We currently have a problem that we cannot support advanced features
of an IOMMU domain (ex. IOMMU_CACHE), because we have no guarantee
that those features will be supported by all of the hardware units
involved with the domain over its lifetime. For instance, the Intel
VT-d architecture does not require that all DRHDs support snoop
control. If we create a domain based on a device behind a DRHD that
does support snoop control and enable SNP support via the IOMMU_CACHE
mapping option, we cannot then add a device behind a DRHD which does
not support snoop control or we'll get reserved bit faults from the
SNP bit in the pagetables. To add to the complexity, we can't know
the properties of a domain until a device is attached.
We could pass this problem off to userspace and require that a
separate vfio container be used, but we don't know how to handle page
accounting in that case. How do we know that a page pinned in one
container is the same page as a different container and avoid double
billing the user for the page.
The solution is therefore to support multiple IOMMU domains per
container. In the majority of cases, only one domain will be required
since hardware is typically consistent within a system. However, this
provides us the ability to validate compatibility of domains and
support mixed environments where page table flags can be different
between domains.
To do this, our DMA tracking needs to change. We currently try to
coalesce user mappings into as few tracking entries as possible. The
problem then becomes that we lose granularity of user mappings. We've
never guaranteed that a user is able to unmap at a finer granularity
than the original mapping, but we must honor the granularity of the
original mapping. This coalescing code is therefore removed, allowing
only unmaps covering complete maps. The change in accounting is
fairly small here, a typical QEMU VM will start out with roughly a
dozen entries, so it's arguable if this coalescing was ever needed.
We also move IOMMU domain creation to the point where a group is
attached to the container. An interesting side-effect of this is that
we now have access to the device at the time of domain creation and
can probe the devices within the group to determine the bus_type.
This finally makes vfio_iommu_type1 completely device/bus agnostic.
In fact, each IOMMU domain can host devices on different buses managed
by different physical IOMMUs, and present a single DMA mapping
interface to the user. When a new domain is created, mappings are
replayed to bring the IOMMU pagetables up to the state of the current
container. And of course, DMA mapping and unmapping automatically
traverse all of the configured IOMMU domains.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Cc: Varun Sethi <Varun.Sethi@freescale.com>
Set the RxRPC header flag to request an ACK packet for every odd-numbered DATA
packet unless it's the last one (which implicitly requests an ACK anyway).
This is similar to how librx appears to work.
If we don't do this, we'll send out a full window of packets and then just sit
there until the other side gets bored and sends an ACK to indicate that it's
been idle for a while and has received no new packets.
Requesting a lot of ACKs shouldn't be a problem as ACKs should be merged when
possible.
As AF_RXRPC currently works, it will schedule an ACK to be generated upon
receipt of a DATA packet with the ACK-request packet set - and in the time
taken to schedule this in a work queue, several other packets are likely to
arrive and then all get ACK'd together.
Signed-off-by: David Howells <dhowells@redhat.com>
Expose RxRPC parameters via sysctls to control the Rx window size, the Rx MTU
maximum size and the number of packets that can be glued into a jumbo packet.
More info added to Documentation/networking/rxrpc.txt.
Signed-off-by: David Howells <dhowells@redhat.com>
Improve ACK production by the following means:
(1) Don't send an ACK_REQUESTED ack immediately even if the RXRPC_MORE_PACKETS
flag isn't set on a data packet that has also has RXRPC_REQUEST_ACK set.
MORE_PACKETS just means that the sender just emptied its Tx data buffer.
More data will be forthcoming unless RXRPC_LAST_PACKET is also flagged.
It is possible to see runs of DATA packets with MORE_PACKETS unset that
aren't waiting for an ACK.
It is therefore better to wait a small instant to see if we can combine an
ACK for several packets.
(2) Don't send an ACK_IDLE ack immediately unless we're responding to the
terminal data packet of a call.
Whilst sending an ACK_IDLE mid-call serves to let the other side know
that we won't be asking it to resend certain Tx buffers and that it can
discard them, spamming it with loads of acks just because we've
temporarily run out of data just distracts it.
(3) Put the ACK_IDLE ack generation timeout up to half a second rather than a
single jiffy. Just because we haven't been given more data immediately
doesn't mean that more isn't forthcoming. The other side may be busily
finding the data to send to us.
Signed-off-by: David Howells <dhowells@redhat.com>
Add sysctls for configuring RxRPC protocol handling, specifically controls on
delays before ack generation, the delay before resending a packet, the maximum
lifetime of a call and the expiration times of calls, connections and
transports that haven't been recently used.
More info added in Documentation/networking/rxrpc.txt.
Signed-off-by: David Howells <dhowells@redhat.com>
AF_RXRPC sends UDP packets with the "Don't Fragment" bit set in an attempt to
determine the maximum packet size between the local socket and the peer by
invoking the generation of ICMP_FRAG_NEEDED packets.
Once a packet is sent with the "Don't Fragment" bit set, it is then
inconvenient to break it up as that requires recalculating all the rxrpc serial
and sequence numbers and reencrypting all the fragments, so we switch off the
"Don't Fragment" service temporarily and send the bounced packet again. Future
packets then use the new MTU.
That's all fine. The problem lies in rxrpc_UDP_error_report() where the code
that deals with ICMP_FRAG_NEEDED packets lives. Packets of this type have a
field (ee_info) to indicate the maximum packet size at the reporting node - but
sometimes ee_info isn't filled in and is just left as 0 and the code must allow
for this.
When ee_info is 0, the code should take the MTU size we're currently using and
reduce it for the next packet we want to send. However, it takes ee_info
(which is known to be 0) and tries to reduce that instead.
This was discovered by Coverity.
Reported-by: Dave Jones <davej@redhat.com>
Signed-off-by: David Howells <dhowells@redhat.com>
The arsta structure wasn't initialized for
non-ap interfaces. This should fix related
warnings/crashes.
Reported-By: Ben Greear <greearb@candelatech.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
The nwifi header is padded to 4 bytes. This wasn't
a problem until one tried to (at least) rx 4addr
frames.
This finally allows managed iface to be used in a
bridge.
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Makes it easier to determine why some failures
happened.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Commit "kvm: Record the preemption status of vcpus using preempt notifiers"
caused a performance regression on s390. It turned out that in the case that
if a former sleeping cpu, that was woken up, this cpu is not a yield candidate
since it gave up the cpu voluntarily. To retain this candiate its preempted
flag is set during wakeup and interrupt delivery time.
Significant performance measurement work and code analysis to solve this
issue was provided by Mao Chuan Li and his team in Beijing.
Signed-off-by: Michael Mueller <mimu@linux.vnet.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Use the arch specific function kvm_arch_vcpu_runnable() to add a further
criterium to identify a suitable vcpu to yield to during undirected yield
processing.
Signed-off-by: Michael Mueller <mimu@linux.vnet.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A vcpu is defined to be runnable if an interrupt is pending.
Signed-off-by: Michael Mueller <mimu@linux.vnet.ibm.com>
Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Implement the get TSF by simply returning 0 so that IBSS
merging is happening. Otherwise, IBSS nodes that have similar
SSID naming won't merge. This is simply fooling the mac80211
that the TSF in the received beacon is higher than the local TSF.
Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>
Read-only large sptes can be created due to read-only faults as
follows:
- QEMU pagetable entry that maps guest memory is read-only
due to COW.
- Guest read faults such memory, COW is not broken, because
it is a read-only fault.
- Enable dirty logging, large spte not nuked because it is read-only.
- Write-fault on such memory causes guest to loop endlessly
(which must go down to level 1 because dirty logging is enabled).
Fix by dropping large spte when necessary.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Implements an I2C-over-AUX I2C adapter on top of the generic drm_dp_aux
infrastructure. It extracts the retry logic from existing drivers, which
should help in porting those drivers to this new helper.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
---
Changes in v5:
- move comments partially to to header file
- keep MOT set between I2C messages
- return -EPROTO on short reads
Changes in v4:
- fix typo "bitrate" -> "bit rate"
Changes in v3:
- add back DRM_DEBUG_KMS and DRM_ERROR messages
- embed i2c_adapter within struct drm_dp_aux
- fix typo in comment
Add a helper to probe a DP link (read out the supported DPCD revision,
maximum rate, link count and capabilities) as well as power up the DP
link and configure it accordingly.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
---
Changes in v5:
- export helpers
Changes in v4:
- fix a couple of typos in comments as pointed out by Alex Deucher
Changes in v3:
- split into drm_dp_link_power_up() and drm_dp_link_configure()
- do not change sink state for DPCD versions earlier than 1.1
- sleep for 1-2 ms after setting local sink to D0 state
- read and write consecutive registers where possible
- read DPCD revision when link is probed
- remove duplicate kerneldoc
The function reads the link status (6 bytes starting at offset 0x202)
from the DPCD so that it can be conveniently passed to other DPCD
helpers.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
This is a superset of the current i2c_dp_aux bus functionality and can
be used to transfer native AUX in addition to I2C-over-AUX messages.
Helpers are provided to read and write the DPCD, either blockwise or
byte-wise. Many of the existing helpers for DisplayPort take a copy of a
portion of the DPCD and operate on that, without a way to write data
back to the DPCD (e.g. for configuration of the link).
Subsequent patches will build upon this infrastructure to provide common
functionality in a generic way.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
---
Changes in v5:
- move comments partially to struct drm_dp_aux_msg in header file
- return -EPROTO on short reads in DPCD helpers
Changes in v4:
- fix a typo in a comment
Changes in v3:
- reorder drm_dp_dpcd_writeb() arguments to be more intuitive
- return number of bytes transferred in drm_dp_dpcd_write()
- factor out drm_dp_dpcd_access()
- describe error codes
Convert with dev_err() and co from snd_printk(), etc.
As there are too deep indirections (e.g. ep->chip->dev->dev),
a few new local macros, usb_audio_err() & co, are introduced.
Also, the device numbers in some messages are dropped, as they are
shown in the prefix automatically.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Convert with dev_err() and co from snd_printk(), etc.
A couple of prints are difficult to convert with dev_err() so they are
converted to pr_err() at least.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Convert with dev_err() and co from snd_printk(), etc.
A couple of prints are difficult to convert with dev_err() so they are
converted to pr_err() at least.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Convert with dev_err() and co from snd_printk(), etc.
The debug prints are also reformatted to suit with dev_dbg().
Signed-off-by: Takashi Iwai <tiwai@suse.de>
