[Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[gerg]

Some small issues are causing problems with running modern versions of
Linux on the m68k/ColdFire 5208 target. These 3 patches fix those problems.

They are all due to use of more advanced architecture features not used
in older Linux kernels.

Regards
Greg


v2:
* rebased to current master head
* constant change from Peter Crosthwaite

 hw/m68k/mcf_intc.c      |   14 ++++++++++++++
 target-m68k/op_helper.c |    7 +++----
 target-m68k/translate.c |    8 ++++----
 3 files changed, 21 insertions(+), 8 deletions(-)

[Qemu-devel] [PATCH v2 1/3] m68k: implement more ColdFire 5208 interrupt controller functionality

[gerg]

From: Greg Ungerer <gerg@uclinux.org>

Implement the SIMR and CIMR registers of the 5208 interrupt controller.
These are used by modern versions of Linux running on ColdFire (not sure
of the exact version they were introduced, but they have been in for quite
a while now).

Without this change when attempting to run a linux-3.5 kernel you will
see:

  qemu: hardware error: mcf_intc_write: Bad write offset 28

and execution will stop and dump out.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
Reviewed-by: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
---
 hw/m68k/mcf_intc.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/hw/m68k/mcf_intc.c b/hw/m68k/mcf_intc.c
index 621423c..f13c7f3 100644
--- a/hw/m68k/mcf_intc.c
+++ b/hw/m68k/mcf_intc.c
@@ -102,6 +102,20 @@ static void mcf_intc_write(void *opaque, hwaddr addr,
     case 0x0c:
         s->imr = (s->imr & 0xffffffff00000000ull) | (uint32_t)val;
         break;
+    case 0x1c:
+        if (val & 0x40) {
+            s->imr = ~0ull;
+        } else {
+            s->imr |= (0x1ull << (val & 0x3f));
+        }
+        break;
+    case 0x1d:
+        if (val & 0x40) {
+            s->imr = 0ull;
+        } else {
+            s->imr &= ~(0x1ull << (val & 0x3f));
+        }
+        break;
     default:
         hw_error("mcf_intc_write: Bad write offset %d\n", offset);
         break;
-- 
1.9.1

[Qemu-devel] [PATCH v2 2/3] m68k: implement move to/from usp register instruction

[gerg]

From: Greg Ungerer <gerg@uclinux.org>

Fill out the code support for the move to/from usp instructions. They are
being decoded, but there is no code to support there actions. So add it.

Current versions of Linux running on the ColdFire 5208 use these instructions.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
---
 target-m68k/translate.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/target-m68k/translate.c b/target-m68k/translate.c
index 4959b97..96d75bf 100644
--- a/target-m68k/translate.c
+++ b/target-m68k/translate.c
@@ -1995,8 +1995,8 @@ DISAS_INSN(move_from_usp)
         gen_exception(s, s->pc - 2, EXCP_PRIVILEGE);
         return;
     }
-    /* TODO: Implement USP.  */
-    gen_exception(s, s->pc - 2, EXCP_ILLEGAL);
+    tcg_gen_ld_i32(AREG(insn, 0), cpu_env,
+                   offsetof(CPUM68KState, sp[M68K_USP]));
 }
 
 DISAS_INSN(move_to_usp)
@@ -2005,8 +2005,8 @@ DISAS_INSN(move_to_usp)
         gen_exception(s, s->pc - 2, EXCP_PRIVILEGE);
         return;
     }
-    /* TODO: Implement USP.  */
-    gen_exception(s, s->pc - 2, EXCP_ILLEGAL);
+    tcg_gen_st_i32(AREG(insn, 0), cpu_env,
+                   offsetof(CPUM68KState, sp[M68K_USP]));
 }
 
 DISAS_INSN(halt)
-- 
1.9.1

Re: [Qemu-devel] [PATCH v2 2/3] m68k: implement move to/from usp register instruction

[Laurent Vivier]

Le 19/06/2015 15:43, gerg@uclinux.org a écrit :
> From: Greg Ungerer <gerg@uclinux.org>
> 
> Fill out the code support for the move to/from usp instructions. They are
> being decoded, but there is no code to support there actions. So add it.
> 
> Current versions of Linux running on the ColdFire 5208 use these instructions.
> 
> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
> Reviewed-by: Richard Henderson <rth@twiddle.net>
> ---
>  target-m68k/translate.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)


Reviewed-by: Laurent Vivier <laurent@vivier.eu>

[Qemu-devel] [PATCH v2 3/3] m68k: fix usp processing on interrupt entry and exception exit

[gerg]

From: Greg Ungerer <gerg@uclinux.org>

The action to potentially switch sp register is not occurring at the correct
point in the interrupt entry or exception exit sequences.

For the interrupt entry case the sp on entry is used to create the stack
exception frame - but this may well be the user stack pointer, since we
haven't done the switch yet. Re-order the flow to switch the sp regs then
use the current sp to create the exception frame.

For the return from exception case the code is unwinding the sp after
switching sp registers. But it should always unwind the supervisor sp
first, then carry out any required sp switch.

Note that these problems don't effect operation unless the user sp bit is
set in the CACR register. Only a single sp is used in the default power up
state. Previously Linux only used this single sp mode. But modern versions
of Linux use the user sp mode now, so we need correct behavior for Linux
to work.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
Reviewed-by: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
---
 target-m68k/op_helper.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/target-m68k/op_helper.c b/target-m68k/op_helper.c
index 06661f5..3a0d16f 100644
--- a/target-m68k/op_helper.c
+++ b/target-m68k/op_helper.c
@@ -63,8 +63,8 @@ static void do_rte(CPUM68KState *env)
     env->pc = cpu_ldl_kernel(env, sp + 4);
     sp |= (fmt >> 28) & 3;
     env->sr = fmt & 0xffff;
-    m68k_switch_sp(env);
     env->aregs[7] = sp + 8;
+    m68k_switch_sp(env);
 }
 
 static void do_interrupt_all(CPUM68KState *env, int is_hw)
@@ -108,10 +108,7 @@ static void do_interrupt_all(CPUM68KState *env, int is_hw)
 
     vector = cs->exception_index << 2;
 
-    sp = env->aregs[7];
-
     fmt |= 0x40000000;
-    fmt |= (sp & 3) << 28;
     fmt |= vector << 16;
     fmt |= env->sr;
 
@@ -121,6 +118,8 @@ static void do_interrupt_all(CPUM68KState *env, int is_hw)
         env->sr &= ~SR_M;
     }
     m68k_switch_sp(env);
+    sp = env->aregs[7];
+    fmt |= (sp & 3) << 28;
 
     /* ??? This could cause MMU faults.  */
     sp &= ~3;
-- 
1.9.1

Re: [Qemu-devel] [PATCH v2 3/3] m68k: fix usp processing on interrupt entry and exception exit

[Laurent Vivier]

Le 19/06/2015 15:43, gerg@uclinux.org a écrit :
> From: Greg Ungerer <gerg@uclinux.org>
> 
> The action to potentially switch sp register is not occurring at the correct
> point in the interrupt entry or exception exit sequences.
> 
> For the interrupt entry case the sp on entry is used to create the stack
> exception frame - but this may well be the user stack pointer, since we
> haven't done the switch yet. Re-order the flow to switch the sp regs then
> use the current sp to create the exception frame.
> 
> For the return from exception case the code is unwinding the sp after
> switching sp registers. But it should always unwind the supervisor sp
> first, then carry out any required sp switch.
> 
> Note that these problems don't effect operation unless the user sp bit is
> set in the CACR register. Only a single sp is used in the default power up
> state. Previously Linux only used this single sp mode. But modern versions
> of Linux use the user sp mode now, so we need correct behavior for Linux
> to work.
> 
> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
> Reviewed-by: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
> ---
>  target-m68k/op_helper.c | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)

Reviewed-by: Laurent Vivier <laurent@vivier.eu>

Re: [Qemu-devel] [PATCH v2 3/3] m68k: fix usp processing on interrupt entry and exception exit

[Greg Ungerer]

Hi Laurent,

On 20/06/15 05:35, Laurent Vivier wrote:
> Le 19/06/2015 15:43, gerg@uclinux.org a écrit :
>> From: Greg Ungerer <gerg@uclinux.org>
>>
>> The action to potentially switch sp register is not occurring at the correct
>> point in the interrupt entry or exception exit sequences.
>>
>> For the interrupt entry case the sp on entry is used to create the stack
>> exception frame - but this may well be the user stack pointer, since we
>> haven't done the switch yet. Re-order the flow to switch the sp regs then
>> use the current sp to create the exception frame.
>>
>> For the return from exception case the code is unwinding the sp after
>> switching sp registers. But it should always unwind the supervisor sp
>> first, then carry out any required sp switch.
>>
>> Note that these problems don't effect operation unless the user sp bit is
>> set in the CACR register. Only a single sp is used in the default power up
>> state. Previously Linux only used this single sp mode. But modern versions
>> of Linux use the user sp mode now, so we need correct behavior for Linux
>> to work.
>>
>> Signed-off-by: Greg Ungerer <gerg@uclinux.org>
>> Reviewed-by: Peter Crosthwaite <peter.crosthwaite@xilinx.com>
>> ---
>>  target-m68k/op_helper.c | 7 +++----
>>  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> Reviewed-by: Laurent Vivier <laurent@vivier.eu>

Thanks for the reviews.

Regards
Greg

Re: [Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[Laurent Vivier]

Le 19/06/2015 15:43, gerg@uclinux.org a écrit :
> Some small issues are causing problems with running modern versions of
> Linux on the m68k/ColdFire 5208 target. These 3 patches fix those problems.
> 
> They are all due to use of more advanced architecture features not used
> in older Linux kernels.

Do you have disk image allowing to test Coldfire platform emulation ?

Laurent

Re: [Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[Greg Ungerer]

Hi Laurent,

On 20/06/15 05:39, Laurent Vivier wrote:
>
> Le 19/06/2015 15:43, gerg@uclinux.org a écrit :
>> Some small issues are causing problems with running modern versions of
>> Linux on the m68k/ColdFire 5208 target. These 3 patches fix those problems.
>>
>> They are all due to use of more advanced architecture features not used
>> in older Linux kernels.
> Do you have disk image allowing to test Coldfire platform emulation ?

Not a disk image, but a small kernel with builtin root filesystem.
This points at one I am running at the moment.

http://www.uclinux.org/ports/coldfire/vmlinux.m5208evb-20150620

It is a linux-4.1.0-rc8 kernel.

I have one more fix still to come for the qemu mcf_fec.c net driver.
It currently doesn't support the mdio actions or any attached phy and
this will cause the linux fec driver to fail probing.

Regards
Greg

Re: [Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[Laurent Vivier]

Le 20/06/2015 06:55, Greg Ungerer a écrit :
> Hi Laurent,

Hi Greg,

> On 20/06/15 05:39, Laurent Vivier wrote:
>>
>> Le 19/06/2015 15:43, gerg@uclinux.org a écrit :
>>> Some small issues are causing problems with running modern versions of
>>> Linux on the m68k/ColdFire 5208 target. These 3 patches fix those
>>> problems.
>>>
>>> They are all due to use of more advanced architecture features not used
>>> in older Linux kernels.
>> Do you have disk image allowing to test Coldfire platform emulation ?
> 
> Not a disk image, but a small kernel with builtin root filesystem.
> This points at one I am running at the moment.
> 
> http://www.uclinux.org/ports/coldfire/vmlinux.m5208evb-20150620
> 
> It is a linux-4.1.0-rc8 kernel.

Thank you.

> I have one more fix still to come for the qemu mcf_fec.c net driver.
> It currently doesn't support the mdio actions or any attached phy and
> this will cause the linux fec driver to fail probing.
> 

I'd like to merge my qemu-m68k tree, at least the non privileged
instructions (https://github.com/vivier/qemu-m68k/tree/680x0-v2.3.0) but
I'm afraid it breaks coldfire support. So I need to test before.

Laurent

Re: [Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[Andreas Färber]

Hi Greg,

Am 20.06.2015 um 06:55 schrieb Greg Ungerer:
> I have one more fix still to come for the qemu mcf_fec.c net driver.
> It currently doesn't support the mdio actions or any attached phy and
> this will cause the linux fec driver to fail probing.

I still have some mcf_* device conversions back from KVM Forum 2012
lying around somewhere, waiting for further cleanups for submission...

Regards,
Andreas

-- 
SUSE Linux GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer, Jane Smithard, Dilip Upmanyu, Graham Norton; HRB
21284 (AG Nürnberg)

Re: [Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[Greg Ungerer]

Hi Andreas,

On 23/06/15 02:49, Andreas Färber wrote:
> Am 20.06.2015 um 06:55 schrieb Greg Ungerer:
>> I have one more fix still to come for the qemu mcf_fec.c net driver.
>> It currently doesn't support the mdio actions or any attached phy and
>> this will cause the linux fec driver to fail probing.
> 
> I still have some mcf_* device conversions back from KVM Forum 2012
> lying around somewhere, waiting for further cleanups for submission...

Please dust of and send if you have the time. Would be
great to have more.

Regards
Greg

Re: [Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[Laurent Vivier]

Le 19/06/2015 15:43, gerg@uclinux.org a écrit :
> Some small issues are causing problems with running modern versions of
> Linux on the m68k/ColdFire 5208 target. These 3 patches fix those problems.
> 
> They are all due to use of more advanced architecture features not used
> in older Linux kernels.
> 
> Regards
> Greg
> 
> 
> v2:
> * rebased to current master head
> * constant change from Peter Crosthwaite
> 
>  hw/m68k/mcf_intc.c      |   14 ++++++++++++++
>  target-m68k/op_helper.c |    7 +++----
>  target-m68k/translate.c |    8 ++++----
>  3 files changed, 21 insertions(+), 8 deletions(-)
> 

Tested-by: Laurent Vivier <laurent@vivier.eu>

Re: [Qemu-devel] [PATCH v2 0/3] m68k: fix ColdFire support

[Peter Maydell]

On 19 June 2015 at 14:43,  <gerg@uclinux.org> wrote:
>
> Some small issues are causing problems with running modern versions of
> Linux on the m68k/ColdFire 5208 target. These 3 patches fix those problems.
>
> They are all due to use of more advanced architecture features not used
> in older Linux kernels.

Applied to master, thanks (and thanks to all for review).

-- PMM