| Message ID | 20210924140912.201481-2-chandan.babu@oracle.com (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
| Series | xfsprogs: generic serialisation primitives | expand |
On 9/24/21 9:09 AM, Chandan Babu R wrote: > From: Dave Chinner <dchinner@redhat.com> > > The upcoming buffer cache rework/kerenl sync-up requires atomic > variables. I could use C++11 atomics build into GCC, but they are a > pain to work with and shoe-horn into the kernel atomic variable API. > > Much easier is to introduce a dependency on liburcu - the userspace > RCU library. This provides atomic variables that very closely match > the kernel atomic variable API, and it provides a very similar > memory model and memory barrier support to the kernel. And we get > RCU support that has an identical interface to the kernel and works > the same way. > > Hence kernel code written with RCU algorithms and atomic variables > will just slot straight into the userspace xfsprogs code without us > having to think about whether the lockless algorithms will work in > userspace or not. This reduces glue and hoop jumping, and gets us > a step closer to having the entire userspace libxfs code MT safe. > > Signed-off-by: Dave Chinner <dchinner@redhat.com> > [chandan.babu@oracle.com: Add m4 macros to detect availability of liburcu] Thanks for fixing that up. I had tried to use rcu_init like Dave originally had, and I like that better in general, but I had trouble with it - I guess maybe it gets redefined based on memory model magic and the actual symbol "rcu_init" maybe isn't available? I didn't dig very much. Also, dumb question from me - how do we know where we need the rcu_[un]register_thread() calls? Will it be obvious if we miss it in the future? "each thread must invoke this function before its first call to rcu_read_lock() or call_rcu()." Thanks, -Eric
On 25 Sep 2021 at 03:21, Eric Sandeen wrote: > On 9/24/21 9:09 AM, Chandan Babu R wrote: >> From: Dave Chinner <dchinner@redhat.com> >> The upcoming buffer cache rework/kerenl sync-up requires atomic >> variables. I could use C++11 atomics build into GCC, but they are a >> pain to work with and shoe-horn into the kernel atomic variable API. >> Much easier is to introduce a dependency on liburcu - the userspace >> RCU library. This provides atomic variables that very closely match >> the kernel atomic variable API, and it provides a very similar >> memory model and memory barrier support to the kernel. And we get >> RCU support that has an identical interface to the kernel and works >> the same way. >> Hence kernel code written with RCU algorithms and atomic variables >> will just slot straight into the userspace xfsprogs code without us >> having to think about whether the lockless algorithms will work in >> userspace or not. This reduces glue and hoop jumping, and gets us >> a step closer to having the entire userspace libxfs code MT safe. >> Signed-off-by: Dave Chinner <dchinner@redhat.com> >> [chandan.babu@oracle.com: Add m4 macros to detect availability of liburcu] > > Thanks for fixing that up. I had tried to use rcu_init like Dave originally > had, and I like that better in general, but I had trouble with it - I guess > maybe it gets redefined based on memory model magic and the actual symbol > "rcu_init" maybe isn't available? I didn't dig very much. Yes, rcu_init() gets defined to one of the variants of userspace rcu. Hence there is no function named rcu_init(). > > Also, dumb question from me - how do we know where we need the > rcu_[un]register_thread() calls? Will it be obvious if we miss it > in the future? "each thread must invoke this function before its > first call to rcu_read_lock() or call_rcu()." Unfortunately, I don't think there is an obvious method to detect if calls to rcu_[un]register_thread() is missing from any code changes implemented in the future. PS: I am no expert at RCU. My knowledge about RCU is limited to what I could learn by reading articles on the world wide web.
On Fri, Sep 24, 2021 at 04:51:32PM -0500, Eric Sandeen wrote: > On 9/24/21 9:09 AM, Chandan Babu R wrote: > > From: Dave Chinner <dchinner@redhat.com> > > > > The upcoming buffer cache rework/kerenl sync-up requires atomic > > variables. I could use C++11 atomics build into GCC, but they are a > > pain to work with and shoe-horn into the kernel atomic variable API. > > > > Much easier is to introduce a dependency on liburcu - the userspace > > RCU library. This provides atomic variables that very closely match > > the kernel atomic variable API, and it provides a very similar > > memory model and memory barrier support to the kernel. And we get > > RCU support that has an identical interface to the kernel and works > > the same way. > > > > Hence kernel code written with RCU algorithms and atomic variables > > will just slot straight into the userspace xfsprogs code without us > > having to think about whether the lockless algorithms will work in > > userspace or not. This reduces glue and hoop jumping, and gets us > > a step closer to having the entire userspace libxfs code MT safe. > > > > Signed-off-by: Dave Chinner <dchinner@redhat.com> > > [chandan.babu@oracle.com: Add m4 macros to detect availability of liburcu] > > Thanks for fixing that up. I had tried to use rcu_init like Dave originally > had, and I like that better in general, but I had trouble with it - I guess > maybe it gets redefined based on memory model magic and the actual symbol > "rcu_init" maybe isn't available? I didn't dig very much. Ah, so I just checked where the m4/package_urcu.m4 file went - forgot to re-add it after it rejected on apply. The diff is this: diff --git a/m4/package_urcu.m4 b/m4/package_urcu.m4 new file mode 100644 index 000000000000..9b0dee35d9a1 --- /dev/null +++ b/m4/package_urcu.m4 @@ -0,0 +1,22 @@ +AC_DEFUN([AC_PACKAGE_NEED_URCU_H], + [ AC_CHECK_HEADERS(urcu.h) + if test $ac_cv_header_urcu_h = no; then + AC_CHECK_HEADERS(urcu.h,, [ + echo + echo 'FATAL ERROR: could not find a valid urcu header.' + exit 1]) + fi + ]) + +AC_DEFUN([AC_PACKAGE_NEED_RCU_INIT], + [ AC_MSG_CHECKING([for liburcu]) + AC_TRY_COMPILE([ +#define _GNU_SOURCE +#include <urcu.h> + ], [ + rcu_init(); + ], liburcu=-lurcu + AC_MSG_RESULT(yes), + AC_MSG_RESULT(no)) + AC_SUBST(liburcu) + ]) And this ends up with the output on a current debian system of: checking urcu.h usability... yes checking urcu.h presence... yes checking for urcu.h... yes checking for liburcu... yes IOWs, the package check I was using directly probed for rcu_init() being present in liburcu. Hence if liburcu is not providing this function via the default linking like we use with xfsprogs, then we fail at the configure step. So it liburcu is not providing rcu_init(), then the configure step should fail, and you need to work out what is different about the liburcu that is installed on the distro you are building on.... > Also, dumb question from me - how do we know where we need the > rcu_[un]register_thread() calls? Will it be obvious if we miss it > in the future? "each thread must invoke this function before its > first call to rcu_read_lock() or call_rcu()." urcu should fire an assert on any attempt to access urcu TLS storage because the "urcu.registered" flag in the TLS area has not been initialised. IOWs, if we miss register/unregister then things will go bad in urcu calls and it should be noticed. Cheers, Dave.
On 9/25/21 6:05 PM, Dave Chinner wrote: > On Fri, Sep 24, 2021 at 04:51:32PM -0500, Eric Sandeen wrote: >> On 9/24/21 9:09 AM, Chandan Babu R wrote: >>> From: Dave Chinner <dchinner@redhat.com> >>> >>> The upcoming buffer cache rework/kerenl sync-up requires atomic >>> variables. I could use C++11 atomics build into GCC, but they are a >>> pain to work with and shoe-horn into the kernel atomic variable API. >>> >>> Much easier is to introduce a dependency on liburcu - the userspace >>> RCU library. This provides atomic variables that very closely match >>> the kernel atomic variable API, and it provides a very similar >>> memory model and memory barrier support to the kernel. And we get >>> RCU support that has an identical interface to the kernel and works >>> the same way. >>> >>> Hence kernel code written with RCU algorithms and atomic variables >>> will just slot straight into the userspace xfsprogs code without us >>> having to think about whether the lockless algorithms will work in >>> userspace or not. This reduces glue and hoop jumping, and gets us >>> a step closer to having the entire userspace libxfs code MT safe. >>> >>> Signed-off-by: Dave Chinner <dchinner@redhat.com> >>> [chandan.babu@oracle.com: Add m4 macros to detect availability of liburcu] >> >> Thanks for fixing that up. I had tried to use rcu_init like Dave originally >> had, and I like that better in general, but I had trouble with it - I guess >> maybe it gets redefined based on memory model magic and the actual symbol >> "rcu_init" maybe isn't available? I didn't dig very much. > > Ah, so I just checked where the m4/package_urcu.m4 file went - > forgot to re-add it after it rejected on apply. The diff is this: > > diff --git a/m4/package_urcu.m4 b/m4/package_urcu.m4 > new file mode 100644 > index 000000000000..9b0dee35d9a1 > --- /dev/null > +++ b/m4/package_urcu.m4 > @@ -0,0 +1,22 @@ > +AC_DEFUN([AC_PACKAGE_NEED_URCU_H], > + [ AC_CHECK_HEADERS(urcu.h) > + if test $ac_cv_header_urcu_h = no; then > + AC_CHECK_HEADERS(urcu.h,, [ > + echo > + echo 'FATAL ERROR: could not find a valid urcu header.' > + exit 1]) > + fi > + ]) > + > +AC_DEFUN([AC_PACKAGE_NEED_RCU_INIT], > + [ AC_MSG_CHECKING([for liburcu]) > + AC_TRY_COMPILE([ > +#define _GNU_SOURCE > +#include <urcu.h> > + ], [ > + rcu_init(); > + ], liburcu=-lurcu > + AC_MSG_RESULT(yes), > + AC_MSG_RESULT(no)) > + AC_SUBST(liburcu) > + ]) Works great here too. My error was not including urcu.h in the test I think, and looking for the symbol directly. I will merge this version rather than Chandan's. Thanks, -Eric
On 9/24/21 9:09 AM, Chandan Babu R wrote: > From: Dave Chinner <dchinner@redhat.com> > > The upcoming buffer cache rework/kerenl sync-up requires atomic > variables. I could use C++11 atomics build into GCC, but they are a > pain to work with and shoe-horn into the kernel atomic variable API. > > Much easier is to introduce a dependency on liburcu - the userspace > RCU library. This provides atomic variables that very closely match > the kernel atomic variable API, and it provides a very similar > memory model and memory barrier support to the kernel. And we get > RCU support that has an identical interface to the kernel and works > the same way. > > Hence kernel code written with RCU algorithms and atomic variables > will just slot straight into the userspace xfsprogs code without us > having to think about whether the lockless algorithms will work in > userspace or not. This reduces glue and hoop jumping, and gets us > a step closer to having the entire userspace libxfs code MT safe. > > Signed-off-by: Dave Chinner <dchinner@redhat.com> > [chandan.babu@oracle.com: Add m4 macros to detect availability of liburcu] > Signed-off-by: Chandan Babu R <chandan.babu@oracle.com> I've added Dave's m4 macros in place of yours, and with that circuitous route I think we have a good set of changes, so: Reviewed-by: Eric Sandeen <sandeen@redhat.com> for the combined set of this-plus-dave's-m4's (elsewhere in replies on this thread) Thanks, -Eric
diff --git a/configure.ac b/configure.ac index 56871745..61c6cf40 100644 --- a/configure.ac +++ b/configure.ac @@ -154,6 +154,9 @@ AC_PACKAGE_NEED_UUIDCOMPARE AC_PACKAGE_NEED_PTHREAD_H AC_PACKAGE_NEED_PTHREADMUTEXINIT +AC_PACKAGE_NEED_URCU_H +AC_PACKAGE_NEED_RCU_SET_POINTER_SYM + AC_HAVE_FADVISE AC_HAVE_MADVISE AC_HAVE_MINCORE diff --git a/copy/Makefile b/copy/Makefile index 449b235f..1b00cd0d 100644 --- a/copy/Makefile +++ b/copy/Makefile @@ -9,7 +9,8 @@ LTCOMMAND = xfs_copy CFILES = xfs_copy.c HFILES = xfs_copy.h -LLDLIBS = $(LIBXFS) $(LIBXLOG) $(LIBFROG) $(LIBUUID) $(LIBPTHREAD) $(LIBRT) +LLDLIBS = $(LIBXFS) $(LIBXLOG) $(LIBFROG) $(LIBUUID) $(LIBPTHREAD) $(LIBRT) \ + $(LIBURCU) LTDEPENDENCIES = $(LIBXFS) $(LIBXLOG) $(LIBFROG) LLDFLAGS = -static-libtool-libs diff --git a/copy/xfs_copy.c b/copy/xfs_copy.c index fc7d225f..f5eff969 100644 --- a/copy/xfs_copy.c +++ b/copy/xfs_copy.c @@ -110,6 +110,7 @@ do_message(int flags, int code, const char *fmt, ...) fprintf(stderr, _("Aborting XFS copy -- logfile error -- reason: %s\n"), strerror(errno)); + rcu_unregister_thread(); pthread_exit(NULL); } } @@ -224,6 +225,7 @@ begin_reader(void *arg) { thread_args *args = arg; + rcu_register_thread(); for (;;) { pthread_mutex_lock(&args->wait); if (do_write(args, NULL)) @@ -243,6 +245,7 @@ handle_error: if (--glob_masks.num_working == 0) pthread_mutex_unlock(&mainwait); pthread_mutex_unlock(&glob_masks.mutex); + rcu_unregister_thread(); pthread_exit(NULL); return NULL; } diff --git a/db/Makefile b/db/Makefile index beafb105..5c017898 100644 --- a/db/Makefile +++ b/db/Makefile @@ -18,7 +18,8 @@ CFILES = $(HFILES:.h=.c) btdump.c btheight.c convert.c info.c namei.c \ timelimit.c LSRCFILES = xfs_admin.sh xfs_ncheck.sh xfs_metadump.sh -LLDLIBS = $(LIBXFS) $(LIBXLOG) $(LIBFROG) $(LIBUUID) $(LIBRT) $(LIBPTHREAD) +LLDLIBS = $(LIBXFS) $(LIBXLOG) $(LIBFROG) $(LIBUUID) $(LIBRT) $(LIBPTHREAD) \ + $(LIBURCU) LTDEPENDENCIES = $(LIBXFS) $(LIBXLOG) $(LIBFROG) LLDFLAGS += -static-libtool-libs diff --git a/debian/control b/debian/control index e4ec897c..71c08167 100644 --- a/debian/control +++ b/debian/control @@ -3,7 +3,7 @@ Section: admin Priority: optional Maintainer: XFS Development Team <linux-xfs@vger.kernel.org> Uploaders: Nathan Scott <nathans@debian.org>, Anibal Monsalve Salazar <anibal@debian.org>, Bastian Germann <bastiangermann@fishpost.de> -Build-Depends: libinih-dev (>= 53), uuid-dev, dh-autoreconf, debhelper (>= 5), gettext, libtool, libedit-dev, libblkid-dev (>= 2.17), linux-libc-dev, libdevmapper-dev, libattr1-dev, libicu-dev, pkg-config +Build-Depends: libinih-dev (>= 53), uuid-dev, dh-autoreconf, debhelper (>= 5), gettext, libtool, libedit-dev, libblkid-dev (>= 2.17), linux-libc-dev, libdevmapper-dev, libattr1-dev, libicu-dev, pkg-config, liburcu-dev Standards-Version: 4.0.0 Homepage: https://xfs.wiki.kernel.org/ diff --git a/growfs/Makefile b/growfs/Makefile index a107d348..08601de7 100644 --- a/growfs/Makefile +++ b/growfs/Makefile @@ -9,7 +9,8 @@ LTCOMMAND = xfs_growfs CFILES = xfs_growfs.c -LLDLIBS = $(LIBXFS) $(LIBXCMD) $(LIBFROG) $(LIBUUID) $(LIBRT) $(LIBPTHREAD) +LLDLIBS = $(LIBXFS) $(LIBXCMD) $(LIBFROG) $(LIBUUID) $(LIBRT) $(LIBPTHREAD) \ + $(LIBURCU) ifeq ($(ENABLE_EDITLINE),yes) LLDLIBS += $(LIBEDITLINE) $(LIBTERMCAP) diff --git a/include/builddefs.in b/include/builddefs.in index e8f447f9..78eddf4a 100644 --- a/include/builddefs.in +++ b/include/builddefs.in @@ -22,6 +22,7 @@ LDFLAGS = LIBRT = @librt@ LIBUUID = @libuuid@ +LIBURCU = @liburcu@ LIBPTHREAD = @libpthread@ LIBTERMCAP = @libtermcap@ LIBEDITLINE = @libeditline@ @@ -125,7 +126,8 @@ CROND_DIR = @crond_dir@ GCCFLAGS = -funsigned-char -fno-strict-aliasing -Wall # -Wbitwise -Wno-transparent-union -Wno-old-initializer -Wno-decl -PCFLAGS = -D_GNU_SOURCE $(GCCFLAGS) +# _LGPL_SOURCE is for liburcu to work correctly with GPL/LGPL programs +PCFLAGS = -D_LGPL_SOURCE -D_GNU_SOURCE $(GCCFLAGS) ifeq ($(HAVE_UMODE_T),yes) PCFLAGS += -DHAVE_UMODE_T endif diff --git a/include/platform_defs.h.in b/include/platform_defs.h.in index 539bdbec..7c6b3ada 100644 --- a/include/platform_defs.h.in +++ b/include/platform_defs.h.in @@ -23,6 +23,7 @@ #include <limits.h> #include <stdbool.h> #include <libgen.h> +#include <urcu.h> typedef struct filldir filldir_t; diff --git a/libfrog/workqueue.c b/libfrog/workqueue.c index 8c1a163e..702a53e2 100644 --- a/libfrog/workqueue.c +++ b/libfrog/workqueue.c @@ -11,6 +11,7 @@ #include <stdbool.h> #include <errno.h> #include <assert.h> +#include <urcu.h> #include "workqueue.h" /* Main processing thread */ @@ -24,6 +25,7 @@ workqueue_thread(void *arg) * Loop pulling work from the passed in work queue. * Check for notification to exit after every chunk of work. */ + rcu_register_thread(); while (1) { pthread_mutex_lock(&wq->lock); @@ -60,6 +62,7 @@ workqueue_thread(void *arg) (wi->function)(wi->queue, wi->index, wi->arg); free(wi); } + rcu_unregister_thread(); return NULL; } diff --git a/libxfs/init.c b/libxfs/init.c index 1ec83791..b06faf8a 100644 --- a/libxfs/init.c +++ b/libxfs/init.c @@ -310,6 +310,8 @@ libxfs_init(libxfs_init_t *a) fd = -1; flags = (a->isreadonly | a->isdirect); + rcu_init(); + rcu_register_thread(); radix_tree_init(); if (a->volname) { @@ -1023,6 +1025,7 @@ libxfs_destroy( libxfs_bcache_free(); cache_destroy(libxfs_bcache); leaked = destroy_zones(); + rcu_unregister_thread(); if (getenv("LIBXFS_LEAK_CHECK") && leaked) exit(1); } diff --git a/libxfs/libxfs_priv.h b/libxfs/libxfs_priv.h index 7181a858..db90e173 100644 --- a/libxfs/libxfs_priv.h +++ b/libxfs/libxfs_priv.h @@ -210,8 +210,7 @@ enum ce { CE_DEBUG, CE_CONT, CE_NOTE, CE_WARN, CE_ALERT, CE_PANIC }; #define spin_unlock(a) ((void) 0) #define likely(x) (x) #define unlikely(x) (x) -#define rcu_read_lock() ((void) 0) -#define rcu_read_unlock() ((void) 0) + /* Need to be able to handle this bare or in control flow */ static inline bool WARN_ON(bool expr) { return (expr); diff --git a/logprint/Makefile b/logprint/Makefile index 758504b3..cdedbd0d 100644 --- a/logprint/Makefile +++ b/logprint/Makefile @@ -12,7 +12,8 @@ CFILES = logprint.c \ log_copy.c log_dump.c log_misc.c \ log_print_all.c log_print_trans.c log_redo.c -LLDLIBS = $(LIBXFS) $(LIBXLOG) $(LIBFROG) $(LIBUUID) $(LIBRT) $(LIBPTHREAD) +LLDLIBS = $(LIBXFS) $(LIBXLOG) $(LIBFROG) $(LIBUUID) $(LIBRT) $(LIBPTHREAD) \ + $(LIBURCU) LTDEPENDENCIES = $(LIBXFS) $(LIBXLOG) $(LIBFROG) LLDFLAGS = -static-libtool-libs diff --git a/m4/Makefile b/m4/Makefile index c6c73dc9..73120530 100644 --- a/m4/Makefile +++ b/m4/Makefile @@ -24,6 +24,7 @@ LSRCFILES = \ package_services.m4 \ package_types.m4 \ package_icu.m4 \ + package_urcu.m4 \ package_utilies.m4 \ package_uuiddev.m4 \ multilib.m4 \ diff --git a/m4/package_urcu.m4 b/m4/package_urcu.m4 new file mode 100644 index 00000000..74a24db9 --- /dev/null +++ b/m4/package_urcu.m4 @@ -0,0 +1,24 @@ +AC_DEFUN([AC_PACKAGE_NEED_URCU_H], + [ AC_CHECK_HEADERS([urcu.h]) + if test $ac_cv_header_urcu_h = no; then + echo + echo 'FATAL ERROR: could not find a valid URCU header.' + echo 'Install the Userspace RCU development package.' + exit 1 + fi + ]) + +AC_DEFUN([AC_PACKAGE_NEED_RCU_SET_POINTER_SYM], + [ AC_CHECK_FUNCS(rcu_set_pointer_sym) + if test $ac_cv_func_rcu_set_pointer_sym = yes; then + liburcu="" + else + AC_CHECK_LIB(urcu, rcu_set_pointer_sym,, [ + echo + echo 'FATAL ERROR: could not find a valid urcu library.' + echo 'Install the Userspace RCU development package.' + exit 1]) + liburcu="-lurcu" + fi + AC_SUBST(liburcu) + ]) diff --git a/mdrestore/Makefile b/mdrestore/Makefile index d946955b..8f28ddab 100644 --- a/mdrestore/Makefile +++ b/mdrestore/Makefile @@ -8,7 +8,8 @@ include $(TOPDIR)/include/builddefs LTCOMMAND = xfs_mdrestore CFILES = xfs_mdrestore.c -LLDLIBS = $(LIBXFS) $(LIBFROG) $(LIBRT) $(LIBPTHREAD) $(LIBUUID) +LLDLIBS = $(LIBXFS) $(LIBFROG) $(LIBRT) $(LIBPTHREAD) $(LIBUUID) \ + $(LIBURCU) LTDEPENDENCIES = $(LIBXFS) $(LIBFROG) LLDFLAGS = -static diff --git a/mkfs/Makefile b/mkfs/Makefile index b8805f7e..811ba9db 100644 --- a/mkfs/Makefile +++ b/mkfs/Makefile @@ -11,7 +11,7 @@ HFILES = CFILES = proto.c xfs_mkfs.c LLDLIBS += $(LIBXFS) $(LIBXCMD) $(LIBFROG) $(LIBRT) $(LIBPTHREAD) $(LIBBLKID) \ - $(LIBUUID) $(LIBINIH) + $(LIBUUID) $(LIBINIH) $(LIBURCU) LTDEPENDENCIES += $(LIBXFS) $(LIBXCMD) $(LIBFROG) LLDFLAGS = -static-libtool-libs diff --git a/repair/Makefile b/repair/Makefile index 5f0764d1..47536ca1 100644 --- a/repair/Makefile +++ b/repair/Makefile @@ -72,7 +72,7 @@ CFILES = \ xfs_repair.c LLDLIBS = $(LIBXFS) $(LIBXLOG) $(LIBXCMD) $(LIBFROG) $(LIBUUID) $(LIBRT) \ - $(LIBPTHREAD) $(LIBBLKID) + $(LIBPTHREAD) $(LIBBLKID) $(LIBURCU) LTDEPENDENCIES = $(LIBXFS) $(LIBXLOG) $(LIBXCMD) $(LIBFROG) LLDFLAGS = -static-libtool-libs diff --git a/repair/prefetch.c b/repair/prefetch.c index 48affa18..22a0c0c9 100644 --- a/repair/prefetch.c +++ b/repair/prefetch.c @@ -660,6 +660,7 @@ pf_io_worker( if (buf == NULL) return NULL; + rcu_register_thread(); pthread_mutex_lock(&args->lock); while (!args->queuing_done || !btree_is_empty(args->io_queue)) { pftrace("waiting to start prefetch I/O for AG %d", args->agno); @@ -682,6 +683,7 @@ pf_io_worker( free(buf); pftrace("finished prefetch I/O for AG %d", args->agno); + rcu_unregister_thread(); return NULL; } @@ -726,6 +728,8 @@ pf_queuing_worker( struct xfs_ino_geometry *igeo = M_IGEO(mp); unsigned long long cluster_mask; + rcu_register_thread(); + cluster_mask = (1ULL << igeo->inodes_per_cluster) - 1; for (i = 0; i < PF_THREAD_COUNT; i++) { @@ -739,7 +743,7 @@ pf_queuing_worker( args->io_threads[i] = 0; if (i == 0) { pf_skip_prefetch_thread(args); - return NULL; + goto out; } /* * since we have at least one I/O thread, use them for @@ -779,7 +783,6 @@ pf_queuing_worker( * Start processing as well, in case everything so * far was already prefetched and the queue is empty. */ - pf_start_io_workers(args); pf_start_processing(args); sem_wait(&args->ra_count); @@ -841,6 +844,8 @@ pf_queuing_worker( if (next_args) pf_create_prefetch_thread(next_args); +out: + rcu_unregister_thread(); return NULL; } diff --git a/repair/progress.c b/repair/progress.c index e5a9c1ef..f6c4d988 100644 --- a/repair/progress.c +++ b/repair/progress.c @@ -182,6 +182,7 @@ progress_rpt_thread (void *p) do_error (_("progress_rpt: cannot malloc progress msg buffer\n")); running = 1; + rcu_register_thread(); /* * Specify a repeating timer that fires each MSG_INTERVAL seconds. @@ -286,7 +287,8 @@ progress_rpt_thread (void *p) do_warn(_("cannot delete timer\n")); free (msgbuf); - return (NULL); + rcu_unregister_thread(); + return NULL; } int diff --git a/scrub/Makefile b/scrub/Makefile index 47c887eb..849e3afd 100644 --- a/scrub/Makefile +++ b/scrub/Makefile @@ -71,7 +71,8 @@ spacemap.c \ vfs.c \ xfs_scrub.c -LLDLIBS += $(LIBHANDLE) $(LIBFROG) $(LIBPTHREAD) $(LIBICU_LIBS) $(LIBRT) +LLDLIBS += $(LIBHANDLE) $(LIBFROG) $(LIBPTHREAD) $(LIBICU_LIBS) $(LIBRT) \ + $(LIBURCU) LTDEPENDENCIES += $(LIBHANDLE) $(LIBFROG) LLDFLAGS = -static diff --git a/scrub/progress.c b/scrub/progress.c index 15247b7c..a3d096f9 100644 --- a/scrub/progress.c +++ b/scrub/progress.c @@ -116,6 +116,7 @@ progress_report_thread(void *arg) struct timespec abstime; int ret; + rcu_register_thread(); pthread_mutex_lock(&pt.lock); while (1) { uint64_t progress_val; @@ -139,6 +140,7 @@ progress_report_thread(void *arg) progress_report(progress_val); } pthread_mutex_unlock(&pt.lock); + rcu_unregister_thread(); return NULL; }