From: junyan.he@gmx.com
To: qemu-devel@nongnu.org
Cc: ehabkost@redhat.com, imammedo@redhat.com, pbonzini@redhat.com,
crosthwaite.peter@gmail.com, rth@twiddle.net,
xiaoguangrong.eric@gmail.com, mst@redhat.com,
quintela@redhat.com, dgilbert@redhat.com, stefanha@redhat.com,
Junyan He <junyan.he@intel.com>,
Haozhong Zhang <haozhong.zhang@intel.com>
Subject: [Qemu-devel] [PATCH 7/9 V5] migration/ram: ensure write persistence on loading compressed pages to PMEM
Date: Thu, 10 May 2018 10:08:56 +0800 [thread overview]
Message-ID: <1525918138-6189-8-git-send-email-junyan.he@gmx.com> (raw)
In-Reply-To: <1525918138-6189-1-git-send-email-junyan.he@gmx.com>
From: Junyan He <junyan.he@intel.com>
When loading a compressed page to persistent memory, flush CPU cache
after the data is decompressed. Combined with a call to pmem_drain()
at the end of memory loading, we can guarantee those compressed pages
are persistently loaded to PMEM.
Signed-off-by: Haozhong Zhang <haozhong.zhang@intel.com>
---
include/qemu/pmem.h | 1 +
migration/ram.c | 10 ++++++++--
stubs/pmem.c | 4 ++++
3 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/include/qemu/pmem.h b/include/qemu/pmem.h
index cb9fa5f..c9140fb 100644
--- a/include/qemu/pmem.h
+++ b/include/qemu/pmem.h
@@ -20,6 +20,7 @@ void *pmem_memcpy_nodrain(void *pmemdest, const void *src, size_t len);
void *pmem_memcpy_persist(void *pmemdest, const void *src, size_t len);
void *pmem_memset_nodrain(void *pmemdest, int c, size_t len);
void pmem_drain(void);
+void pmem_flush(const void *addr, size_t len);
#endif /* CONFIG_LIBPMEM */
diff --git a/migration/ram.c b/migration/ram.c
index 2a180bc..e0f3dbc 100644
--- a/migration/ram.c
+++ b/migration/ram.c
@@ -286,6 +286,7 @@ struct DecompressParam {
uint8_t *compbuf;
int len;
z_stream stream;
+ bool is_pmem;
};
typedef struct DecompressParam DecompressParam;
@@ -2591,6 +2592,9 @@ static void *do_data_decompress(void *opaque)
error_report("decompress data failed");
qemu_file_set_error(decomp_file, ret);
}
+ if (param->is_pmem) {
+ pmem_flush(des, len);
+ }
qemu_mutex_lock(&decomp_done_lock);
param->done = true;
@@ -2702,7 +2706,8 @@ exit:
}
static void decompress_data_with_multi_threads(QEMUFile *f,
- void *host, int len)
+ void *host, int len,
+ bool is_pmem)
{
int idx, thread_count;
@@ -2716,6 +2721,7 @@ static void decompress_data_with_multi_threads(QEMUFile *f,
qemu_get_buffer(f, decomp_param[idx].compbuf, len);
decomp_param[idx].des = host;
decomp_param[idx].len = len;
+ decomp_param[idx].is_pmem = is_pmem;
qemu_cond_signal(&decomp_param[idx].cond);
qemu_mutex_unlock(&decomp_param[idx].mutex);
break;
@@ -3073,7 +3079,7 @@ static int ram_load(QEMUFile *f, void *opaque, int version_id)
ret = -EINVAL;
break;
}
- decompress_data_with_multi_threads(f, host, len);
+ decompress_data_with_multi_threads(f, host, len, is_pmem);
break;
case RAM_SAVE_FLAG_XBZRLE:
diff --git a/stubs/pmem.c b/stubs/pmem.c
index b50c35e..9e7d86a 100644
--- a/stubs/pmem.c
+++ b/stubs/pmem.c
@@ -31,3 +31,7 @@ void *pmem_memcpy_nodrain(void *pmemdest, const void *src, size_t len)
{
return memcpy(pmemdest, src, len);
}
+
+void pmem_flush(const void *addr, size_t len)
+{
+}
--
2.7.4
next prev parent reply other threads:[~2018-05-10 2:09 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-10 2:08 [Qemu-devel] [PATCH V5 0/9] nvdimm: guarantee persistence of QEMU writes to persistent memory junyan.he
2018-05-10 2:08 ` [Qemu-devel] [PATCH 1/9 V5] memory, exec: switch file ram allocation functions to 'flags' parameters junyan.he
2018-05-10 21:08 ` Murilo Opsfelder Araujo
2018-05-11 7:20 ` Junyan He
2018-05-31 12:46 ` Stefan Hajnoczi
2018-05-10 2:08 ` [Qemu-devel] [PATCH 2/9 V5] hostmem-file: add the 'pmem' option junyan.he
2018-05-31 12:35 ` Stefan Hajnoczi
2018-05-31 13:09 ` Stefan Hajnoczi
2018-06-01 5:14 ` Junyan He
2018-05-10 2:08 ` [Qemu-devel] [PATCH 3/9 V5] configure: add libpmem support junyan.he
2018-05-31 12:54 ` Stefan Hajnoczi
2018-05-10 2:08 ` [Qemu-devel] [PATCH 4/9 V5] mem/nvdimm: ensure write persistence to PMEM in label emulation junyan.he
2018-05-10 2:08 ` [Qemu-devel] [PATCH 5/9 V5] migration/ram: ensure write persistence on loading zero pages to PMEM junyan.he
2018-05-10 2:08 ` [Qemu-devel] [PATCH 6/9 V5] migration/ram: ensure write persistence on loading normal " junyan.he
2018-05-10 2:08 ` junyan.he [this message]
2018-05-10 2:08 ` [Qemu-devel] [PATCH 8/9 V5] migration/ram: ensure write persistence on loading xbzrle " junyan.he
2018-05-10 2:08 ` [Qemu-devel] [PATCH 9/9 V5] migration/ram: Add check and info message to nvdimm post copy junyan.he
2018-05-10 2:21 ` [Qemu-devel] [PATCH V5 0/9] nvdimm: guarantee persistence of QEMU writes to persistent memory He, Junyan
2018-05-10 2:22 ` no-reply
2018-05-21 3:19 ` Junyan He
2018-05-28 5:26 ` Junyan He
2018-05-31 13:18 ` Stefan Hajnoczi
2018-05-31 14:28 ` Dr. David Alan Gilbert
2018-05-31 14:36 ` Junyan He
2018-05-31 14:42 ` Dr. David Alan Gilbert
2018-05-31 15:04 ` Junyan He
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1525918138-6189-8-git-send-email-junyan.he@gmx.com \
--to=junyan.he@gmx.com \
--cc=crosthwaite.peter@gmail.com \
--cc=dgilbert@redhat.com \
--cc=ehabkost@redhat.com \
--cc=haozhong.zhang@intel.com \
--cc=imammedo@redhat.com \
--cc=junyan.he@intel.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.com \
--cc=rth@twiddle.net \
--cc=stefanha@redhat.com \
--cc=xiaoguangrong.eric@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).