Re: [PATCH v13 14/24] virt: gunyah: Add Qualcomm Gunyah platform ops

[Elliot Berman <quic_eberman@quicinc.com>]

On 6/5/2023 12:48 PM, Alex Elder wrote:
> On 5/9/23 3:47 PM, Elliot Berman wrote:
>> Qualcomm platforms have a firmware entity which performs access control
>> to physical pages. Dynamically started Gunyah virtual machines use the
>> QCOM_SCM_RM_MANAGED_VMID for access. Linux thus needs to assign access
>> to the memory used by guest VMs. Gunyah doesn't do this operation for us
>> since it is the current VM (typically VMID_HLOS) delegating the access
>> and not Gunyah itself. Use the Gunyah platform ops to achieve this so
>> that only Qualcomm platforms attempt to make the needed SCM calls.
>>
>> Co-developed-by: Prakruthi Deepak Heragu <quic_pheragu@quicinc.com>
>> Signed-off-by: Prakruthi Deepak Heragu <quic_pheragu@quicinc.com>
>> Signed-off-by: Elliot Berman <quic_eberman@quicinc.com>
> 
> Minor suggestions below.  Please consider them, but either way:
> 
> Reviewed-by: Alex Elder <elder@linaro.org>
> 
>> ---
>>   drivers/virt/gunyah/Kconfig       |  13 +++
>>   drivers/virt/gunyah/Makefile      |   1 +
>>   drivers/virt/gunyah/gunyah_qcom.c | 147 ++++++++++++++++++++++++++++++
>>   3 files changed, 161 insertions(+)
>>   create mode 100644 drivers/virt/gunyah/gunyah_qcom.c
>>
>> diff --git a/drivers/virt/gunyah/Kconfig b/drivers/virt/gunyah/Kconfig
>> index de815189dab6..0421b751aad4 100644
>> --- a/drivers/virt/gunyah/Kconfig
>> +++ b/drivers/virt/gunyah/Kconfig
>> @@ -5,6 +5,7 @@ config GUNYAH
>>       depends on ARM64
>>       depends on MAILBOX
>>       select GUNYAH_PLATFORM_HOOKS
>> +    imply GUNYAH_QCOM_PLATFORM if ARCH_QCOM
>>       help
>>         The Gunyah drivers are the helper interfaces that run in a 
>> guest VM
>>         such as basic inter-VM IPC and signaling mechanisms, and 
>> higher level
>> @@ -15,3 +16,15 @@ config GUNYAH
>>   config GUNYAH_PLATFORM_HOOKS
>>       tristate
>> +
>> +config GUNYAH_QCOM_PLATFORM
>> +    tristate "Support for Gunyah on Qualcomm platforms"
>> +    depends on GUNYAH
>> +    select GUNYAH_PLATFORM_HOOKS
>> +    select QCOM_SCM
>> +    help
>> +      Enable support for interacting with Gunyah on Qualcomm
>> +      platforms. Interaction with Qualcomm firmware requires
>> +      extra platform-specific support.
>> +
>> +      Say Y/M here to use Gunyah on Qualcomm platforms.
>> diff --git a/drivers/virt/gunyah/Makefile b/drivers/virt/gunyah/Makefile
>> index 4fbeee521d60..2aa9ff038ed0 100644
>> --- a/drivers/virt/gunyah/Makefile
>> +++ b/drivers/virt/gunyah/Makefile
>> @@ -1,6 +1,7 @@
>>   # SPDX-License-Identifier: GPL-2.0
>>   obj-$(CONFIG_GUNYAH_PLATFORM_HOOKS) += gunyah_platform_hooks.o
>> +obj-$(CONFIG_GUNYAH_QCOM_PLATFORM) += gunyah_qcom.o
>>   gunyah-y += rsc_mgr.o rsc_mgr_rpc.o vm_mgr.o vm_mgr_mm.o
>>   obj-$(CONFIG_GUNYAH) += gunyah.o
>> diff --git a/drivers/virt/gunyah/gunyah_qcom.c 
>> b/drivers/virt/gunyah/gunyah_qcom.c
>> new file mode 100644
>> index 000000000000..18acbda8fcbd
>> --- /dev/null
>> +++ b/drivers/virt/gunyah/gunyah_qcom.c
>> @@ -0,0 +1,147 @@
>> +// SPDX-License-Identifier: GPL-2.0-only
>> +/*
>> + * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights 
>> reserved.
>> + */
>> +
>> +#include <linux/arm-smccc.h>
>> +#include <linux/gunyah_rsc_mgr.h>
>> +#include <linux/module.h>
>> +#include <linux/firmware/qcom/qcom_scm.h>
>> +#include <linux/types.h>
>> +#include <linux/uuid.h>
>> +
>> +#define QCOM_SCM_RM_MANAGED_VMID    0x3A
>> +#define QCOM_SCM_MAX_MANAGED_VMID    0x3F
> 
> Is this limited to 63 because there are at most 64 VMIDs
> that can be represented in a 64-bit unsigned?
> 

It's a limitation imposed by QC firmware, but I speculate that's why 63 
is selected.

>> +
>> +static int qcom_scm_gh_rm_pre_mem_share(struct gh_rm *rm, struct 
>> gh_rm_mem_parcel *mem_parcel)
>> +{
>> +    struct qcom_scm_vmperm *new_perms;
>> +    u64 src, src_cpy;
>> +    int ret = 0, i, n;
>> +    u16 vmid;
>> +
>> +    new_perms = kcalloc(mem_parcel->n_acl_entries, 
>> sizeof(*new_perms), GFP_KERNEL);
>> +    if (!new_perms)
>> +        return -ENOMEM;
>> +
>> +    for (n = 0; n < mem_parcel->n_acl_entries; n++) {
>> +        vmid = le16_to_cpu(mem_parcel->acl_entries[n].vmid);
>> +        if (vmid <= QCOM_SCM_MAX_MANAGED_VMID)
>> +            new_perms[n].vmid = vmid;
>> +        else
>> +            new_perms[n].vmid = QCOM_SCM_RM_MANAGED_VMID;
> 
> So any out-of-range VM ID will cause the hunk of memory to
> be assigned to the resource manager.  Is it expected that
> this can occur (and not be an error)?
> 

Yes, that's the expectation for these virtual machines. This is for the 
access control implemented in QC firmware, so it's not that we're 
assigning memory to resource manager, but rather assigning memory to a 
resource manager-managed VM. This is done to de-escalate the access 
level of guest VMs.

>> +        if (mem_parcel->acl_entries[n].perms & GH_RM_ACL_X)
>> +            new_perms[n].perm |= QCOM_SCM_PERM_EXEC;
>> +        if (mem_parcel->acl_entries[n].perms & GH_RM_ACL_W)
>> +            new_perms[n].perm |= QCOM_SCM_PERM_WRITE;
>> +        if (mem_parcel->acl_entries[n].perms & GH_RM_ACL_R)
>> +            new_perms[n].perm |= QCOM_SCM_PERM_READ;
>> +    }
>> +
>> +    src = (1ull << QCOM_SCM_VMID_HLOS);
> 
>      src = BIT_ULL(QCOM_SCM_VMID_HLOS);
> 
>> +
>> +    for (i = 0; i < mem_parcel->n_mem_entries; i++) {
>> +        src_cpy = src;
>> +        ret = 
>> qcom_scm_assign_mem(le64_to_cpu(mem_parcel->mem_entries[i].phys_addr),
>> +                        le64_to_cpu(mem_parcel->mem_entries[i].size),
>> +                        &src_cpy, new_perms, mem_parcel->n_acl_entries);
> 
> Loops like this can look simpler if you jump to error handling
> at the end that does this unwind activity, rather than incorporating
> it inside the loop itself.  Or even just breaking if ret != 0, e.g.:
> 
>          if (ret)
>              break;
>      }
> 
>      if (!ret)
>          return 0;
> 
>      /* And do the following block here, "outdented" twice */
> 
>> +        if (ret) {
>> +            src = 0;
>> +            for (n = 0; n < mem_parcel->n_acl_entries; n++) {
>> +                vmid = le16_to_cpu(mem_parcel->acl_entries[n].vmid);
>> +                if (vmid <= QCOM_SCM_MAX_MANAGED_VMID)
>> +                    src |= (1ull << vmid);
> 
>      src |= BIT_ULL(vmid);
> 
>> +                else
>> +                    src |= (1ull << QCOM_SCM_RM_MANAGED_VMID);
> 
>      src |= BIT_ULL(QCOM_SCM_RM_MANAGED_VMID);
> 
>> +            }
>> +
>> +            new_perms[0].vmid = QCOM_SCM_VMID_HLOS;
>> +
>> +            for (i--; i >= 0; i--) {
>> +                src_cpy = src;
>> +                WARN_ON_ONCE(qcom_scm_assign_mem(
>> +                        
>> le64_to_cpu(mem_parcel->mem_entries[i].phys_addr),
>> +                        le64_to_cpu(mem_parcel->mem_entries[i].size),
>> +                        &src_cpy, new_perms, 1));
>> +            }
>> +            break;
>> +        }
>> +    }
>> +
>> +    kfree(new_perms);
>> +    return ret;
>> +}
>> +
>> +static int qcom_scm_gh_rm_post_mem_reclaim(struct gh_rm *rm, struct 
>> gh_rm_mem_parcel *mem_parcel)
>> +{
>> +    struct qcom_scm_vmperm new_perms;
>> +    u64 src = 0, src_cpy;
>> +    int ret = 0, i, n;
>> +    u16 vmid;
>> +
>> +    new_perms.vmid = QCOM_SCM_VMID_HLOS;
>> +    new_perms.perm = QCOM_SCM_PERM_EXEC | QCOM_SCM_PERM_WRITE | 
>> QCOM_SCM_PERM_READ;
>> +
>> +    for (n = 0; n < mem_parcel->n_acl_entries; n++) {
>> +        vmid = le16_to_cpu(mem_parcel->acl_entries[n].vmid);
>> +        if (vmid <= QCOM_SCM_MAX_MANAGED_VMID)
>> +            src |= (1ull << vmid);
>> +        else
>> +            src |= (1ull << QCOM_SCM_RM_MANAGED_VMID);
>> +    }
>> +
>> +    for (i = 0; i < mem_parcel->n_mem_entries; i++) {
>> +        src_cpy = src;
>> +        ret = 
>> qcom_scm_assign_mem(le64_to_cpu(mem_parcel->mem_entries[i].phys_addr),
>> +                        le64_to_cpu(mem_parcel->mem_entries[i].size),
>> +                        &src_cpy, &new_perms, 1);
>> +        WARN_ON_ONCE(ret);
>> +    }
>> +
>> +    return ret;
>> +}
>> +
>> +static struct gh_rm_platform_ops qcom_scm_gh_rm_platform_ops = {
>> +    .pre_mem_share = qcom_scm_gh_rm_pre_mem_share,
>> +    .post_mem_reclaim = qcom_scm_gh_rm_post_mem_reclaim,
>> +};
>> +
>> +/* {19bd54bd-0b37-571b-946f-609b54539de6} */
>> +static const uuid_t QCOM_EXT_UUID =
>> +    UUID_INIT(0x19bd54bd, 0x0b37, 0x571b, 0x94, 0x6f, 0x60, 0x9b, 
>> 0x54, 0x53, 0x9d, 0xe6);
>> +
>> +#define GH_QCOM_EXT_CALL_UUID_ID    
>> ARM_SMCCC_CALL_VAL(ARM_SMCCC_FAST_CALL, ARM_SMCCC_SMC_32, \
>> +                               ARM_SMCCC_OWNER_VENDOR_HYP, 0x3f01)
>> +
>> +static bool gh_has_qcom_extensions(void)
>> +{
>> +    struct arm_smccc_res res;
>> +    uuid_t uuid;
>> +
>> +    arm_smccc_1_1_smc(GH_QCOM_EXT_CALL_UUID_ID, &res);
>> +
>> +    ((u32 *)&uuid.b[0])[0] = lower_32_bits(res.a0);
>> +    ((u32 *)&uuid.b[0])[1] = lower_32_bits(res.a1);
>> +    ((u32 *)&uuid.b[0])[2] = lower_32_bits(res.a2);
>> +    ((u32 *)&uuid.b[0])[3] = lower_32_bits(res.a3);
> 
> I said this elsewhere.  I'd rather see:
> 
>      u32 *u = (u32 *)&uuid;        /* Or &uuid.b? */
> 
>      *u++ = lower_32_bits(res.a0);
>          . . .
> 
>> +
>> +    return uuid_equal(&uuid, &QCOM_EXT_UUID);
>> +}
>> +
>> +static int __init qcom_gh_platform_hooks_register(void)
>> +{
>> +    if (!gh_has_qcom_extensions())
>> +        return -ENODEV;
>> +
>> +    return gh_rm_register_platform_ops(&qcom_scm_gh_rm_platform_ops);
>> +}
>> +
>> +static void __exit qcom_gh_platform_hooks_unregister(void)
>> +{
>> +    gh_rm_unregister_platform_ops(&qcom_scm_gh_rm_platform_ops);
>> +}
>> +
>> +module_init(qcom_gh_platform_hooks_register);
>> +module_exit(qcom_gh_platform_hooks_unregister);
>> +MODULE_DESCRIPTION("Qualcomm Technologies, Inc. Platform Hooks for 
>> Gunyah");
>> +MODULE_LICENSE("GPL");
>