[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v12 10/21] migration: Create multipage support
From: |
Juan Quintela |
Subject: |
Re: [Qemu-devel] [PATCH v12 10/21] migration: Create multipage support |
Date: |
Wed, 09 May 2018 12:53:37 +0200 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/25.3 (gnu/linux) |
"Dr. David Alan Gilbert" <address@hidden> wrote:
> * Juan Quintela (address@hidden) wrote:
>> We only create/destry the page list here. We will use it later.
>>
>> Signed-off-by: Juan Quintela <address@hidden>
>> ---
>> migration/ram.c | 56 +++++++++++++++++++++++++++++++++++++++++++++++++
>> 1 file changed, 56 insertions(+)
>>
>> diff --git a/migration/ram.c b/migration/ram.c
>> index ffefa73099..b19300992e 100644
>> --- a/migration/ram.c
>> +++ b/migration/ram.c
>> @@ -412,6 +412,20 @@ typedef struct {
>> uint8_t id;
>> } __attribute__((packed)) MultiFDInit_t;
>>
>> +typedef struct {
>> + /* number of used pages */
>> + uint32_t used;
>> + /* number of allocated pages */
>> + uint32_t allocated;
>> + /* global number of generated multifd packets */
>> + uint32_t seq;
>
> Is that sufficiently large?
> Consider a 40Gbps link; that's ~4GByte/second,
> so if each packet is a 4K page, that's just over one hour at
> that link speed; that's a long migration on a fast link, but
> it's not impossible is it - and what happens when it wraps?
Nothing really, it is just a counter that is used for traces.
Later, Juan.