| Message ID | 20200531162427.57410-1-ysato@users.sourceforge.jp (mailing list archive) |
|---|---|
| Headers | show |
| Series | Add RX hardware emulation | expand |
On 5/31/20 6:24 PM, Yoshinori Sato wrote: > Hello. > > This series add to hardware emulation module for RX target. > > Details below. > Interrupt controller, 8bit timer, 16bit comapare match timer and > SCI is RX62N integrated peripheral. > rx-virt - RX62N MCU and external RAM. It like gdb simulator. > > The compare match timer has a CPU interface similar to the SH4 timer. > sh_timer will be deprecated and integrated into this module. > SCI is also implemented in sh_serial, but the functionality is omitted. > I implemented the complete one as renesas_sci. > > git repository here. > git://git.pf.osdn.net/gitroot/y/ys/ysato/qemu.git tags/hw-rx-20200601 Sigh. It is a bit unfortunate and demotivating that we can not coordinate this. I was going to send a pre-merge for this today. What is positive is you unified some SH4/RX peripherals. I'll try to merge both series. If it takes too long, I might send my current queue for you to review, then go back to this one you just sent. Regards, Phil.
On Mon, 01 Jun 2020 01:45:47 +0900, Philippe Mathieu-Daudé wrote: > > On 5/31/20 6:24 PM, Yoshinori Sato wrote: > > Hello. > > > > This series add to hardware emulation module for RX target. > > > > Details below. > > Interrupt controller, 8bit timer, 16bit comapare match timer and > > SCI is RX62N integrated peripheral. > > rx-virt - RX62N MCU and external RAM. It like gdb simulator. > > > > The compare match timer has a CPU interface similar to the SH4 timer. > > sh_timer will be deprecated and integrated into this module. > > SCI is also implemented in sh_serial, but the functionality is omitted. > > I implemented the complete one as renesas_sci. > > > > git repository here. > > git://git.pf.osdn.net/gitroot/y/ys/ysato/qemu.git tags/hw-rx-20200601 > > Sigh. It is a bit unfortunate and demotivating that we can not > coordinate this. I was going to send a pre-merge for this today. > > What is positive is you unified some SH4/RX peripherals. > > I'll try to merge both series. If it takes too long, I might send my > current queue for you to review, then go back to this one you just sent. > > Regards, > > Phil. > Oh. Sorry. The old code had some spec differences. I have fixed those mistakes here. I would like to merge this if possible.
Hi Yoshinori, On 6/1/20 4:32 PM, Yoshinori Sato wrote: > On Mon, 01 Jun 2020 01:45:47 +0900, > Philippe Mathieu-Daudé wrote: >> >> On 5/31/20 6:24 PM, Yoshinori Sato wrote: >>> Hello. >>> >>> This series add to hardware emulation module for RX target. >>> >>> Details below. >>> Interrupt controller, 8bit timer, 16bit comapare match timer and >>> SCI is RX62N integrated peripheral. >>> rx-virt - RX62N MCU and external RAM. It like gdb simulator. >>> >>> The compare match timer has a CPU interface similar to the SH4 timer. >>> sh_timer will be deprecated and integrated into this module. >>> SCI is also implemented in sh_serial, but the functionality is omitted. >>> I implemented the complete one as renesas_sci. >>> >>> git repository here. >>> git://git.pf.osdn.net/gitroot/y/ys/ysato/qemu.git tags/hw-rx-20200601 >> >> Sigh. It is a bit unfortunate and demotivating that we can not >> coordinate this. I was going to send a pre-merge for this today. >> >> What is positive is you unified some SH4/RX peripherals. >> >> I'll try to merge both series. If it takes too long, I might send my >> current queue for you to review, then go back to this one you just sent. >> >> Regards, >> >> Phil. >> > > Oh. Sorry. > > The old code had some spec differences. > I have fixed those mistakes here. Yes, I am glad to see how it evolved, in particular how you merged the SH4/RX common peripherals :) > > I would like to merge this if possible. I posted the v33 of your v32: https://www.mail-archive.com/qemu-devel@nongnu.org/msg708317.html As it is mostly reviewed and tested, it seems more reasonable to merge what is reviewed, and then apply your latest patches on top. It took us already a long time to get there, and I'd not start again to review 30 more times this version. I believe if would be easier this way, you review/ack the v33 and we get the peripheral cleanup on top. My 2 cents, feel free to disagree and insist on restarting the whole review process. Regards, Phil.