- 02 Jul, 2019 7 commits
-
-
Pierre Morel authored
We prepare the interception of the PQAP/AQIC instruction for the case the AQIC facility is enabled in the guest. First of all we do not want to change existing behavior when intercepting AP instructions without the SIE allowing the guest to use AP instructions. In this patch we only handle the AQIC interception allowed by facility 65 which will be enabled when the complete interception infrastructure will be present. We add a callback inside the KVM arch structure for s390 for a VFIO driver to handle a specific response to the PQAP instruction with the AQIC command and only this command. But we want to be able to return a correct answer to the guest even there is no VFIO AP driver in the kernel. Therefor, we inject the correct exceptions from inside KVM for the case the callback is not initialized, which happens when the vfio_ap driver is not loaded. We do consider the responsibility of the driver to always initialize the PQAP callback if it defines queues by initializing the CRYCB for a guest. If the callback has been setup we call it. If not we setup an answer considering that no queue is available for the guest when no callback has been setup. Signed-off-by:
Pierre Morel <pmorel@linux.ibm.com> Reviewed-by:
Tony Krowiak <akrowiak@linux.ibm.com> Acked-by:
Harald Freudenberger <freude@linux.ibm.com> Acked-by:
Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by:
Halil Pasic <pasic@linux.ibm.com> Signed-off-by:
Vasily Gorbik <gor@linux.ibm.com>
-
Vasily Gorbik authored
Kasan instrumentation of backchain unwinder stack reads is disabled completely and simply uses READ_ONCE_NOCHECK now. READ_ONCE_TASK_STACK macro is unused and could be removed. Signed-off-by: -
Vasily Gorbik authored
Avoid kasan false positive when current task is interrupted in-between stack frame allocation and backchain write instructions leaving new stack frame backchain invalid. In particular if backchain is 0 the unwinder tries to read pt_regs from the stack and might hit kasan poisoned bytes, leading to kasan "stack-out-of-bounds" report. Disable kasan instrumentation of unwinder stack reads, since this limitation couldn't be handled otherwise with current backchain unwinder implementation. Fixes: 78c98f90 ("s390/unwind: introduce stack unwind API") Reported-by:
Julian Wiedmann <jwi@linux.ibm.com> Tested-by:
Benjamin Block <bblock@linux.ibm.com> Signed-off-by:
-
Julian Wiedmann authored
Current code sets the dsci to 0x00000080. Which doesn't make any sense, as the indicator area is located in the _left-most_ byte. Worse: if the dsci is the _shared_ indicator, this potentially clears the indication of activity for a _different_ device. tiqdio_thinint_handler() will then have no reason to call that device's IRQ handler, and the device ends up stalling. Fixes: d0c9d4a8 ("[S390] qdio: set correct bit in dsci") Cc: <stable@vger.kernel.org> Signed-off-by:
-
Julian Wiedmann authored
When tiqdio_remove_input_queues() removes a queue from the tiq_list as part of qdio_shutdown(), it doesn't re-initialize the queue's list entry and the prev/next pointers go stale. If a subsequent qdio_establish() fails while sending the ESTABLISH cmd, it calls qdio_shutdown() again in QDIO_IRQ_STATE_ERR state and tiqdio_remove_input_queues() will attempt to remove the queue entry a second time. This dereferences the stale pointers, and bad things ensue. Fix this by re-initializing the list entry after removing it from the list. For good practice also initialize the list entry when the queue is first allocated, and remove the quirky checks that papered over this omission. Note that prior to commit e5218134 ("s390/qdio: fix access to uninitialized qdio_q fields"), these checks were bogus anyway. setup_queues_misc() clears the whole queue struct, and thus needs to re-init the prev/next pointers as well. Fixes: 779e6e1c ("[S390] qdio: new qdio driver.") Cc: <stable@vger.kernel.org> Signed-off-by:
-
Dan Carpenter authored
The "len" variable is the length of the option up to the next option or to the end of the string which ever first. We want to print the invalid option so we want precision "%.*s" but the format is width "%*s" so it prints up to the end of the string. Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Tested-by:
Stefan Haberland <sth@linux.ibm.com> Signed-off-by:
-
Cornelia Huck authored
Sometimes, we want to control which of the matching drivers binds to a subchannel device (e.g. for subchannels we want to handle via vfio-ccw). For pci devices, a mechanism to do so has been introduced in 782a985d ("PCI: Introduce new device binding path using pci_dev.driver_override"). It makes sense to introduce the driver_override attribute for subchannel devices as well, so that we can easily extend the 'driverctl' tool (which makes use of the driver_override attribute for pci). Note that unlike pci we still require a driver override to match the subchannel type; matching more than one subchannel type is probably not useful anyway. Signed-off-by:
Cornelia Huck <cohuck@redhat.com> Reviewed-by:
Sebastian Ott <sebott@linux.ibm.com> Signed-off-by:
-
- 24 Jun, 2019 2 commits
-
-
Cornelia Huck authored
Reported by sparse. Fixes: 7f8e89a8 ("vfio-ccw: Factor out the ccw0-to-ccw1 transition") Signed-off-by:
-
Vasily Gorbik authored
Merge tag 'vfio-ccw-20190621' of https://git.kernel.org/pub/scm/linux/kernel/git/kvms390/vfio-ccw into features Refactoring of the vfio-ccw cp handling, simplifying the code and avoiding unneeded allocating/copying. * tag 'vfio-ccw-20190621' of https://git.kernel.org/pub/scm/linux/kernel/git/kvms390/vfio-ccw: vfio-ccw: Remove copy_ccw_from_iova() vfio-ccw: Factor out the ccw0-to-ccw1 transition vfio-ccw: Copy CCW data outside length calculation vfio-ccw: Skip second copy of guest cp to host vfio-ccw: Move guest_cp storage into common struct s390/cio: Combine direct and indirect CCW paths vfio-ccw: Rearrange IDAL allocation in direct CCW vfio-ccw: Remove pfn_array_table vfio-ccw: Adjust the first IDAW outside of the nested loops vfio-ccw: Rearrange pfn_array and pfn_array_table arrays s390/cio: Use generalized CCW handler in cp_init() s390/cio: Generalize the TIC handler s390/cio: Refactor the routine that handles TIC CCWs s390/cio: Squash cp_free() and cp_unpin_free() Signed-off-by:
-
- 21 Jun, 2019 5 commits
-
-
Eric Farman authored
Just to keep things tidy. Signed-off-by:
Eric Farman <farman@linux.ibm.com> Message-Id: <20190618202352.39702-6-farman@linux.ibm.com> Reviewed-by:
Farhan Ali <alifm@linux.ibm.com> Signed-off-by:
-
Eric Farman authored
This is a really useful function, but it's buried in the copy_ccw_from_iova() routine so that ccwchain_calc_length() can just work with Format-1 CCWs while doing its counting. But it means we're translating a full 2K of "CCWs" to Format-1, when in reality there's probably far fewer in that space. Let's factor it out, so maybe we can do something with it later. Signed-off-by:
-
Eric Farman authored
It doesn't make much sense to "hide" the copy to the channel_program struct inside a routine that calculates the length of the chain. Let's move it to the calling routine, which will later copy from channel_program to the memory it allocated itself. Signed-off-by:
-
Eric Farman authored
We already pinned/copied/unpinned 2K (256 CCWs) of guest memory to the host space anchored off vfio_ccw_private. There's no need to do that again once we have the length calculated, when we could just copy the section we need to the "permanent" space for the I/O. Signed-off-by:
-
Eric Farman authored
Rather than allocating/freeing a piece of memory every time we try to figure out how long a CCW chain is, let's use a piece of memory allocated for each device. The io_mutex added with commit 4f766173 ("vfio-ccw: protect the I/O region") is held for the duration of the VFIO_CCW_EVENT_IO_REQ event that accesses/uses this space, so there should be no race concerns with another CPU attempting an (unexpected) SSCH for the same device. Suggested-by:
-
- 19 Jun, 2019 4 commits
-
-
Julian Wiedmann authored
This allows device drivers (eg. qeth) to use the struct when processing information retrieved via RCD. Signed-off-by:
-
Heiko Carstens authored
The stfle inline assembly returns the number of double words written (condition code 0) or the double words it would have written (condition code 3), if the memory array it got as parameter would have been large enough. The current stfle implementation assumes that the array is always large enough and clears those parts of the array that have not been written to with a subsequent memset call. If however the array is not large enough memset will get a negative length parameter, which means that memset clears memory until it gets an exception and the kernel crashes. To fix this simply limit the maximum length. Move also the inline assembly to an extra function to avoid clobbering of register 0, which might happen because of the added min_t invocation together with code instrumentation. The bug was introduced with commit 14375bc4 ("[S390] cleanup facility list handling") but was rather harmless, since it would only write to a rather large array. It became a potential problem with commit 3ab121ab ("[S390] kernel: Add z/VM LGR detection"). Since then it writes to an array with only four double words, while some machines already deliver three double words. As soon as machines have a facility bit within the fifth double a crash on IPL would happen. Fixes: 14375bc4 ("[S390] cleanup facility list handling") Cc: <stable@vger.kernel.org> # v2.6.37+ Reviewed-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by:
-
Heiko Carstens authored
This feature has never been used, so remove it. Acked-by:
Hendrik Brueckner <brueckner@linux.ibm.com> Signed-off-by:
-
Heiko Carstens authored
Replace defconfig with performance_defconfig. defconfig had some more or less random debug options enabled, where nobody knows why anymore. Just remove the old defconfig and replace it with performance_defconfig, which reduces the number of configs to maintain. A config with debugging options enabled is debug_defconfig which is supposed to be rather close to performance_defconfig except that is has debug options enabled. Acked-by:
-
- 17 Jun, 2019 9 commits
-
-
Eric Farman authored
With both the direct-addressed and indirect-addressed CCW paths simplified to this point, the amount of shared code between them is (hopefully) more easily visible. Move the processing of IDA-specific bits into the direct-addressed path, and add some useful commentary of what the individual pieces are doing. This allows us to remove the entire ccwchain_fetch_idal() routine and maintain a single function for any non-TIC CCW. Signed-off-by:
-
Eric Farman authored
This is purely deck furniture, to help understand the merge of the direct and indirect handlers. Signed-off-by:
-
Eric Farman authored
Now that both CCW codepaths build this nested array: ccwchain->pfn_array_table[1]->pfn_array[#idaws/#pages] We can collapse this into simply: ccwchain->pfn_array[#idaws/#pages] Let's do that, so that we don't have to continually navigate two nested arrays when the first array always has a count of one. Signed-off-by:
-
Eric Farman authored
Now that pfn_array_table[] is always an array of 1, it seems silly to check for the very first entry in an array in the middle of two nested loops, since we know it'll only ever happen once. Let's move this outside the loops to simplify things, even though the "k" variable is still necessary. Signed-off-by:
-
Eric Farman authored
While processing a channel program, we currently have two nested arrays that carry a slightly different structure. The direct CCW path creates this: ccwchain->pfn_array_table[1]->pfn_array[#pages] while an IDA CCW creates: ccwchain->pfn_array_table[#idaws]->pfn_array[1] The distinction appears to state that each pfn_array_table entry points to an array of contiguous pages, represented by a pfn_array, um, array. Since the direct-addressed scenario can ONLY represent contiguous pages, it makes the intermediate array necessary but difficult to recognize. Meanwhile, since an IDAL can contain non-contiguous pages and there is no logic in vfio-ccw to detect adjacent IDAWs, it is the second array that is necessary but appearing to be superfluous. I am not aware of any documentation that states the pfn_array[] needs to be of contiguous pages; it is just what the code does today. I don't see any reason for this either, let's just flip the IDA codepath around so that it generates: ch_pat->pfn_array_table[1]->pfn_array[#idaws] This will bring it in line with the direct-addressed codepath, so that we can understand the behavior of this memory regardless of what type of CCW is being processed. And it means the casual observer does not need to know/care whether the pfn_array[] represents contiguous pages or not. NB: The existing vfio-ccw code only supports 4K-block Format-2 IDAs, so that "#pages" == "#idaws" in this area. This means that we will have difficulty with this overlap in terminology if support for Format-1 or 2K-block Format-2 IDAs is ever added. I don't think that this patch changes our ability to make that distinction. Signed-off-by:
-
Eric Farman authored
It is now pretty apparent that ccwchain_handle_ccw() (nee ccwchain_handle_tic()) does everything that cp_init() wants to do. Let's remove that duplicated code from cp_init() and let ccwchain_handle_ccw() handle it itself. Signed-off-by:
-
Eric Farman authored
Refactor ccwchain_handle_tic() into a routine that handles a channel program address (which itself is a CCW pointer), rather than a CCW pointer that is only a TIC CCW. This will make it easier to reuse this code for other CCW commands. Signed-off-by:
-
Eric Farman authored
Extract the "does the target of this TIC already exist?" check from ccwchain_handle_tic(), so that it's easier to refactor that function into one that cp_init() is able to use. Signed-off-by:
-
Eric Farman authored
The routine cp_free() does nothing but call cp_unpin_free(), and while most places call cp_free() there is one caller of cp_unpin_free() used when the cp is guaranteed to have not been marked initialized. This seems like a dubious way to make a distinction, so let's combine these routines and make cp_free() do all the work. Signed-off-by:
-
- 15 Jun, 2019 13 commits
-
-
Martin Schwidefsky authored
Signed-off-by:
Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by:
-
Heiko Carstens authored
stop_machine is the only user left of cpu_relax_yield. Given that it now has special semantics which are tied to stop_machine introduce a weak stop_machine_yield function which architectures can override, and get rid of the generic cpu_relax_yield implementation. Acked-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Martin Schwidefsky authored
The stop_machine loop to advance the state machine and to wait for all affected CPUs to check-in calls cpu_relax_yield in a tight loop until the last missing CPUs acknowledged the state transition. On a virtual system where not all logical CPUs are backed by real CPUs all the time it can take a while for all CPUs to check-in. With the current definition of cpu_relax_yield a diagnose 0x44 is done which tells the hypervisor to schedule *some* other CPU. That can be any CPU and not necessarily one of the CPUs that need to run in order to advance the state machine. This can lead to a pretty bad diagnose 0x44 storm until the last missing CPU finally checked-in. Replace the undirected cpu_relax_yield based on diagnose 0x44 with a directed yield. Each CPU in the wait loop will pick up the next CPU in the cpumask of stop_machine. The diagnose 0x9c is used to tell the hypervisor to run this next CPU instead of the current one. If there is only a limited number of real CPUs backing the virtual CPUs we end up with the real CPUs passed around in a round-robin fashion. [heiko.carstens@de.ibm.com]: Use cpumask_next_wrap as suggested by Peter Zijlstra. Signed-off-by: -
Heiko Carstens authored
spin_cpu_yield is unused, therefore remove it. Acked-by:
-
Vasily Gorbik authored
Simplify conditions and remove unnecessary variable in data exception handler. Signed-off-by:
-
Halil Pasic authored
The hypervisor needs to interact with the summary indicators, so these need to be DMA memory as well (at least for protected virtualization guests). Signed-off-by:
Michael Mueller <mimu@linux.ibm.com> Tested-by:
-
Halil Pasic authored
Before virtio-ccw could get away with not using DMA API for the pieces of memory it does ccw I/O with. With protected virtualization this has to change, since the hypervisor needs to read and sometimes also write these pieces of memory. The hypervisor is supposed to poke the classic notifiers, if these are used, out of band with regards to ccw I/O. So these need to be allocated as DMA memory (which is shared memory for protected virtualization guests). Let us factor out everything from struct virtio_ccw_device that needs to be DMA memory in a satellite that is allocated as such. Note: The control blocks of I/O instructions do not need to be shared. These are marshalled by the ultravisor. Signed-off-by:
-
Halil Pasic authored
This will come in handy soon when we pull out the indicators from virtio_ccw_device to a memory area that is shared with the hypervisor (in particular for protected virtualization guests). Signed-off-by:
-
Halil Pasic authored
The flag AIRQ_IV_CACHELINE was recently added to airq_iv_create(). Let us use it! We actually wanted the vector to span a cacheline all along. Signed-off-by:
-
Halil Pasic authored
Protected virtualization guests have to use shared pages for airq notifier bit vectors, because the hypervisor needs to write these bits. Let us make sure we allocate DMA memory for the notifier bit vectors by replacing the kmem_cache with a dma_cache and kalloc() with cio_dma_zalloc(). Signed-off-by:
-
Halil Pasic authored
As virtio-ccw devices are channel devices, we need to use the dma area within the common I/O layer for any communication with the hypervisor. Note that we do not need to use that area for control blocks directly referenced by instructions, e.g. the orb. It handles neither QDIO in the common code, nor any device type specific stuff (like channel programs constructed by the DASD driver). An interesting side effect is that virtio structures are now going to get allocated in 31 bit addressable storage. Signed-off-by:
-
Halil Pasic authored
To support protected virtualization cio will need to make sure the memory used for communication with the hypervisor is DMA memory. Let us introduce one global pool for cio. Our DMA pools are implemented as a gen_pool backed with DMA pages. The idea is to avoid each allocation effectively wasting a page, as we typically allocate much less than PAGE_SIZE. Signed-off-by:
-
Halil Pasic authored
On s390, protected virtualization guests have to use bounced I/O buffers. That requires some plumbing. Let us make sure, any device that uses DMA API with direct ops correctly is spared from the problems, that a hypervisor attempting I/O to a non-shared page would bring. Signed-off-by:
Claudio Imbrenda <imbrenda@linux.ibm.com> Reviewed-by:
-
