aboutsummaryrefslogtreecommitdiffstats
path: root/config
Commit message (Collapse)AuthorAgeFilesLines
* autoconf: Pretend `CONFIG_MODULES` is always onDamian Szuberski2022-04-262-26/+62
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Unconditionally inject `CONFIG_MODULES` make variable and `#define CONFIG_MODULES` to Kbuild in `ZFS_LINUX_COMPILE` autoconf function to emulate loadable kernel modules support. This allows OpenZFS to perform Linux checks despite `CONFIG_MODULES=n` in the actual Linux config. - Add `ZFS_AC_KERNEL_CONFIG_MODULES` check which encompasses the logic from `ZFS_AC_KERNEL_TEST_MODULE` with additional diagnostic messages to the user - Removed `ZFS_AC_KERNEL_TEST_MODULE` as it merely duplicates every check in `ZFS_AC_KERNEL_CONFIG_DEFINED` - Moved `ZFS_AC_MODULE_SYMVERS` after `ZFS_AC_KERNEL_CONFIG_DEFINED` so the user has a chance to see the proper diagnostic from the steps before. A workaround for Linux's ``` commit 3e3005df73b535cb849cf4ec8075d6aa3c460f68 Author: Masahiro Yamada <[email protected]> Date: Wed Mar 31 22:38:03 2021 +0900 kbuild: unify modules(_install) for in-tree and external modules If you attempt to build or install modules ('make modules(_install)' with CONFIG_MODULES disabled, you will get a clear error message, but nothing for external module builds. Factor out the modules and modules_install rules into the common part, so you will get the same error message when you try to build external modules with CONFIG_MODULES=n. Signed-off-by: Masahiro Yamada <[email protected]> ``` Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #10832 Closes #13361
* Linux 5.18 compat: kobj_type.default_attrs replaced with default_groupsнаб2022-04-222-0/+39
| | | | | | | | Upstream-commit: cdb4f26a63c391317e335e6e683a614358e70aeb ("kobject: kobj_type: remove default_attrs") Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13357
* Strengthen Linux kernel capabilities detectionDamian Szuberski2022-04-212-4/+31
| | | | | | | | | | | | | | - Add `CONFIG_BLOCK` Linux config requirement to `ZFS_AC_KERNEL_CONFIG_DEFINED`. OpenZFS won't compile without that block device support due to large amount of functional dependencies on it. - Remove dependency on `groups_alloc()` in `ZFS_AC_KERNEL_SRC_GROUP_INFO_GID` to circumvent the missing stub in Linux 4.X kernel headers. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #13351
* rpm -> deb doesn't fail when optional packages are missingDamian Szuberski2022-04-201-2/+2
| | | | | | Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #13331 Closes #13336
* Linux 5.17 compat: GENHD_FL_EXT_DEVT / GENHD_FL_NO_PART_SCANBrian Behlendorf2022-04-192-0/+60
| | | | | | | | | | | | | | As of the 5.17 kernel the GENHD_FL_EXT_DEVT flag has been removed and the GENHD_FL_NO_PART_SCAN flag renamed GENHD_FL_NO_PART. Update zvol_alloc() to set GENHD_FL_NO_PART for the newer kernels which is sufficient. The behavior for prior kernels remains unchanged. 1ebe2e5f ("block: remove GENHD_FL_EXT_DEVT") 46e7eac6 ("block: rename GENHD_FL_NO_PART_SCAN to GENHD_FL_NO_PART") Reviewed-by: Tony Hutter <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Closes #13294 Closes #13297
* Forbid asctime{,_r}(), gmtime(), localtime()наб2022-04-051-0/+4
| | | | | | | | ctime() is only used in binary main threads, which is fine Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13284
* Linux 5.18 compat: use address_space_operations->readaheadRiccardo Schirone2022-04-042-0/+27
| | | | | | | | | ->readpages was removed and replaced by ->readahead. Define zpl_readahead for kernels that don't have ->readpages. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Riccardo Schirone <[email protected]> Closes #13278
* fgrep -> grep -Fнаб2022-04-012-4/+4
| | | | | | | | Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: John Kennedy <[email protected]> Reviewed-by: Ryan Moeller <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13259
* egrep -> grep -Eнаб2022-04-011-1/+1
| | | | | | | | Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: John Kennedy <[email protected]> Reviewed-by: Ryan Moeller <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13259
* Linux optimize access checks when ACL is trivialAndrew2022-04-012-0/+31
| | | | | | | | | | | | | | | | | | | | | Bypass check of ZFS aces if the ACL is trivial. When an ACL is trivial its permissions are represented by the mode without any loss of information. In this case, it is safe to convert the access request into equivalent mode and then pass desired mask and inode to generic_permission(). This has the added benefit of also checking whether entries in a POSIX ACL on the file grant the desired access. This commit also skips the ACL check on looking up the xattr dir since such restrictions don't exist in Linux kernel and it makes xattr lookup behavior inconsistent between SA and file-based xattrs. We also don't want to perform a POSIX ACL check while looking up the POSIX ACL if for some reason it is located in the xattr dir rather than an SA. Reviewed-by: Brian Behlendorf <[email protected]> Co-authored-by: Ryan Moeller <[email protected]> Signed-off-by: Andrew Walker <[email protected]> Closes #13237
* config: user: check for <aio.h>наб2022-03-283-1/+9
| | | | | | | | | And always zpool_read_label_slow() on non-conformant libcs Reviewed-by: Brian Behlendorf <[email protected]> Co-authored-by: José Luis Salvador Rufo <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13207 Closes #13254
* Linux 5.18 compat: replace genhd.h with blkdev.h includesнаб2022-03-282-5/+4
| | | | | | | | | | | | | blkdev.h includes genhd.h since dawn of upstream git, so this is globally safe Upstream-commit: 322cbb50de711814c42fb088f6d31901502c711a ("block: remove genhd.h") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13251
* Linux 5.18 compat: 4-argument bio_alloc()наб2022-03-281-0/+37
| | | | | | | | | | | | | | | | | | bio_alloc(gfp_t gfp_mask, unsigned short nr_iovecs) became bio_alloc(struct block_device *bdev, unsigned short nr_vecs, unsigned int opf, gfp_t gfp_mask) passing NULL/0 continues previous behaviour Upstream-commit: 07888c665b405b1cd3577ddebfeb74f4717a84c4 ("block: pass a block_device and opf to bio_alloc") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13251
* RPM: Split out pam_zfs_key into separate packageColMelvin2022-03-181-2/+3
| | | | | | | | | | | | | Create a separate `pam_zfs_key` package for the PAM module components, an optional addition to the deliverables, in much the same way as the Python bindings are released as a separate `python#-pyzfs` package. This makes it clear when the PAM module is shipped with the package, since it's now in its own package. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Tony Hutter <[email protected]> Signed-off-by: Chris Lindee <[email protected]> Closes: #13026
* config: always-arch: don't subst TARGET_CPUнаб2022-03-181-6/+0
| | | | | | Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13225
* config: always-arch: prune unused TARGET_CPU_*, add TARGET_CPU catch-allнаб2022-03-181-3/+3
| | | | | | | | | This fixes (harmless) error spew from configuring on, e.g., armv6l Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13193 Closes #13225
* Forbid b{copy,zero,cmp}(). Don't include <strings.h> for <string.h>наб2022-03-151-0/+3
| | | | | | Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #12996
* Adding ZERO_PAGE detectionBrian Atkinson2022-03-142-0/+29
| | | | | | | | | | | On some architectures ZERO_PAGE is unavailable because it references a GPL exported symbol of empty_zero_page. Originally e08b993 removed the call to PAGE_ZERO(0) for assignment to the abd_zero_page. However, a simple check can be done to avoid a kernel allocation and free for the abd_zero_page if ZERO_PAGE is available. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Brian Atkinson <[email protected]> Closes #13199
* Linux x86 SIMD: factor out unneeded kernel dependenciesAttila Fülöp2022-03-081-99/+6
| | | | | | | | | | | | | | | | | Cleanup the kernel SIMD code by removing kernel dependencies. - Replace XSTATE_XSAVE with our own XSAVE implementation for all kernels not exporting kernel_fpu{begin,end}(), see #13059 - Replace union fpregs_state by a uint8_t * buffer and get the size of the buffer from the hardware via the CPUID instruction - Replace kernels xgetbv() by our own implementation which was already there for userspace. Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Attila Fülöp <[email protected]> Closes #13102
* Use _Noreturn (C11; GNU89) properlyAlejandro Colomar2022-03-042-27/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | A function that returns with no value is a different thing from a function that doesn't return at all. Those are two orthogonal concepts, commonly confused. pthread_create(3) expects a pointer to a start routine that has a very precise prototype: void *(*start_routine)(void *); However, other thread functions, such as kernel ones, expect: void (*start_routine)(void *); Providing a different one is incorrect, and has only been working because the ABIs happen to produce a compatible function. We should use '_Noreturn void', since it's the natural type, and then provide a '_Noreturn void *' wrapper for pthread functions. For consistency, replace most cases of __NORETURN or __attribute__((noreturn)) by _Noreturn. _Noreturn is understood by -std=gnu89, so it should be safe to use everywhere. Ref: https://github.com/openzfs/zfs/pull/13110#discussion_r808450136 Ref: https://software.codidact.com/posts/285972 Reviewed-by: Brian Behlendorf <[email protected]> Co-authored-by: Ahelenia Ziemiańska <[email protected]> Signed-off-by: Alejandro Colomar <[email protected]> Closes #13120
* config: gcc != ccнаб2022-02-261-11/+11
| | | | | | | | Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Rich Ercolani <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13145 Closes #13152
* config: check for -Wno-cast-function-typeнаб2022-02-262-0/+27
| | | | | | | | Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Rich Ercolani <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13145 Closes #13152
* Linux 5.11 compat: x86 SIMD: fix kernel_fpu_{begin,end}() detectionAttila Fülöp2022-02-241-2/+4
| | | | | | | | | | | | | | | | Linux 5.11 changed kernel_fpu_begin() to an inlined function and moved the functionality to kernel_fpu_begin_mask(). This breaks the existing detection mechanism since it checks if kernel_fpu_begin is an exported kernel symbol, which isn't the case for an inlined function. To avoid assumptions about internal implementation, replace ZFS_LINUX_TEST_RESULT_SYMBOL in favor of ZFS_LINUX_TEST_RESULT which already makes sure kernel_fpu_{begin,end}() is usable by us. Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Attila Fülöp <[email protected]> Closes #13147
* config: add -Wextra (sans sign-compare and missing-field-initializers)наб2022-02-181-1/+1
| | | | | | | Reviewed-by: Alejandro Colomar <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13110
* config: prune unused -Wno-bool-compare checksнаб2022-02-182-28/+0
| | | | | | | Reviewed-by: Alejandro Colomar <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13110
* libtpool: -Wno-clobberedнаб2022-02-185-25/+32
| | | | | | | | | | Also remove -Wno-unused-but-set-variable Upstream-bug: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61118 Reviewed-by: Alejandro Colomar <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13110
* Fix Linux kernel directories detectionDamian Szuberski2022-02-161-30/+75
| | | | | | | | | | | | | | | | | | | | | | | | | Most modern Linux distributions have separate locations for bare source and prebuilt ("build") files. Additionally, there are `source` and `build` symlinks in `/lib/modules/$(KERNEL_VERSION)` pointing to them. The order of directory search is now: - `configure` command line values if both `--with-linux` and `--with-linux-obj` were defined - If only `--with-linux` was defined, `--with-linux-obj` is assumed to have the same value as `--with-linux` - If neither `--with-linux` nor `--with-linux-obj` were defined autodetection is used: - `/lib/modules/$(uname -r)/{source,build}` respectively, if exist - The first directory in `/lib/modules` with the highest version number according to `sort -V` which contains `source` and `build` symlinks/directories - The first directory matching `/usr/src/kernels/*` and `/usr/src/linux-*` with the highest version number according to `sort -V`. Here the source and prebuilt directories are assumed to be the same. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #9935 Closes #13096
* `checkabi`/`storeabi` relevant only to x86_64Damian Szuberski2022-02-161-3/+22
| | | | | | | | | The stored ABI files are for the x86_64 architecture. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: George Melikov <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #11345 Closes #13104
* Propagate KERNEL_* to *.specDamian Szuberski2022-02-091-0/+3
| | | | | | | Reviewed-by: Brian Behlendorf <[email protected]> Authored-by: Damian Szuberski <[email protected]> Signed-off-by: Peter Levine <[email protected]> Closes #13046
* Add support for $KERNEL_{CC,LD,LLVM} variablesPeter Levine2022-02-091-1/+7
| | | | | | | | | | | | | | | | | | Currently, $(CC), $(LD), and $(LLVM) variables aren't passed to kbuild while building modules. This causes modules to build with the default GNU GCC toolchain and prevents experimenting with other toolchains such as CLANG/LLVM. It can also lead to build failure if the CFLAGS/LDFLAGS passed are incompatible with gcc/ld. Pass $KERNEL_CC, $KERNEL_LD, and $KERNEL_LLVM as $(CC), $(LD), and $(LLVM), respectively, to kbuild for each that is defined in the environment. This should take care of the majority of alternative toolchain use cases. Reviewed-by: Damian Szuberski <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Peter Levine <[email protected]> Closes #13046
* Linux 5.16 compat: don't use XSTATE_XSAVE to save FPU stateAttila Fülöp2022-02-092-2/+107
| | | | | | | | | | | | | | | | | | | | | | | | | | | Linux 5.16 moved XSTATE_XSAVE and XSTATE_XRESTORE out of our reach, so add our own XSAVE{,OPT,S} code and use it for Linux 5.16. Please note that this differs from previous behavior in that it won't handle exceptions created by XSAVE an XRSTOR. This is sensible for three reasons. - Exceptions during XSAVE and XRSTOR can only occur if the feature is not supported or enabled or the memory operand isn't aligned on a 64 byte boundary. If this happens something else went terribly wrong, and it may be better to stop execution. - Previously we just printed a warning and didn't handle the fault, this is arguable for the above reason. - All other *SAVE instruction also don't handle exceptions, so this at least aligns behavior. Finally add a test to catch such a regression in the future. Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Attila Fülöp <[email protected]> Closes #13042 Closes #13059
* Add `--enable=all` to ShellCheck by defaultDamian Szuberski2022-02-071-1/+9
| | | | | | | | | | | Change enforced shell type from `dash` to `sh` and excluded `SC2039` and `SC3043` by default. `local` keyword is accepted by all POSIX shells from practical point of view. There is no need anymore to enforce dash so `local` is accepted. Reviewed-by: George Melikov <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #13020
* Add `--enable-asan` and `--enable-ubsan` switchesDamian Szuberski2022-02-034-1/+62
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | `configure` now accepts `--enable-asan` and `--enable-ubsan` switches which results in passing `-fsanitize=address` and `-fsanitize=undefined`, respectively, to the compiler. Those flags are enabled in GitHub workflows for ZTS and zloop. Errors reported by both instrumentations are corrected, except for: - Memory leak reporting is (temporarily) suppressed. The cost of fixing them is relatively high compared to the gains. - Checksum computing functions in `module/zcommon/zfs_fletcher*` have UBSan errors suppressed. It is completely impractical to enforce 64-byte payload alignment there due to performance impact. - There's no ASan heap poisoning in `module/zstd/lib/zstd.c`. A custom memory allocator is used there rendering that measure unfeasible. - Memory leaks detection has to be suppressed for `cmd/zvol_id`. `zvol_id` is run by udev with the help of `ptrace(2)`. Tracing is incompatible with memory leaks detection. Reviewed-by: Ahelenia Ziemiańska <[email protected]> Reviewed-by: George Melikov <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #12928
* Linux 5.17 compat: PDE_DATA() renamed to pde_data()наб2022-01-251-7/+9
| | | | | | | | | | | | | Upstream commit 359745d78351c6f5442435f81549f0207ece28aa ("proc: remove PDE_DATA() completely") Link: https://lore.kernel.org/all/[email protected]/T/#u Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #13004 Closes #12989
* Linux 5.17 compat: dequeue_signal() takes a 4th argumentнаб2022-01-251-2/+43
| | | | | | | | | | | | | Linux 5.17's dequeue_signal() takes an additional enum pid_type * output argument Upstream commit 5768d8906bc23d512b1a736c1e198aa833a6daa4 ("signal: Requeue signals in the appropriate queue") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #12989
* Linux 5.17 compat: detect complete_and_exit() renameнаб2022-01-252-0/+29
| | | | | | | | | | | | | Linux 5.17 sees a rename from complete_and_exit() to kthread complete_and_exit() Upstream commit cead18552660702a4a46f58e65188fe5f36e9dfe ("exit: Rename complete_and_exit to kthread_complete_and_exit") Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #12989
* Add support for FALLOC_FL_ZERO_RANGERich Ercolani2022-01-241-0/+17
| | | | | | | | | | For us, I think it's always just FALLOC_FL_PUNCH_HOLE with a fake mustache on. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Coleman Kane <[email protected]> Signed-off-by: Rich Ercolani <[email protected]> Closes #12975
* Linux 5.16 compat: Added mapping for iov_iter_fault_in_readableRich Ercolani2022-01-242-3/+20
| | | | | | | | | | Linux decided to rename this for some reason. At some point, we should probably invert this mapping, but for now... Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Coleman Kane <[email protected]> Signed-off-by: Rich Ercolani <[email protected]> Closes #12975
* Linux 5.16 compat: Added add_disk check for returnRich Ercolani2022-01-242-0/+29
| | | | | | | | | add_disk went from void to must-check int return. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Coleman Kane <[email protected]> Signed-off-by: Rich Ercolani <[email protected]> Closes #12975
* Linux 5.16 compat: Check slab.h for kvmallocRich Ercolani2022-01-241-0/+1
| | | | | | | | | As it says on the tin - the folio work moved a bunch out of mm.h. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Coleman Kane <[email protected]> Signed-off-by: Rich Ercolani <[email protected]> Closes #12975
* Removed Python 2 and Python 3.5- supportDamian Szuberski2022-01-132-23/+9
| | | | | | | | | | | | Deprecation of Python versions below 3.6 gives opportunity to unify the build and install requirements for OpenZFS packages. The minimal supported Python version is 3.6 as this is the most recent Python package CentOS/RHEL 7 users can get. Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Rich Ercolani <[email protected]> Reviewed-by: John Kennedy <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #12925
* Updated the lz4 decompressorRich Ercolani2022-01-071-0/+1
| | | | | | | | | | | | | | | | As an experiment, I stole the lz4 decompressor from upstream lz4 (1.9.3), and landed it. Feedback suggested that keeping the vendor lz4 code isolated and unlinted was probably reasonable, so I lobbed it into its own file. It also seemed reasonable to put the mostly-untouched* code into lz4.c proper, and relegate the integrated and ZFS-specific code to lz4_zfs.c. Reviewed-by: Matthew Ahrens <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Rich Ercolani <[email protected]> Closes #12805
* config: check for parallel(1), use it for cstyleнаб2022-01-062-0/+9
| | | | | | | | | | | | | | | | | | Before: $ time make cstyle real 0m23.118s user 0m23.002s sys 0m0.114s After: $ time make cstyle real 0m4.577s user 0m31.487s sys 0m0.699s Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Issue #12899
* libfetch: unquote @LIBFETCH_SONAME@ substнаб2022-01-061-3/+3
| | | | | | | | | | @LIBFETCH_SONAME@ is no longer quoted. The C define still is. Ref: 153f7c9f72082d7ef5ee27fcbec1bcb94ba88151 Ref: https://github.com/openzfs/zfs/pull/12835#discussion_r776833743 Reviewed-by: Brian Behlendorf <[email protected]> Reviewed-by: Damian Szuberski <[email protected]> Signed-off-by: Ahelenia Ziemiańska <[email protected]> Closes #12922
* Linux 5.16: The blk-cgroup.h header is where struct blkcg_gq is definedColeman Kane2021-12-071-0/+27
| | | | | | | | | | | | | The definition of struct blkcg_gq was moved into blk-cgroup.h, which is a header that's been in Linux since 2015. This is used by vdev_blkg_tryget() in module/os/linux/zfs/vdev_disk.c. Since the kernel for CentOS 7 and similar-generation releases doesn't have this header, its inclusion is guarded by a configure test. Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Coleman Kane <[email protected]> Closes #12819
* Linux 5.16: bio_set_dev is no longer a helper macroColeman Kane2021-12-071-0/+28
| | | | | | | | | | | | | | | | | | This change adds a confiugre check to determine if bio_set_dev is a helper macro or not. If not, then the attempt to override its internal call to bio_associate_blkg(), with a macro definition to our own version, is no longer possible, as the compiler won't use it when compiling the new inline function replacement implemented in the header. This change also creates a new vdev_bio_set_dev() function that performs the same work, and also performs the work implemented in vdev_bio_associate_blkg(), as it is the only thing calling that function in our code. Our custom vdev_bio_associate_blkg() is now only compiled if the bio_set_dev() is a macro in the Linux headers. Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Coleman Kane <[email protected]> Closes #12819
* Linux 5.16: type member of iov_iter renamed iter_typeColeman Kane2021-12-071-0/+22
| | | | | | | | | | | | | The iov_iter->type member was renamed iov_iter->iter_type. However, while looking into this, realized that in 2018 a iov_iter_type(*iov) accessor function was introduced. So if that is present, use it, otherwise fall back to trying the existing behavior of directly accessing type from iov_iter. Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Coleman Kane <[email protected]> Closes #12819
* Linux 5.16: block_device_operations->submit_bio now returns voidColeman Kane2021-12-071-0/+29
| | | | | | | | | | The return type for the submit_bio member of struct block_device_operations was changed to no longer return a value. Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Coleman Kane <[email protected]> Closes #12819
* Linux 5.13 compat: retry zvol_open() when contendedBrian Behlendorf2021-12-011-0/+22
| | | | | | | | | | | | | | | | | | | | | | | Due to a possible lock inversion the zvol open call path on Linux needs to be able to retry in the case where the spa_namespace_lock cannot be acquired. For Linux 5.12 an older kernel this was accomplished by returning -ERESTARTSYS from zvol_open() to request that blkdev_get() drop the bdev->bd_mutex lock, reaquire it, then call the open callback again. However, as of the 5.13 kernel this behavior was removed. Therefore, for 5.12 and older kernels we preserved the existing retry logic, but for 5.13 and newer kernels we retry internally in zvol_open(). This should always succeed except in the case where a pool's vdev are layed on zvols, in which case it may fail. To handle this case vdev_disk_open() has been updated to retry when opening a device when -ERESTARTSYS is returned. Reviewed-by: Tony Hutter <[email protected]> Reviewed-by: Tony Nguyen <[email protected]> Signed-off-by: Brian Behlendorf <[email protected]> Issue #12301 Closes #12759
* Pass `--enable=all` to shellcheck within contrib/Damian Szuberski2021-11-301-1/+1
| | | | | | | | | | | | | | | - Remove `SHELLCHECK_IGNORE` in favor of inline suppressions and more general `SHELLCHECK_OPTS`. - Exclude `SC2250` (turned on by `--enable=all`) globally - Pass `--enable=all` to shellcheck for scripts in contrib/: it's very important to catch errors early in areas that are not easily testable. Reviewed-by: George Melikov <[email protected]> Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: szubersk <[email protected]> Closes #12760
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-29 11:43:52 +0000