- 08 Jun, 2020 40 commits
-
-
Christoph Hellwig authored
Rename the current flush_icache_range to flush_icache_user_range as per commit ae92ef8a ("PATCH] flush icache in correct context") there seems to be an assumption that it operates on user addresses. Add a flush_icache_range around it that for now is a no-op. Signed-off-by:
Christoph Hellwig <hch@lst.de> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Acked-by:
Geert Uytterhoeven <geert@linux-m68k.org> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Link: http://lkml.kernel.org/r/20200515143646.3857579-25-hch@lst.deSigned-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Christoph Hellwig authored
flush_icache_user_range will be the name for a generic primitive. Move the arm name so that arm already has an implementation. Signed-off-by:
-
Christoph Hellwig authored
The Xtensa implementation of flush_icache_range seems to be able to cope with user addresses. Just define flush_icache_user_range to flush_icache_range. [jcmvbkbc@gmail.com: fix flush_icache_user_range in noMMU configs] Link: http://lkml.kernel.org/r/20200525221556.4270-1-jcmvbkbc@gmail.comSigned-off-by:
Max Filippov <jcmvbkbc@gmail.com> Signed-off-by:
-
Christoph Hellwig authored
The SuperH implementation of flush_icache_range seems to be able to cope with user addresses. Just define flush_icache_user_range to flush_icache_range. Signed-off-by:
-
Christoph Hellwig authored
Define flush_icache_user_range to flush_icache_range unless the architecture provides its own implementation. Signed-off-by:
-
Christoph Hellwig authored
The function currently known as flush_icache_user_range only operates on a single page. Rename it to flush_icache_user_page as we'll need the name flush_icache_user_range for something else soon. Signed-off-by:
-
Christoph Hellwig authored
flush_icache_user_range is only used by <asm-generic/cacheflush.h>, so remove it from the architectures that implement it, but don't use <asm-generic/cacheflush.h>. Signed-off-by:
-
Christoph Hellwig authored
RISC-V needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Also remove the pointless __KERNEL__ ifdef while we're at it. Signed-off-by:
Palmer Dabbelt <palmerdabbelt@google.com> Acked-by:
-
Christoph Hellwig authored
Power needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Also remove the pointless __KERNEL__ ifdef while we're at it. Signed-off-by:
-
Christoph Hellwig authored
OpenRISC needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Signed-off-by:
-
Christoph Hellwig authored
m68knommu needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Signed-off-by:
Greg Ungerer <gerg@linux-m68k.org> Cc: Greg Ungerer <gerg@linux-m68k.org> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Link: http://lkml.kernel.org/r/20200515143646.3857579-15-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
Microblaze needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Signed-off-by:
-
Christoph Hellwig authored
IA64 needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Signed-off-by:
-
Christoph Hellwig authored
Hexagon needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Signed-off-by:
Brian Cain <bcain@codeaurora.org> Link: http://lkml.kernel.org/r/20200515143646.3857579-12-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
C6x needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Signed-off-by:
Mark Salter <msalter@redhat.com> Cc: Aurelien Jacquiot <jacquiot.aurelien@gmail.com> Link: http://lkml.kernel.org/r/20200515143646.3857579-11-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
ARM64 needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Signed-off-by:
Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will@kernel.org> Link: http://lkml.kernel.org/r/20200515143646.3857579-10-hch@lst.deSigned-off-by:
-
Christoph Hellwig authored
Alpha needs almost no cache flushing routines of its own. Rely on asm-generic/cacheflush.h for the defaults. Signed-off-by:
-
Christoph Hellwig authored
There is a magic ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE cpp symbol that guards non-stub availability of flush_dcache_pagge. Use that to check if flush_dcache_pagg is implemented. Signed-off-by:
-
Christoph Hellwig authored
This seems to lead to some crazy include loops when using asm-generic/cacheflush.h on more architectures, so leave it to the arch header for now. [hch@lst.de: fix warning] Link: http://lkml.kernel.org/r/20200520173520.GA11199@lst.deSigned-off-by:
-
Christoph Hellwig authored
cacheflush.h uses a somewhat to generic include guard name that clashes with various arch files. Use a more specific one. Signed-off-by:
-
Christoph Hellwig authored
flush_cache_user_range is an ARMism not used by any generic or unicore32 specific code. Signed-off-by:
-
Christoph Hellwig authored
flush_icache_user_range is only used by copy_to_user_page, which is only used by core VM code. Signed-off-by:
-
Christoph Hellwig authored
flush_icache_page is only used by mm/memory.c. Signed-off-by:
-
Christoph Hellwig authored
Patch series "sort out the flush_icache_range mess", v2. flush_icache_range is mostly used for kernel address, except for the following cases: - the nommu brk and mmap implementations - the read_code helper that is only used for binfmt_flat, binfmt_elf_fdpic, and binfmt_aout including the broken ia32 compat version - binfmt_flat itself none of which really are used by a typical MMU enabled kernel, as a.out can only be build for alpha and m68k to start with. But strangely enough commit ae92ef8a ("PATCH] flush icache in correct context") added a "set_fs(KERNEL_DS)" around the flush_icache_range call in the module loader, because apparently m68k assumed user pointers. This series first cleans up the cacheflush implementations, largely by switching as much as possible to the asm-generic version after a few preparations, then moves the misnamed current flush_icache_user_range to a new name, to finally introduce a real flush_icache_user_range to be used for the above use cases to flush the instruction cache for a userspace address range. The last patch then drops the set_fs in the module code and moves it into the m68k implementation. This patch (of 29): The arguments passed look bogus, try to fix them to something that seems to make sense. Signed-off-by:
-
John Hubbard authored
This code was using get_user_pages*(), in approximately a "Case 5" scenario (accessing the data within a page), using the categorization from [1]. That means that it's time to convert the get_user_pages*() + put_page() calls to pin_user_pages*() + unpin_user_pages() calls. There is some helpful background in [2]: basically, this is a small part of fixing a long-standing disconnect between pinning pages, and file systems' use of those pages. [1] Documentation/core-api/pin_user_pages.rst [2] "Explicit pinning of user-space pages": https://lwn.net/Articles/807108/Signed-off-by:John Hubbard <jhubbard@nvidia.com> Signed-off-by:
Jan Kara <jack@suse.cz> Acked-by:
Michael S. Tsirkin <mst@redhat.com> Acked-by:
Pankaj Gupta <pankaj.gupta.linux@gmail.com> Cc: Jason Wang <jasowang@redhat.com> Cc: Dave Chinner <david@fromorbit.com> Cc: Jérôme Glisse <jglisse@redhat.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Souptick Joarder <jrdr.linux@gmail.com> Cc: Vlastimil Babka <vbabka@suse.cz> Link: http://lkml.kernel.org/r/20200529234309.484480-3-jhubbard@nvidia.comSigned-off-by:
-
John Hubbard authored
Patch series "vhost, docs: convert to pin_user_pages(), new "case 5"" It recently became clear to me that there are some get_user_pages*() callers that don't fit neatly into any of the four cases that are so far listed in pin_user_pages.rst. vhost.c is one of those. Add a Case 5 to the documentation, and refer to that when converting vhost.c. Thanks to Jan Kara for helping me (again) in understanding the interaction between get_user_pages() and page writeback [1]. This is based on today's mmotm, which has a nearby patch to pin_user_pages.rst that rewords cases 3 and 4. Note that I have only compile-tested the vhost.c patch, although that does also include cross-compiling for a few other arches. Any run-time testing would be greatly appreciated. [1] https://lore.kernel.org/r/20200529070343.GL14550@quack2.suse.cz This patch (of 2): There are four cases listed in pin_user_pages.rst. These are intended to help developers figure out whether to use get_user_pages*(), or pin_user_pages*(). However, the four cases do not cover all the situations. For example, drivers/vhost/vhost.c has a "pin, write to page, set page dirty, unpin" case. Add a fifth case, to help explain that there is a general pattern that requires pin_user_pages*() API calls. [jhubbard@nvidia.com: v2] Link: http://lkml.kernel.org/r/20200601052633.853874-2-jhubbard@nvidia.comSigned-off-by:
-
John Hubbard authored
All of the pin_user_pages*() API calls will cause pages to be dma-pinned. As such, they are all suitable for either DMA, RDMA, and/or Direct IO. The documentation should say so, but it was instead saying that three of the API calls were only suitable for Direct IO. This was discovered when a reviewer wondered why an API call that specifically recommended against Case 2 (DMA/RDMA) was being used in a DMA situation [1]. Fix this by simply deleting those claims. The gup.c comments already refer to the more extensive Documentation/core-api/pin_user_pages.rst, which does have the correct guidance. So let's just write it once, there. [1] https://lore.kernel.org/r/20200529074658.GM30374@kadamSigned-off-by:
David Hildenbrand <david@redhat.com> Acked-by:
Souptick Joarder <jrdr.linux@gmail.com> Cc: Dan Carpenter <dan.carpenter@oracle.com> Cc: Jan Kara <jack@suse.cz> Cc: Vlastimil Babka <vbabka@suse.cz> Link: http://lkml.kernel.org/r/20200529084515.46259-1-jhubbard@nvidia.comSigned-off-by:
-
John Hubbard authored
This code was using get_user_pages*(), and all of the callers so far were in a "Case 2" scenario (DMA/RDMA), using the categorization from [1]. That means that it's time to convert the get_user_pages*() + put_page() calls to pin_user_pages*() + unpin_user_pages() calls. There is some helpful background in [2]: basically, this is a small part of fixing a long-standing disconnect between pinning pages, and file systems' use of those pages. [1] Documentation/core-api/pin_user_pages.rst [2] "Explicit pinning of user-space pages": https://lwn.net/Articles/807108/Signed-off-by: -
John Hubbard authored
Patch series "mm/gup: introduce pin_user_pages_locked(), use it in frame_vector.c", v2. This adds yet one more pin_user_pages*() variant, and uses that to convert mm/frame_vector.c. With this, along with maybe 20 or 30 other recent patches in various trees, we are close to having the relevant gup call sites converted--with the notable exception of the bio/block layer. This patch (of 2): Introduce pin_user_pages_locked(), which is nearly identical to get_user_pages_locked() except that it sets FOLL_PIN and rejects FOLL_GET. As with other pairs of get_user_pages*() and pin_user_pages() API calls, it's prudent to assert that FOLL_PIN is *not* set in the get_user_pages*() call, so add that as part of this. [jhubbard@nvidia.com: v2] Link: http://lkml.kernel.org/r/20200531234131.770697-2-jhubbard@nvidia.comSigned-off-by:
-
John Hubbard authored
Update case 3 so that it covers the use of mmu notifiers, for hardware that does, or does not have replayable page faults. Also, elaborate case 4 slightly, as it was quite cryptic. Signed-off-by:
-
Souptick Joarder authored
API __get_user_pages_fast() renamed to get_user_pages_fast_only() to align with pin_user_pages_fast_only(). As part of this we will get rid of write parameter. Instead caller will pass FOLL_WRITE to get_user_pages_fast_only(). This will not change any existing functionality of the API. All the callers are changed to pass FOLL_WRITE. Also introduce get_user_page_fast_only(), and use it in a few places that hard-code nr_pages to 1. Updated the documentation of the API. Signed-off-by:
-
Rafael Aquini authored
Users with SYS_ADMIN capability can add arbitrary taint flags to the running kernel by writing to /proc/sys/kernel/tainted or issuing the command 'sysctl -w kernel.tainted=...'. This interface, however, is open for any integer value and this might cause an invalid set of flags being committed to the tainted_mask bitset. This patch introduces a simple way for proc_taint() to ignore any eventual invalid bit coming from the user input before committing those bits to the kernel tainted_mask. Signed-off-by:
Rafael Aquini <aquini@redhat.com> Signed-off-by:
Luis Chamberlain <mcgrof@kernel.org> Cc: Kees Cook <keescook@chromium.org> Cc: Iurii Zaikin <yzaikin@google.com> Cc: "Theodore Ts'o" <tytso@mit.edu> Link: http://lkml.kernel.org/r/20200512223946.888020-1-aquini@redhat.comSigned-off-by:
-
Guilherme G. Piccoli authored
Usually when the kernel reaches an oops condition, it's a point of no return; in case not enough debug information is available in the kernel splat, one of the last resorts would be to collect a kernel crash dump and analyze it. The problem with this approach is that in order to collect the dump, a panic is required (to kexec-load the crash kernel). When in an environment of multiple virtual machines, users may prefer to try living with the oops, at least until being able to properly shutdown their VMs / finish their important tasks. This patch implements a way to collect a bit more debug details when an oops event is reached, by printing all the CPUs backtraces through the usage of NMIs (on architectures that support that). The sysctl added (and documented) here was called "oops_all_cpu_backtrace", and when set will (as the name suggests) dump all CPUs backtraces. Far from ideal, this may be the last option though for users that for some reason cannot panic on oops. Most of times oopses are clear enough to indicate the kernel portion that must be investigated, but in virtual environments it's possible to observe hypervisor/KVM issues that could lead to oopses shown in other guests CPUs (like virtual APIC crashes). This patch hence aims to help debug such complex issues without resorting to kdump. Signed-off-by:
Guilherme G. Piccoli <gpiccoli@canonical.com> Signed-off-by:
Kees Cook <keescook@chromium.org> Cc: Luis Chamberlain <mcgrof@kernel.org> Cc: Iurii Zaikin <yzaikin@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Randy Dunlap <rdunlap@infradead.org> Cc: Matthew Wilcox <willy@infradead.org> Link: http://lkml.kernel.org/r/20200327224116.21030-1-gpiccoli@canonical.comSigned-off-by:
-
Guilherme G. Piccoli authored
Commit 401c636a ("kernel/hung_task.c: show all hung tasks before panic") introduced a change in that we started to show all CPUs backtraces when a hung task is detected _and_ the sysctl/kernel parameter "hung_task_panic" is set. The idea is good, because usually when observing deadlocks (that may lead to hung tasks), the culprit is another task holding a lock and not necessarily the task detected as hung. The problem with this approach is that dumping backtraces is a slightly expensive task, specially printing that on console (and specially in many CPU machines, as servers commonly found nowadays). So, users that plan to collect a kdump to investigate the hung tasks and narrow down the deadlock definitely don't need the CPUs backtrace on dmesg/console, which will delay the panic and pollute the log (crash tool would easily grab all CPUs traces with 'bt -a' command). Also, there's the reciprocal scenario: some users may be interested in seeing the CPUs backtraces but not have the system panic when a hung task is detected. The current approach hence is almost as embedding a policy in the kernel, by forcing the CPUs backtraces' dump (only) on hung_task_panic. This patch decouples the panic event on hung task from the CPUs backtraces dump, by creating (and documenting) a new sysctl called "hung_task_all_cpu_backtrace", analog to the approach taken on soft/hard lockups, that have both a panic and an "all_cpu_backtrace" sysctl to allow individual control. The new mechanism for dumping the CPUs backtraces on hung task detection respects "hung_task_warnings" by not dumping the traces in case there's no warnings left. Signed-off-by:
-
Guilherme G. Piccoli authored
After a recent change introduced by Vlastimil's series [0], kernel is able now to handle sysctl parameters on kernel command line; also, the series introduced a simple infrastructure to convert legacy boot parameters (that duplicate sysctls) into sysctl aliases. This patch converts the watchdog parameters softlockup_panic and {hard,soft}lockup_all_cpu_backtrace to use the new alias infrastructure. It fixes the documentation too, since the alias only accepts values 0 or 1, not the full range of integers. We also took the opportunity here to improve the documentation of the previously converted hung_task_panic (see the patch series [0]) and put the alias table in alphabetical order. [0] http://lkml.kernel.org/r/20200427180433.7029-1-vbabka@suse.czSigned-off-by:Vlastimil Babka <vbabka@suse.cz> Cc: Kees Cook <keescook@chromium.org> Cc: Iurii Zaikin <yzaikin@google.com> Cc: Luis Chamberlain <mcgrof@kernel.org> Link: http://lkml.kernel.org/r/20200507214624.21911-1-gpiccoli@canonical.comSigned-off-by:
-
Vlastimil Babka authored
Testing is done by a new parameter debug.test_sysctl.boot_int which defaults to 0 and it's expected that the tester passes a boot parameter that sets it to 1. The test checks if it's set to 1. To distinguish true failure from parameter not being set, the test checks /proc/cmdline for the expected parameter, and whether test_sysctl is built-in and not a module. [vbabka@suse.cz: skip the new test if boot_int sysctl is not present] Link: http://lkml.kernel.org/r/305af605-1e60-cf84-fada-6ce1ca37c102@suse.czSigned-off-by:
-
Vlastimil Babka authored
The testing script recommends CONFIG_TEST_SYSCTL=y, but actually only works with CONFIG_TEST_SYSCTL=m. Testing of sysctl setting via boot param however requires the test to be built-in, so make sure the test script supports it. Signed-off-by:
-
Vlastimil Babka authored
We can now handle sysctl parameters on kernel command line and have infrastructure to convert legacy command line options that duplicate sysctl to become a sysctl alias. This patch converts the hung_task_panic parameter. Note that the sysctl handler is more strict and allows only 0 and 1, while the legacy parameter allowed any non-zero value. But there is little reason anyone would not be using 1. Signed-off-by:
Michal Hocko <mhocko@suse.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Christian Brauner <christian.brauner@ubuntu.com> Cc: David Rientjes <rientjes@google.com> Cc: "Eric W . Biederman" <ebiederm@xmission.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Guilherme G . Piccoli" <gpiccoli@canonical.com> Cc: Iurii Zaikin <yzaikin@google.com> Cc: Ivan Teterevkov <ivan.teterevkov@nutanix.com> Cc: Luis Chamberlain <mcgrof@kernel.org> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Matthew Wilcox <willy@infradead.org> Cc: Michal Hocko <mhocko@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/20200427180433.7029-4-vbabka@suse.czSigned-off-by:
-
Vlastimil Babka authored
We can now handle sysctl parameters on kernel command line, but historically some parameters introduced their own command line equivalent, which we don't want to remove for compatibility reasons. We can, however, convert them to the generic infrastructure with a table translating the legacy command line parameters to their sysctl names, and removing the one-off param handlers. This patch adds the support and makes the first conversion to demonstrate it, on the (deprecated) numa_zonelist_order parameter. Signed-off-by:
-
Vlastimil Babka authored
Patch series "support setting sysctl parameters from kernel command line", v3. This series adds support for something that seems like many people always wanted but nobody added it yet, so here's the ability to set sysctl parameters via kernel command line options in the form of sysctl.vm.something=1 The important part is Patch 1. The second, not so important part is an attempt to clean up legacy one-off parameters that do the same thing as a sysctl. I don't want to remove them completely for compatibility reasons, but with generic sysctl support the idea is to remove the one-off param handlers and treat the parameters as aliases for the sysctl variants. I have identified several parameters that mention sysctl counterparts in Documentation/admin-guide/kernel-parameters.txt but there might be more. The conversion also has varying level of success: - numa_zonelist_order is converted in Patch 2 together with adding the necessary infrastructure. It's easy as it doesn't really do anything but warn on deprecated value these days. - hung_task_panic is converted in Patch 3, but there's a downside that now it only accepts 0 and 1, while previously it was any integer value - nmi_watchdog maps to two sysctls nmi_watchdog and hardlockup_panic, so there's no straighforward conversion possible - traceoff_on_warning is a flag without value and it would be required to handle that somehow in the conversion infractructure, which seems pointless for a single flag This patch (of 5): A recently proposed patch to add vm_swappiness command line parameter in addition to existing sysctl [1] made me wonder why we don't have a general support for passing sysctl parameters via command line. Googling found only somebody else wondering the same [2], but I haven't found any prior discussion with reasons why not to do this. Settings the vm_swappiness issue aside (the underlying issue might be solved in a different way), quick search of kernel-parameters.txt shows there are already some that exist as both sysctl and kernel parameter - hung_task_panic, nmi_watchdog, numa_zonelist_order, traceoff_on_warning. A general mechanism would remove the need to add more of those one-offs and might be handy in situations where configuration by e.g. /etc/sysctl.d/ is impractical. Hence, this patch adds a new parse_args() pass that looks for parameters prefixed by 'sysctl.' and tries to interpret them as writes to the corresponding sys/ files using an temporary in-kernel procfs mount. This mechanism was suggested by Eric W. Biederman [3], as it handles all dynamically registered sysctl tables, even though we don't handle modular sysctls. Errors due to e.g. invalid parameter name or value are reported in the kernel log. The processing is hooked right before the init process is loaded, as some handlers might be more complicated than simple setters and might need some subsystems to be initialized. At the moment the init process can be started and eventually execute a process writing to /proc/sys/ then it should be also fine to do that from the kernel. Sysctls registered later on module load time are not set by this mechanism - it's expected that in such scenarios, setting sysctl values from userspace is practical enough. [1] https://lore.kernel.org/r/BL0PR02MB560167492CA4094C91589930E9FC0@BL0PR02MB5601.namprd02.prod.outlook.com/ [2] https://unix.stackexchange.com/questions/558802/how-to-set-sysctl-using-kernel-command-line-parameter [3] https://lore.kernel.org/r/87bloj2skm.fsf@x220.int.ebiederm.org/Signed-off-by:
Masami Hiramatsu <mhiramat@kernel.org> Acked-by:
-
