On 2012-03-16 14:46, Lee Essen wrote:
>
> On 16 Mar 2012, at 13:14, Jan Kiszka wrote:
>
>> On 2012-03-16 10:23, Lee Essen wrote:
>>> +#ifdef __sun__
>>> +#include
>>> +#else
>>> #include
>>> #include
>>> +#endif
>>
>> As Paolo already said, this should somehow be centralised.
>
> Yep, fair po
Il 16/03/2012 18:25, Lee Essen ha scritto:
> Correct … but actually this raises another question … why create a separate
> pointer to kvm_state…
>
> KVMState *s = kvm_state;
>
> … why not just use kvm_state throughout the function?
>
> This seems to be a common approach in many of the funct
On 16 Mar 2012, at 11:56, Paolo Bonzini wrote:
> Il 16/03/2012 10:23, Lee Essen ha scritto:
>> +while (mlock(base, (nbytes = step * ps)) == -1) {
>> +if (errno != EAGAIN) {
>> +return -1;
>> +}
>> +
>> +if (waiting == 0) {
>> +waiting = gethrtim
On 16 Mar 2012, at 16:28, Paolo Bonzini wrote:
> Il 16/03/2012 14:46, Lee Essen ha scritto:
>> It does indeed … this came from the original Joyent code, I must
>> admit I did wonder whether gcc would optimise it away. I did consider
>> adding something to stop gcc complaining, but I don't fully
>
Il 16/03/2012 14:46, Lee Essen ha scritto:
> It does indeed … this came from the original Joyent code, I must
> admit I did wonder whether gcc would optimise it away. I did consider
> adding something to stop gcc complaining, but I don't fully
> understand why this is necessary given the mlock() bi
On 16 Mar 2012, at 13:14, Jan Kiszka wrote:
> On 2012-03-16 10:23, Lee Essen wrote:
>> +#ifdef __sun__
>> +#include
>> +#else
>> #include
>> #include
>> +#endif
>
> As Paolo already said, this should somehow be centralised.
Yep, fair point. I'll address this one.
> Also, CONFIG_SOLARIS vs.
On 2012-03-16 10:23, Lee Essen wrote:
> This fixes a number of issues with the build process (namely ensuring the use
> of bash), adds specific support for the Illumos port of KVM and fixes a few
> general Solaris compatibility issues.
>
> There are still some things outstanding:
>
> - there's
Il 16/03/2012 10:23, Lee Essen ha scritto:
> diff --git a/Makefile.objs b/Makefile.objs
> index 226b01d..c2a440a 100644
> --- a/Makefile.objs
> +++ b/Makefile.objs
> @@ -373,12 +373,12 @@ else
> trace.h: trace.h-timestamp
> endif
> trace.h-timestamp: $(SRC_PATH)/trace-events $(BUILD_DIR)/config-
Am 16.03.2012 10:23, schrieb Lee Essen:
> This fixes a number of issues with the build process (namely ensuring
> the use of bash), adds specific support for the Illumos port of KVM
> and fixes a few general Solaris compatibility issues.
>
> There are still some things outstanding:
>
> - there's a
