Re: [Patch] (big) real mode emulation - emulator_write_std

[Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org>]

Nitin A Kamble wrote:
> Hi Avi,
>   I have been collected quiet a few patches in my tree now. They are 
> all in dirty(style, debug) state. Now I started cleaning some of the 
> code in my tree, and cutting patches again. I plan to send you steady 
> stream of patches now. so you can expect lots of patches coming from 
> me in coming days now.
>
> About the status of the big real mode; In my private tree I have 
> evolved the segment registers support for the emulator to next level 
> now. I think I can also cut that patch also and send it to you. now it 
> is much more complete than earlier. As of today the SuSe Linux 10.1 is 
> proceeding in the big-real mode into the int 10 call. that is with the 
> help emulation of about 20 new instructions, and also the 
> infrastructure support for those instructions.  Probably I should keep 
> a count of instructions emulated in the big real mode to gauge my 
> progress.
>
> I am cutting my tree into simple patches, so it will be easy for you 
> to understand and apply/comment.
>     

> diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
> index f7ff231..4a213e2 100644
> --- a/drivers/kvm/kvm_main.c
> +++ b/drivers/kvm/kvm_main.c
> @@ -962,8 +962,35 @@ static int emulator_write_std(unsigned long addr,
>  			      unsigned int bytes,
>  			      struct kvm_vcpu *vcpu)
>  {
> -	pr_unimpl(vcpu, "emulator_write_std: addr %lx n %d\n", addr, bytes);
> -	return X86EMUL_UNHANDLEABLE;
> +	void *data = (void *) val;
> +
> +	while (bytes) {
> +		gpa_t gpa;
> +		unsigned offset = addr & (PAGE_SIZE-1);
> +		unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset);
> +		unsigned long pfn;
> +		struct page *page;
> +		void *page_virt;
> +
> +		gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
> +		if (gpa == UNMAPPED_GVA)
> +			return X86EMUL_PROPAGATE_FAULT;
> +		pfn = gpa >> PAGE_SHIFT;
> +		page = gfn_to_page(vcpu->kvm, pfn);
> +		if (!page)
> +			return X86EMUL_UNHANDLEABLE;
> +		page_virt = kmap_atomic(page, KM_USER0);
> +
> +		memcpy(page_virt + offset, data, tocopy);
> +
> +		kunmap_atomic(page_virt, KM_USER0);
> +
> +		bytes -= tocopy;
> +		data += tocopy;
> +		addr += tocopy;
> +	}
> +
> +	return X86EMUL_CONTINUE;
>  }
>  


This doesn't do the right thing if the write is targeted at a page 
table.  Just use emulator_write_emulated.


-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/