- 23 Nov, 2016 5 commits
-
-
Heiko Carstens authored
Enable the contiguous memory allocator but set the default size to zero. If somebody wants to use the cma allocator the "cma=" kernel parameter has to be used. Signed-off-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by:
Martin Schwidefsky <schwidefsky@de.ibm.com>
-
Heiko Carstens authored
In order to make the cma infrastructure usable we need to add a small architecture backend which calls dma_contiguous_reserve. Otherwise we would end up with the cma allocator enabled, but no pool where memory can be allocated from. Signed-off-by:
-
Heiko Carstens authored
Use the psw_bits macro and simplify the code. The generated code is also better since it doesn't contain any conditional branches anymore. Reviewed-by:
Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by:
-
Heiko Carstens authored
We have the s390 specific THREAD_ORDER define and the THREAD_SIZE_ORDER define which is also used in common code. Both have exactly the same semantics. Therefore get rid of THREAD_ORDER and always use THREAD_SIZE_ORDER instead. Signed-off-by:
-
Heiko Carstens authored
For an unknown (historic) reason the s390 specific implementation of set_fs returns whatever the __ctl_load would return. The set_fs macro however is supposed to return void. Change the macro to do that. Signed-off-by:
-
- 17 Nov, 2016 5 commits
-
-
Sebastian Ott authored
Get rid of a useless memset from dma_alloc. Users of dma_alloc who want zero initialized memory can get it by specifying __GFP_ZERO or use one of the zalloc variants. Signed-off-by:
Sebastian Ott <sebott@linux.vnet.ibm.com> Reviewed-by:
Gerald Schaefer <gerald.schaefer@de.ibm.com> Signed-off-by:
-
Sebastian Ott authored
We have 2 strategies to reduce the number of RPCIT instructions: * A HW feature indicated via the tlb_refresh bit allows us to omit RPCIT for invalid -> valid translation-table entry updates. * With "lazy flush" we omit RPCIT for valid -> invalid updates until we run out of dma addresses. When we have to reuse dma addresses we issue a global tlb flush using only one RPCIT instruction. Currently lazy flushing depends on tlb_refresh. Since there is no technical reason for this remove this dependency. Signed-off-by:
-
Sebastian Ott authored
__s390_dma_map_sg maps a dma-contiguous area. Although we only map whole pages we have to take into account that the area doesn't start or stop at a page boundary because we use the dma address to loop over the individual sg entries. Failing to do that might lead to an access of the wrong sg entry. Fixes: ee877b81 ("s390/pci_dma: improve map_sg") Reported-and-tested-by:
Christoph Raisch <raisch@de.ibm.com> Signed-off-by:
-
Martin Schwidefsky authored
The TOD clock offset injected by an STP sync check can be negative. If the resulting total tod_steering_delta gets negative the kernel will panic. Change the type of tod_steering_delta to a signed type. Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Fixes: 75c7b6f3 ("s390/time: steer clocksource on STP sync events") Signed-off-by:
-
Michael Holzheu authored
Signed-off-by:
Michael Holzheu <holzheu@linux.vnet.ibm.com> Signed-off-by:
-
- 15 Nov, 2016 1 commit
-
-
Martin Schwidefsky authored
Move the last two architecture specific fields from the thread_info structure to the thread_struct. All that is left in thread_info is the flags field. Reviewed-by:
-
- 11 Nov, 2016 6 commits
-
-
Martin Schwidefsky authored
The user_timer and system_timer fields are used for the per-thread cputime accounting code. The access to these values is simpler if they are moved to the thread_struct as the task_thread_info(tsk) indirection is not needed anymore. Reviewed-by:
-
Martin Schwidefsky authored
The system_call field in thread_info structure is used by the signal code to store the number of the current system call while the debugger interacts with its inferior. A better location for the system_call field is with the other debugger related information in the thread_struct. Reviewed-by:
-
Heiko Carstens authored
This is the s390 variant of commit 15f4eae7 ("x86: Move thread_info into task_struct"). Signed-off-by:
-
Martin Schwidefsky authored
Convert s390 to use a field in the struct lowcore for the CPU preemption count. It is a bit cheaper to access a lowcore field compared to a thread_info variable and it removes the depencency on a task related structure. bloat-o-meter on the vmlinux image for the default configuration (CONFIG_PREEMPT_NONE=y) reports a small reduction in text size: add/remove: 0/0 grow/shrink: 18/578 up/down: 228/-5448 (-5220) A larger improvement is achieved with the default configuration but with CONFIG_PREEMPT=y and CONFIG_DEBUG_PREEMPT=n: add/remove: 2/6 grow/shrink: 59/4477 up/down: 1618/-228762 (-227144) Reviewed-by:
-
Martin Schwidefsky authored
Replace the bitops specific atomic update code by the functions from atomic_ops.h. This saves a few lines of non-trivial code. Signed-off-by: -
Martin Schwidefsky authored
Rework atomic.h to make the low level functions avaible for use in other headers without using atomic_t, e.g. in bitops.h. Reviewed-by:
-
- 08 Nov, 2016 1 commit
-
-
Masahiro Yamada authored
The dependency between the object and the source is handled by scripts/Makefile.host, so only "hostprogs-y += gen_facilities" is fine. Signed-off-by:
Masahiro Yamada <yamada.masahiro@socionext.com> Acked-by:
-
- 07 Nov, 2016 2 commits
-
-
Masahiro Yamada authored
We generally expect headers in arch/$(ARCH)/include/asm directory are included from kernel sources, but facilities_src.h is not; it is included from the arch/s390/tools/gen_facilities.c tool. There is no reason to expose this header to the public include path. Furthermore, facilities_src.h makes sure to be included only from gen_facilities.c by the following: #ifndef S390_GEN_FACILITIES_C #error "This file can only be included by gen_facilities.c" #endif This check can be removed by merging the two files. Signed-off-by:
-
Masahiro Yamada authored
The header facilities_src.h is only included from gen_facilities.c and the tool is compiled with the following extra options: HOSTCFLAGS_gen_facilities.o += -Wall $(LINUXINCLUDE) Please note $(LINUXINCLUDE) is expanded into build options including: -include $(srctree)/include/linux/kconfig.h So, the Makefile always forces the tool to include kconfig.h, i.e., the #include <linux/kconfig.h> directive in the header is redundant. Signed-off-by:
-
- 31 Oct, 2016 9 commits
-
-
Paul Gortmaker authored
The Kconfig currently controlling compilation of this code is: arch/s390/Kconfig:config S390_GUEST arch/s390/Kconfig: def_bool y ...meaning that it currently is not being built as a module by anyone. Lets remove the modular code that is essentially orphaned, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. Also note that MODULE_DEVICE_TABLE is a no-op for non-modular code. We replace module.h with moduleparam.h since the file does declare some module_param() and leaving that as-is is currently the easiest way to remain compatible with existing boot arg use cases. Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Cornelia Huck <cornelia.huck@de.ibm.com> Cc: linux-s390@vger.kernel.org Signed-off-by:
Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by:
-
Paul Gortmaker authored
The Kconfig currently controlling compilation of this code is: drivers/pci/hotplug/Kconfig:config HOTPLUG_PCI_S390 drivers/pci/hotplug/Kconfig: bool "System z PCI Hotplug Support" ...meaning that it currently is not being built as a module by anyone. Lets remove the couple traces of modular infrastructure use, so that when reading the driver there is no doubt it is builtin-only. We also delete the MODULE_LICENSE tag etc. since all that information was (or is now) contained at the top of the file in the comments. We don't exchange module.h for init.h or export.h since the file does not contain any initcalls or EXPORT of symbols. Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com> Cc: Bjorn Helgaas <bhelgaas@google.com> Cc: linux-s390@vger.kernel.org Signed-off-by:
-
Paul Gortmaker authored
The Makefile currently controlling compilation of this code is obj-y meaning that it currently is not being built as a module by anyone. Lets remove the couple traces of modular infrastructure use, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. We replace module.h with init.h and export.h since the file does export some symbols. Cc: linux-s390@vger.kernel.org Signed-off-by:
-
Paul Gortmaker authored
The Kconfig currently controlling compilation of this code is: arch/s390/Kconfig:config S390_HYPFS_FS arch/s390/Kconfig: def_bool y ...meaning that it currently is not being built as a module by anyone. Lets remove the couple traces of modular infrastructure use, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. Also note that MODULE_ALIAS is a no-op for non-modular code. We also delete the MODULE_LICENSE tag etc. since all that information was (or is now) contained at the top of the file in the comments. Build testing indicated the presence of module.h was masking an implicit include of kobject.h, hence the addition of that. Cc: linux-s390@vger.kernel.org Signed-off-by:
-
Paul Gortmaker authored
The Makefile currently controlling compilation of this code is obj-y, meaning that it currently is not being built as a module by anyone. Lets remove the couple traces of modular infrastructure, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. Cc: linux-s390@vger.kernel.org Signed-off-by:
-
Paul Gortmaker authored
The Kconfig currently controlling compilation of this code is: drivers/s390/char/Kconfig:config SCLP_TTY drivers/s390/char/Kconfig: def_bool y ...meaning that it currently is not being built as a module by anyone. Lets remove the couple traces of modular infrastructure use, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. We don't replace module.h with init.h since the file already has that. Cc: Peter Oberparleiter <oberpar@linux.vnet.ibm.com> Cc: linux-s390@vger.kernel.org Signed-off-by:
-
Paul Gortmaker authored
The Kconfig currently controlling compilation of this code is: drivers/s390/char/Kconfig:config TN3215 drivers/s390/char/Kconfig: def_bool y ...meaning that it currently is not being built as a module by anyone. Lets remove the modular code that is essentially orphaned, so that when reading the driver there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. We don't replace module.h with init.h since the file already has that. Cc: linux-s390@vger.kernel.org Signed-off-by:
-
Paul Gortmaker authored
The Kconfig currently controlling compilation of this code is: arch/s390/Kconfig:config CRASH_DUMP arch/s390/Kconfig: bool "kernel crash dumps" ...meaning that it currently is not being built as a module by anyone. Lets remove the modular code that is essentially orphaned, so that when reading the driver there is no doubt it is builtin-only. Since module_init wasn't even being used by this file, the init ordering remains unchanged with this commit. We also delete the MODULE_LICENSE tag etc. since all that information was (or is now) contained at the top of the file in the comments. We don't replace module.h with init.h since the file already has that. Cc: linux-s390@vger.kernel.org Signed-off-by:
-
Paul Gortmaker authored
The Makefile currently controlling compilation of this code is: obj-y += airq.o blacklist.o chsc.o cio.o css.o chp.o idset.o isc.o \ fcx.o itcw.o crw.o ccwreq.o trace.o ioasm.o ccw_device-objs += device.o device_fsm.o device_ops.o ccw_device-objs += device_id.o device_pgid.o device_status.o obj-y += ccw_device.o cmf.o ...meaning that the files here are not being built as modular. Lets remove the couple traces of modular infrastructure use, so that when reading the code there is no doubt it is builtin-only. Since module_init translates to device_initcall in the non-modular case, the init ordering remains unchanged with this commit. Also note that MODULE_DEVICE_TABLE is a no-op for non-modular code. We delete the MODULE_LICENSE tag etc. since all that information was (or is now) contained at the top of the file in the comments. We replace module.h with export.h where the file does export some symbols. Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Peter Oberparleiter <oberpar@linux.vnet.ibm.com> Cc: Cornelia Huck <cornelia.huck@de.ibm.com> Cc: Arnd Bergmann <arndb@de.ibm.com> Cc: linux-s390@vger.kernel.org Signed-off-by:
-
- 28 Oct, 2016 11 commits
-
-
Jan Höppner authored
The function dasd_ro_store() calls set_disk_ro() to set the device in question read-only. Since set_disk_ro() might sleep, we can't call it while holding a lock. However, we also can't simply check if the device, block, and gdp references are valid before we call set_disk_ro() because an offline processing might have been started in the meanwhile which will destroy those references. In order to reliably call set_disk_ro() we have to ensure several things: - Still check validity of the mentioned references but additionally check if offline processing is running and bail out accordingly. Also, do this while holding the device lock. - To ensure that the block device is still safe after the lock, increase the open_count while still holding the device lock. Reviewed-by:
Stefan Haberland <sth@linux.vnet.ibm.com> Signed-off-by:
Jan Höppner <hoeppner@linux.vnet.ibm.com> Signed-off-by:
-
Jan Höppner authored
The reference to a device in question may get lost when the extended error reporting (EER) attribute is being enabled/disabled while the device is set offline at the same time. This is due to missing refcounting and incorrect locking. Fix this by the following: - In dasd_eer_store() get the device directly and handle the refcount accordingly. - Move the lock in dasd_eer_enable() up so we can ensure safe processing. - Check if the device is being set offline and return with -EBUSY if so. - While at it, change the return code from -EPERM to -EMEDIUMTYPE as suggested by a FIXME, since that is what we're actually checking. Reviewed-by:
-
Jan Höppner authored
Before we set a device offline, the open_count for the block device is checked and certain flags are checked and set as well. However, this is all done without holding any lock. Potentially, if the open_count was checked but the DASD_FLAG_OFFLINE wasn't set yet, a different process might want to increase the open_count depending on whether DASD_FLAG_OFFLINE is set or not in the meanwhile. This is quite racy and can lead to the loss of the device for that process and subsequently lead to a panic. Fix this by checking the open_count and setting the offline flags while holding the ccwdev lock. Reviewed-by:
-
Jan Höppner authored
When setting certain attributes, we actually set the according feature flag. Do this by using dasd_set_feature() at a few occurrences and remove duplicate code. In dasd_set_feature() dasd_find_busid() is used to retrieve the devmap for the device in question. Combined with the change above, this would require the device to be set online at least once so that a devmap is being created. Change that by using dasd_devmap_from_cdev() instead, which uses dasd_find_busid() first and will create a devmap accordingly if there is none yet. Reviewed-by:
-
Jan Höppner authored
simple_strtoul() has been marked obsolete for quite some time now. Replace a few last occurrences with kstrtouint(). Reviewed-by:
-
Dong Jia Shi authored
Clean up DEV_STATE_SENSE_PGID related code, since it's not used anymore. Everything related to path verification is handled within DEV_STATE_VERIFY. Signed-off-by:
Dong Jia Shi <bjsdjshi@linux.vnet.ibm.com> Acked-by:
-
Martin Schwidefsky authored
On STP sync events the TOD clock will jump in time, either forward or backward. The TOD clocksource claims to be continuous but in case of an STP sync with a negative offset it is not. Subtract the offset injected by the STP sync check from the result of the TOD clocksource to make it continuous again. Add code to drift the offset towards zero with a fixed rate, steering 1 second in ~9 hours. Suggested-by:
David Hildenbrand <dahi@linux.vnet.ibm.com> Signed-off-by:
-
Martin Schwidefsky authored
The last_update_clock time stamp in the lowcore should be adjusted by the TOD clock delta that is created by the clock synchronization. Otherwise the calculation of the steal time will be incorrect. Signed-off-by: -
Martin Schwidefsky authored
Merge clock_sync_cpu into stp_sync_clock and split out the update of the global and per-CPU clock fields into clock_sync_global and clock_sync_local. Signed-off-by: -
Jan Höppner authored
block->request_queue is used many times in dasd_setup_queue. Define a separate variable to increase readability a bit and to make it better reusable. Reviewed-by:
-
Jan Höppner authored
Currently the block queue value max_segments is set to -1L, which is then implicitly casted to unsigned short in blk_queue_max_segments. This results in 65535 (64k) max_segments. Even though the resulting value is correct, setting it implicitly using -1L is rather confusing. Set the value explicitly using the USHRT_MAX macro instead. Reviewed-by:
-
