From: Alexander Graf <agraf@suse.de>
To: Alexey Kardashevskiy <aik@ozlabs.ru>, qemu-devel@nongnu.org
Cc: qemu-ppc@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 1/2] vmstate: Add helper to enable GHashTable migration
Date: Thu, 22 May 2014 12:57:31 +0200 [thread overview]
Message-ID: <537DD81B.4080007@suse.de> (raw)
In-Reply-To: <1400756006-26297-2-git-send-email-aik@ozlabs.ru>
On 22.05.14 12:53, Alexey Kardashevskiy wrote:
> This adds a VMSTATE_HASH_V macro. This implements put/get callbacks for it.
> This implements a qemu_hash_init() wrapper to save key/value sizes.
>
> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
> ---
> include/migration/vmstate.h | 10 +++++++++
> include/qemu-common.h | 13 +++++++++++
> vmstate.c | 54 +++++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 77 insertions(+)
>
> diff --git a/include/migration/vmstate.h b/include/migration/vmstate.h
> index 7e45048..6af599d 100644
> --- a/include/migration/vmstate.h
> +++ b/include/migration/vmstate.h
> @@ -166,6 +166,7 @@ extern const VMStateInfo vmstate_info_timer;
> extern const VMStateInfo vmstate_info_buffer;
> extern const VMStateInfo vmstate_info_unused_buffer;
> extern const VMStateInfo vmstate_info_bitmap;
> +static const VMStateInfo vmstate_info_hash;
>
> #define type_check_2darray(t1,t2,n,m) ((t1(*)[n][m])0 - (t2*)0)
> #define type_check_array(t1,t2,n) ((t1(*)[n])0 - (t2*)0)
> @@ -740,6 +741,15 @@ extern const VMStateInfo vmstate_info_bitmap;
> #define VMSTATE_BUFFER_UNSAFE(_field, _state, _version, _size) \
> VMSTATE_BUFFER_UNSAFE_INFO(_field, _state, _version, vmstate_info_buffer, _size)
>
> +#define VMSTATE_HASH_V(_field, _state, _version) { \
> + .name = (stringify(_field)), \
> + .version_id = (_version), \
> + .size = sizeof(qemu_hash), \
> + .info = &vmstate_info_hash, \
> + .flags = VMS_SINGLE, \
> + .offset = vmstate_offset_value(_state, _field, qemu_hash), \
> +}
> +
> #define VMSTATE_UNUSED_V(_v, _size) \
> VMSTATE_UNUSED_BUFFER(NULL, _v, _size)
>
> diff --git a/include/qemu-common.h b/include/qemu-common.h
> index 3f3fd60..ee973e7 100644
> --- a/include/qemu-common.h
> +++ b/include/qemu-common.h
> @@ -462,4 +462,17 @@ int parse_debug_env(const char *name, int max, int initial);
>
> const char *qemu_ether_ntoa(const MACAddr *mac);
>
> +typedef struct qemu_hash {
> + GHashTable *hash;
> + int key_size;
> + int value_size;
> +} qemu_hash;
> +
> +static inline void qemu_hash_init(qemu_hash *h, int key_size, int value_size)
> +{
> + h->key_size = key_size;
> + h->value_size = value_size;
> + h->hash = g_hash_table_new_full(g_int_hash, g_int_equal, g_free, g_free);
> +}
> +
> #endif
> diff --git a/vmstate.c b/vmstate.c
> index b5882fa..2148e73 100644
> --- a/vmstate.c
> +++ b/vmstate.c
> @@ -667,3 +667,57 @@ const VMStateInfo vmstate_info_bitmap = {
> .get = get_bitmap,
> .put = put_bitmap,
> };
> +
> +/* Save restore for qemu_hash which is a wrapper over GHashTable */
> +static int get_hash(QEMUFile *f, void *pv, size_t size)
> +{
> + qemu_hash *h = pv;
> + uint32_t num = g_hash_table_size(h->hash);
> + gpointer key, value;
> +
> + num = qemu_get_be32(f);
> +
> + for ( ; num; --num) {
> + int i;
> +
> + key = g_malloc0(h->key_size);
> + for (i = 0; i < h->key_size; ++i) {
> + ((uint8_t *)key)[i] = qemu_get_byte(f);
> + }
> + value = g_malloc0(h->value_size);
> + for (i = 0; i < h->value_size; ++i) {
> + ((uint8_t *)value)[i] = qemu_get_byte(f);
> + }
> + g_hash_table_insert(h->hash, key, value);
> + }
> +
> + return 0;
> +}
> +
> +static void put_hash(QEMUFile *f, void *pv, size_t size)
> +{
> + qemu_hash *h = pv;
> + uint32_t num = g_hash_table_size(h->hash);
> + GHashTableIter iter;
> + gpointer key, value;
> +
> + qemu_put_be32(f, num);
> +
> + g_hash_table_iter_init(&iter, h->hash);
> + while (g_hash_table_iter_next (&iter, &key, &value)) {
> + int i;
> +
> + for (i = 0; i < h->key_size; ++i) {
> + qemu_put_byte(f, ((uint8_t *)key)[i]);
> + }
> + for (i = 0; i < h->value_size; ++i) {
> + qemu_put_byte(f, ((uint8_t *)value)[i]);
I think the only way to do this safely would be to do a recursive
introspective walk through the hash table's key and value properties.
Every key and every property can be an object again, right?
We would basically impose a limit onto ourselves that every member of
the hash table would have to be a GObject again.
Alex
next prev parent reply other threads:[~2014-05-22 10:57 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-15 9:59 [Qemu-devel] [PATCH v2 0/8] Move interrupts from spapr to xics Alexey Kardashevskiy
2014-05-15 9:59 ` [Qemu-devel] [PATCH v2 1/8] xics: Add flags for interrupts Alexey Kardashevskiy
2014-05-21 6:30 ` Alexey Kardashevskiy
2014-05-21 8:40 ` Alexander Graf
2014-05-15 9:59 ` [Qemu-devel] [PATCH v2 2/8] xics: Add xics_find_source() Alexey Kardashevskiy
2014-05-15 9:59 ` [Qemu-devel] [PATCH v2 3/8] xics: Disable flags reset on xics reset Alexey Kardashevskiy
2014-05-15 9:59 ` [Qemu-devel] [PATCH v2 4/8] spapr: Move interrupt allocator to xics Alexey Kardashevskiy
2014-05-21 8:34 ` Alexander Graf
2014-05-21 8:46 ` Alexey Kardashevskiy
2014-05-21 8:47 ` Alexander Graf
2014-05-15 9:59 ` [Qemu-devel] [PATCH v2 5/8] xics: Remove obsolete xics_set_irq_type() Alexey Kardashevskiy
2014-05-15 9:59 ` [Qemu-devel] [PATCH v2 6/8] spapr: Remove @next_irq Alexey Kardashevskiy
2014-05-15 9:59 ` [Qemu-devel] [PATCH v2 7/8] xics: Implement xics_ics_free() Alexey Kardashevskiy
2014-05-15 9:59 ` [Qemu-devel] [PATCH v2 8/8] spapr_pci: Use XICS interrupt allocator and do not cache interrupts in PHB Alexey Kardashevskiy
2014-05-21 8:40 ` Alexander Graf
2014-05-21 8:52 ` Alexey Kardashevskiy
2014-05-21 9:06 ` Alexander Graf
2014-05-21 9:11 ` Michael S. Tsirkin
2014-05-21 9:13 ` Alexander Graf
2014-05-21 9:33 ` Alexey Kardashevskiy
2014-05-21 9:38 ` Michael S. Tsirkin
2014-05-21 10:03 ` Alexey Kardashevskiy
2014-05-21 9:50 ` Alexander Graf
2014-05-21 10:13 ` Alexey Kardashevskiy
2014-05-21 10:35 ` Alexander Graf
2014-05-21 12:42 ` Alexey Kardashevskiy
2014-05-22 6:53 ` Alexey Kardashevskiy
2014-05-22 7:16 ` Alexander Graf
2014-05-22 10:53 ` Alexey Kardashevskiy
2014-05-22 10:53 ` [Qemu-devel] [PATCH 1/2] vmstate: Add helper to enable GHashTable migration Alexey Kardashevskiy
2014-05-22 10:57 ` Alexander Graf [this message]
2014-05-22 11:04 ` Alexey Kardashevskiy
2014-05-22 10:53 ` [Qemu-devel] [PATCH 2/2] spapr_pci: Use XICS interrupt allocator and do not cache interrupts in PHB Alexey Kardashevskiy
2014-05-22 10:57 ` [Qemu-devel] [PATCH v2 8/8] " Alexander Graf
2014-05-22 14:25 ` Alexey Kardashevskiy
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=537DD81B.4080007@suse.de \
--to=agraf@suse.de \
--cc=aik@ozlabs.ru \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
/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).