Re: [Xen-devel] [PATCH 0/2] block/xen-blkfront: Support non-indirect with 64KB page granularity

[Bob Liu]

On 09/13/2015 08:06 PM, Julien Grall wrote:
> 
> 
> On 12/09/2015 10:46, Bob Liu wrote:
>> Hi Julien,
> 
> Hi Bob,
> 
> 
>> On 09/12/2015 03:31 AM, Julien Grall wrote:
>>> Hi all,
>>>
>>> This is a follow-up on the previous discussion [1] related to guest using 64KB
>>> page granularity not booting with backend using non-indirect grant.
>>>
>>> This has been successly tested on ARM64 with both 64KB and 4KB page granularity
>>> guests and QEMU as the backend. Indeed QEMU is not supported indirect.
>>>
>>
>> What's the linux kernel page granularity of the backend(dom0) in your test?
> 
> The PV protocol and the grant are always based on 4KB page granularity. So the backend page granularity doesn't matter.
> 
>>
>> Did you test if using xen-blkback as the backend? Especially when linux kernel page
>> granularity of dom0 is 4kB while domU is 64KB.
> 
> Sorry, but I don't understand what you are trying to know with those 2 questions.
> 
> xen-blkback is always supporting indirect grant mapping (AFAICT it's not possible for the user to disable it). 
> The issue is only happening when the backend is not supporting non-indirect grant (such as QEMU).
> What matters while testing this series is having a backend which doesn't not support indirect grant. 
> All the code added should never be called if the backend is support indirect grant and/or the frontend is using 4KB page granularity.

I may misunderstood here.
But I think same changes are also required even if backend supports indirect grant when frontend is using 64KB page granularity.
Else
1) How to set up the grant map for requests in domU?
The minimum segment buffer size in a request is PAGE_SIZE(64KB) while grant is 4KB based.

2) Codes like below in blkback.c may not work correctly?
if ((req->u.rw.seg[i].last_sect >= (PAGE_SIZE >> 9)) ||

Because PAGE_SIZE in backend is 4KB, while the written value by domU is 64KB based.

Thanks,
-Bob

> There is some safeguard in patch #2 to ensure that this extra request is never created in those situations (see the definition of HAS_EXTRA_REQ).
> That means that there is no difference for x86 and arm32 given that the page granularity is always 4KB.
> 
> Nonetheless, this patchset has been tested with DOM0 4KB using both QEMU and xen-blkback for the backend.
> 
> I have done testing on backend using 64KB page but with some patches not yet sent upstream. This is because gnttdev doesn't yet support 64KB pages.
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/