[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH V6 0/7] nvdimm: guarantee persistence of QEMU wr
From: |
Eduardo Habkost |
Subject: |
Re: [Qemu-devel] [PATCH V6 0/7] nvdimm: guarantee persistence of QEMU writes to persistent memory |
Date: |
Wed, 13 Jun 2018 17:05:34 -0300 |
User-agent: |
Mutt/1.9.2 (2017-12-15) |
On Tue, Jun 12, 2018 at 01:38:08PM +0000, Junyan He wrote:
> He have pmem_persist and pmem_memcpy_persist stub functions.
>
> If no libpmem and user really specify pmem=on, we just do nothing or just
> memcpy.
>
> Real persistent memory always require libpmem support its load/save.
>
> If pmem=on and without libpmem, we can think that user want to imitate
>
> pmem=on while the HW environment is without real persistent memory existing.
>
> It may help debug on some machine without real pmem.
What exactly it would help debug? Is there any difference
between testing the code with pmem=off and pmem=on if libptest is
unavailable?
I would prefer to simply not let the user set pmem=on if libptest
is not available.
--
Eduardo
- Re: [Qemu-devel] [PATCH V6 7/7] migration/ram: ensure write persistence on loading all data to PMEM., (continued)
- Re: [Qemu-devel] [PATCH V6 0/7] nvdimm: guarantee persistence of QEMU writes to persistent memory,
Eduardo Habkost <=