From: Greg Kurz <groug@kaod.org>
To: David Gibson <david@gibson.dropbear.id.au>
Cc: surajjs@au1.ibm.com, lvivier@redhat.com, qemu-ppc@nongnu.org,
qemu-devel@nongnu.org, mdroth@linux.vnet.ibm.com,
abologna@redhat.com
Subject: Re: [Qemu-devel] [PATCH 3/6] spapr: Validate capabilities on migration
Date: Mon, 18 Dec 2017 12:31:26 +0100 [thread overview]
Message-ID: <20171218123126.7580bb28@bahia.lan> (raw)
In-Reply-To: <20171218092024.21645-4-david@gibson.dropbear.id.au>
On Mon, 18 Dec 2017 20:20:21 +1100
David Gibson <david@gibson.dropbear.id.au> wrote:
> Now that the "pseries" machine type implements optional capabilities (well,
> one so far) there's the possibility of having different capabilities
> available at either end of a migration. Although arguably a user error,
> it would be nice to catch this situation and fail as gracefully as we can.
>
> This adds code to migrate the capabilities flags. These aren't pulled
> directly into the destination's configuration since what the user has
> specified on the destination command line should take precedence. However,
> they are checked against the destination capabilities.
>
> If the source was using a capability which is absent on the destination,
> we fail the migration, since that could easily cause a guest crash or other
> bad behaviour. If the source lacked a capability which is present on the
> destination we warn, but allow the migration to proceed.
>
> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
> ---
Reviewed-by: Greg Kurz <groug@kaod.org>
> hw/ppc/spapr.c | 6 ++++
> hw/ppc/spapr_caps.c | 96 ++++++++++++++++++++++++++++++++++++++++++++++++--
> include/hw/ppc/spapr.h | 6 ++++
> 3 files changed, 105 insertions(+), 3 deletions(-)
>
> diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
> index f8fee8ebcf..86fc83f9c2 100644
> --- a/hw/ppc/spapr.c
> +++ b/hw/ppc/spapr.c
> @@ -1589,6 +1589,11 @@ static int spapr_post_load(void *opaque, int version_id)
> sPAPRMachineState *spapr = (sPAPRMachineState *)opaque;
> int err = 0;
>
> + err = spapr_caps_post_migration(spapr);
> + if (err) {
> + return err;
> + }
> +
> if (!object_dynamic_cast(OBJECT(spapr->ics), TYPE_ICS_KVM)) {
> CPUState *cs;
> CPU_FOREACH(cs) {
> @@ -1755,6 +1760,7 @@ static const VMStateDescription vmstate_spapr = {
> &vmstate_spapr_ov5_cas,
> &vmstate_spapr_patb_entry,
> &vmstate_spapr_pending_events,
> + &vmstate_spapr_caps,
> NULL
> }
> };
> diff --git a/hw/ppc/spapr_caps.c b/hw/ppc/spapr_caps.c
> index 2f0ef98670..a8c726a88b 100644
> --- a/hw/ppc/spapr_caps.c
> +++ b/hw/ppc/spapr_caps.c
> @@ -22,6 +22,7 @@
> * THE SOFTWARE.
> */
> #include "qemu/osdep.h"
> +#include "qemu/error-report.h"
> #include "qapi/error.h"
> #include "qapi/visitor.h"
> #include "sysemu/hw_accel.h"
> @@ -83,6 +84,93 @@ static sPAPRCapabilities default_caps_with_cpu(sPAPRMachineState *spapr,
> return caps;
> }
>
> +static bool spapr_caps_needed(void *opaque)
> +{
> + sPAPRMachineState *spapr = opaque;
> +
> + return (spapr->forced_caps.mask != 0) || (spapr->forbidden_caps.mask != 0);
> +}
> +
> +/* This has to be called from the top-level spapr post_load, not the
> + * caps specific one. Otherwise it wouldn't be called when the source
> + * caps are all defaults, which could still conflict with overridden
> + * caps on the destination */
> +int spapr_caps_post_migration(sPAPRMachineState *spapr)
> +{
> + uint64_t allcaps = 0;
> + int i;
> + bool ok = true;
> + sPAPRCapabilities dstcaps = spapr->effective_caps;
> + sPAPRCapabilities srccaps;
> +
> + srccaps = default_caps_with_cpu(spapr, first_cpu);
> + srccaps.mask |= spapr->mig_forced_caps.mask;
> + srccaps.mask &= ~spapr->mig_forbidden_caps.mask;
> +
> + for (i = 0; i < ARRAY_SIZE(capability_table); i++) {
> + sPAPRCapabilityInfo *info = &capability_table[i];
> +
> + allcaps |= info->flag;
> +
> + if ((srccaps.mask & info->flag) && !(dstcaps.mask & info->flag)) {
> + error_report("cap-%s=on in incoming stream, but off in destination",
> + info->name);
> + ok = false;
> + }
> +
> + if (!(srccaps.mask & info->flag) && (dstcaps.mask & info->flag)) {
> + warn_report("cap-%s=off in incoming stream, but on in destination",
> + info->name);
> + }
> + }
> +
> + if (spapr->mig_forced_caps.mask & ~allcaps) {
> + error_report(
> + "Unknown capabilities 0x%"PRIx64" enabled in incoming stream",
> + spapr->mig_forced_caps.mask & ~allcaps);
> + ok = false;
> + }
> + if (spapr->mig_forbidden_caps.mask & ~allcaps) {
> + warn_report(
> + "Unknown capabilities 0x%"PRIx64" disabled in incoming stream",
> + spapr->mig_forbidden_caps.mask & ~allcaps);
> + }
> +
> + return ok ? 0 : -EINVAL;
> +}
> +
> +static int spapr_caps_pre_save(void *opaque)
> +{
> + sPAPRMachineState *spapr = opaque;
> +
> + spapr->mig_forced_caps = spapr->forced_caps;
> + spapr->mig_forbidden_caps = spapr->forbidden_caps;
> + return 0;
> +}
> +
> +static int spapr_caps_pre_load(void *opaque)
> +{
> + sPAPRMachineState *spapr = opaque;
> +
> + spapr->mig_forced_caps = spapr_caps(0);
> + spapr->mig_forbidden_caps = spapr_caps(0);
> + return 0;
> +}
> +
> +const VMStateDescription vmstate_spapr_caps = {
> + .name = "spapr/caps",
> + .version_id = 1,
> + .minimum_version_id = 1,
> + .needed = spapr_caps_needed,
> + .pre_save = spapr_caps_pre_save,
> + .pre_load = spapr_caps_pre_load,
> + .fields = (VMStateField[]) {
> + VMSTATE_UINT64(mig_forced_caps.mask, sPAPRMachineState),
> + VMSTATE_UINT64(mig_forbidden_caps.mask, sPAPRMachineState),
> + VMSTATE_END_OF_LIST()
> + },
> +};
> +
> void spapr_caps_reset(sPAPRMachineState *spapr)
> {
> Error *local_err = NULL;
> @@ -92,6 +180,11 @@ void spapr_caps_reset(sPAPRMachineState *spapr)
> /* First compute the actual set of caps we're running with.. */
> caps = default_caps_with_cpu(spapr, first_cpu);
>
> + /* Remove unnecessary forced/forbidden bits (this will help us
> + * with migration) */
> + spapr->forced_caps.mask &= ~caps.mask;
> + spapr->forbidden_caps.mask &= caps.mask;
> +
> caps.mask |= spapr->forced_caps.mask;
> caps.mask &= ~spapr->forbidden_caps.mask;
>
> @@ -176,9 +269,6 @@ void spapr_caps_validate(sPAPRMachineState *spapr, Error **errp)
> error_setg(errp, "Some sPAPR capabilities set both on and off");
> return;
> }
> -
> - /* Check for any caps incompatible with other caps. Nothing to do
> - * yet */
> }
>
> void spapr_caps_add_properties(sPAPRMachineClass *smc, Error **errp)
> diff --git a/include/hw/ppc/spapr.h b/include/hw/ppc/spapr.h
> index dc64f4ebcb..5c85f39c3b 100644
> --- a/include/hw/ppc/spapr.h
> +++ b/include/hw/ppc/spapr.h
> @@ -54,6 +54,8 @@ typedef enum {
> * Capabilities
> */
>
> +/* These bits go in the migration stream, so they can't be reassigned */
> +
> /* Hardware Transactional Memory */
> #define SPAPR_CAP_HTM 0x0000000000000001ULL
>
> @@ -142,6 +144,7 @@ struct sPAPRMachineState {
> const char *icp_type;
>
> sPAPRCapabilities forced_caps, forbidden_caps;
> + sPAPRCapabilities mig_forced_caps, mig_forbidden_caps;
> sPAPRCapabilities effective_caps;
> };
>
> @@ -743,6 +746,8 @@ qemu_irq spapr_qirq(sPAPRMachineState *spapr, int irq);
> /*
> * Handling of optional capabilities
> */
> +extern const VMStateDescription vmstate_spapr_caps;
> +
> static inline sPAPRCapabilities spapr_caps(uint64_t mask)
> {
> sPAPRCapabilities caps = { mask };
> @@ -757,5 +762,6 @@ static inline bool spapr_has_cap(sPAPRMachineState *spapr, uint64_t cap)
> void spapr_caps_reset(sPAPRMachineState *spapr);
> void spapr_caps_validate(sPAPRMachineState *spapr, Error **errp);
> void spapr_caps_add_properties(sPAPRMachineClass *smc, Error **errp);
> +int spapr_caps_post_migration(sPAPRMachineState *spapr);
>
> #endif /* HW_SPAPR_H */
next prev parent reply other threads:[~2017-12-18 11:31 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-18 9:20 [Qemu-devel] [PATCH 0/6] spapr: Add optional capabilities David Gibson
2017-12-18 9:20 ` [Qemu-devel] [PATCH 1/6] spapr: Capabilities infrastructure David Gibson
2017-12-18 9:58 ` Greg Kurz
2017-12-18 10:15 ` David Gibson
2017-12-18 9:20 ` [Qemu-devel] [PATCH 2/6] spapr: Treat Hardware Transactional Memory (HTM) as an optional capability David Gibson
2017-12-18 11:10 ` Greg Kurz
2017-12-19 0:35 ` David Gibson
2017-12-18 9:20 ` [Qemu-devel] [PATCH 3/6] spapr: Validate capabilities on migration David Gibson
2017-12-18 11:31 ` Greg Kurz [this message]
2017-12-18 9:20 ` [Qemu-devel] [PATCH 4/6] target/ppc: Clean up probing of VMX, VSX and DFP availability on KVM David Gibson
2017-12-18 11:45 ` Greg Kurz
2017-12-18 9:20 ` [Qemu-devel] [PATCH 5/6] spapr: Handle VMX/VSX presence as an spapr capability flag David Gibson
2017-12-18 11:47 ` Greg Kurz
2017-12-18 9:20 ` [Qemu-devel] [PATCH 6/6] spapr: Handle Decimal Floating Point (DFP) as an optional capability David Gibson
2017-12-18 11:51 ` Greg Kurz
2017-12-19 0:37 ` [Qemu-devel] [PATCH 0/6] spapr: Add optional capabilities David Gibson
2017-12-20 22:32 ` Benjamin Herrenschmidt
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=20171218123126.7580bb28@bahia.lan \
--to=groug@kaod.org \
--cc=abologna@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=lvivier@redhat.com \
--cc=mdroth@linux.vnet.ibm.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
--cc=surajjs@au1.ibm.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).