Re: [PATCH v2 20/30] KVM: s390: pci: provide routines for enabling/disabling IOAT assist

[Matthew Rosato <mjrosato@linux.ibm.com>]

On 1/25/22 8:29 AM, Pierre Morel wrote:
> 
> 
> On 1/14/22 21:31, Matthew Rosato wrote:
>> These routines will be wired into the vfio_pci_zdev ioctl handlers to
>> respond to requests to enable / disable a device for PCI I/O Address
>> Translation assistance.
>>
>> Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
>> ---
>>   arch/s390/include/asm/kvm_pci.h |  15 ++++
>>   arch/s390/include/asm/pci_dma.h |   2 +
>>   arch/s390/kvm/pci.c             | 139 ++++++++++++++++++++++++++++++++
>>   arch/s390/kvm/pci.h             |   2 +
>>   4 files changed, 158 insertions(+)
>>
>> diff --git a/arch/s390/include/asm/kvm_pci.h 
>> b/arch/s390/include/asm/kvm_pci.h
>> index 01fe14fffd7a..770849f13a70 100644
>> --- a/arch/s390/include/asm/kvm_pci.h
>> +++ b/arch/s390/include/asm/kvm_pci.h
>> @@ -16,11 +16,21 @@
>>   #include <linux/kvm_host.h>
>>   #include <linux/kvm.h>
>>   #include <linux/pci.h>
>> +#include <linux/mutex.h>
>>   #include <asm/pci_insn.h>
>> +#include <asm/pci_dma.h>
>> +
>> +struct kvm_zdev_ioat {
>> +    unsigned long *head[ZPCI_TABLE_PAGES];
>> +    unsigned long **seg;
>> +    unsigned long ***pt;
>> +    struct mutex lock;
> 
> Can we please rename the mutex ioat_lock to have a unique name easy to 
> follow for maintenance.
> Can you please add a description about when the lock should be used?
> 

OK.  The lock is meant to protect the contents of kvm_zdev_ioat -- I'll 
think of something to describe it.

>> +};
>>   struct kvm_zdev {
>>       struct zpci_dev *zdev;
>>       struct kvm *kvm;
>> +    struct kvm_zdev_ioat ioat;
>>       struct zpci_fib fib;
>>   };
>> @@ -33,6 +43,11 @@ int kvm_s390_pci_aif_enable(struct zpci_dev *zdev, 
>> struct zpci_fib *fib,
>>                   bool assist);
>>   int kvm_s390_pci_aif_disable(struct zpci_dev *zdev);
>> +int kvm_s390_pci_ioat_probe(struct zpci_dev *zdev);
>> +int kvm_s390_pci_ioat_enable(struct zpci_dev *zdev, u64 iota);
>> +int kvm_s390_pci_ioat_disable(struct zpci_dev *zdev);
>> +u8 kvm_s390_pci_get_dtsm(struct zpci_dev *zdev);
>> +
>>   int kvm_s390_pci_interp_probe(struct zpci_dev *zdev);
>>   int kvm_s390_pci_interp_enable(struct zpci_dev *zdev);
>>   int kvm_s390_pci_interp_disable(struct zpci_dev *zdev);
>> diff --git a/arch/s390/include/asm/pci_dma.h 
>> b/arch/s390/include/asm/pci_dma.h
>> index 91e63426bdc5..69e616d0712c 100644
>> --- a/arch/s390/include/asm/pci_dma.h
>> +++ b/arch/s390/include/asm/pci_dma.h
>> @@ -50,6 +50,8 @@ enum zpci_ioat_dtype {
>>   #define ZPCI_TABLE_ALIGN        ZPCI_TABLE_SIZE
>>   #define ZPCI_TABLE_ENTRY_SIZE        (sizeof(unsigned long))
>>   #define ZPCI_TABLE_ENTRIES        (ZPCI_TABLE_SIZE / 
>> ZPCI_TABLE_ENTRY_SIZE)
>> +#define ZPCI_TABLE_PAGES        (ZPCI_TABLE_SIZE >> PAGE_SHIFT)
>> +#define ZPCI_TABLE_ENTRIES_PAGES    (ZPCI_TABLE_ENTRIES * 
>> ZPCI_TABLE_PAGES)
>>   #define ZPCI_TABLE_BITS            11
>>   #define ZPCI_PT_BITS            8
>> diff --git a/arch/s390/kvm/pci.c b/arch/s390/kvm/pci.c
>> index 7ed9abc476b6..39c13c25a700 100644
>> --- a/arch/s390/kvm/pci.c
>> +++ b/arch/s390/kvm/pci.c
>> @@ -13,12 +13,15 @@
>>   #include <asm/pci.h>
>>   #include <asm/pci_insn.h>
>>   #include <asm/pci_io.h>
>> +#include <asm/pci_dma.h>
>>   #include <asm/sclp.h>
>>   #include "pci.h"
>>   #include "kvm-s390.h"
>>   struct zpci_aift *aift;
>> +#define shadow_ioat_init zdev->kzdev->ioat.head[0]
>> +
>>   static inline int __set_irq_noiib(u16 ctl, u8 isc)
>>   {
>>       union zpci_sic_iib iib = {{0}};
>> @@ -344,6 +347,135 @@ int kvm_s390_pci_aif_disable(struct zpci_dev *zdev)
>>   }
>>   EXPORT_SYMBOL_GPL(kvm_s390_pci_aif_disable);
>> +int kvm_s390_pci_ioat_probe(struct zpci_dev *zdev)
>> +{
>> +    /* Must have a KVM association registered */
> 
> may be add something like : "The ioat structure is embeded in kzdev"
> 
>> +    if (!zdev->kzdev || !zdev->kzdev->kvm)
> 
> Why do we need to check for kvm ?
> Having kzdev is already tested by the unique caller.
> 

We probably don't need to check for the kzdev because the caller already 
did this, agreed there.

But as for checking the kvm association, Alex asked for this in a 
comment to v1 (comment was against one of the vfio patches that call 
these routines) -- The reason being the probe comes from a userspace 
request and can be against any vfio-pci(-zdev) device at any time, and 
there's no point in proceeding if this device is not associated with a 
KVM guest -- It's possible for the KVM notifier to also pass a null KVM 
address -- so I think it's better to just be sure here.  In a 
well-behaved environment we would never see this (so, another case for 
an s390dbf entry)

>> +        return -EINVAL;
>> +
>> +    return 0;
>> +}
>> +EXPORT_SYMBOL_GPL(kvm_s390_pci_ioat_probe);
>> +
>> +int kvm_s390_pci_ioat_enable(struct zpci_dev *zdev, u64 iota)
>> +{
>> +    gpa_t gpa = (gpa_t)(iota & ZPCI_RTE_ADDR_MASK);
>> +    struct kvm_zdev_ioat *ioat;
>> +    struct page *page;
>> +    struct kvm *kvm;
>> +    unsigned int idx;
>> +    void *iaddr;
>> +    int i, rc = 0;
> 
> no need to initialize rc

Agree based on the changes below

> 
>> +
>> +    if (shadow_ioat_init)
>> +        return -EINVAL;
>> +
>> +    /* Ensure supported type specified */
>> +    if ((iota & ZPCI_IOTA_RTTO_FLAG) != ZPCI_IOTA_RTTO_FLAG)
>> +        return -EINVAL;
>> +
>> +    kvm = zdev->kzdev->kvm;
>> +    ioat = &zdev->kzdev->ioat;
>> +    mutex_lock(&ioat->lock);
>> +    idx = srcu_read_lock(&kvm->srcu);
>> +    for (i = 0; i < ZPCI_TABLE_PAGES; i++) {
>> +        page = gfn_to_page(kvm, gpa_to_gfn(gpa));
>> +        if (is_error_page(page)) {
>> +            srcu_read_unlock(&kvm->srcu, idx);
>> +            rc = -EIO;
>> +            goto out;
> 
>              goto unpin ?

Ah, right, in case we hit this error somewhere in the middle of the loop.

> 
>> +        }
>> +        iaddr = page_to_virt(page) + (gpa & ~PAGE_MASK);
>> +        ioat->head[i] = (unsigned long *)iaddr;
>> +        gpa += PAGE_SIZE;
>> +    }
>> +    srcu_read_unlock(&kvm->srcu, idx);
>> +
>> +    zdev->kzdev->ioat.seg = kcalloc(ZPCI_TABLE_ENTRIES_PAGES,
>> +                    sizeof(unsigned long *), GFP_KERNEL);
> 
> What about:
> 
>          ioat->seg = kcalloc(ZPCI_TABLE_ENTRIES_PAGES,
>                              sizeof(*ioat->seg), GFP_KERNEL);
>      if (!ioat->seg)
> ...
>      ioat->pt = ...
> ?

Yep, would be fine (seems I forgot about the local *ioat here)

> 
>> +    if (!zdev->kzdev->ioat.seg)
>> +        goto unpin;
>> +    zdev->kzdev->ioat.pt = kcalloc(ZPCI_TABLE_ENTRIES,
>> +                       sizeof(unsigned long **), GFP_KERNEL);
>> +    if (!zdev->kzdev->ioat.pt)
>> +        goto free_seg;
>> +
>> +out:
>> +    mutex_unlock(&ioat->lock);
>> +    return rc;
> 
>      return 0 ?

Yes, we can do that now that we don't goto out: after is_error_page

> 
>> +
>> +free_seg:
>> +    kfree(zdev->kzdev->ioat.seg);
> 
> kfree(ioat->seg) ?
> rc = -ENOMEM;
> 
>> +unpin:
>> +    for (i = 0; i < ZPCI_TABLE_PAGES; i++) {
>> +        kvm_release_pfn_dirty((u64)ioat->head[i] >> PAGE_SHIFT);
>> +        ioat->head[i] = 0;
>> +    }
>> +    mutex_unlock(&ioat->lock);
>> +    return -ENOMEM;
> 
>      return rc;

And yes, agreed, now that we come here for other reasons (-EIO) we must 
return rc here and also set rc=-ENOMEM as you say for the kfree case above.