Re: [PATCH 2/7] mini-os: Tidied up code

[Andrew Cooper <andrew.cooper3@citrix.com>]

On 04/06/14 17:06, Thomas Leonard wrote:
> From: Karim Raslan <karim.allah.ahmed@gmail.com>
>
> Signed-off-by: Karim Allah Ahmed <karim.allah.ahmed@gmail.com>
> [talex5@gmail.com: separated from big ARM commit]
> Acked-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
> [talex5@gmail.com: use __func__ in DEBUG macro]
> Signed-off-by: Thomas Leonard <talex5@gmail.com>
> ---
>  extras/mini-os/README                      |  7 ++++--
>  extras/mini-os/arch/x86/time.c             |  2 +-
>  extras/mini-os/console/console.c           |  2 +-
>  extras/mini-os/events.c                    |  3 ++-
>  extras/mini-os/gntmap.c                    | 35 ++++++++++++++----------------
>  extras/mini-os/include/x86/arch_spinlock.h |  2 +-
>  extras/mini-os/sched.c                     |  3 +++
>  extras/mini-os/xenbus/xenbus.c             |  2 +-
>  8 files changed, 30 insertions(+), 26 deletions(-)
>
> diff --git a/extras/mini-os/README b/extras/mini-os/README
> index 710a303..bf49700 100644
> --- a/extras/mini-os/README
> +++ b/extras/mini-os/README
> @@ -25,8 +25,11 @@ This includes:
>  
>  - to build it with much better libc support, see the stubdom/ directory
>  
> -- to start it do the following in domain0 (assuming xend is running)
> -  # xm create domain_config
> +- to start it do the following in domain0
> +  # xm create domain_config # old style xend ( assuming xend is running )
> +or
> +  # xl create domain_config # using toolstack
> +

Xend has been rm -rf'd from the codebase.  All new patches should drop
all references to it and just refer to xl.

>  
>  This starts the kernel and prints out a bunch of stuff and then once every
>  second the system time.
> diff --git a/extras/mini-os/arch/x86/time.c b/extras/mini-os/arch/x86/time.c
> index 89bc382..2c8d033 100644
> --- a/extras/mini-os/arch/x86/time.c
> +++ b/extras/mini-os/arch/x86/time.c
> @@ -212,7 +212,7 @@ void block_domain(s_time_t until)
>  
>  
>  /*
> - * Just a dummy 
> + * Just a dummy
>   */
>  static void timer_handler(evtchn_port_t ev, struct pt_regs *regs, void *ign)
>  {
> diff --git a/extras/mini-os/console/console.c b/extras/mini-os/console/console.c
> index 380f53c..5538bd4 100644
> --- a/extras/mini-os/console/console.c
> +++ b/extras/mini-os/console/console.c
> @@ -124,7 +124,7 @@ void print(int direct, const char *fmt, va_list args)
>      static char   buf[1024];
>      
>      (void)vsnprintf(buf, sizeof(buf), fmt, args);
> - 
> +
>      if(direct)
>      {
>          (void)HYPERVISOR_console_io(CONSOLEIO_write, strlen(buf), buf);
> diff --git a/extras/mini-os/events.c b/extras/mini-os/events.c
> index 036b84b..2da9b01 100644
> --- a/extras/mini-os/events.c
> +++ b/extras/mini-os/events.c
> @@ -138,7 +138,8 @@ evtchn_port_t bind_virq(uint32_t virq, evtchn_handler_t handler, void *data)
>  	op.virq = virq;
>  	op.vcpu = smp_processor_id();
>  
> -	if ( (rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq, &op)) != 0 )
> +	rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq, &op);
> +	if (rc != 0)
>  	{
>  		printk("Failed to bind virtual IRQ %d with rc=%d\n", virq, rc);
>  		return -1;
> diff --git a/extras/mini-os/gntmap.c b/extras/mini-os/gntmap.c
> index 22ed450..f6ab3ad 100644
> --- a/extras/mini-os/gntmap.c
> +++ b/extras/mini-os/gntmap.c
> @@ -38,6 +38,15 @@
>  #include <inttypes.h>
>  #include <mini-os/gntmap.h>
>  
> +//#define GNTMAP_DEBUG
> +#ifdef GNTMAP_DEBUG
> +#define DEBUG(_f, _a...) \
> +    printk("MINI_OS(gntmap.c:%d): %s" _f "\n", __LINE__, __func__, ## _a)

Is the "MINI_OS" prefix useful?  I suspect not so much.

~Andrew