From: Igor Mitsyanko <i.mitsyanko@samsung.com>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: Juan Quintela <quintela@redhat.com>,
qemu-devel@nongnu.org, patches@linaro.org
Subject: Re: [Qemu-devel] [PATCH] vmstate: Add support for saving/loading bitmaps
Date: Fri, 10 Aug 2012 20:22:15 +0400 [thread overview]
Message-ID: <50253537.1070003@samsung.com> (raw)
In-Reply-To: <1344513263-22971-1-git-send-email-peter.maydell@linaro.org>
On 08/09/2012 03:54 PM, Peter Maydell wrote:
> Add support for saving/loading bitmap.h bitmaps in vmstate.
>
> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
> ---
> This will be needed for saving/restoring the bitmap in sd.c which
> is introduced by Igor's latest patchset; the relevant VMSTATE line is:
> VMSTATE_BITMAP(wp_groups, SDState, 1, wpgrps_size),
> (and you'll need to make wpgrps_size an int32_t, not uint32_t).
>
> Igor: I've only tested this fairly lightly, you'll probably want to
> do things like testing save on 32 bit and load on 64 bit and
> vice-versa.
>
> savevm.c | 41 +++++++++++++++++++++++++++++++++++++++++
> vmstate.h | 13 +++++++++++++
> 2 files changed, 54 insertions(+)
>
> diff --git a/savevm.c b/savevm.c
> index 6e82b2d..0e2de97 100644
> --- a/savevm.c
> +++ b/savevm.c
> @@ -86,6 +86,7 @@
> #include "memory.h"
> #include "qmp-commands.h"
> #include "trace.h"
> +#include "bitops.h"
>
> #define SELF_ANNOUNCE_ROUNDS 5
>
> @@ -1159,6 +1160,46 @@ const VMStateInfo vmstate_info_unused_buffer = {
> .put = put_unused_buffer,
> };
>
> +/* bitmaps (as defined by bitmap.h). Note that size here is the size
> + * of the bitmap in bits. The on-the-wire format of a bitmap is 64
> + * bit words with the bits in big endian order. The in-memory format
> + * is an array of 'unsigned long', which may be either 32 or 64 bits.
> + */
> +/* This is the number of 64 bit words sent over the wire */
> +#define BITS_TO_U64S(nr) DIV_ROUND_UP(nr, 64)
> +static int get_bitmap(QEMUFile *f, void *pv, size_t size)
> +{
> + unsigned long *bmp = pv;
> + int i, idx = 0;
> + for (i = 0; i < BITS_TO_U64S(size); i++) {
> + uint64_t w = qemu_get_be64(f);
> + bmp[idx++] = w;
> + if (sizeof(unsigned long) == 4 && idx < BITS_TO_LONGS(size)) {
> + bmp[idx++] = w >> 32;
> + }
> + }
> + return 0;
> +}
> +
> +static void put_bitmap(QEMUFile *f, void *pv, size_t size)
> +{
> + unsigned long *bmp = pv;
> + int i, idx = 0;
> + for (i = 0; i < BITS_TO_U64S(size); i++) {
> + uint64_t w = bmp[idx++];
> + if (sizeof(unsigned long) == 4 && idx < BITS_TO_LONGS(size)) {
> + w |= ((uint64_t)bmp[idx++]) << 32;
> + }
> + qemu_put_be64(f, w);
> + }
> +}
> +
> +const VMStateInfo vmstate_info_bitmap = {
> + .name = "bitmap",
> + .get = get_bitmap,
> + .put = put_bitmap,
> +};
> +
> typedef struct CompatEntry {
> char idstr[256];
> int instance_id;
> diff --git a/vmstate.h b/vmstate.h
> index 5bd2b76..c45f46e 100644
> --- a/vmstate.h
> +++ b/vmstate.h
> @@ -139,6 +139,7 @@ extern const VMStateInfo vmstate_info_uint64;
> 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;
>
> #define type_check_array(t1,t2,n) ((t1(*)[n])0 - (t2*)0)
> #define type_check_pointer(t1,t2) ((t1**)0 - (t2*)0)
> @@ -411,6 +412,18 @@ extern const VMStateInfo vmstate_info_unused_buffer;
> .flags = VMS_BUFFER, \
> }
>
> +/* _field_size should be a uint32_t field in the _state struct giving the
"..should be an int32_t.."
> + * size of the bitmap _field in bits.
> + */
> +#define VMSTATE_BITMAP(_field, _state, _version, _field_size) { \
> + .name = (stringify(_field)), \
> + .version_id = (_version), \
> + .size_offset = vmstate_offset_value(_state, _field_size, int32_t),\
> + .info = &vmstate_info_bitmap, \
> + .flags = VMS_VBUFFER|VMS_POINTER, \
> + .offset = offsetof(_state, _field), \
> +}
> +
> /* _f : field name
> _f_n : num of elements field_name
> _n : num of elements
I've successfully tested this patch with migration from 32bit to 64bit
little endian host and vice versa. Haven’t tested with
bigendian-littleendian migration since I don't have a bigendian machine
at my disposal.
Tested-by: Igor Mitsyanko <i.mitsyanko@samsung.com>
next prev parent reply other threads:[~2012-08-10 16:22 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-08-09 11:54 [Qemu-devel] [PATCH] vmstate: Add support for saving/loading bitmaps Peter Maydell
2012-08-10 16:22 ` Igor Mitsyanko [this message]
2012-08-10 16:30 ` Peter Maydell
2012-08-10 16:35 ` Igor Mitsyanko
2012-08-13 11:12 ` 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=50253537.1070003@samsung.com \
--to=i.mitsyanko@samsung.com \
--cc=patches@linaro.org \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=quintela@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).