Re: [edk2] [PATCH 0/3] unify FVP Base and Foundation model support

2015-08-18 Thread Leif Lindholm
On Tue, Aug 18, 2015 at 01:15:54PM +0200, Ard Biesheuvel wrote:
 On 12 August 2015 at 08:45, Ard Biesheuvel ard.biesheu...@linaro.org wrote:
  Instead of omitting some drivers that are known to break the Foundation
  model when ARM_FOUNDATION_FVP is defined, fix those drivers so that they
  simply fail to load without interfering with the boot.
 
  This way, we can use the same boot image for both models.
 
 
 Ping?

Basicallt looks good to me, but I thought I'd wait for Ryan to comment
(and he's on holiday through this week).

Just to confirm - the Mmio probes just return zeroes in the Foundation
model, rather than aborting?

/
Leif

  Ard Biesheuvel (3):
ArmPlatformPkg/PL180MciDxe: check PrimeCell ID before initializing
ArmPlatformPkg/LcdGraphicsOutputDxe: check PrimeCell ID before
  initializing
ArmPlatformPkg/FVP: unify support for Foundation and Base models
 
   ArmPlatformPkg/ArmVExpressPkg/ArmVExpress-FVP-AArch64.dsc  | 13 
  -
   ArmPlatformPkg/ArmVExpressPkg/ArmVExpress-FVP-AArch64.fdf  |  4 
  
   ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/LcdGraphicsOutputDxe.c |  5 
  +
   ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/LcdGraphicsOutputDxe.h |  2 +-
   ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/PL111Lcd.c | 19 
  +++
   ArmPlatformPkg/Drivers/PL180MciDxe/PL180Mci.c  | 13 
  +
   ArmPlatformPkg/Drivers/PL180MciDxe/PL180Mci.h  | 17 
  +
   ArmPlatformPkg/Include/Drivers/PL111Lcd.h  |  9 
  +
   8 files changed, 64 insertions(+), 18 deletions(-)
 
  --
  1.9.1
 
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel


Re: [edk2] [PATCH 0/3] unify FVP Base and Foundation model support

2015-08-18 Thread Ard Biesheuvel
On 18 August 2015 at 14:12, Leif Lindholm leif.lindh...@linaro.org wrote:
 On Tue, Aug 18, 2015 at 01:15:54PM +0200, Ard Biesheuvel wrote:
 On 12 August 2015 at 08:45, Ard Biesheuvel ard.biesheu...@linaro.org wrote:
  Instead of omitting some drivers that are known to break the Foundation
  model when ARM_FOUNDATION_FVP is defined, fix those drivers so that they
  simply fail to load without interfering with the boot.
 
  This way, we can use the same boot image for both models.
 

 Ping?

 Basicallt looks good to me, but I thought I'd wait for Ryan to comment
 (and he's on holiday through this week).

 Just to confirm - the Mmio probes just return zeroes in the Foundation
 model, rather than aborting?


Foundation model output:

WARNING: Attempted to access VE CS3 space: Read 1 bytes from 0x1c050fe0.
WARNING: Attempted to access VE CS3 space: Read 1 bytes from 0x1c1f0fe0.

Note sure what these reads return, but in each case, the first byte
already differs from the expected value so the init aborts.

I know that this would still be sufficient to blow up on a real
system, but thankfully the model does not model /that/ :-)


  Ard Biesheuvel (3):
ArmPlatformPkg/PL180MciDxe: check PrimeCell ID before initializing
ArmPlatformPkg/LcdGraphicsOutputDxe: check PrimeCell ID before
  initializing
ArmPlatformPkg/FVP: unify support for Foundation and Base models
 
   ArmPlatformPkg/ArmVExpressPkg/ArmVExpress-FVP-AArch64.dsc  | 13 
  -
   ArmPlatformPkg/ArmVExpressPkg/ArmVExpress-FVP-AArch64.fdf  |  4 
  
   ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/LcdGraphicsOutputDxe.c |  5 
  +
   ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/LcdGraphicsOutputDxe.h |  2 +-
   ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/PL111Lcd.c | 19 
  +++
   ArmPlatformPkg/Drivers/PL180MciDxe/PL180Mci.c  | 13 
  +
   ArmPlatformPkg/Drivers/PL180MciDxe/PL180Mci.h  | 17 
  +
   ArmPlatformPkg/Include/Drivers/PL111Lcd.h  |  9 
  +
   8 files changed, 64 insertions(+), 18 deletions(-)
 
  --
  1.9.1
 
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel


Re: [edk2] [PATCH 0/3] unify FVP Base and Foundation model support

2015-08-18 Thread Ard Biesheuvel
On 12 August 2015 at 08:45, Ard Biesheuvel ard.biesheu...@linaro.org wrote:
 Instead of omitting some drivers that are known to break the Foundation
 model when ARM_FOUNDATION_FVP is defined, fix those drivers so that they
 simply fail to load without interfering with the boot.

 This way, we can use the same boot image for both models.


Ping?

 Ard Biesheuvel (3):
   ArmPlatformPkg/PL180MciDxe: check PrimeCell ID before initializing
   ArmPlatformPkg/LcdGraphicsOutputDxe: check PrimeCell ID before
 initializing
   ArmPlatformPkg/FVP: unify support for Foundation and Base models

  ArmPlatformPkg/ArmVExpressPkg/ArmVExpress-FVP-AArch64.dsc  | 13 
 -
  ArmPlatformPkg/ArmVExpressPkg/ArmVExpress-FVP-AArch64.fdf  |  4 
  ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/LcdGraphicsOutputDxe.c |  5 +
  ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/LcdGraphicsOutputDxe.h |  2 +-
  ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/PL111Lcd.c | 19 
 +++
  ArmPlatformPkg/Drivers/PL180MciDxe/PL180Mci.c  | 13 
 +
  ArmPlatformPkg/Drivers/PL180MciDxe/PL180Mci.h  | 17 
 +
  ArmPlatformPkg/Include/Drivers/PL111Lcd.h  |  9 
 +
  8 files changed, 64 insertions(+), 18 deletions(-)

 --
 1.9.1

___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel


Re: [edk2] [PATCH 0/3] unify FVP Base and Foundation model support

2015-08-18 Thread Leif Lindholm
On Tue, Aug 18, 2015 at 02:14:53PM +0200, Ard Biesheuvel wrote:
 On 18 August 2015 at 14:12, Leif Lindholm leif.lindh...@linaro.org wrote:
  On Tue, Aug 18, 2015 at 01:15:54PM +0200, Ard Biesheuvel wrote:
  On 12 August 2015 at 08:45, Ard Biesheuvel ard.biesheu...@linaro.org 
  wrote:
   Instead of omitting some drivers that are known to break the Foundation
   model when ARM_FOUNDATION_FVP is defined, fix those drivers so that they
   simply fail to load without interfering with the boot.
  
   This way, we can use the same boot image for both models.
  
 
  Ping?
 
  Basicallt looks good to me, but I thought I'd wait for Ryan to comment
  (and he's on holiday through this week).
 
  Just to confirm - the Mmio probes just return zeroes in the Foundation
  model, rather than aborting?
 
 
 Foundation model output:
 
 WARNING: Attempted to access VE CS3 space: Read 1 bytes from 0x1c050fe0.
 WARNING: Attempted to access VE CS3 space: Read 1 bytes from 0x1c1f0fe0.
 
 Note sure what these reads return, but in each case, the first byte
 already differs from the expected value so the init aborts.
 
 I know that this would still be sufficient to blow up on a real
 system, but thankfully the model does not model /that/ :-)

No, my main concern would be if future versions of the model delete
this trapping and instead signal an abort (or do something else
entirely), to better mimic hardware behaviour.

Do we have output console available at this point?
If so, maybe put in a printout - at least in DEBUG build - that this
is about to happen, to reduce debug time if the model behaviour
changes?

Regardless, unless Ryan objects on Monday:
Reviewed-by: Leif Lindholm leif.lindh...@linaro.org

/
Leif

   Ard Biesheuvel (3):
 ArmPlatformPkg/PL180MciDxe: check PrimeCell ID before initializing
 ArmPlatformPkg/LcdGraphicsOutputDxe: check PrimeCell ID before
   initializing
 ArmPlatformPkg/FVP: unify support for Foundation and Base models
  
ArmPlatformPkg/ArmVExpressPkg/ArmVExpress-FVP-AArch64.dsc  | 13 
   -
ArmPlatformPkg/ArmVExpressPkg/ArmVExpress-FVP-AArch64.fdf  |  4 
   
ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/LcdGraphicsOutputDxe.c |  5 
   +
ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/LcdGraphicsOutputDxe.h |  2 
   +-
ArmPlatformPkg/Drivers/LcdGraphicsOutputDxe/PL111Lcd.c | 19 
   +++
ArmPlatformPkg/Drivers/PL180MciDxe/PL180Mci.c  | 13 
   +
ArmPlatformPkg/Drivers/PL180MciDxe/PL180Mci.h  | 17 
   +
ArmPlatformPkg/Include/Drivers/PL111Lcd.h  |  9 
   +
8 files changed, 64 insertions(+), 18 deletions(-)
  
   --
   1.9.1
  
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel


Re: [edk2] [PATCH 0/3] unify FVP Base and Foundation model support

2015-08-18 Thread Ard Biesheuvel
On 18 August 2015 at 15:50, Leif Lindholm leif.lindh...@linaro.org wrote:
 On Tue, Aug 18, 2015 at 02:14:53PM +0200, Ard Biesheuvel wrote:
 On 18 August 2015 at 14:12, Leif Lindholm leif.lindh...@linaro.org wrote:
  On Tue, Aug 18, 2015 at 01:15:54PM +0200, Ard Biesheuvel wrote:
  On 12 August 2015 at 08:45, Ard Biesheuvel ard.biesheu...@linaro.org 
  wrote:
   Instead of omitting some drivers that are known to break the Foundation
   model when ARM_FOUNDATION_FVP is defined, fix those drivers so that they
   simply fail to load without interfering with the boot.
  
   This way, we can use the same boot image for both models.
  
 
  Ping?
 
  Basicallt looks good to me, but I thought I'd wait for Ryan to comment
  (and he's on holiday through this week).
 
  Just to confirm - the Mmio probes just return zeroes in the Foundation
  model, rather than aborting?
 

 Foundation model output:

 WARNING: Attempted to access VE CS3 space: Read 1 bytes from 0x1c050fe0.
 WARNING: Attempted to access VE CS3 space: Read 1 bytes from 0x1c1f0fe0.

 Note sure what these reads return, but in each case, the first byte
 already differs from the expected value so the init aborts.

 I know that this would still be sufficient to blow up on a real
 system, but thankfully the model does not model /that/ :-)

 No, my main concern would be if future versions of the model delete
 this trapping and instead signal an abort (or do something else
 entirely), to better mimic hardware behaviour.

 Do we have output console available at this point?
 If so, maybe put in a printout - at least in DEBUG build - that this
 is about to happen, to reduce debug time if the model behaviour
 changes?


I suppose that makes sense. These are DXEs being launched, so it is
definitely feasible to print some diagnostic on a DEBUG build.

 Regardless, unless Ryan objects on Monday:
 Reviewed-by: Leif Lindholm leif.lindh...@linaro.org


Thanks,
Ard.
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel