pci_mmio, get_system_io(),
                                      0x28, 32, TYPE_PCI_BUS);
-    memory_region_add_subregion(get_system_memory(), BONITO_PCILO_BASE,
mr);
+    memory_region_add_subregion(get_system_memory(),
+                                BONITO_PCILO_BASE, &s->pci_mmio);
 }
---

Using snippet:
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>