| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <c709b37c-e8ce-59ba-1283-d7f0d7f3265f@linux.vnet.ibm.com>
Date: Wed, 28 Feb 2018 13:50:48 -0500
From: Tony Krowiak <akrowiak@...ux.vnet.ibm.com>
To: linux-s390@...r.kernel.org, linux-kernel@...r.kernel.org,
kvm@...r.kernel.org
Cc: freude@...ibm.com, schwidefsky@...ibm.com,
heiko.carstens@...ibm.com, borntraeger@...ibm.com,
cohuck@...hat.com, kwankhede@...dia.com,
bjsdjshi@...ux.vnet.ibm.com, pbonzini@...hat.com,
alex.williamson@...hat.com, pmorel@...ux.vnet.ibm.com,
alifm@...ux.vnet.ibm.com, mjrosato@...ux.vnet.ibm.com,
jjherne@...ux.vnet.ibm.com, thuth@...hat.com,
pasic@...ux.vnet.ibm.com, fiuczy@...ux.vnet.ibm.com,
buendgen@...ibm.com
Subject: Re: [PATCH v2 07/15] KVM: s390: Interfaces to configure/deconfigure
guest's AP matrix
On 02/27/2018 09:28 AM, Tony Krowiak wrote:
> Provides interfaces to assign AP adapters, usage domains
> and control domains to a KVM guest.
>
> A KVM guest is started by executing the Start Interpretive Execution (SIE)
> instruction. The SIE state description is a control block that contains the
> state information for a KVM guest and is supplied as input to the SIE
> instruction. The SIE state description has a satellite structure called the
> Crypto Control Block (CRYCB). The CRYCB contains three bitmask fields
> identifying the adapters, queues (domains) and control domains assigned to
> the KVM guest:
>
> * The AP Adapter Mask (APM) field identifies the AP adapters assigned to
> the KVM guest
>
> * The AP Queue Mask (AQM) field identifies the AP queues assigned to
> the KVM guest. Each AP queue is connected to a usage domain within
> an AP adapter.
>
> * The AP Domain Mask (ADM) field identifies the control domains
> assigned to the KVM guest.
>
> Each adapter, queue (usage domain) and control domain are identified by
> a number from 0 to 255. The bits in each mask, from most significant to
> least significant bit, correspond to the numbers 0-255. When a bit is
> set, the corresponding adapter, queue (usage domain) or control domain
> is assigned to the KVM guest.
>
> This patch will set the bits in the APM, AQM and ADM fields of the
> CRYCB referenced by the KVM guest's SIE state description. The process
> used is:
>
> 1. Verify that the bits to be set do not exceed the maximum bit
> number for the given mask.
>
> 2. Verify that the APQNs that can be derived from the intersection
> of the bits set in the APM and AQM fields of the KVM guest's CRYCB
> are not assigned to any other KVM guest running on the same linux
> host.
>
> 3. Set the APM, AQM and ADM in the CRYCB according to the matrix
> configured for the mediated matrix device via its sysfs
> adapter, domain and control domain attribute files respectively.
>
> Signed-off-by: Tony Krowiak <akrowiak@...ux.vnet.ibm.com>
> ---
> arch/s390/include/asm/kvm-ap.h | 36 +++++
> arch/s390/kvm/kvm-ap.c | 257 +++++++++++++++++++++++++++++++++
> drivers/s390/crypto/vfio_ap_ops.c | 19 +++
> drivers/s390/crypto/vfio_ap_private.h | 4 +
> 4 files changed, 316 insertions(+), 0 deletions(-)
>
> diff --git a/arch/s390/include/asm/kvm-ap.h b/arch/s390/include/asm/kvm-ap.h
> index ef749e7..46e7c5b 100644
> --- a/arch/s390/include/asm/kvm-ap.h
> +++ b/arch/s390/include/asm/kvm-ap.h
> @@ -10,9 +10,45 @@
> #define _ASM_KVM_AP
> #include <linux/types.h>
> #include <linux/kvm_host.h>
> +#include <linux/types.h>
> +#include <linux/kvm_host.h>
> +#include <linux/bitops.h>
> +
> +#define KVM_AP_MASK_BYTES(n)(n / BITS_PER_BYTE)
I don't know how I missed it, but there is a BITS_TO_BYTES macro in
linux/bitops.h.
It makes no sense to reinvent the wheel. Also, the above will only produce
a valid value if (n % BITS_PER_BYTE) == 0. That would most likely hold
true for
the cases in which the macro is used, but the BITS_TO_BYTES macro
compensates for
the case where n is not evenly divisible by 8.
> +
> +/**
> + * The AP matrix is comprised of three bit masks identifying the adapters,
> + * queues (domains) and control domains that belong to an AP matrix. The bits in
> + * each mask, from least significant to most significant bit, correspond to IDs
> + * 0 to the maximum ID allowed for a given mask. When a bit is set, the
> + * corresponding ID belongs to the matrix.
> + *
> + * @apm_max: max number of bits in @apm
> + * @apm identifies the AP adapters in the matrix
> + * @aqm_max: max number of bits in @aqm
> + * @aqm identifies the AP queues (domains) in the matrix
> + * @adm_max: max number of bits in @adm
> + * @adm identifies the AP control domains in the matrix
> + */
> +struct kvm_ap_matrix {
> + int apm_max;
> + unsigned long *apm;
> + int aqm_max;
> + unsigned long *aqm;
> + int adm_max;
> + unsigned long *adm;
> +};
>
> void kvm_ap_set_crycb_format(struct kvm *kvm, __u32 *crycbd);
>
> int kvm_ap_get_crycb_format(struct kvm *kvm);
>
> +int kvm_ap_matrix_create(struct kvm_ap_matrix **ap_matrix);
> +
> +void kvm_ap_matrix_destroy(struct kvm_ap_matrix *ap_matrix);
> +
> +int kvm_ap_configure_matrix(struct kvm *kvm, struct kvm_ap_matrix *matrix);
> +
> +void kvm_ap_deconfigure_matrix(struct kvm *kvm);
> +
> #endif /* _ASM_KVM_AP */
> diff --git a/arch/s390/kvm/kvm-ap.c b/arch/s390/kvm/kvm-ap.c
> index bafe63b..bb29045 100644
> --- a/arch/s390/kvm/kvm-ap.c
> +++ b/arch/s390/kvm/kvm-ap.c
> @@ -8,6 +8,7 @@
>
> #include <asm/kvm-ap.h>
> #include <asm/ap.h>
> +#include <linux/bitops.h>
>
> #include "kvm-s390.h"
>
> @@ -16,6 +17,125 @@ int kvm_ap_get_crycb_format(struct kvm *kvm)
> return kvm->arch.crypto.crycbd & CRYCB_FORMAT_MASK;
> }
>
> +static inline void kvm_ap_clear_crycb_masks(struct kvm *kvm)
> +{
> + int crycb_fmt = kvm_ap_get_crycb_format(kvm);
> +
> + if (crycb_fmt == CRYCB_FORMAT2)
> + memset(&kvm->arch.crypto.crycb->apcb1, 0,
> + sizeof(kvm->arch.crypto.crycb->apcb1));
> + else
> + memset(&kvm->arch.crypto.crycb->apcb0, 0,
> + sizeof(kvm->arch.crypto.crycb->apcb0));
> +}
> +
> +static inline unsigned long *kvm_ap_get_crycb_apm(struct kvm *kvm)
> +{
> + unsigned long *apm;
> + int crycb_fmt = kvm_ap_get_crycb_format(kvm);
> +
> + if (crycb_fmt == CRYCB_FORMAT2)
> + apm = (unsigned long *)kvm->arch.crypto.crycb->apcb1.apm;
> + else
> + apm = (unsigned long *)kvm->arch.crypto.crycb->apcb0.apm;
> +
> + return apm;
> +}
> +
> +static inline unsigned long *kvm_ap_get_crycb_aqm(struct kvm *kvm)
> +{
> + unsigned long *aqm;
> + int crycb_fmt = kvm_ap_get_crycb_format(kvm);
> +
> + if (crycb_fmt == CRYCB_FORMAT2)
> + aqm = (unsigned long *)kvm->arch.crypto.crycb->apcb1.aqm;
> + else
> + aqm = (unsigned long *)kvm->arch.crypto.crycb->apcb0.aqm;
> +
> + return aqm;
> +}
> +
> +static inline unsigned long *kvm_ap_get_crycb_adm(struct kvm *kvm)
> +{
> + unsigned long *adm;
> + int crycb_fmt = kvm_ap_get_crycb_format(kvm);
> +
> + if (crycb_fmt == CRYCB_FORMAT2)
> + adm = (unsigned long *)kvm->arch.crypto.crycb->apcb1.adm;
> + else
> + adm = (unsigned long *)kvm->arch.crypto.crycb->apcb0.adm;
> +
> + return adm;
> +}
> +
> +static void kvm_ap_set_crycb_masks(struct kvm *kvm,
> + struct kvm_ap_matrix *matrix)
> +{
> + unsigned long *apm = kvm_ap_get_crycb_apm(kvm);
> + unsigned long *aqm = kvm_ap_get_crycb_aqm(kvm);
> + unsigned long *adm = kvm_ap_get_crycb_adm(kvm);
> +
> + kvm_ap_clear_crycb_masks(kvm);
> + memcpy(apm, matrix->apm, KVM_AP_MASK_BYTES(matrix->apm_max));
> + memcpy(aqm, matrix->aqm, KVM_AP_MASK_BYTES(matrix->aqm_max));
Replace KVM_AP_MASK_BYTES macro with BITS_TO_BYTES macro
> +
> + /*
> + * Merge the AQM and ADM since the ADM is a superset of the
> + * AQM by architectural convention.
> + */
> + bitmap_or(adm, adm, aqm, matrix->adm_max);
> +}
> +
> +static void kvm_ap_log_sharing_err(struct kvm *kvm, unsigned long apid,
> + unsigned long apqi)
> +{
> + pr_err("%s: AP queue %02lx.%04lx is registered to guest %s", __func__,
> + apid, apqi, kvm->arch.dbf->name);
> +}
> +
> +/**
> + * kvm_ap_validate_queue_sharing
> + *
> + * Verifies that the APQNs derived from the intersection of the AP adapter IDs
> + * and AP queue indexes comprising the AP matrix are not configured for
> + * another guest. AP queue sharing is not allowed.
> + *
> + * @kvm: the KVM guest
> + * @matrix: the AP matrix
> + *
> + * Returns 0 if the APQNs are valid, otherwise; returns -EBUSY.
> + */
> +static int kvm_ap_validate_queue_sharing(struct kvm *kvm,
> + struct kvm_ap_matrix *matrix)
> +{
> + struct kvm *vm;
> + unsigned long *apm, *aqm;
> + unsigned long apid, apqi;
> +
> +
> + /* No other VM may share an AP Queue with the input VM */
> + list_for_each_entry(vm, &vm_list, vm_list) {
> + if (kvm == vm)
> + continue;
> +
> + apm = kvm_ap_get_crycb_apm(vm);
> + if (!bitmap_and(apm, apm, matrix->apm, matrix->apm_max))
> + continue;
> +
> + aqm = kvm_ap_get_crycb_aqm(vm);
> + if (!bitmap_and(aqm, aqm, matrix->aqm, matrix->aqm_max))
> + continue;
> +
> + for_each_set_bit_inv(apid, apm, matrix->apm_max)
> + for_each_set_bit_inv(apqi, aqm, matrix->aqm_max)
> + kvm_ap_log_sharing_err(kvm, apid, apqi);
> +
> + return -EBUSY;
> + }
> +
> + return 0;
> +}
> +
> static int kvm_ap_apxa_installed(void)
> {
> int ret;
> @@ -50,3 +170,140 @@ void kvm_ap_set_crycb_format(struct kvm *kvm, __u32 *crycbd)
> *crycbd |= CRYCB_FORMAT1;
> }
> }
> +
> +static int kvm_ap_matrix_apm_create(struct kvm_ap_matrix *ap_matrix, int apxa)
> +{
> + if (apxa)
> + ap_matrix->apm_max = 256;
> + else
> + ap_matrix->apm_max = 64;
> +
> + ap_matrix->apm = kzalloc(KVM_AP_MASK_BYTES(ap_matrix->apm_max),
> + GFP_KERNEL);
Replace KVM_AP_MASK_BYTES macro with BITS_TO_BYTES macro
> + if (!ap_matrix->apm)
> + return -ENOMEM;
> +
> + return 0;
> +}
> +
> +static int kvm_ap_matrix_aqm_create(struct kvm_ap_matrix *ap_matrix, int apxa)
> +{
> + if (apxa)
> + ap_matrix->aqm_max = 256;
> + else
> + ap_matrix->aqm_max = 16;
> +
> + ap_matrix->aqm = kzalloc(KVM_AP_MASK_BYTES(ap_matrix->aqm_max),
> + GFP_KERNEL);
Replace KVM_AP_MASK_BYTES macro with BITS_TO_BYTES macro
> + if (!ap_matrix->aqm)
> + return -ENOMEM;
> +
> + return 0;
> +}
> +
> +static int kvm_ap_matrix_adm_create(struct kvm_ap_matrix *ap_matrix, int apxa)
> +{
> + if (apxa)
> + ap_matrix->adm_max = 256;
> + else
> + ap_matrix->adm_max = 16;
> +
> + ap_matrix->adm = kzalloc(KVM_AP_MASK_BYTES(ap_matrix->adm_max),
> + GFP_KERNEL);
Replace KVM_AP_MASK_BYTES macro with BITS_TO_BYTES macro
> + if (!ap_matrix->adm)
> + return -ENOMEM;
> +
> + return 0;
> +}
> +
> +static void kvm_ap_matrix_masks_destroy(struct kvm_ap_matrix *ap_matrix)
> +{
> + kfree(ap_matrix->apm);
> + kfree(ap_matrix->aqm);
> + kfree(ap_matrix->adm);
> +}
> +
> +int kvm_ap_matrix_create(struct kvm_ap_matrix **ap_matrix)
> +{
> + int ret;
> + int apxa = kvm_ap_apxa_installed();
> + struct kvm_ap_matrix *matrix;
> +
> + matrix = kzalloc(sizeof(*matrix), GFP_KERNEL);
> + if (!matrix)
> + return -ENOMEM;
> +
> + ret = kvm_ap_matrix_apm_create(matrix, apxa);
> + if (ret)
> + goto mask_create_err;
> +
> + ret = kvm_ap_matrix_aqm_create(matrix, apxa);
> + if (ret)
> + goto mask_create_err;
> +
> + ret = kvm_ap_matrix_adm_create(matrix, apxa);
> + if (ret)
> + goto mask_create_err;
> +
> + *ap_matrix = matrix;
> +
> + return 0;
> +
> +mask_create_err:
> + kvm_ap_matrix_masks_destroy(matrix);
> + kfree(matrix);
> + return ret;
> +}
> +EXPORT_SYMBOL(kvm_ap_matrix_create);
> +
> +void kvm_ap_matrix_destroy(struct kvm_ap_matrix *ap_matrix)
> +{
> + kvm_ap_matrix_masks_destroy(ap_matrix);
> + kfree(ap_matrix);
> +}
> +EXPORT_SYMBOL(kvm_ap_matrix_destroy);
> +
> +/**
> + * kvm_ap_configure_matrix
> + *
> + * Configure the AP matrix for a KVM guest.
> + *
> + * @kvm: the KVM guest
> + * @matrix: the matrix configuration information
> + *
> + * Returns 0 if the APQNs derived from the intersection of the set of adapter
> + * IDs (APM) and queue indexes (AQM) in @matrix are not configured for any
> + * other KVM guest running on the same linux host. Otherwise returns an error
> + * code.
> + */
> +int kvm_ap_configure_matrix(struct kvm *kvm, struct kvm_ap_matrix *matrix)
> +{
> + int ret = 0;
> +
> + mutex_lock(&kvm->lock);
> +
> + ret = kvm_ap_validate_queue_sharing(kvm, matrix);
> + if (ret)
> + return ret;
> +
> + kvm_ap_set_crycb_masks(kvm, matrix);
> +
> + mutex_unlock(&kvm->lock);
> +
> + return 0;
> +}
> +EXPORT_SYMBOL(kvm_ap_configure_matrix);
> +
> +/**
> + * kvm_ap_deconfigure_matrix
> + *
> + * Deconfigure the AP matrix for a KVM guest. Clears all of the bits in the
> + * APM, AQM and ADM in the guest's CRYCB.
> + *
> + * @kvm: the KVM guest
> + */
> +void kvm_ap_deconfigure_matrix(struct kvm *kvm)
> +{
> + kvm_ap_clear_crycb_masks(kvm);
> +}
> +EXPORT_SYMBOL(kvm_ap_deconfigure_matrix);
> diff --git a/drivers/s390/crypto/vfio_ap_ops.c b/drivers/s390/crypto/vfio_ap_ops.c
> index 4292a5e..4fda44e 100644
> --- a/drivers/s390/crypto/vfio_ap_ops.c
> +++ b/drivers/s390/crypto/vfio_ap_ops.c
> @@ -10,6 +10,7 @@
> #include <linux/device.h>
> #include <linux/list.h>
> #include <linux/ctype.h>
> +#include <asm/kvm-ap.h>
>
> #include "vfio_ap_private.h"
>
> @@ -18,8 +19,23 @@
>
> static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
> {
> + int ret;
> + struct ap_matrix_mdev *matrix_mdev;
> struct ap_matrix *ap_matrix = to_ap_matrix(mdev_parent_dev(mdev));
> + struct kvm_ap_matrix *matrix;
> +
> + ret = kvm_ap_matrix_create(&matrix);
> + if (ret)
> + return ret;
> +
> + matrix_mdev = kzalloc(sizeof(*matrix_mdev), GFP_KERNEL);
> + if (!matrix_mdev) {
> + kvm_ap_matrix_destroy(matrix);
> + return -ENOMEM;
> + }
>
> + matrix_mdev->matrix = matrix;
> + mdev_set_drvdata(mdev, matrix_mdev);
> ap_matrix->available_instances--;
>
> return 0;
> @@ -28,7 +44,10 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
> static int vfio_ap_mdev_remove(struct mdev_device *mdev)
> {
> struct ap_matrix *ap_matrix = to_ap_matrix(mdev_parent_dev(mdev));
> + struct ap_matrix_mdev *matrix_mdev = mdev_get_drvdata(mdev);
>
> + kvm_ap_matrix_destroy(matrix_mdev->matrix);
> + kfree(matrix_mdev);
> ap_matrix->available_instances++;
>
> return 0;
> diff --git a/drivers/s390/crypto/vfio_ap_private.h b/drivers/s390/crypto/vfio_ap_private.h
> index c264415..522564e 100644
> --- a/drivers/s390/crypto/vfio_ap_private.h
> +++ b/drivers/s390/crypto/vfio_ap_private.h
> @@ -27,6 +27,10 @@ struct ap_matrix {
> int available_instances;
> };
>
> +struct ap_matrix_mdev {
> + struct kvm_ap_matrix *matrix;
> +};
> +
> static inline struct ap_matrix *to_ap_matrix(struct device *dev)
> {
> return container_of(dev, struct ap_matrix, device);
Powered by blists - more mailing lists