| Message ID | 20181220152126.18741-4-julien.grall@arm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | arm: Allow the user to specify where the RAM is placed in the memory | expand |
On Thu, Dec 20, 2018 at 03:21:20PM +0000, Julien Grall wrote: > At the moment, virtio scsi only register a bank starting at 0. On some > architectures this may not be true and the guest may have multiple > memory region. > > Register all the memory regions to vhost by browsing kvm->mem_banks. The > code is based on the virtio_net__vhost_init implementation. > > Signed-off-by: Julien Grall <julien.grall@arm.com> > > --- > > This code was not tested as I don't have any setup with scsi. > > Changes in v3: > - Don't forget to increment i in the loop. > --- > virtio/scsi.c | 21 +++++++++++++-------- > 1 file changed, 13 insertions(+), 8 deletions(-) > > diff --git a/virtio/scsi.c b/virtio/scsi.c > index a429ac8..e541ab5 100644 > --- a/virtio/scsi.c > +++ b/virtio/scsi.c > @@ -178,24 +178,29 @@ static struct virtio_ops scsi_dev_virtio_ops = { > > static void virtio_scsi_vhost_init(struct kvm *kvm, struct scsi_dev *sdev) > { > + struct kvm_mem_bank *bank; > struct vhost_memory *mem; > u64 features; > - int r; > + int r, i; > > sdev->vhost_fd = open("/dev/vhost-scsi", O_RDWR); > if (sdev->vhost_fd < 0) > die_perror("Failed openning vhost-scsi device"); > > - mem = calloc(1, sizeof(*mem) + sizeof(struct vhost_memory_region)); > + mem = calloc(1, sizeof(*mem) + kvm->mem_slots * sizeof(struct vhost_memory_region)); > if (mem == NULL) > die("Failed allocating memory for vhost memory map"); > > - mem->nregions = 1; > - mem->regions[0] = (struct vhost_memory_region) { > - .guest_phys_addr = 0, > - .memory_size = kvm->ram_size, > - .userspace_addr = (unsigned long)kvm->ram_start, > - }; > + i = 0; > + list_for_each_entry(bank, &kvm->mem_banks, list) { > + mem->regions[0] = (struct vhost_memory_region) { > + .guest_phys_addr = bank->guest_phys_addr, > + .memory_size = bank->size, > + .userspace_addr = (unsigned long)bank->host_addr, > + }; Should this be mem->regions[i] ? Maybe we should just remove this altogether if nobody is using it. Will
On 22/01/2019 05:44, Will Deacon wrote: > On Thu, Dec 20, 2018 at 03:21:20PM +0000, Julien Grall wrote: >> At the moment, virtio scsi only register a bank starting at 0. On some >> architectures this may not be true and the guest may have multiple >> memory region. >> >> Register all the memory regions to vhost by browsing kvm->mem_banks. The >> code is based on the virtio_net__vhost_init implementation. >> >> Signed-off-by: Julien Grall <julien.grall@arm.com> >> >> diff --git a/virtio/scsi.c b/virtio/scsi.c >> index a429ac8..e541ab5 100644 >> --- a/virtio/scsi.c >> +++ b/virtio/scsi.c >> @@ -178,24 +178,29 @@ static struct virtio_ops scsi_dev_virtio_ops = { >> >> static void virtio_scsi_vhost_init(struct kvm *kvm, struct scsi_dev *sdev) >> { >> + struct kvm_mem_bank *bank; >> struct vhost_memory *mem; >> u64 features; >> - int r; >> + int r, i; >> >> sdev->vhost_fd = open("/dev/vhost-scsi", O_RDWR); >> if (sdev->vhost_fd < 0) >> die_perror("Failed openning vhost-scsi device"); >> >> - mem = calloc(1, sizeof(*mem) + sizeof(struct vhost_memory_region)); >> + mem = calloc(1, sizeof(*mem) + kvm->mem_slots * sizeof(struct vhost_memory_region)); >> if (mem == NULL) >> die("Failed allocating memory for vhost memory map"); >> >> - mem->nregions = 1; >> - mem->regions[0] = (struct vhost_memory_region) { >> - .guest_phys_addr = 0, >> - .memory_size = kvm->ram_size, >> - .userspace_addr = (unsigned long)kvm->ram_start, >> - }; >> + i = 0; >> + list_for_each_entry(bank, &kvm->mem_banks, list) { >> + mem->regions[0] = (struct vhost_memory_region) { >> + .guest_phys_addr = bank->guest_phys_addr, >> + .memory_size = bank->size, >> + .userspace_addr = (unsigned long)bank->host_addr, >> + }; > > Should this be mem->regions[i] ? > > Maybe we should just remove this altogether if nobody is using it. And also incrementing i in the loop. I assume we are adding the support for this by specifying the memory placement option later in the series. I had raised this in one of the previous iterations. Cheers Suzuki
Hi, On 1/22/19 9:22 AM, Suzuki K Poulose wrote: > > > On 22/01/2019 05:44, Will Deacon wrote: >> On Thu, Dec 20, 2018 at 03:21:20PM +0000, Julien Grall wrote: >>> At the moment, virtio scsi only register a bank starting at 0. On some >>> architectures this may not be true and the guest may have multiple >>> memory region. >>> >>> Register all the memory regions to vhost by browsing kvm->mem_banks. The >>> code is based on the virtio_net__vhost_init implementation. >>> >>> Signed-off-by: Julien Grall <julien.grall@arm.com> >>> > >>> diff --git a/virtio/scsi.c b/virtio/scsi.c >>> index a429ac8..e541ab5 100644 >>> --- a/virtio/scsi.c >>> +++ b/virtio/scsi.c >>> @@ -178,24 +178,29 @@ static struct virtio_ops scsi_dev_virtio_ops = { >>> static void virtio_scsi_vhost_init(struct kvm *kvm, struct scsi_dev >>> *sdev) >>> { >>> + struct kvm_mem_bank *bank; >>> struct vhost_memory *mem; >>> u64 features; >>> - int r; >>> + int r, i; >>> sdev->vhost_fd = open("/dev/vhost-scsi", O_RDWR); >>> if (sdev->vhost_fd < 0) >>> die_perror("Failed openning vhost-scsi device"); >>> - mem = calloc(1, sizeof(*mem) + sizeof(struct vhost_memory_region)); >>> + mem = calloc(1, sizeof(*mem) + kvm->mem_slots * sizeof(struct >>> vhost_memory_region)); >>> if (mem == NULL) >>> die("Failed allocating memory for vhost memory map"); >>> - mem->nregions = 1; >>> - mem->regions[0] = (struct vhost_memory_region) { >>> - .guest_phys_addr = 0, >>> - .memory_size = kvm->ram_size, >>> - .userspace_addr = (unsigned long)kvm->ram_start, >>> - }; >>> + i = 0; >>> + list_for_each_entry(bank, &kvm->mem_banks, list) { >>> + mem->regions[0] = (struct vhost_memory_region) { >>> + .guest_phys_addr = bank->guest_phys_addr, >>> + .memory_size = bank->size, >>> + .userspace_addr = (unsigned long)bank->host_addr, >>> + }; >> >> Should this be mem->regions[i] ? >> >> Maybe we should just remove this altogether if nobody is using it. > > And also incrementing i in the loop. I assume we are adding the support > for this > by specifying the memory placement option later in the series. I had > raised this > in one of the previous iterations. I have incremented 'i' in this version but forgot to update mem->regions[i] :/. Sorry for that. Cheers,
diff --git a/virtio/scsi.c b/virtio/scsi.c index a429ac8..e541ab5 100644 --- a/virtio/scsi.c +++ b/virtio/scsi.c @@ -178,24 +178,29 @@ static struct virtio_ops scsi_dev_virtio_ops = { static void virtio_scsi_vhost_init(struct kvm *kvm, struct scsi_dev *sdev) { + struct kvm_mem_bank *bank; struct vhost_memory *mem; u64 features; - int r; + int r, i; sdev->vhost_fd = open("/dev/vhost-scsi", O_RDWR); if (sdev->vhost_fd < 0) die_perror("Failed openning vhost-scsi device"); - mem = calloc(1, sizeof(*mem) + sizeof(struct vhost_memory_region)); + mem = calloc(1, sizeof(*mem) + kvm->mem_slots * sizeof(struct vhost_memory_region)); if (mem == NULL) die("Failed allocating memory for vhost memory map"); - mem->nregions = 1; - mem->regions[0] = (struct vhost_memory_region) { - .guest_phys_addr = 0, - .memory_size = kvm->ram_size, - .userspace_addr = (unsigned long)kvm->ram_start, - }; + i = 0; + list_for_each_entry(bank, &kvm->mem_banks, list) { + mem->regions[0] = (struct vhost_memory_region) { + .guest_phys_addr = bank->guest_phys_addr, + .memory_size = bank->size, + .userspace_addr = (unsigned long)bank->host_addr, + }; + i++; + } + mem->nregions = i; r = ioctl(sdev->vhost_fd, VHOST_SET_OWNER); if (r != 0)
At the moment, virtio scsi only register a bank starting at 0. On some architectures this may not be true and the guest may have multiple memory region. Register all the memory regions to vhost by browsing kvm->mem_banks. The code is based on the virtio_net__vhost_init implementation. Signed-off-by: Julien Grall <julien.grall@arm.com> --- This code was not tested as I don't have any setup with scsi. Changes in v3: - Don't forget to increment i in the loop. --- virtio/scsi.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-)