- 12 Mar, 2024 2 commits
-
-
Thorsten Blum authored
Fixes Coccinelle/coccicheck warnings reported by do_div.cocci. Compared to do_div(), div64_ul() does not implicitly cast the divisor and does not unnecessarily calculate the remainder. Link: https://lkml.kernel.org/r/20240229210456.63234-2-thorsten.blum@toblux.com Link: https://lkml.kernel.org/r/20240306142547.4612-1-konishi.ryusuke@gmail.comSigned-off-by:
Thorsten Blum <thorsten.blum@toblux.com> Signed-off-by:
Ryusuke Konishi <konishi.ryusuke@gmail.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org>
-
Uwe Kleine-König authored
As indicated in the added comment, the algorithm works better if b is big. As multiplication is commutative, a and b can be swapped. Do this if a is bigger than b. Link: https://lkml.kernel.org/r/20240303092408.662449-2-u.kleine-koenig@pengutronix.deSigned-off-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Tested-by:
Biju Das <biju.das.jz@bp.renesas.com> Signed-off-by:
-
- 06 Mar, 2024 7 commits
-
-
yang.zhang authored
When loading segments, ubytes is <= mbytes. When ubytes is exhausted, there could be remaining mbytes. Then in the while loop, the buf pointer advancing with mchunk will causing meaningless reading even though it doesn't harm. So let's change to make sure that all of the copying and the rest only happens before uchunk goes to zero. Link: https://lkml.kernel.org/r/20240222092119.5602-1-gaoshanliukou@163.comSigned-off-by:
yang.zhang <yang.zhang@hexintek.com> Acked-by:
Baoquan He <bhe@redhat.com> Signed-off-by:
-
Oleg Nesterov authored
This initialization is incomplete and unnecessary, neither do_group_exit() nor PF_USER_WORKER need ksig->info. Link: https://lkml.kernel.org/r/20240226165653.GA20834@redhat.comSigned-off-by:
Oleg Nesterov <oleg@redhat.com> Cc: Christian Brauner <brauner@kernel.org> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: Peter Collingbourne <pcc@google.com> Cc: Wen Yang <wenyang.linux@foxmail.com> Signed-off-by:
-
Oleg Nesterov authored
ksig->ka and ksig->info are not initialized if get_signal() returns 0 or if the caller is PF_USER_WORKER. Check signr != 0 before SA_EXPOSE_TAGBITS and move the "out" label down. The latter means that ksig->sig won't be initialized if a PF_USER_WORKER thread gets a fatal signal but this is fine, PF_USER_WORKER's don't use ksig. And there is nothing new, in this case ksig->ka and ksig-info are not initialized anyway. Add a comment. Link: https://lkml.kernel.org/r/20240226165650.GA20829@redhat.comSigned-off-by:
-
Oleg Nesterov authored
Patch series "get_signal: minor cleanups and fix". Lets remove this clear_siginfo() right now. It is incomplete (and thus looks confusing) and unnecessary. Also, PF_USER_WORKER's already don't get a fully initialized ksig anyway. This patch (of 3): Cleanup and preparation for the next changes. get_signal() uses signr or ksig->info.si_signo or ksig->sig in a chaotic way, this looks confusing. Change it to always use signr. Link: https://lkml.kernel.org/r/20240226165612.GA20787@redhat.com Link: https://lkml.kernel.org/r/20240226165647.GA20826@redhat.comSigned-off-by:
-
Ricardo B. Marliere authored
Since commit aed65af1 ("drivers: make device_type const"), the driver core can properly handle constant struct device_type. Make sure that new usages of the struct already enter the tree as const. Link: https://lkml.kernel.org/r/20240218-device_cleanup-checkpatch-v1-1-8b0b89c4f6b1@marliere.netSigned-off-by:
Ricardo B. Marliere <ricardo@marliere.net> Reviewed-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Joe Perches <joe@perches.com> Signed-off-by:
-
Ahelenia Ziemiańska authored
Found with git grep 'MODULE_AUTHOR(".*([^)]*@' Fixed with sed -i '/MODULE_AUTHOR(".*([^)]*@/{s/ (/ </g;s/)"/>"/;s/)and/> and/}' \ $(git grep -l 'MODULE_AUTHOR(".*([^)]*@') Also: in drivers/media/usb/siano/smsusb.c normalise ", INC" to ", Inc"; this is what every other MODULE_AUTHOR for this company says, and it's what the header says in drivers/sbus/char/openprom.c normalise a double-spaced separator; this is clearly copied from the copyright header, where the names are aligned on consecutive lines thusly: * Linux/SPARC PROM Configuration Driver * Copyright (C) 1996 Thomas K. Dyas (tdyas@noc.rutgers.edu) * Copyright (C) 1996 Eddie C. Dost (ecd@skynet.be) but the authorship branding is single-line Link: https://lkml.kernel.org/r/mk3geln4azm5binjjlfsgjepow4o73domjv6ajybws3tz22vb3@tarta.nabijaczleweli.xyzSigned-off-by:Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz> Cc: Joe Perches <joe@perches.com> Signed-off-by:
-
Andy Shevchenko authored
Replace open coded functionalify of kstrdup_and_replace() with a call. Link: https://lkml.kernel.org/r/20240213162741.3102810-1-andriy.shevchenko@linux.intel.comSigned-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by:
Luis Chamberlain <mcgrof@kernel.org> Cc: Jason Baron <jbaron@akamai.com> Cc: Jim Cromie <jim.cromie@gmail.com> Signed-off-by:
-
- 22 Feb, 2024 31 commits
-
-
Wei Yang authored
This is what list_is_head() exactly do. Link: https://lkml.kernel.org/r/20240208021423.15704-1-richard.weiyang@gmail.comSigned-off-by:
Wei Yang <richard.weiyang@gmail.com> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by:
-
Ryusuke Konishi authored
The hosting site where the nilfs project had a mirror site continues to be in trouble, so we have decided not to use that site. This will reflect it in the MAINTAINERS file. Link: https://lkml.kernel.org/r/20240208093018.6334-1-konishi.ryusuke@gmail.comSigned-off-by:
-
Alexey Dobriyan authored
smp_processor_id family of macros never accepted any arguments. #define __smp_processor_id(x) works by accident (see C99 6.10.3 §4). __smp_processor_id() gets 1 (empty) argument and passes it down to raw_smp_processor_id() which doesn't accept arguments. Link: https://lkml.kernel.org/r/0037d1f2-8153-4b33-b43e-f4b6ecd710ac@p183Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Cc: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by:
-
Jan Kara authored
When fat_encode_fh_nostale() encodes file handle without a parent it stores only first 10 bytes of the file handle. However the length of the file handle must be a multiple of 4 so the file handle is actually 12 bytes long and the last two bytes remain uninitialized. This is not great at we potentially leak uninitialized information with the handle to userspace. Properly initialize the full handle length. Link: https://lkml.kernel.org/r/20240205122626.13701-1-jack@suse.cz Reported-by: syzbot+3ce5dea5b1539ff36769@syzkaller.appspotmail.com Fixes: ea3983ac ("fat: restructure export_operations") Signed-off-by:
Jan Kara <jack@suse.cz> Acked-by:
OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Cc: Amir Goldstein <amir73il@gmail.com> Cc: <stable@vger.kernel.org> Signed-off-by:
-
Ricardo B. Marliere authored
Since commit d492cc25 ("driver core: device.h: make struct bus_type a const *"), the driver core can properly handle constant struct bus_type. Make sure that new usages of the struct already enter the tree as const. Link: https://lkml.kernel.org/r/20240204-bus_cleanup-checkpatch-v1-1-8d51dcecda20@marliere.netSigned-off-by:
Thomas Weißschuh <linux@weissschuh.net> Reviewed-by:
-
Feng Tang authored
For debugging kernel panics and other bugs, there is already an option of panic_print to dump all tasks' call stacks. On today's large servers running many containers, there could be thousands of tasks or more, and this will print out huge amount of call stacks, taking a lot of time (for serial console which is main target user case of panic_print). And in many cases, only those several tasks being blocked are key for the panic, so add an option to only dump blocked tasks' call stacks. [akpm@linux-foundation.org: clarify documentation a little] Link: https://lkml.kernel.org/r/20240202132042.3609657-1-feng.tang@intel.comSigned-off-by:
Feng Tang <feng.tang@intel.com> Tested-by:
Guilherme G. Piccoli <gpiccoli@igalia.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Josh Poimboeuf <jpoimboe@kernel.org> Cc: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Randy Dunlap <rdunlap@infradead.org> Signed-off-by:
-
Geert Uytterhoeven authored
Since commit 3570ee04 ("s390/smp: keep the original lowcore for CPU 0"), there is no longer any architecture that needs to override arch_call_rest_init(). Remove the weak wrapper around rest_init(), call rest_init() directly, and make rest_init() static. Link: https://lkml.kernel.org/r/aa10868bfb176eef4abb8bb4a710b85330792694.1706106183.git.geert@linux-m68k.orgSigned-off-by:
Geert Uytterhoeven <geert+renesas@glider.be> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Ilya Leoshkevich <iii@linux.ibm.com> Cc: Josh Poimboeuf <jpoimboe@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Signed-off-by:
-
Muhammad Usama Anjum authored
Add missing tests to run_vmtests.sh. The mm kselftests are run through run_vmtests.sh. If a test isn't present in this script, it'll not run with run_tests or `make -C tools/testing/selftests/mm run_tests`. [usama.anjum@collabora.com: use correct flag in the code] Link: https://lkml.kernel.org/r/20240201130538.1404897-2-usama.anjum@collabora.com Link: https://lkml.kernel.org/r/20240125154608.720072-6-usama.anjum@collabora.com Cc: Ryan Roberts <ryan.roberts@arm.com> Signed-off-by:
Muhammad Usama Anjum <usama.anjum@collabora.com> Cc: Shuah Khan <shuah@kernel.org> Signed-off-by:
-
Muhammad Usama Anjum authored
Save and restore nr_hugepages before changing it during the test. A test should not change system wide settings. Link: https://lkml.kernel.org/r/20240125154608.720072-5-usama.anjum@collabora.com Fixes: 5f23f6d0 ("x86/pkeys: Add self-tests") Signed-off-by:
-
Muhammad Usama Anjum authored
Save and restore nr_hugepages before changing it during the test. A test should not change system wide settings. Link: https://lkml.kernel.org/r/20240125154608.720072-4-usama.anjum@collabora.comSigned-off-by:
-
Muhammad Usama Anjum authored
Remove sudo as some test running environments may not have sudo available. Instead skip the test if root privileges aren't available in the test. [usama.anjum@collabora.com: on-fault-limit: run test without root privileges otherwise skip] Link: https://lkml.kernel.org/r/20240201130538.1404897-1-usama.anjum@collabora.com Link: https://lkml.kernel.org/r/20240125154608.720072-3-usama.anjum@collabora.comSigned-off-by:
-
Muhammad Usama Anjum authored
Patch series "selftests/mm: Improve run_vmtests.sh", v3. In this series, I'm trying to add 3 missing tests to vm_runtests.sh which is used to run all the tests in mm suite. These tests weren't running by CIs. While enabling them and through review feedback, I've fixed some problems in tests as well. I've found more flakiness in more tests which I'll be fixing with future patches. hugetlb-read-hwpoison test is being added where it can only run with newly added "-d" (destructive) flag only. Not sure why it is failing again. So once it become stable, we can think of moving it to default set of tests if it doesn't have any side-effect to them. This patch (of 5): Do not unmount the cgroup if it wasn't mounted by the test. The earlier patch had fixed this for charge_reserved_hugetlb, but not for this test. I'm adding fixes tag to that earlier patch. Link: https://lkml.kernel.org/r/20240125154608.720072-1-usama.anjum@collabora.com Link: https://lkml.kernel.org/r/20240125154608.720072-2-usama.anjum@collabora.com Fixes: 209376ed ("selftests/vm: make charge_reserved_hugetlb.sh work with existing cgroup setting") Signed-off-by:
-
Nathan Chancellor authored
The minimum supported version of LLVM for building the kernel has been bumped to 13.0.1. Update the __diag_clang() macros for this bump. Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-11-f5ff9bda41c5@kernel.orgSigned-off-by:
Nathan Chancellor <nathan@kernel.org> Reviewed-by:
Kees Cook <keescook@chromium.org> Cc: Albert Ou <aou@eecs.berkeley.edu> Cc: "Aneesh Kumar K.V (IBM)" <aneesh.kumar@kernel.org> Cc: Ard Biesheuvel <ardb@kernel.org> Cc: Borislav Petkov (AMD) <bp@alien8.de> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Conor Dooley <conor@kernel.org> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Masahiro Yamada <masahiroy@kernel.org> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: "Naveen N. Rao" <naveen.n.rao@linux.ibm.com> Cc: Nicholas Piggin <npiggin@gmail.com> Cc: Nicolas Schier <nicolas@fjasle.eu> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Russell King <linux@armlinux.org.uk> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Will Deacon <will@kernel.org> Signed-off-by:
-
Nathan Chancellor authored
Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, this condition can be changed to just CONFIG_CC_IS_CLANG, as the build will fail during the configuration stage for older LLVM versions. Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-10-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, this condition is always true, as the build will fail during the configuration stage for older LLVM versions. Remove it. Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-9-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, this condition is always true, as the build will fail during the configuration stage for older LLVM versions. Remove it. Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-8-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, the condition for using _mcount as MCOUNT_NAME is always true, as the build will fail during the configuration stage for older LLVM versions. Replace MCOUNT_NAME with _mcount directly. This effectively reverts commit 7ce04771 ("riscv: Workaround mcount name prior to clang-13"). Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-7-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
This reverts commit 6fcb5741 ("powerpc: Kconfig: disable CONFIG_COMPAT for clang < 12"). Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, this condition is always true, as the build will fail during the configuration stage for older LLVM versions. Remove it. Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-6-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, several conditions become tautologies, as they will always be true because the build will fail during the configuration stage for older LLVM versions. Drop them, as they are unnecessary. Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-5-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, the conditional expression added to get_current() by commit c1e42efa ("ARM: 9151/1: Thumb2: avoid __builtin_thread_pointer() on Clang") is always true, as the build will fail during the configuration stage for older LLVM versions. Remove it, effectively reverting the aforementioned change. Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-4-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, the inner ifeq statement is always false, as the build will fail during the configuration stage for older LLVM versions. This effectively reverts part of commit b33fff07 ("x86, build: allow LTO to be selected") and its follow up fix, commit 2398ce80 ("x86, lto: Pass -stack-alignment only on LLD < 13.0.0"). Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-3-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
Now that the minimum supported version of LLVM for building the kernel has been bumped to 13.0.1, the inner ifeq statement is always false, as the build will fail during the configuration stage for older LLVM versions. This effectively reverts commit 24845dcb ("Makefile: LTO: have linker check -Wframe-larger-than") and its follow up fix, commit 0236526d ("Makefile: lto: Pass -warn-stack-size only on LLD < 13.0.0"). Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-2-f5ff9bda41c5@kernel.orgSigned-off-by:
-
Nathan Chancellor authored
Patch series "Bump the minimum supported version of LLVM to 13.0.1". This series bumps the minimum supported version of LLVM for building the kernel to 13.0.1. The first patch does the bump and all subsequent patches clean up all the various workarounds and checks for earlier versions. Quoting the first patch's commit message for those that were only on CC for the clean ups: When __builtin_mul_overflow() has arguments that differ in terms of signedness and width, LLVM may generate a libcall to __muloti4 because it performs the checks in terms of 65-bit multiplication. This issue becomes harder to hit (but still possible) after LLVM 12.0.0, which includes a special case for matching widths but different signs. To gain access to this special case, which the kernel can take advantage of when calls to __muloti4 appear, bump the minimum supported version of LLVM for building the kernel to 13.0.1. 13.0.1 was chosen because there is minimal impact to distribution support while allowing a few more workarounds to be dropped in the kernel source than if 12.0.0 were chosen. Looking at container images of up to date distribution versions: archlinux:latest clang version 16.0.6 debian:oldoldstable-slim clang version 7.0.1-8+deb10u2 (tags/RELEASE_701/final) debian:oldstable-slim Debian clang version 11.0.1-2 debian:stable-slim Debian clang version 14.0.6 debian:testing-slim Debian clang version 16.0.6 (19) debian:unstable-slim Debian clang version 16.0.6 (19) fedora:38 clang version 16.0.6 (Fedora 16.0.6-3.fc38) fedora:latest clang version 17.0.6 (Fedora 17.0.6-1.fc39) fedora:rawhide clang version 17.0.6 (Fedora 17.0.6-1.fc40) opensuse/leap:latest clang version 15.0.7 opensuse/tumbleweed:latest clang version 17.0.6 ubuntu:focal clang version 10.0.0-4ubuntu1 ubuntu:latest Ubuntu clang version 14.0.0-1ubuntu1.1 ubuntu:rolling Ubuntu clang version 16.0.6 (15) ubuntu:devel Ubuntu clang version 17.0.6 (3) The only distribution that gets left behind is Debian Bullseye, as the default version is 11.0.1; other distributions either have a newer version than 13.0.1 or one older than the current minimum of 11.0.0. Debian has easy access to more recent LLVM versions through apt.llvm.org, so this is not as much of a concern. There are also the kernel.org LLVM toolchains, which should work with distributions with glibc 2.28 and newer. Another benefit of slimming up the number of supported versions of LLVM for building the kernel is reducing the build capacity needed to support a matrix that builds with each supported version, which allows a matrix to reallocate the freed up build capacity towards something else, such as more configuration combinations. This passes my build matrix with all supported versions. This is based on Andrew's mm-nonmm-unstable to avoid trivial conflicts with my series to update the LLVM links across the repository [1] but I can easily rebase it to linux-kbuild if Masahiro would rather these patches go through there (and defer the conflict resolution to the merge window). [1]: https://lore.kernel.org/20240109-update-llvm-links-v1-0-eb09b59db071@kernel.org/ This patch (of 11): When __builtin_mul_overflow() has arguments that differ in terms of signedness and width, LLVM may generate a libcall to __muloti4 because it performs the checks in terms of 65-bit multiplication. This issue becomes harder to hit (but still possible) after LLVM 12.0.0, which includes a special case for matching widths but different signs. To gain access to this special case, which the kernel can take advantage of when calls to __muloti4 appear, bump the minimum supported version of LLVM for building the kernel to 13.0.1. 13.0.1 was chosen because there is minimal impact to distribution support while allowing a few more workarounds to be dropped in the kernel source than if 12.0.0 were chosen. Looking at container images of up to date distribution versions: archlinux:latest clang version 16.0.6 debian:oldoldstable-slim clang version 7.0.1-8+deb10u2 (tags/RELEASE_701/final) debian:oldstable-slim Debian clang version 11.0.1-2 debian:stable-slim Debian clang version 14.0.6 debian:testing-slim Debian clang version 16.0.6 (19) debian:unstable-slim Debian clang version 16.0.6 (19) fedora:38 clang version 16.0.6 (Fedora 16.0.6-3.fc38) fedora:latest clang version 17.0.6 (Fedora 17.0.6-1.fc39) fedora:rawhide clang version 17.0.6 (Fedora 17.0.6-1.fc40) opensuse/leap:latest clang version 15.0.7 opensuse/tumbleweed:latest clang version 17.0.6 ubuntu:focal clang version 10.0.0-4ubuntu1 ubuntu:latest Ubuntu clang version 14.0.0-1ubuntu1.1 ubuntu:rolling Ubuntu clang version 16.0.6 (15) ubuntu:devel Ubuntu clang version 17.0.6 (3) The only distribution that gets left behind is Debian Bullseye, as the default version is 11.0.1; other distributions either have a newer version than 13.0.1 or one older than the current minimum of 11.0.0. Debian has easy access to more recent LLVM versions through apt.llvm.org, so this is not as much of a concern. There are also the kernel.org LLVM toolchains, which should work with distributions with glibc 2.28 and newer. Another benefit of slimming up the number of supported versions of LLVM for building the kernel is reducing the build capacity needed to support a matrix that builds with each supported version, which allows a matrix to reallocate the freed up build capacity towards something else, such as more configuration combinations. Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-0-f5ff9bda41c5@kernel.org Closes: https://github.com/ClangBuiltLinux/linux/issues/1975 Link: https://github.com/llvm/llvm-project/issues/38013 Link: https://github.com/llvm/llvm-project/commit/3203143f1356a4e4e3ada231156fc6da6e1a9f9d Link: https://mirrors.edge.kernel.org/pub/tools/llvm/ Link: https://lkml.kernel.org/r/20240125-bump-min-llvm-ver-to-13-0-1-v1-1-f5ff9bda41c5@kernel.orgSigned-off-by: -
Ryusuke Konishi authored
Convert all remaining usages of kmap_atomic in cpfile to kmap_local. Link: https://lkml.kernel.org/r/20240122140202.6950-16-konishi.ryusuke@gmail.comSigned-off-by:
-
Ryusuke Konishi authored
All calls to nilfs_cpfile_get_checkpoint() and nilfs_cpfile_put_checkpoint() that call kmap() and kunmap() separately are now gone, so remove these methods. Link: https://lkml.kernel.org/r/20240122140202.6950-15-konishi.ryusuke@gmail.comSigned-off-by:
-
Ryusuke Konishi authored
Move the code for reading from a checkpoint entry that is performed in nilfs_attach_checkpoint() to the cpfile side, and make the page mapping local and temporary. And use kmap_local instead of kmap to access the checkpoint entry page. In order to load the ifile inode information included in the checkpoint entry within the inode lock section of nilfs_ifile_read(), the newly added checkpoint reading method nilfs_cpfile_read_checkpoint() is called indirectly via nilfs_ifile_read() instead of from nilfs_attach_checkpoint(). Link: https://lkml.kernel.org/r/20240122140202.6950-14-konishi.ryusuke@gmail.comSigned-off-by:
-
Ryusuke Konishi authored
Move the checkpoint finalization routine to the cpfile side, and make the page mapping local and temporary. And use kmap_local instead of kmap to access the checkpoint entry page when finalizing a checkpoint. In this conversion, some of the information on the checkpoint entry being rewritten is passed through the arguments of the newly added method nilfs_cpfile_finalize_checkpoint(). Link: https://lkml.kernel.org/r/20240122140202.6950-13-konishi.ryusuke@gmail.comSigned-off-by:
-
Ryusuke Konishi authored
In order to convert kmap() used in cpfile to kmap_local, first move the checkpoint creation routine, which is one of the places where kmap is used, to the cpfile side and make the page mapping local and temporary. And use kmap_local instead of kmap to access the checkpoint entry page (and header block page) when generating a checkpoint. Link: https://lkml.kernel.org/r/20240122140202.6950-12-konishi.ryusuke@gmail.comSigned-off-by:
-
Ryusuke Konishi authored
Convert deprecated kmap() and kmap_atomic() to use kmap_local for the ifile metadata file used to manage disk inodes. In some usages, calls to kmap_local and kunmap_local are split into different helpers, but those usages can be safely changed to local thread kmap. Link: https://lkml.kernel.org/r/20240122140202.6950-11-konishi.ryusuke@gmail.comSigned-off-by:
-
Ryusuke Konishi authored
It is now clear that nilfs_bmap_write() is only used to finalize logs written to disk. Concurrent bmap modification operations are not performed on bmaps in this context. Additionally, this function does not modify data used in read-only operations such as bmap lookups. Therefore, there is no need to acquire bmap->b_sem in nilfs_bmap_write(), so delete it. Link: https://lkml.kernel.org/r/20240122140202.6950-10-konishi.ryusuke@gmail.comSigned-off-by:
-
Ryusuke Konishi authored
Before converting the disk inode management metadata file ifile, the call to nilfs_bmap_write(), the i_device_code setting, and the zero-fill code for inodes on the super root block are moved from nilfs_write_inode_common() to its callers. This cleanup simplifies the role and arguments of nilfs_write_inode_common() and collects calls to nilfs_bmap_write() to the log writing code. Also, add and use a new helper nilfs_write_root_mdt_inode() to avoid code duplication in the data export routine nilfs_segctor_fill_in_super_root() to the super root block's buffer. Link: https://lkml.kernel.org/r/20240122140202.6950-9-konishi.ryusuke@gmail.comSigned-off-by:
-
