mbox series

[rc1,00/15] hw: Renesas patches (SH4 and RX)

Message ID 20200617191519.14842-1-f4bug@amsat.org
Headers show
Series hw: Renesas patches (SH4 and RX) | expand

Message

Philippe Mathieu-Daudé June 17, 2020, 7:15 p.m. UTC 
Hi Richard, Yoshinori,

This is a merger of the latest SH4 patches (add a common entry for
Renesas hardware in MAINTAINERS, an acceptance test from Thomas),
and Yoshinori's v32 [*] of the RX machine.
Call this release, or merge, candidate 1.

[*] https://www.mail-archive.com/qemu-devel@nongnu.org/msg682290.html

---

Richard,

A few months ago, I volunteered to integrate the RX port posted last
year on the list, to help you ease the burden of your heavy workload.

As it took a long time, at some point I suggested doing the integration
in two parts, first the architecture part (target/ directory), then the
hardware (hw/ directory). You disagreed, arguing that this approach
could be dangerous, as we don’t want to maintain a partly integrated
port. In hindsight I learned the hard way that you were right!

When I decided to send the architectural part first, I made the
commitment to you to finish the full integration before the following
release. The second part has been a painful experience.

With this series, I have now fulfilled this commitment. I don't plan
to continue working on the RX port.

Thank you for your support during this experiment, I learned a lot,
and I’d be grateful to work with you again in the future in other
parts of the project.

---

Yoshinori,

I appreciated integrating your work, which is of very good quality
overall. Despite the numerous iterations you had to go through, the
outcome has proven worthwhile.
Working with you has been a great experience.

I volunteered to help maintain the RX hardware in my leisure time.
Such hobby projects are supposed to be fun, and not have the same
level of pressure as paid work.

Sadly, my experience trying to get your work merged has not been fun:
too much aggressive criticism, and constant demands for more work.
The pressure has become too great for what is, essentially, unpaid
work.

---

Missing review: patches 2 and 12
- 02/15 MAINTAINERS: Add an entry for common Renesas peripherals
- 12/15 hw/rx: Register R5F562N7 and R5F562N8 MCUs

Changes since v32:

- Rebased on DeviceReset API change
- Renamed CPU -> MCU
- Renamed device -> microcontroller
- Rebased on device_class_set_parent_realize API change
- Split timer patch in two distincts: CMT and TMR
- Rebased on ramdev API change
- Use TYPE_RX62N_CPU definition
- Use RX62N_NR_TMR/CMT/SCI definitions
- Replace obsolete set_machine() by machine tag
- Split rx-virt patch in two distinct, MCU first, machine next
- Renamed machined as gdbsim
- Make TYPE_RX62N_MCU an abstract class,
  add TYPE_R5F562N7_MCU and TYPE_R5F562N8_MCU models
- Rename the machine gdbsim-r5f562n8, also add the gdbsim-r5f562n7
- Filled various VMStateField for migration
- Rebase on qdev/sysbus API change

CI:
https://travis-ci.org/github/philmd/qemu/builds/698974425

Philippe Mathieu-Daudé (7):
  MAINTAINERS: Cover sh_intc files in the R2D/Shix machine sections
  MAINTAINERS: Add an entry for common Renesas peripherals
  hw/sh4: Use MemoryRegion typedef
  hw/sh4: Extract timer definitions to 'hw/timer/tmu012.h'
  hw/timer/sh_timer: Remove unused 'qemu/timer.h' include
  hw/rx: Register R5F562N7 and R5F562N8 MCUs
  BootLinuxConsoleTest: Test the RX GDB simulator

Richard Henderson (1):
  hw/rx: Honor -accel qtest

Yoshinori Sato (7):
  hw/intc: RX62N interrupt controller (ICUa)
  hw/timer: RX62N 8-Bit timer (TMR)
  hw/timer: RX62N compare match timer (CMT)
  hw/char: RX62N serial communication interface (SCI)
  hw/rx: RX62N microcontroller (MCU)
  hw/rx: Add RX GDB simulator
  docs: Document the RX target

 docs/system/target-rx.rst             |  36 ++
 docs/system/targets.rst               |   1 +
 default-configs/rx-softmmu.mak        |   1 +
 include/hw/char/renesas_sci.h         |  51 +++
 include/hw/intc/rx_icu.h              |  76 ++++
 include/hw/rx/rx62n.h                 |  78 +++++
 include/hw/sh4/sh.h                   |  12 +-
 include/hw/timer/renesas_cmt.h        |  40 +++
 include/hw/timer/renesas_tmr.h        |  55 +++
 include/hw/timer/tmu012.h             |  23 ++
 hw/char/renesas_sci.c                 | 350 +++++++++++++++++++
 hw/intc/rx_icu.c                      | 397 +++++++++++++++++++++
 hw/rx/rx-gdbsim.c                     | 196 +++++++++++
 hw/rx/rx62n.c                         | 318 +++++++++++++++++
 hw/sh4/sh7750.c                       |   1 +
 hw/timer/renesas_cmt.c                | 283 +++++++++++++++
 hw/timer/renesas_tmr.c                | 477 ++++++++++++++++++++++++++
 hw/timer/sh_timer.c                   |   3 +-
 MAINTAINERS                           |  33 +-
 hw/Kconfig                            |   1 +
 hw/char/Kconfig                       |   3 +
 hw/char/Makefile.objs                 |   1 +
 hw/intc/Kconfig                       |   3 +
 hw/intc/Makefile.objs                 |   1 +
 hw/rx/Kconfig                         |  10 +
 hw/rx/Makefile.objs                   |   2 +
 hw/timer/Kconfig                      |   6 +
 hw/timer/Makefile.objs                |   2 +
 tests/acceptance/machine_rx_gdbsim.py |  68 ++++
 29 files changed, 2515 insertions(+), 13 deletions(-)
 create mode 100644 docs/system/target-rx.rst
 create mode 100644 include/hw/char/renesas_sci.h
 create mode 100644 include/hw/intc/rx_icu.h
 create mode 100644 include/hw/rx/rx62n.h
 create mode 100644 include/hw/timer/renesas_cmt.h
 create mode 100644 include/hw/timer/renesas_tmr.h
 create mode 100644 include/hw/timer/tmu012.h
 create mode 100644 hw/char/renesas_sci.c
 create mode 100644 hw/intc/rx_icu.c
 create mode 100644 hw/rx/rx-gdbsim.c
 create mode 100644 hw/rx/rx62n.c
 create mode 100644 hw/timer/renesas_cmt.c
 create mode 100644 hw/timer/renesas_tmr.c
 create mode 100644 hw/rx/Kconfig
 create mode 100644 hw/rx/Makefile.objs
 create mode 100644 tests/acceptance/machine_rx_gdbsim.py