- 26 Mar, 2018 5 commits
-
-
Julian Wiedmann authored
Get rid of the confusing two-stage translation in a hot path, and only handle CCQs that we anticipate for the respective command. Any unexpected value (such as CCQ 97 (rc == 1) for SQBS) should be considered a severe HW/driver bug, and traced as such. Signed-off-by:
Julian Wiedmann <jwi@linux.vnet.ibm.com> Reviewed-by:
Benjamin Block <bblock@linux.vnet.ibm.com> Signed-off-by:
Martin Schwidefsky <schwidefsky@de.ibm.com>
-
Julian Wiedmann authored
Immediate retry of EQBS after CCQ 96 means that we potentially misreport the state of buffers inspected during the first EQBS call. This occurs when 1. the first EQBS finds all inspected buffers still in the initial state set by the driver (ie INPUT EMPTY or OUTPUT PRIMED), 2. the EQBS terminates early with CCQ 96, and 3. by the time that the second EQBS comes around, the state of those previously inspected buffers has changed. If the state reported by the second EQBS is 'driver-owned', all we know is that the previous buffers are driver-owned now as well. But we can't tell if they all have the same state. So for instance - the second EQBS reports OUTPUT EMPTY, but any number of the previous buffers could be OUTPUT ERROR by now, - the second EQBS reports OUTPUT ERROR, but any number of the previous buffers could be OUTPUT EMPTY by now. Effectively, this can result in both over- and underreporting of errors. If the state reported by the second EQBS is 'HW-owned', that doesn't guarantee that the previous buffers have not been switched to driver-owned in the mean time. So for instance - the second EQBS reports INPUT EMPTY, but any number of the previous buffers could be INPUT PRIMED (or INPUT ERROR) by now. This would result in failure to process pending work on the queue. If it's the final check before yielding initiative, this can cause a (temporary) queue stall due to IRQ avoidance. Fixes: 25f269f1 ("[S390] qdio: EQBS retry after CCQ 96") Cc: <stable@vger.kernel.org> #v3.2+ Signed-off-by:
-
Julian Wiedmann authored
Only attempt to merge PENDING into EMPTY buffers for devices where the PENDING state is actually expected (ie. IQD with CQ). This might speed up the hot path a little bit. Signed-off-by:
Ursula Braun <ubraun@linux.vnet.ibm.com> Reviewed-by:
-
Julian Wiedmann authored
On an Output queue, both EMPTY and PENDING buffer states imply that the buffer is ready for completion-processing by the upper-layer drivers. So for a non-QEBSM Output queue, get_buf_states() merges mixed batches of PENDING and EMPTY buffers into one large batch of EMPTY buffers. The upper-layer driver (ie. qeth) later distuingishes PENDING from EMPTY by inspecting the slsb_state for QDIO_OUTBUF_STATE_FLAG_PENDING. But the merge logic in get_buf_states() contains a bug that causes us to erronously also merge ERROR buffers into such a batch of EMPTY buffers (ERROR is 0xaf, EMPTY is 0xa1; so ERROR & EMPTY == EMPTY). Effectively, most outbound ERROR buffers are currently discarded silently and processed as if they had succeeded. Note that this affects _all_ non-QEBSM device types, not just IQD with CQ. Fix it by explicitly spelling out the exact conditions for merging. For extracting the "get initial state" part out of the loop, this relies on the fact that get_buf_states() is never called with a count of 0. The QEBSM path already strictly requires this, and the two callers with variable 'count' make sure of it. Fixes: 104ea556 ("qdio: support asynchronous delivery of storage blocks") Cc: <stable@vger.kernel.org> #v3.2+ Signed-off-by:
-
Julian Wiedmann authored
When determining the buffer count that get_buf_states() should be queried for, 'count' is capped at 127 buffers. So the check q->first_to_check == (q->first_to_check + count) % 128 can be reduced to count == 0 This helps to emphasize that get_buf_states() is really only called with count > 0. Signed-off-by:
-
- 19 Mar, 2018 8 commits
-
-
Martin Schwidefsky authored
Merge tag 'vfio-ccw-20180305' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/vfio-ccw into features Pull vfio-ccw patches from Cornelia Huck: A small documentation update, and reject transport mode requests (which we don't support).
-
Vasily Gorbik authored
It seems to be suboptimal to compress the entire image, so that we could decompress and throw away the first 68k (0x11000) during the boot process. This patch trims 0x11000 bytes during the build process and adjusts symbols in size.h accordingly. Besides making bzImage smaller, this change would also be a benefit for kaslr. It allows to decompress the kernel right to desired offset, without worrying about dead weight in front of the decompressed image. Signed-off-by:
Vasily Gorbik <gor@linux.vnet.ibm.com> Acked-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by:
-
Harald Freudenberger authored
A kernel crash occurs when the aes_s390 kernel module is removed on machines < z14. This only happens on kernel version 4.15 and higher on machines not supporting MSA 8. The reason for the crash is a unconditional crypto_unregister_aead() invocation where no previous crypto_register_aead() had been called. The fix now remembers if there has been a successful registration and only then calls the unregister function upon kernel module remove. The code now crashing has been introduced with "bf7fa038 s390/crypto: add s390 platform specific aes gcm support." Signed-off-by:
Harald Freudenberger <freude@linux.vnet.ibm.com> Signed-off-by:
-
Christian Borntraeger authored
Signed-off-by:
Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by:
-
Christian Borntraeger authored
commit 8f50af49 ("s390/console: Make preferred console handling more consistent") created a separate console state for the ascii console. This has the side effect that we register no tty for the line mode interface as soon as there an ascii interface as default console. Under KVM this results in no getty program on the line mode tty if the guest has both types of interfaces. As we can have multiple ttys at the same time we do not want to disable the tty on sclp_line0 under KVM. So instead of checking for the console mode, we now check for the presence of the sclp line mode interface. As z/VM multiplexes the line mode interface on the 32xx screen we continue to disable the line mode tty for the z/VM case. CC: Peter Oberparleiter <oberpar@linux.vnet.ibm.com> Fixes: 8f50af49 ("s390/console: Make preferred console handling more consistent") Signed-off-by:
-
Farhan Ali authored
The S390 architecture does not support any graphics hardware, but with the latest support for Virtio GPU in Linux and Virtio GPU emulation in QEMU, it's possible to enable graphics for S390 using the Virtio GPU device. To enable display we need to enable the Linux Virtual Terminal (VT) layer for S390. But the VT subsystem initializes quite early at boot so we need a dummy console driver till the Virtio GPU driver is initialized and we can run the framebuffer console. The framebuffer console over a Virtio GPU device can be run in combination with the serial SCLP console (default on S390). The SCLP console can still be accessed by management applications (eg: via Libvirt's virsh console). Signed-off-by:
Farhan Ali <alifm@linux.vnet.ibm.com> Acked-by:
Thomas Huth <thuth@redhat.com> Message-Id: <e23b61f4f599ba23881727a1e8880e9d60cc6a48.1519315352.git.alifm@linux.vnet.ibm.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Farhan Ali authored
The Linux Virtual Terminal (VT) layer provides a default keymap which is compiled when VT layer is enabled. But at the same time we are also compiling the EBCDIC keymap and this causes the linker to complain. So let's rename the EBCDIC keymap variables to prevent linker conflict. Signed-off-by:
-
Farhan Ali authored
The 'commit e25df120 ("[S390] Kconfig: menus with depends on HAS_IOMEM.")' added the HAS_IOMEM dependecy for "Graphics support". This disabled the "Graphics support" menu for S390. But if we enable VT layer for S390, we would also need to enable the dummy console. So let's remove the HAS_IOMEM dependency. Move this dependency to sub menu items and console drivers that use io memory. Signed-off-by:
Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by:
-
- 01 Mar, 2018 2 commits
-
-
Cornelia Huck authored
vfio-ccw only supports command mode for channel programs, not transport mode. User space is supposed to already take care of that and pass us command-mode ORBs only, but better make sure and return an error to the caller instead of trying to process tcws as ccws. Reviewed-by:
Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com> Acked-by:
Halil Pasic <pasic@linux.vnet.ibm.com> Signed-off-by:
Cornelia Huck <cohuck@redhat.com>
-
Cornelia Huck authored
The vfio-ccw documentation comes from the cover letter of the original patch submission, which shows in some parts. Give it some love; in particular: - Remove/rework statements that make sense in a cover letter, but not in regular documentation. - Fix some typos. - Describe the current limitations in more detail. Acked-by:
-
- 27 Feb, 2018 6 commits
-
-
Stefan Haberland authored
Set the XRC timestamps even if XRC is not supported by the storage server to help debugging the storage server firmware. Do not advertise valid time stamps if the system time could not be obtained. Reviewed-by:
Sebastian Ott <sebott@linux.vnet.ibm.com> Reviewed-by:
Peter Oberparleiter <oberpar@linux.vnet.ibm.com> Signed-off-by:
Stefan Haberland <sth@linux.vnet.ibm.com> Signed-off-by:
-
Stefan Haberland authored
Reported by smatch that the usage of cqr->block is inconsistent. The sanity check is not needed because _dasd_requeue_request already checks for a valid cqr->block pointer and all referenced ERP requests have a valid cqr->block pointer as well since it is copied during ERP process. Signed-off-by:
Jan Hoeppner <hoeppner@linux.vnet.ibm.com> Signed-off-by:
-
Julian Wiedmann authored
s/sene/sense Signed-off-by:
-
Vasily Gorbik authored
Even though s390 compressed kernel support has been introduced long ago, by default uncompressed kernel image is used by packaging targets like: "make rpm-pkg", "make deb-pkg" as well as "make image_name". At the same time "make install" would use compressed bzImage by default. To avoid ambiguity, set bzImage as default image for packaging. Signed-off-by:
-
Vasily Gorbik authored
Common code defines linker symbols which denote sections start/end in a form of char []. Referencing those symbols as _symbol or &_symbol yields the same result, but "_symbol" form is more widespread across newly written code. Convert s390 specific code to this style. Also removes unused _text symbol definition in boot/compressed/misc.c. Signed-off-by:
-
Vasily Gorbik authored
Exception table (__ex_table section) is not used during the decompressor phase and could be discarded to save the memory. It is currently generated due to sclp_service_call function (sclp_early_core.c). An assumption is that decompressor usage of sclp_service_call via sclp_early_printk should never trigger exceptions. Signed-off-by:
-
- 22 Feb, 2018 13 commits
-
-
Cornelia Huck authored
This function is checking for the suspend control, not the function control. Signed-off-by:
-
Claudio Imbrenda authored
Change the size of the sclp mask to 64 bits. Reviewed-by:
Claudio Imbrenda <imbrenda@linux.vnet.ibm.com> Signed-off-by:
-
Claudio Imbrenda authored
Qemu before version 2.11 does not implement the architecture correctly, and does not allow for a mask size of size different than 4. This patch introduces a compatibility mode for such systems, forcing the mask sizes to 4. Since the mask size is currently still 4 anyway, this patch should have no impact whatsoever by itself, but it will be needed when the mask size is increased to 64 bits in the next patch. Reviewed-by:
-
Claudio Imbrenda authored
Switch the layout of the event masks to be a generic buffer, and implement accessors to retrieve the values of the masks. This will be needed in the next patches, where we will eventually switch the mask size to 64 bits. Reviewed-by:
-
Claudio Imbrenda authored
Replace hardcoded instances where 32 or unsigned int (or long) is used for SCLP event masks, and replace with sizeof(sccb_mask_t) and sccb_mask_t respectively. This improves readability and prepares for when we will increase sccb_mask_t to 64 bits. Reviewed-by:
-
Stefan Haberland authored
Make the behavior in case of constant IFCC/CCC errors configurable. Add a sysfs attribute to switch between path disabled after threshold exceeded (default) and message only. Reviewed-by:
-
Peter Oberparleiter authored
Add functions to retrieve data associated with an SCLP Store Data entity. Automatically retrieve data for the "config" entity during boot and make that data available to user-space via sysfs: /sys/firmware/sclp_sd/config/data Reading from this file will return config data contents. /sys/firmware/sclp_sd/config/reload Writing to this file will cause the latest version of data related to the config entity to be read from the SCLP interface. Generate a KOBJ_CHANGE whenever new data is retrieved. Signed-off-by: -
Heiko Carstens authored
Provide base_asce_alloc() and base_asce_free() helper functions which can be used to allocate an ASCE and all required region, segment and page tables required to access memory regions of the virtual kernel address space. Both, the ASCE and all tables, do not use any features that correspond to e.g. enhanced DAT features. This is required for some I/O functions that pass an ASCE, like e.g. some service call requests, but which may not use any enhanced features. Acked-by:
Janosch Frank <frankja@linux.vnet.ibm.com> Signed-off-by:
-
Martin Schwidefsky authored
The system call path can be interrupted before the switch back to the standard branch prediction with BPENTER has been done. The critical section cleanup code skips forward to .Lsysc_do_svc and bypasses the BPENTER. In this case the kernel and all subsequent code will run with the limited branch prediction. Fixes: eacf67eb9b32 ("s390: run user space and KVM guests with modified branch prediction") Signed-off-by: -
Sebastian Ott authored
When we terminate driver I/O (because we need to stop using a certain channel path) we also need to ensure that a timer (which may have been set up using ccw_device_start_timeout) is cleared. Signed-off-by:
-
Sebastian Ott authored
When a timeout occurs for users of ccw_device_start_timeout we will stop the IO and call the drivers int handler with the irb pointer set to ERR_PTR(-ETIMEDOUT). Sometimes however we'd set the irb pointer to ERR_PTR(-EIO) which is not intended. Just set the correct value in all codepaths. Reported-by:
-
Sebastian Ott authored
There are cases a device driver can't start IO because the device is currently in use by cio. In this case the device driver is notified when the device is usable again. Using ccw_device_start_timeout we would set the timeout (and change an existing timeout) before we test for internal usage. Worst case this could lead to an unexpected timer deletion. Fix this by setting the timeout after we test for internal usage. Signed-off-by:
-
Hendrik Brueckner authored
Commit f19fbd5e ("s390: introduce execute-trampolines for branches") introduces .cfi_* assembler directives. Instead of using the directives directly, use the macros from asm/dwarf.h. This also ensures that the dwarf debug information are created in the .debug_frame section. Fixes: f19fbd5e ("s390: introduce execute-trampolines for branches") Signed-off-by:
Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by:
-
- 19 Feb, 2018 3 commits
-
-
Eugeniu Rosca authored
I've accidentally stumbled upon the IS_ENABLED(EXPOLINE_*) lines, which obviously always evaluate to false. Fix this. Fixes: f19fbd5e ("s390: introduce execute-trampolines for branches") Signed-off-by:
Eugeniu Rosca <erosca@de.adit-jv.com> Signed-off-by:
-
Stefan Haberland authored
Internal DASD device driver I/O such as query host access count or path verification is started using the _sleep_on() function. To mark a request as started or ended the callback_data is set to either DASD_SLEEPON_START_TAG or DASD_SLEEPON_END_TAG. In cases where the request has to be stopped unconditionally the status is set to DASD_SLEEPON_END_TAG as well which leads to immediate clearing of the request. But the request might still be on a device request queue for normal operation which might lead to a panic because of a BUG() statement in __dasd_device_process_final_queue() or a list corruption of the device request queue. Fix by removing the setting of DASD_SLEEPON_END_TAG in the dasd_cancel_req() and dasd_generic_requeue_all_requests() functions and ensure that the request is not deleted in the requeue function. Trigger the device tasklet in the requeue function and let the normal processing cleanup the request. Signed-off-by:
-
Linus Torvalds authored
-
- 18 Feb, 2018 3 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull x86 Kconfig fixes from Thomas Gleixner: "Three patchlets to correct HIGHMEM64G and CMPXCHG64 dependencies in Kconfig when CPU selections are explicitely set to M586 or M686" * 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/Kconfig: Explicitly enumerate i686-class CPUs in Kconfig x86/Kconfig: Exclude i586-class CPUs lacking PAE support from the HIGHMEM64G Kconfig group x86/Kconfig: Add missing i586-class CPUs to the X86_CMPXCHG64 Kconfig group
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull perf updates from Thomas Gleixner: "Perf tool updates and kprobe fixes: - perf_mmap overwrite mode fixes/overhaul, prep work to get 'perf top' using it, making it bearable to use it in large core count systems such as Knights Landing/Mill Intel systems (Kan Liang) - s/390 now uses syscall.tbl, just like x86-64 to generate the syscall table id -> string tables used by 'perf trace' (Hendrik Brueckner) - Use strtoull() instead of home grown function (Andy Shevchenko) - Synchronize kernel ABI headers, v4.16-rc1 (Ingo Molnar) - Document missing 'perf data --force' option (Sangwon Hong) - Add perf vendor JSON metrics for ARM Cortex-A53 Processor (William Cohen) - Improve error handling and error propagation of ftrace based kprobes so failures when installing kprobes are not silently ignored and create disfunctional tracepoints" * 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (27 commits) kprobes: Propagate error from disarm_kprobe_ftrace() kprobes: Propagate error from arm_kprobe_ftrace() Revert "tools include s390: Grab a copy of arch/s390/include/uapi/asm/unistd.h" perf s390: Rework system call table creation by using syscall.tbl perf s390: Grab a copy of arch/s390/kernel/syscall/syscall.tbl tools/headers: Synchronize kernel ABI headers, v4.16-rc1 perf test: Fix test trace+probe_libc_inet_pton.sh for s390x perf data: Document missing --force option perf tools: Substitute yet another strtoull() perf top: Check the latency of perf_top__mmap_read() perf top: Switch default mode to overwrite mode perf top: Remove lost events checking perf hists browser: Add parameter to disable lost event warning perf top: Add overwrite fall back perf evsel: Expose the perf_missing_features struct perf top: Check per-event overwrite term perf mmap: Discard legacy interface for mmap read perf test: Update mmap read functions for backward-ring-buffer test perf mmap: Introduce perf_mmap__read_event() perf mmap: Introduce perf_mmap__read_done() ... -
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull irq updates from Thomas Gleixner: "A small set of updates mostly for irq chip drivers: - MIPS GIC fix for spurious, masked interrupts - fix for a subtle IPI bug in GICv3 - do not probe GICv3 ITSs that are marked as disabled - multi-MSI support for GICv2m - various small cleanups" * 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: irqdomain: Re-use DEFINE_SHOW_ATTRIBUTE() macro irqchip/bcm: Remove hashed address printing irqchip/gic-v2m: Add PCI Multi-MSI support irqchip/gic-v3: Ignore disabled ITS nodes irqchip/gic-v3: Use wmb() instead of smb_wmb() in gic_raise_softirq() irqchip/gic-v3: Change pr_debug message to pr_devel irqchip/mips-gic: Avoid spuriously handling masked interrupts
-
