search for: _illegal_

...t MUST NOT report > > any error." > > > Absolutely. So what happens if you drop these code lines? > There's no driver registered for this ID, so it's just ignored. > Seems like what spec is asking for, no? Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > > > + } > > > > > > Need to also > > > 1. validate that feature bit VIRTIO_1 is set > > > 2. validate that ID is not for a legacy device > > > > > > otherwise device specific drivers might get invoked &...

...t MUST NOT report > > any error." > > > Absolutely. So what happens if you drop these code lines? > There's no driver registered for this ID, so it's just ignored. > Seems like what spec is asking for, no? Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > > > + } > > > > > > Need to also > > > 1. validate that feature bit VIRTIO_1 is set > > > 2. validate that ID is not for a legacy device > > > > > > otherwise device specific drivers might get invoked &...

...this. > > > Absolutely. So what happens if you drop these code lines? > > > There's no driver registered for this ID, so it's just ignored. > > > Seems like what spec is asking for, no? > > > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > Yes - there will be a device, but no driver will drive it. A device with an *illegal* ID. > > > > > @@ -496,7 +531,8 @@ static int virtio_mmio_remove(struct platform_device *pdev) > > > > > > { > > > > > > struct virtio_mmio...

...this. > > > Absolutely. So what happens if you drop these code lines? > > > There's no driver registered for this ID, so it's just ignored. > > > Seems like what spec is asking for, no? > > > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > Yes - there will be a device, but no driver will drive it. A device with an *illegal* ID. > > > > > @@ -496,7 +531,8 @@ static int virtio_mmio_remove(struct platform_device *pdev) > > > > > > { > > > > > > struct virtio_mmio...

...to license of the receiver. Another example -- recently there was a court ruling that entering a website automatically subjected you to a lengthy "N"-page license -- and that gathering prices from the website for a price comparison report (like Consumer reports might come up with) was _illegal_ because it violated the license that you accepted by entering the website. Compare this to the fine print on my store that says you accept the store's license conditions when you enter the store. The license conditions are available to read, or you can just 'accept' the license by ent...

...t happens if you drop these code lines? > > > > > There's no driver registered for this ID, so it's just ignored. > > > > > Seems like what spec is asking for, no? > > > > > > > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > > > > > Yes - there will be a device, but no driver will drive it. > > > > A device with an *illegal* ID. > > So? The "device" is just a bunch of allocated memory. There's no driver > and no one touches is, which is what the spec requi...

...t happens if you drop these code lines? > > > > > There's no driver registered for this ID, so it's just ignored. > > > > > Seems like what spec is asking for, no? > > > > > > > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > > > > > Yes - there will be a device, but no driver will drive it. > > > > A device with an *illegal* ID. > > So? The "device" is just a bunch of allocated memory. There's no driver > and no one touches is, which is what the spec requi...

...; Absolutely. So what happens if you drop these code lines? > > > > There's no driver registered for this ID, so it's just ignored. > > > > Seems like what spec is asking for, no? > > > > > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > > > Yes - there will be a device, but no driver will drive it. > > A device with an *illegal* ID. So? The "device" is just a bunch of allocated memory. There's no driver and no one touches is, which is what the spec requires. > > > > > >...

...; Absolutely. So what happens if you drop these code lines? > > > > There's no driver registered for this ID, so it's just ignored. > > > > Seems like what spec is asking for, no? > > > > > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > > > Yes - there will be a device, but no driver will drive it. > > A device with an *illegal* ID. So? The "device" is just a bunch of allocated memory. There's no driver and no one touches is, which is what the spec requires. > > > > > >...

...or." > > > > > > Absolutely. So what happens if you drop these code lines? > > There's no driver registered for this ID, so it's just ignored. > > Seems like what spec is asking for, no? > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. Yes - there will be a device, but no driver will drive it. > > > > > + } > > > > > > > > Need to also > > > > 1. validate that feature bit VIRTIO_1 is set > > > > 2. validate that ID is not for a legacy device...

...or." > > > > > > Absolutely. So what happens if you drop these code lines? > > There's no driver registered for this ID, so it's just ignored. > > Seems like what spec is asking for, no? > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. Yes - there will be a device, but no driver will drive it. > > > > > + } > > > > > > > > Need to also > > > > 1. validate that feature bit VIRTIO_1 is set > > > > 2. validate that ID is not for a legacy device...

...p these code lines? > > > > > > There's no driver registered for this ID, so it's just ignored. > > > > > > Seems like what spec is asking for, no? > > > > > > > > > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > > > > > > > Yes - there will be a device, but no driver will drive it. > > > > > > A device with an *illegal* ID. > > > > So? The "device" is just a bunch of allocated memory. There's no driver > > and no one touch...

...p these code lines? > > > > > > There's no driver registered for this ID, so it's just ignored. > > > > > > Seems like what spec is asking for, no? > > > > > > > > > > Not to me, no. There will be a vm_dev registered with an _illegal_ ID. > > > > > > > > Yes - there will be a device, but no driver will drive it. > > > > > > A device with an *illegal* ID. > > > > So? The "device" is just a bunch of allocated memory. There's no driver > > and no one touch...

...rted by SLAs, and then sell add-ons with SLAs for a more specific configuration. If you want an industry-specific Linux, you're not going to find it from a generic distribution vendor. And you shouldn't wonder why they don't throw in things -- especially not things like Java which are _illegal_ to bundle freely. -- Bryan P.S. As always, maintain your own APT/YUM repository internally, and mix in any required RPMs. I virtually _never_ install from CD/DVD, almost always via NFS -- possibly with some post-install APT/YUM. -- Bryan J. Smith mailto:b.j.smith at ieee.org

On Thu, 2015-01-15 at 16:51 +0000, Michael S. Tsirkin wrote: > > + uint64_t addr = virt_to_phys(info->queue); > > Kernel normally uses u64 for this type. Sure, well spotted. > > + > > + writel(addr & 0xffffffff, > > + vm_dev->base + VIRTIO_MMIO_QUEUE_DESC_LOW); > > + writel((addr

On Thu, 2015-01-15 at 16:51 +0000, Michael S. Tsirkin wrote: > > + uint64_t addr = virt_to_phys(info->queue); > > Kernel normally uses u64 for this type. Sure, well spotted. > > + > > + writel(addr & 0xffffffff, > > + vm_dev->base + VIRTIO_MMIO_QUEUE_DESC_LOW); > > + writel((addr