From: Michael Roth <mdroth@linux.vnet.ibm.com>
To: Juan Quintela <quintela@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
Umesh Deshpande <udeshpan@redhat.com>,
qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 04/27] add a version number to ram_list
Date: Wed, 25 Jul 2012 18:27:57 -0500 [thread overview]
Message-ID: <20120725232757.GB2708@illuin> (raw)
In-Reply-To: <1343155012-26316-5-git-send-email-quintela@redhat.com>
On Tue, Jul 24, 2012 at 08:36:29PM +0200, Juan Quintela wrote:
> From: Umesh Deshpande <udeshpan@redhat.com>
>
> This will be used to detect if last_block might have become invalid
> across different calls to ram_save_live.
>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> Signed-off-by: Umesh Deshpande <udeshpan@redhat.com>
> Signed-off-by: Juan Quintela <quintela@redhat.com>
> ---
> arch_init.c | 6 ++++++
> cpu-all.h | 1 +
> exec.c | 4 ++++
> 3 files changed, 11 insertions(+)
>
> diff --git a/arch_init.c b/arch_init.c
> index 02d36ce..9e05aae 100644
> --- a/arch_init.c
> +++ b/arch_init.c
> @@ -340,6 +340,7 @@ static int save_xbzrle_page(QEMUFile *f, uint8_t *current_data,
>
> static RAMBlock *last_block;
> static ram_addr_t last_offset;
> +static uint32_t last_version;
>
> /*
> * ram_save_block: Writes a page of memory to the stream f
> @@ -496,6 +497,7 @@ static void reset_ram_globals(void)
> {
> last_block = NULL;
> last_offset = 0;
> + last_version = ram_list.version;
> sort_ram_list();
> }
>
> @@ -554,6 +556,10 @@ static int ram_save_iterate(QEMUFile *f, void *opaque)
> int i;
> uint64_t expected_time;
>
> + if (ram_list.version != last_version) {
> + reset_ram_globals();
> + }
> +
Was it a bug in the old code to not do this? In the unthreaded version
it looks like last_block could still become invalid between iterations if it
was removed in the meantime, so I'm trying to understand why it's needed
now.
And why do we re-sort when this happens? The old implementation didn't
seem to place any guarantees on the block ordering beyond the initial
sorting by block->idstr. After that they were sent in MRU order.
> bytes_transferred_last = bytes_transferred;
> bwidth = qemu_get_clock_ns(rt_clock);
>
> diff --git a/cpu-all.h b/cpu-all.h
> index ca3bb24..429b2c6 100644
> --- a/cpu-all.h
> +++ b/cpu-all.h
> @@ -486,6 +486,7 @@ typedef struct RAMBlock {
>
> typedef struct RAMList {
> uint8_t *phys_dirty;
> + uint32_t version;
> QLIST_HEAD(, RAMBlock) blocks_mru;
> QLIST_HEAD(, RAMBlock) blocks;
> uint64_t dirty_pages;
> diff --git a/exec.c b/exec.c
> index afc472f..ceffa4f 100644
> --- a/exec.c
> +++ b/exec.c
> @@ -2555,6 +2555,8 @@ ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
> QLIST_INSERT_HEAD(&ram_list.blocks, new_block, next);
> QLIST_INSERT_HEAD(&ram_list.blocks_mru, new_block, next_mru);
>
> + ram_list.version++;
> +
> ram_list.phys_dirty = g_realloc(ram_list.phys_dirty,
> last_ram_offset() >> TARGET_PAGE_BITS);
> cpu_physical_memory_set_dirty_range(new_block->offset, size, 0xff);
> @@ -2578,6 +2580,7 @@ void qemu_ram_free_from_ptr(ram_addr_t addr)
> if (addr == block->offset) {
> QLIST_REMOVE(block, next);
> QLIST_REMOVE(block, next_mru);
> + ram_list.version++;
> g_free(block);
> return;
> }
> @@ -2592,6 +2595,7 @@ void qemu_ram_free(ram_addr_t addr)
> if (addr == block->offset) {
> QLIST_REMOVE(block, next);
> QLIST_REMOVE(block, next_mru);
> + ram_list.version++;
> if (block->flags & RAM_PREALLOC_MASK) {
> ;
> } else if (mem_path) {
> --
> 1.7.10.4
>
>
next prev parent reply other threads:[~2012-07-25 23:28 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-07-24 18:36 [Qemu-devel] [RFC 00/27] Migration thread (WIP) Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 01/27] buffered_file: g_realloc() can't fail Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 02/27] split MRU ram list Juan Quintela
2012-07-25 20:20 ` Michael Roth
2012-07-26 13:19 ` Avi Kivity
2012-07-24 18:36 ` [Qemu-devel] [PATCH 03/27] savevm: Factorize ram globals reset in its own function Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 04/27] add a version number to ram_list Juan Quintela
2012-07-25 23:27 ` Michael Roth [this message]
2012-07-26 9:19 ` Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 05/27] protect the ramlist with a separate mutex Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 06/27] ram: introduce migration_bitmap_set_dirty() Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 07/27] ram: Introduce migration_bitmap_test_and_reset_dirty() Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 08/27] ram: Export last_ram_offset() Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 09/27] ram: introduce migration_bitmap_sync() Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 10/27] Separate migration bitmap Juan Quintela
2012-07-25 9:16 ` Avi Kivity
2012-07-26 9:22 ` Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 11/27] BufferedFile: append, then flush Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 12/27] buffered_file: rename opaque to migration_state Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 13/27] buffered_file: opaque is MigrationState Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 14/27] buffered_file: unfold migrate_fd_put_buffer Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 15/27] buffered_file: unfold migrate_fd_put_ready Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 16/27] buffered_file: unfold migrate_fd_put_buffer Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 17/27] " Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 18/27] buffered_file: We can access directly to bandwidth_limit Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 19/27] buffered_file: Move from using a timer to use a thread Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 20/27] migration: make qemu_fopen_ops_buffered() return void Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 21/27] migration: stop all cpus correctly Juan Quintela
2012-07-26 12:54 ` Eric Blake
2012-07-24 18:36 ` [Qemu-devel] [PATCH 22/27] migration: make writes blocking Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 23/27] migration: remove unfreeze logic Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 24/27] migration: take finer locking Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 25/27] buffered_file: Unfold the trick to restart generating migration data Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 26/27] buffered_file: don't flush on put buffer Juan Quintela
2012-07-24 18:36 ` [Qemu-devel] [PATCH 27/27] buffered_file: unfold buffered_append in buffered_put_buffer Juan Quintela
2012-07-25 9:55 ` [Qemu-devel] [RFC 00/27] Migration thread (WIP) Orit Wasserman
2012-07-26 10:57 ` Jan Kiszka
2012-07-26 11:16 ` Juan Quintela
2012-07-26 11:56 ` Jan Kiszka
[not found] ` <500EF579.5040607@redhat.com>
2012-07-26 18:41 ` [Qemu-devel] Fwd: " Chegu Vinod
2012-07-26 21:26 ` Chegu Vinod
2012-07-27 11:05 ` Juan Quintela
[not found] ` <4168C988EBDF2141B4E0B6475B6A73D1165CDD@G6W2493.americas.hpqcorp.net>
2012-07-27 14:21 ` [Qemu-devel] FW: " Chegu Vinod
2012-07-26 21:36 ` [Qemu-devel] " Michael Roth
2012-08-02 12:01 ` Juan Quintela
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=20120725232757.GB2708@illuin \
--to=mdroth@linux.vnet.ibm.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.com \
--cc=udeshpan@redhat.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).