- 18 Apr, 2023 16 commits
-
-
Kevin Brodsky authored
typeof is (still) a GNU extension, which means that it cannot be used when building ISO C (e.g. -std=c99). It should therefore be avoided in uapi headers in favour of the ISO-friendly __typeof__. Unfortunately this issue could not be detected by CONFIG_UAPI_HEADER_TEST=y as the __ALIGN_KERNEL() macro is not expanded in any uapi header. This matters from a userspace perspective, not a kernel one. uapi headers and their contents are expected to be usable in a variety of situations, and in particular when building ISO C applications (with -std=c99 or similar). This particular problem can be reproduced by trying to use the __ALIGN_KERNEL macro directly in application code, say: #include <linux/const.h> int align(int x, int a) { return __KERNEL_ALIGN(x, a); } and trying to build that with -std=c99. Link: https://lkml.kernel.org/r/20230411092747.3759032-1-kevin.brodsky@arm.com Fixes: a79ff731 ("netfilter: xtables: make XT_ALIGN() usable in exported headers by exporting __ALIGN_KERNEL()") Signed-off-by:Kevin Brodsky <kevin.brodsky@arm.com> Reported-by:
Ruben Ayrapetyan <ruben.ayrapetyan@arm.com> Tested-by:
Petr Vorel <pvorel@suse.cz> Tested-by:
Masahiro Yamada <masahiroy@kernel.org> Cc: Sam Ravnborg <sam@ravnborg.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org>
-
Yang Yang authored
Delay accounting does not track the delay of IRQ/SOFTIRQ. While IRQ/SOFTIRQ could have obvious impact on some workloads productivity, such as when workloads are running on system which is busy handling network IRQ/SOFTIRQ. Get the delay of IRQ/SOFTIRQ could help users to reduce such delay. Such as setting interrupt affinity or task affinity, using kernel thread for NAPI etc. This is inspired by "sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure"[1]. Also fix some code indent problems of older code. And update tools/accounting/getdelays.c: / # ./getdelays -p 156 -di print delayacct stats ON printing IO accounting PID 156 CPU count real total virtual total delay total delay average 15 15836008 16218149 275700790 18.380ms IO count delay total delay average 0 0 0.000ms SWAP count delay total delay average 0 0 0.000ms RECLAIM count delay total delay average 0 0 0.000ms THRASHING count delay total delay average 0 0 0.000ms COMPACT count delay total delay average 0 0 0.000ms WPCOPY count delay total delay average 36 7586118 0.211ms IRQ count delay total delay average 42 929161 0.022ms [1] commit 52b1364b("sched/psi: Add PSI_IRQ to track IRQ/SOFTIRQ pressure") Link: https://lkml.kernel.org/r/202304081728353557233@zte.com.cnSigned-off-by:Yang Yang <yang.yang29@zte.com.cn> Cc: Jiang Xuexin <jiang.xuexin@zte.com.cn> Cc: wangyong <wang.yong12@zte.com.cn> Cc: junhua huang <huang.junhua@zte.com.cn> Cc: Balbir Singh <bsingharora@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Juri Lelli <juri.lelli@redhat.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by:
-
Amjad Ouled-Ameur authored
join() expects strings but integers are given. Convert chunks list to strings before passing it to join() Link: https://lkml.kernel.org/r/20230406221217.1585486-4-f.fainelli@gmail.comSigned-off-by:
Amjad Ouled-Ameur <aouledameur@baylibre.com> Signed-by:
Florian Fainelli <f.fainelli@gmail.com> Tested-by:
-
Florian Fainelli authored
This GDB script prints the interrupts in the system in the same way that /proc/interrupts does. This does include the architecture specific part done by arch_show_interrupts() for x86, ARM, ARM64 and MIPS. Example output from an ARM64 system: (gdb) lx-interruptlist CPU0 CPU1 CPU2 CPU3 10: 3167 1225 1276 2629 GICv2 30 Level arch_timer 13: 0 0 0 0 GICv2 36 Level arm-pmu 14: 0 0 0 0 GICv2 37 Level arm-pmu 15: 0 0 0 0 GICv2 38 Level arm-pmu 16: 0 0 0 0 GICv2 39 Level arm-pmu 28: 0 0 0 0 interrupt-controller@8410640 5 Edge brcmstb-gpio-wake 30: 125 0 0 0 GICv2 128 Level ttyS0 31: 0 0 0 0 interrupt-controller@8416000 0 Level mspi_done 32: 0 0 0 0 interrupt-controller@8410640 3 Edge brcmstb-waketimer 33: 0 0 0 0 interrupt-controller@8418580 8 Edge brcmstb-waketimer-rtc 34: 872 0 0 0 GICv2 230 Level brcm_scmi@0 35: 0 0 0 0 interrupt-controller@8410640 10 Edge 8d0f200.usb-phy 37: 0 0 0 0 GICv2 97 Level PCIe PME 42: 0 0 0 0 GICv2 145 Level xhci-hcd:usb1 43: 94 0 0 0 GICv2 71 Level mmc1 44: 0 0 0 0 GICv2 70 Level mmc0 IPI0: 23 666 154 98 Rescheduling interrupts IPI1: 247 1053 1701 634 Function call interrupts IPI2: 0 0 0 0 CPU stop interrupts IPI3: 0 0 0 0 CPU stop (for crash dump) interrupts IPI4: 0 0 0 0 Timer broadcast interrupts IPI5: 7 9 5 0 IRQ work interrupts IPI6: 0 0 0 0 CPU wake-up interrupts ERR: 0 Link: https://lkml.kernel.org/r/20230406220451.1583239-1-f.fainelli@gmail.comSigned-off-by: -
Florian Fainelli authored
If CONFIG_DEBUG_INFO_REDUCED is enabled in the kernel configuration, we will typically not be able to load vmlinux-gdb.py and will fail with: Traceback (most recent call last): File "/home/fainelli/work/buildroot/output/arm64/build/linux-custom/vmlinux-gdb.py", line 25, in <module> import linux.utils File "/home/fainelli/work/buildroot/output/arm64/build/linux-custom/scripts/gdb/linux/utils.py", line 131, in <module> atomic_long_counter_offset = atomic_long_type.get_type()['counter'].bitpos KeyError: 'counter' Rather be left wondering what is happening only to find out that reduced debug information is the cause, raise an eror. This was not typically a problem until e3c8d33e ("scripts/gdb: fix 'lx-dmesg' on 32 bits arch") but it has since then. Link: https://lkml.kernel.org/r/20230406215252.1580538-1-f.fainelli@gmail.com Fixes: e3c8d33e ("scripts/gdb: fix 'lx-dmesg' on 32 bits arch") Signed-off-by: -
Kieran Bingham authored
Linux makes use of the Radix Tree data structure to store pointers indexed by integer values. This structure is utilised across many structures in the kernel including the IRQ descriptor tables, and several filesystems. This module provides a method to lookup values from a structure given its head node. Usage: The function lx_radix_tree_lookup, must be given a symbol of type struct radix_tree_root, and an index into that tree. The object returned is a generic integer value, and must be cast correctly to the type based on the storage in the data structure. For example, to print the irq descriptor in the sparse irq_desc_tree at index 18, try the following: (gdb) print (struct irq_desc)$lx_radix_tree_lookup(irq_desc_tree, 18) This script previously existed under commit e127a73d ("scripts/gdb: add a Radix Tree Parser") and was later reverted with b447e025i (Revert "scripts/gdb: add a Radix Tree Parser"). This version expects the XArray based radix tree implementation and has been verified using QEMU/x86 on Linux 6.3-rc5. [f.fainelli@gmail.com: revive and update for xarray implementation] [f.fainelli@gmail.com: guard against a NULL node in the while loop] Link: https://lkml.kernel.org/r/20230405222743.1191674-1-f.fainelli@gmail.com Link: https://lkml.kernel.org/r/20230404214049.1016811-1-f.fainelli@gmail.comSigned-off-by:
Kieran Bingham <kieran.bingham@linaro.org> Signed-off-by:
-
Noah Goldstein authored
This has a slight benefit for x86 and has no effect on other targets. The benefit to x86 is it change the codegen for setting a node to block from `mov %r0, %r1; or $RB_BLACK, %r1` to `lea RB_BLACK(%r0), %r1` which saves an instructions. In all other cases it just replace ALU with ALU (or -> and) which perform the same on all machines I am aware of. Total instructions in rbtree.o: Before - 802 After - 782 so it saves about 20 `mov` instructions. Link: https://lkml.kernel.org/r/20230404221350.3806566-1-goldstein.w.n@gmail.comSigned-off-by:Noah Goldstein <goldstein.w.n@gmail.com> Cc: Michel Lespinasse <michel@lespinasse.org> Signed-off-by:
-
Heiko Carstens authored
The last (only) architecture specific arch_idle_time() implementation was removed with commit be76ea61 ("s390/idle: remove arch_cpu_idle_time() and corresponding code"). Therefore remove the now dead code in fs/proc/stat.c as well. Link: https://lkml.kernel.org/r/20230405143452.2677172-1-hca@linux.ibm.comSigned-off-by:
Heiko Carstens <hca@linux.ibm.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
-
Matthieu Baerts authored
"Link:" and "Closes:" tags have to be used with public URLs. It is difficult to make sure the link is public but at least we can verify the tag is followed by 'http(s)://'. With that, we avoid such a tag that is not allowed [1]: Closes: <number> Now that we check the "link" tags are followed by a URL, we can relax the check linked to "Reported-by being followed by a link tag" to only verify if a "link" tag is present after the "Reported-by" one. Link: https://lore.kernel.org/linux-doc/CAHk-=wh0v1EeDV3v8TzK81nDC40=XuTdY2MCr0xy3m3FiBV3+Q@mail.gmail.com/ [1] Link: https://lkml.kernel.org/r/20230314-doc-checkpatch-closes-tag-v4-5-d26d1fa66f9f@tessares.netSigned-off-by:
Matthieu Baerts <matthieu.baerts@tessares.net> Acked-by:
Joe Perches <joe@perches.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Bagas Sanjaya <bagasdotme@gmail.com> Cc: Daniel Vetter <daniel@ffwll.ch> Cc: David Airlie <airlied@gmail.com> Cc: Dwaipayan Ray <dwaipayanray1@gmail.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Kai Wasserbäch <kai@dev.carbon-project.org> Cc: Konstantin Ryabitsev <konstantin@linuxfoundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Lukas Bulwahn <lukas.bulwahn@gmail.com> Cc: Thorsten Leemhuis <linux@leemhuis.info> Signed-off-by:
-
Matthieu Baerts authored
As a follow-up of a previous patch modifying the documentation to allow using the "Closes:" tag, checkpatch.pl is updated accordingly. checkpatch.pl now no longer complain when the "Closes:" tag is used by itself: commit 76f381bb ("checkpatch: warn when unknown tags are used for links") ... or after the "Reported-by:" tag: commit d7f1d71e ("checkpatch: warn when Reported-by: is not followed by Link:") Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/373 Link: https://lkml.kernel.org/r/20230314-doc-checkpatch-closes-tag-v4-4-d26d1fa66f9f@tessares.netSigned-off-by:
-
Matthieu Baerts authored
The following commit will allow the use of a similar "link" tag. Because there is a possibility that other similar tags will be added in the future and to reduce the number of places where the code will be modified to allow this new tag, a list with all these "link" tags is now used. Two variables are created from it: one to search for such tags and one to print all tags in a warning message. Link: https://lkml.kernel.org/r/20230314-doc-checkpatch-closes-tag-v4-3-d26d1fa66f9f@tessares.netSigned-off-by:
-
Matthieu Baerts authored
When checking if "Reported-by" tag is followed by "Link:", there is no need to print the next line if there is no next line. While at it, also mention in this case that the "Link:" tag should be followed by a URL, similar to the next warning. By doing that, the code is now similar to what is done above when checking if the Co-developed-by tag is properly used. Link: https://lkml.kernel.org/r/20230314-doc-checkpatch-closes-tag-v4-2-d26d1fa66f9f@tessares.net Fixes: d7f1d71e ("checkpatch: warn when Reported-by: is not followed by Link:") Signed-off-by:
-
Matthieu Baerts authored
Since v6.3, checkpatch.pl now complains about the use of "Closes:" tags followed by a link [1]. It also complains if a "Reported-by:" tag is followed by a "Closes:" one [2]. As detailed in the first patch, this "Closes:" tag is used for a bit of time, mainly by DRM and MPTCP subsystems. It is used by some bug trackers to automate the closure of issues when a patch is accepted. It is even planned to use this tag with bugzilla.kernel.org [3]. The first patch updates the documentation to explain what is this "Closes:" tag and how/when to use it. The second patch modifies checkpatch.pl to stop complaining about it. The DRM maintainers and their mailing list have been added in Cc as they are probably interested by these two patches as well. [1] https://lore.kernel.org/all/3b036087d80b8c0e07a46a1dbaaf4ad0d018f8d5.1674217480.git.linux@leemhuis.info/ [2] https://lore.kernel.org/all/bb5dfd55ea2026303ab2296f4a6df3da7dd64006.1674217480.git.linux@leemhuis.info/ [3] https://lore.kernel.org/linux-doc/20230315181205.f3av7h6owqzzw64p@meerkat.local/ This patch (of 5): Making sure a bug tracker is up to date is not an easy task. For example, a first version of a patch fixing a tracked issue can be sent a long time after having created the issue. But also, it can take some time to have this patch accepted upstream in its final form. When it is done, someone -- probably not the person who accepted the patch -- has to remember about closing the corresponding issue. This task of closing and tracking the patch can be done automatically by bug trackers like GitLab [1], GitHub [2] and hopefully soon [3] bugzilla.kernel.org when the appropriated tag is used. The two first ones accept multiple tags but it is probably better to pick one. According to commit 76f381bb ("checkpatch: warn when unknown tags are used for links"), the "Closes" tag seems to have been used in the past by a few people and it is supported by popular bug trackers. Here is how it has been used in the past: $ git log --no-merges --format=email -P --grep='^Closes: http' | \ grep '^Closes: http' | cut -d/ -f3-5 | sort | uniq -c | sort -rn 391 gitlab.freedesktop.org/drm/intel 79 github.com/multipath-tcp/mptcp_net-next 8 gitlab.freedesktop.org/drm/msm 3 gitlab.freedesktop.org/drm/amd 2 gitlab.freedesktop.org/mesa/mesa 1 patchwork.freedesktop.org/series/73320 1 gitlab.freedesktop.org/lima/linux 1 gitlab.freedesktop.org/drm/nouveau 1 github.com/ClangBuiltLinux/linux 1 bugzilla.netfilter.org/show_bug.cgi?id=1579 1 bugzilla.netfilter.org/show_bug.cgi?id=1543 1 bugzilla.netfilter.org/show_bug.cgi?id=1436 1 bugzilla.netfilter.org/show_bug.cgi?id=1427 1 bugs.debian.org/625804 Likely here, the "Closes" tag was only properly used with GitLab and GitHub. We can also see that it has been used quite a few times (and still used recently) and this is then not a "random tag that makes no sense" like it was the case with "BugLink" recently [4]. It has also been misused but that was a long time ago, when it was common to use many different random tags. checkpatch.pl script should then stop complaining about this "Closes" tag. As suggested by Thorsten [5], if this tag is accepted, it should first be described in the documentation. This is what is done here in this patch. To avoid confusion, the "Closes" should be used with any public bug report. No need to check if the underlying bug tracker supports automations. Having this tag with any kind of public bug reports allows bots like regzbot to clearly identify patches fixing a specific bug and avoid false-positives, e.g. patches mentioning it is related to an issue but not fixing it. As suggested by Thorsten [6] again, if we follow the same logic, the "Closes" tag should then be used after a "Reported-by" one. Note that thanks to this "Closes" tag, the mentioned bug trackers can also locate where a patch has been applied in different branches and repositories. If only the "Link" tag is used, the tracking can also be done but the ticket will not be closed and a manual operation will be needed. Also, these bug trackers have some safeguards: the closure is only done if a commit having the "Closes:" tag is applied in a specific branch. It will then not be closed if a random commit having the same tag is published elsewhere. Also in case of closure, a notification is sent to the owners. Link: https://lkml.kernel.org/r/20230314-doc-checkpatch-closes-tag-v4-0-d26d1fa66f9f@tessares.net Link: https://docs.gitlab.com/ee/user/project/issues/managing_issues.html#default-closing-pattern [1] Link: https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/using-keywords-in-issues-and-pull-requests [2] Link: https://lore.kernel.org/linux-doc/20230315181205.f3av7h6owqzzw64p@meerkat.local/ [3] Link: https://lore.kernel.org/all/CAHk-=wgs38ZrfPvy=nOwVkVzjpM3VFU1zobP37Fwd_h9iAD5JQ@mail.gmail.com/ [4] Link: https://lore.kernel.org/all/688cd6cb-90ab-6834-a6f5-97080e39ca8e@leemhuis.info/ [5] Link: https://lore.kernel.org/linux-doc/2194d19d-f195-1a1e-41fc-7827ae569351@leemhuis.info/ [6] Link: https://github.com/multipath-tcp/mptcp_net-next/issues/373 Link: https://lkml.kernel.org/r/20230314-doc-checkpatch-closes-tag-v4-1-d26d1fa66f9f@tessares.netSigned-off-by:
Thorsten Leemhuis <linux@leemhuis.info> Acked-by:
Konstantin Ryabitsev <konstantin@linuxfoundation.org> Acked-by:
-
Peng Liu authored
HRTIMER_MAX_CLOCK_BASES is of enum type hrtimer_base_type. To print it as an integer, HRTIMER_MAX_CLOCK_BASES should be converted first. Link: https://lkml.kernel.org/r/TYCP286MB214640FF0E7F04AC3926A39EC6819@TYCP286MB2146.JPNP286.PROD.OUTLOOK.COMSigned-off-by:
Peng Liu <liupeng17@lenovo.com> Reviewed-by:
Jan Kiszka <jan.kiszka@siemens.com> Cc: Florian Fainelli <f.fainelli@gmail.com> Cc: Kieran Bingham <kbingham@kernel.org> Signed-off-by:
-
Peng Liu authored
Below incompatibilities between Python2 and Python3 made lx-timerlist fail to run under Python3. o xrange() is replaced by range() in Python3 o bytes and str are different types in Python3 o the return value of Inferior.read_memory() is memoryview object in Python3 akpm: cc stable so that older kernels are properly debuggable under newer Python. Link: https://lkml.kernel.org/r/TYCP286MB2146EE1180A4D5176CBA8AB2C6819@TYCP286MB2146.JPNP286.PROD.OUTLOOK.COMSigned-off-by:
-
Peng Liu authored
commit 511885d7 ("lib/timerqueue: Rely on rbtree semantics for next timer") changed struct timerqueue_head, and so print_active_timers() should be changed accordingly with its way to interpret the structure. Link: https://lkml.kernel.org/r/TYCP286MB21463BD277330B26DDC18903C6819@TYCP286MB2146.JPNP286.PROD.OUTLOOK.COMSigned-off-by:
-
- 08 Apr, 2023 24 commits
-
-
Andy Shevchenko authored
When we get a random number to generate a flag in the valid range of UNESCAPE flags, use UNESCAPE_ALL_MASK, It's more correct and prevents from missed updates of the test coverage in the future if any. Link: https://lkml.kernel.org/r/20230327142604.48213-1-andriy.shevchenko@linux.intel.comSigned-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk> Signed-off-by:
-
Florian Fainelli authored
Avoid generating an exception if there are no generic power domain(s) registered: (gdb) lx-genpd-summary domain status children /device runtime status ---------------------------------------------------------------------- Python Exception <class 'gdb.error'>: No symbol "gpd_list" in current context. Error occurred in Python: No symbol "gpd_list" in current context. (gdb) quit [f.fainelli@gmail.com: correctly invoke gdb_eval_or_none] Link: https://lkml.kernel.org/r/20230327185746.3856407-1-f.fainelli@gmail.com Link: https://lkml.kernel.org/r/20230323231659.3319941-1-f.fainelli@gmail.com Fixes: 8207d4a8 ("scripts/gdb: add lx-genpd-summary command") Signed-off-by: -
Florian Fainelli authored
Avoid generating an exception if there are no clocks registered: (gdb) lx-clk-summary enable prepare protect clock count count count rate ------------------------------------------------------------------------ Python Exception <class 'gdb.error'>: No symbol "clk_root_list" in current context. Error occurred in Python: No symbol "clk_root_list" in current context. Link: https://lkml.kernel.org/r/20230323225246.3302977-1-f.fainelli@gmail.com Fixes: d1e9710b ("scripts/gdb: initial clk support: lx-clk-summary") Signed-off-by: -
Bjorn Helgaas authored
arch_kexec_kernel_image_load() only calls kexec_image_load_default(), and there are no arch-specific implementations. Remove the unnecessary arch_kexec_kernel_image_load() and make kexec_image_load_default() static. No functional change intended. Link: https://lkml.kernel.org/r/20230307224416.907040-3-helgaas@kernel.orgSigned-off-by:
Bjorn Helgaas <bhelgaas@google.com> Reviewed-by:
Simon Horman <horms@kernel.org> Acked-by:
Baoquan He <bhe@redhat.com> Cc: Borislav Petkov (AMD) <bp@alien8.de> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Eric Biederman <ebiederm@xmission.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Bjorn Helgaas authored
Patch series "kexec: Remove unnecessary arch hook", v2. There are no arch-specific things in arch_kexec_kernel_image_load(), so remove it and just use the generic version. This patch (of 2): The x86 implementation of arch_kexec_kernel_image_load() is functionally identical to the generic arch_kexec_kernel_image_load(): arch_kexec_kernel_image_load # x86 if (!image->fops || !image->fops->load) return ERR_PTR(-ENOEXEC); return image->fops->load(image, image->kernel_buf, ...) arch_kexec_kernel_image_load # generic kexec_image_load_default if (!image->fops || !image->fops->load) return ERR_PTR(-ENOEXEC); return image->fops->load(image, image->kernel_buf, ...) Remove the x86-specific version and use the generic arch_kexec_kernel_image_load(). No functional change intended. Link: https://lkml.kernel.org/r/20230307224416.907040-1-helgaas@kernel.org Link: https://lkml.kernel.org/r/20230307224416.907040-2-helgaas@kernel.orgSigned-off-by: -
Cai Huoqing authored
Remove pci_clear_master to simplify the code, the bus-mastering is also cleared in do_pci_disable_device, like this: ./drivers/pci/pci.c:2197 static void do_pci_disable_device(struct pci_dev *dev) { u16 pci_command; pci_read_config_word(dev, PCI_COMMAND, &pci_command); if (pci_command & PCI_COMMAND_MASTER) { pci_command &= ~PCI_COMMAND_MASTER; pci_write_config_word(dev, PCI_COMMAND, pci_command); } pcibios_disable_device(dev); }. And dev->is_busmaster is set to 0 in pci_disable_device. Link: https://lkml.kernel.org/r/20230323113711.10523-1-cai.huoqing@linux.devSigned-off-by:Cai Huoqing <cai.huoqing@linux.dev> Cc: Alexandre Bounine <alex.bou9@gmail.com> Cc: Matt Porter <mporter@kernel.crashing.org> Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Signed-off-by:
-
Andy Shevchenko authored
For the sake of cleaning up the kernel.h split the hexadecimal related helpers to own header called 'hex.h'. Link: https://lkml.kernel.org/r/20230323155029.40000-1-andriy.shevchenko@linux.intel.comSigned-off-by:
-
Paolo Abeni authored
We are observing huge contention on the epmutex during an http connection/rate test: 83.17% 0.25% nginx [kernel.kallsyms] [k] entry_SYSCALL_64_after_hwframe [...] |--66.96%--__fput |--60.04%--eventpoll_release_file |--58.41%--__mutex_lock.isra.6 |--56.56%--osq_lock The application is multi-threaded, creates a new epoll entry for each incoming connection, and does not delete it before the connection shutdown - that is, before the connection's fd close(). Many different threads compete frequently for the epmutex lock, affecting the overall performance. To reduce the contention this patch introduces explicit reference counting for the eventpoll struct. Each registered event acquires a reference, and references are released at ep_remove() time. The eventpoll struct is released by whoever - among EP file close() and and the monitored file close() drops its last reference. Additionally, this introduces a new 'dying' flag to prevent races between the EP file close() and the monitored file close(). ep_eventpoll_release() marks, under f_lock spinlock, each epitem as dying before removing it, while EP file close() does not touch dying epitems. The above is needed as both close operations could run concurrently and drop the EP reference acquired via the epitem entry. Without the above flag, the monitored file close() could reach the EP struct via the epitem list while the epitem is still listed and then try to put it after its disposal. An alternative could be avoiding touching the references acquired via the epitems at EP file close() time, but that could leave the EP struct alive for potentially unlimited time after EP file close(), with nasty side effects. With all the above in place, we can drop the epmutex usage at disposal time. Overall this produces a significant performance improvement in the mentioned connection/rate scenario: the mutex operations disappear from the topmost offenders in the perf report, and the measured connections/rate grows by ~60%. To make the change more readable this additionally renames ep_free() to ep_clear_and_put(), and moves the actual memory cleanup in a separate ep_free() helper. Link: https://lkml.kernel.org/r/4a57788dcaf28f5eb4f8dfddcc3a8b172a7357bb.1679504153.git.pabeni@redhat.comSigned-off-by:Paolo Abeni <pabeni@redhat.com> Co-developed-by:
Eric Dumazet <edumazet@google.com> Signed-off-by:
Xiumei Mu <xmu@redhiat.com> Acked-by:
Soheil Hassas Yeganeh <soheil@google.com> Reviewed-by:
Davidlohr Bueso <dave@stgolabs.net> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Carlos Maiolino <cmaiolino@redhat.com> Cc: Christian Brauner <brauner@kernel.org> Cc: Eric Biggers <ebiggers@kernel.org> Cc: Jacob Keller <jacob.e.keller@intel.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by:
-
Guilherme G. Piccoli authored
Currently there is no way to show the callback names for registered, unregistered or executed notifiers. This is very useful for debug purposes, hence add this functionality here in the form of notifiers' tracepoints, one per operation. [akpm@linux-foundation.org: coding-style cleanups] Link: https://lkml.kernel.org/r/20230314200058.1326909-1-gpiccoli@igalia.comSigned-off-by:
Guilherme G. Piccoli <gpiccoli@igalia.com> Cc: Arjan van de Ven <arjan@linux.intel.com> Cc: Michael Kelley <mikelley@microsoft.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Xiaoming Ni <nixiaoming@huawei.com> Cc: Baoquan He <bhe@redhat.com> Cc: Cong Wang <xiyou.wangcong@gmail.com> Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com> Cc: Guilherme G. Piccoli <gpiccoli@igalia.com> Cc: Guilherme G. Piccoli <kernel@gpiccoli.net> Cc: Petr Mladek <pmladek@suse.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Valentin Schneider <valentin.schneider@arm.com> Signed-off-by:
-
Tom Rix authored
smatch reports several warnings kernel/hung_task.c:31:19: warning: symbol 'sysctl_hung_task_check_count' was not declared. Should it be static? kernel/hung_task.c:50:29: warning: symbol 'sysctl_hung_task_check_interval_secs' was not declared. Should it be static? kernel/hung_task.c:52:19: warning: symbol 'sysctl_hung_task_warnings' was not declared. Should it be static? kernel/hung_task.c:75:28: warning: symbol 'sysctl_hung_task_panic' was not declared. Should it be static? These variables are only used in hung_task.c, so they should be static Link: https://lkml.kernel.org/r/20230312164645.471259-1-trix@redhat.comSigned-off-by:
Tom Rix <trix@redhat.com> Cc: Ben Dooks <ben.dooks@sifive.com> Cc: fuyuanli <fuyuanli@didiglobal.com> Cc: John Ogness <john.ogness@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Petr Mladek <pmladek@suse.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk> Signed-off-by:
-
Lukas Bulwahn authored
By now, many developers are working on Linux for embedded systems. There is no need to point out single developers. The linux-embedded mailing list has only little traffic, and most of it is just spam. Remove this obsolete section. Link: https://lkml.kernel.org/r/20230308150625.28732-1-lukas.bulwahn@gmail.comSigned-off-by:
Lukas Bulwahn <lukas.bulwahn@gmail.com> Acked-by:
David Woodhouse <dwmw2@infradead.org> Cc: Olivia Mackall <olivia@selenic.com> Signed-off-by:
-
Gerhard Engleder authored
Since commit 4104a206 ("checkpatch: ignore generated CamelCase defines and enum values") enum values like ETHTOOL_LINK_MODE_Asym_Pause_BIT are ignored. But there are other enums like ETHTOOL_LINK_MODE_1000baseT_Full_BIT, which are not ignored because of the not matching '1000baseT' substring. Add regex to match all ETHTOOL_LINK_MODE enums. Link: https://lkml.kernel.org/r/20230104201524.28078-1-gerhard@engleder-embedded.comSigned-off-by:
Gerhard Engleder <gerhard@engleder-embedded.com> Cc: Andy Whitcroft <apw@canonical.com> Cc: Dwaipayan Ray <dwaipayanray1@gmail.com> Cc: Gerhard Engleder <gerhard@engleder-embedded.com> Cc: Joe Perches <joe@perches.com> Cc: Lukas Bulwahn <lukas.bulwahn@gmail.com> Signed-off-by:
-
Andrew Morton authored
This comes out as Try make KALLSYMS_EXTRA_PASS=1 as a workaround but we want quotes: Try "make KALLSYMS_EXTRA_PASS=1" as a workaround Link: https://lkml.kernel.org/r/202303042034.Cjc7JTd0-lkp@intel.com Cc: kernel test robot <lkp@intel.com> Signed-off-by:
-
Alexey Dobriyan authored
ELF is acronym and therefore should be spelled in all caps. I left one exception at Documentation/arm/nwfpe/nwfpe.rst which looks like being written in the first person. Link: https://lkml.kernel.org/r/Y/3wGWQviIOkyLJW@p183Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by:
-
Alexey Dobriyan authored
gcc inlines kstrdup into kstrdup_const() but it can very efficiently tail call into it instead: $ ./scripts/bloat-o-meter ../vmlinux-000 ../obj/vmlinux add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-84 (-84) Function old new delta kstrdup_const 119 35 -84 Link: https://lkml.kernel.org/r/Y/4fDlbIhTLNLFHz@p183Signed-off-by:
-
Glenn Washburn authored
A running x86 UML kernel reports with architecture "i386:x86-64" as it is a sub-architecture. However, a difference with bare-metal x86 kernels is in how it manages tasks and the current task struct. To identify that the inferior is a UML kernel and not bare-metal, check for the existence of the UML specific symbol "cpu_tasks" which contains the current task struct. Link: https://lkml.kernel.org/r/b839d611e2906ccef2725c34d8e353fab35fe75e.1677469905.git.development@efficientek.comSigned-off-by:
Glenn Washburn <development@efficientek.com> Reviewed-by:
-
Glenn Washburn authored
Patch series "scripts/gdb: Support getting current task struct in UML", v3. A running x86 UML kernel reports with architecture "i386:x86-64" as it is a sub-architecture. However, a difference with bare-metal x86 kernels is in how it manages tasks and the current task struct. To identify that the inferior is a UML kernel and not bare-metal, check for the existence of the UML specific symbol "cpu_tasks" which contains the current task struct. This patch (of 3): There is an extra space in a couple blocks in get_current_task. Though python does not care, let's make the spacing consistent. Also, format better an if expression, removing unneeded parenthesis. Link: https://lkml.kernel.org/r/cover.1677469905.git.development@efficientek.com Link: https://lkml.kernel.org/r/2e117b82240de6893f27cb6507242ce455ed7b5b.1677469905.git.development@efficientek.comSigned-off-by:
-
Dan Carpenter authored
It's more readable to just pass NULL directly instead of using a variable for that. Link: https://lkml.kernel.org/r/Y/yAlDytLH0ZNLNz@kiliSigned-off-by:
Dan Carpenter <error27@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Andrey Konovalov authored
Improve KCOV documentation: - Use KCOV instead of kcov, as the former is more widely-used. - Mention Clang in compiler requirements. - Use ``annotations`` for inline code. - Rework remote coverage collection documentation for better clarity. - Various smaller changes. [andreyknvl@google.com: v2] Link: https://lkml.kernel.org/r/583f41c49eef15210fa813e8229730d11427efa7.1677614637.git.andreyknvl@google.com [andreyknvl@google.com: fix ``annotation`` for KCOV_REMOTE_ENABLE] Link: https://lkml.kernel.org/r/72be5c215c275f35891229b90622ed859f196a46.1677684837.git.andreyknvl@google.com Link: https://lkml.kernel.org/r/0b5efd70e31bba7912cf9a6c951f0e76a8df27df.1677517724.git.andreyknvl@google.comSigned-off-by:
Andrey Konovalov <andreyknvl@google.com> Reviewed-by:
Dmitry Vyukov <dvyukov@google.com> Reviewed-by:
Bagas Sanjaya <bagasdotme@gmail.com> Cc: Alexander Potapenko <glider@google.com> Cc: Marco Elver <elver@google.com> Signed-off-by:
-
Uros Bizjak authored
Remove empty if statement from nfs3_prepare_get_acl and update comment to follow the one from the referred fs/posix_acl.c:get_acl(). No functional change intended. Link: https://lkml.kernel.org/r/20221130151231.3654-1-ubizjak@gmail.comSigned-off-by:
Uros Bizjak <ubizjak@gmail.com> Cc: Trond Myklebust <trond.myklebust@hammerspace.com> Cc: Anna Schumaker <anna@kernel.org> Signed-off-by:
-
Chao Yu authored
procfs' .setattr() has updated i_uid, i_gid and i_mode into proc dirent, we don't need to call mark_inode_dirty() for delayed update, remove it. Link: https://lkml.kernel.org/r/20230131150840.34726-1-chao@kernel.orgSigned-off-by:
Chao Yu <chao@kernel.org> Reviewed-by:
-
Randy Dunlap authored
When CONFIG_PROC_FS is not set, proc_salinfo_show() is not used. Mark the function as __maybe_unused to quieten the warning message. ../arch/ia64/kernel/salinfo.c:584:12: warning: 'proc_salinfo_show' defined but not used [-Wunused-function] 584 | static int proc_salinfo_show(struct seq_file *m, void *v) | ^~~~~~~~~~~~~~~~~ Link: https://lkml.kernel.org/r/20230223034309.13375-1-rdunlap@infradead.org Fixes: 3f3942ac ("proc: introduce proc_create_single{,_data}") Signed-off-by:Randy Dunlap <rdunlap@infradead.org> Cc: Christoph Hellwig <hch@lst.de> Signed-off-by:
-
Randy Dunlap authored
alloc_per_cpu_data() is called by find_memory(), which is marked as __init. Therefore alloc_per_cpu_data() can also be marked as __init to remedy this modpost problem. WARNING: modpost: vmlinux.o: section mismatch in reference: alloc_per_cpu_data (section: .text) -> memblock_alloc_try_nid (section: .init.text) Link: https://lkml.kernel.org/r/20230223034258.12917-1-rdunlap@infradead.org Fixes: 4b9ddc7c ("[IA64] Fix section mismatch in contig.c version of per_cpu_init()") Signed-off-by:
-
Wang Yong authored
Improve the average delay precision of getdelay tool to microsecond. When using the getdelay tool, it is sometimes found that the average delay except CPU is not 0, but display is 0, because the precison is too low. For example, see delay average of SWAP below when using ZRAM. print delayacct stats ON PID 32915 CPU count real total virtual total delay total delay average 339202 2793871936 9233585504 7951112 0.000ms IO count delay total delay average 41 419296904 10ms SWAP count delay total delay average 242589 1045792384 0ms This wrong display is misleading, so improve the millisecond precision of the average delay to microsecond just like CPU. Then user would get more accurate information of delay time. Link: https://lkml.kernel.org/r/202302131408087983857@zte.com.cnSigned-off-by:Wang Yong <wang.yong12@zte.com.cn> Reviewed-by:
-
