As part of an upcoming 40p patchset I have a requirement to change the PCI
configuration of the LSI SCSI. However since commits a64aa5785d "hw: Deprecate -drive
if=scsi with non-onboard HBAs" and b891538e81 "hw/ppc/prep: Fix implicit creation of
"-drive if=scsi", the lsi53c8*_create() wrapper functions don't return the device
state itself.

This patchset replaces the lsi53c895a_create() and lsi53c810_create() functions
with a single lsi53c8xx_handle_legacy_cmdline() function as suggested by Thomas,
which makes the caller responsible for initing the LSI SCSI device and hence
allowing it to be configured as required.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>

v2:
- Don't split LSIState into separate lsi53c895a.h header but instead use a new
  lsi53c8xx_handle_legacy_cmdline() function as suggested by Thomas


Mark Cave-Ayland (3):
  scsi: add lsi53c8xx_handle_legacy_cmdline() function
  scsi: move lsi53c8xx_create() callers to
    lsi53c8xx_handle_legacy_cmdline()
  scsi: remove unused lsi53c895a_create() and lsi53c810_create()
    functions

 hw/arm/realview.c    |  3 ++-
 hw/arm/versatilepb.c |  3 ++-
 hw/hppa/machine.c    |  4 +++-
 hw/ppc/prep.c        |  4 +++-
 hw/scsi/lsi53c895a.c | 11 ++---------
 include/hw/pci/pci.h |  3 +--
 6 files changed, 13 insertions(+), 15 deletions(-)

-- 
2.11.0

On 07/09/2018 14:56, Mark Cave-Ayland wrote:
> As part of an upcoming 40p patchset I have a requirement to change the PCI
> configuration of the LSI SCSI. However since commits a64aa5785d "hw: Deprecate -drive
> if=scsi with non-onboard HBAs" and b891538e81 "hw/ppc/prep: Fix implicit creation of
> "-drive if=scsi", the lsi53c8*_create() wrapper functions don't return the device
> state itself.
> 
> This patchset replaces the lsi53c895a_create() and lsi53c810_create() functions
> with a single lsi53c8xx_handle_legacy_cmdline() function as suggested by Thomas,
> which makes the caller responsible for initing the LSI SCSI device and hence
> allowing it to be configured as required.
> 
> Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
> 
> v2:
> - Don't split LSIState into separate lsi53c895a.h header but instead use a new
>   lsi53c8xx_handle_legacy_cmdline() function as suggested by Thomas
> 
> 
> Mark Cave-Ayland (3):
>   scsi: add lsi53c8xx_handle_legacy_cmdline() function
>   scsi: move lsi53c8xx_create() callers to
>     lsi53c8xx_handle_legacy_cmdline()
>   scsi: remove unused lsi53c895a_create() and lsi53c810_create()
>     functions
> 
>  hw/arm/realview.c    |  3 ++-
>  hw/arm/versatilepb.c |  3 ++-
>  hw/hppa/machine.c    |  4 +++-
>  hw/ppc/prep.c        |  4 +++-
>  hw/scsi/lsi53c895a.c | 11 ++---------
>  include/hw/pci/pci.h |  3 +--
>  6 files changed, 13 insertions(+), 15 deletions(-)
> 

Queued, thanks.

Paolo

Le 07/09/2018 à 14:56, Mark Cave-Ayland a écrit :
> As part of an upcoming 40p patchset I have a requirement to change the PCI
> configuration of the LSI SCSI. However since commits a64aa5785d "hw: Deprecate -drive
> if=scsi with non-onboard HBAs" and b891538e81 "hw/ppc/prep: Fix implicit creation of
> "-drive if=scsi", the lsi53c8*_create() wrapper functions don't return the device
> state itself.
> 
> This patchset replaces the lsi53c895a_create() and lsi53c810_create() functions
> with a single lsi53c8xx_handle_legacy_cmdline() function as suggested by Thomas,
> which makes the caller responsible for initing the LSI SCSI device and hence
> allowing it to be configured as required.
> 
> Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
> 
> v2:
> - Don't split LSIState into separate lsi53c895a.h header but instead use a new
>    lsi53c8xx_handle_legacy_cmdline() function as suggested by Thomas
> 
> 
> Mark Cave-Ayland (3):
>    scsi: add lsi53c8xx_handle_legacy_cmdline() function
>    scsi: move lsi53c8xx_create() callers to
>      lsi53c8xx_handle_legacy_cmdline()
>    scsi: remove unused lsi53c895a_create() and lsi53c810_create()
>      functions
> 
>   hw/arm/realview.c    |  3 ++-
>   hw/arm/versatilepb.c |  3 ++-
>   hw/hppa/machine.c    |  4 +++-
>   hw/ppc/prep.c        |  4 +++-
>   hw/scsi/lsi53c895a.c | 11 ++---------
>   include/hw/pci/pci.h |  3 +--
>   6 files changed, 13 insertions(+), 15 deletions(-)
> 

Reviewed-by: Hervé Poussineau <hpoussin@reactos.org>

On 9/7/18 9:56 AM, Mark Cave-Ayland wrote:
> As part of an upcoming 40p patchset I have a requirement to change the PCI
> configuration of the LSI SCSI. However since commits a64aa5785d "hw: Deprecate -drive
> if=scsi with non-onboard HBAs" and b891538e81 "hw/ppc/prep: Fix implicit creation of
> "-drive if=scsi", the lsi53c8*_create() wrapper functions don't return the device
> state itself.
> 
> This patchset replaces the lsi53c895a_create() and lsi53c810_create() functions
> with a single lsi53c8xx_handle_legacy_cmdline() function as suggested by Thomas,
> which makes the caller responsible for initing the LSI SCSI device and hence
> allowing it to be configured as required.
> 
> Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>

Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>

> 
> v2:
> - Don't split LSIState into separate lsi53c895a.h header but instead use a new
>   lsi53c8xx_handle_legacy_cmdline() function as suggested by Thomas
> 
> 
> Mark Cave-Ayland (3):
>   scsi: add lsi53c8xx_handle_legacy_cmdline() function
>   scsi: move lsi53c8xx_create() callers to
>     lsi53c8xx_handle_legacy_cmdline()
>   scsi: remove unused lsi53c895a_create() and lsi53c810_create()
>     functions
> 
>  hw/arm/realview.c    |  3 ++-
>  hw/arm/versatilepb.c |  3 ++-
>  hw/hppa/machine.c    |  4 +++-
>  hw/ppc/prep.c        |  4 +++-
>  hw/scsi/lsi53c895a.c | 11 ++---------
>  include/hw/pci/pci.h |  3 +--
>  6 files changed, 13 insertions(+), 15 deletions(-)
>