From a8306f2d4dcea03538c70c26d2948483f70254ff Mon Sep 17 00:00:00 2001 From: Nick Desaulniers Date: Thu, 31 Aug 2023 09:33:40 -0700 Subject: compiler.h: unify __UNIQUE_ID commit 6f33d58794ef ("__UNIQUE_ID()") added a fallback definition of __UNIQUE_ID because gcc 4.2 and older did not support __COUNTER__. Also, this commit is effectively a revert of commit b41c29b0527c ("Kbuild: provide a __UNIQUE_ID for clang") which mentions clang 2.6+ supporting __COUNTER__. Documentation/process/changes.rst currently lists the minimum supported version of these compilers as: - gcc: 5.1 - clang: 11.0.0 It should be safe to say that __COUNTER__ is well supported by this point. Link: https://lkml.kernel.org/r/20230831-unique_id-v1-1-28bacd18eb1d@google.com Signed-off-by: Nick Desaulniers Cc: Arnd Bergmann Cc: Jan Beulich Cc: Luc Van Oostenryck Cc: Michal rarek Cc: Nathan Chancellor Cc: Paul Russel Cc: Tom Rix Signed-off-by: Andrew Morton --- include/linux/compiler-clang.h | 5 ----- include/linux/compiler-gcc.h | 2 -- include/linux/compiler.h | 5 +---- 3 files changed, 1 insertion(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h index 9b673fefcef8..ddab1ef22bee 100644 --- a/include/linux/compiler-clang.h +++ b/include/linux/compiler-clang.h @@ -14,11 +14,6 @@ #undef __cleanup #define __cleanup(func) __maybe_unused __attribute__((__cleanup__(func))) -/* same as gcc, this was present in clang-2.6 so we can assume it works - * with any version that can compile the kernel - */ -#define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__) - /* all clang versions usable with the kernel support KASAN ABI version 5 */ #define KASAN_ABI_VERSION 5 diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h index 7af9e34ec261..2ceba3fe4ec1 100644 --- a/include/linux/compiler-gcc.h +++ b/include/linux/compiler-gcc.h @@ -39,8 +39,6 @@ #define __noretpoline __attribute__((__indirect_branch__("keep"))) #endif -#define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__) - #if defined(LATENT_ENTROPY_PLUGIN) && !defined(__CHECKER__) #define __latent_entropy __attribute__((latent_entropy)) #endif diff --git a/include/linux/compiler.h b/include/linux/compiler.h index d7779a18b24f..174099fdc485 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -177,10 +177,7 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val, __asm__ ("" : "=r" (var) : "0" (var)) #endif -/* Not-quite-unique ID. */ -#ifndef __UNIQUE_ID -# define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __LINE__) -#endif +#define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__) /** * data_race - mark an expression as containing intentional data races -- cgit v1.2.3 From 33a9813825710fdc2b980d566ee391fd093a36c6 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 24 Aug 2023 16:31:42 +0200 Subject: introduce __next_thread(), fix next_tid() vs exec() race Patch series "introduce __next_thread(), change next_thread()". After commit dce8f8ed1de1 ("document while_each_thread(), change first_tid() to use for_each_thread()") + this series 1. We have only one lockless user of next_thread(), task_group_seq_get_next(). I think it should be changed too. 2. We have only one user of task_struct->thread_group, thread_group_empty(). The next patches will change thread_group_empty() and kill ->thread_group. This patch (of 2): next_tid(start) does: rcu_read_lock(); if (pid_alive(start)) { pos = next_thread(start); if (thread_group_leader(pos)) pos = NULL; else get_task_struct(pos); it should return pos = NULL when next_thread() wraps to the 1st thread in the thread group, group leader, and the thread_group_leader() check tries to detect this case. But this can race with exec. To simplify, suppose we have a main thread M and a single sub-thread T, next_tid(T) should return NULL. Now suppose that T execs. If next_tid(T) is called after T changes the leadership and before it does release_task() which removes the old leader from list, then next_thread() returns M and thread_group_leader(M) = F. Lockless use of next_thread() should be avoided. After this change only task_group_seq_get_next() does this, and I believe it should be changed as well. Link: https://lkml.kernel.org/r/20230824143112.GA31208@redhat.com Link: https://lkml.kernel.org/r/20230824143142.GA31222@redhat.com Signed-off-by: Oleg Nesterov Cc: Eric W. Biederman Cc: Peter Zijlstra Signed-off-by: Andrew Morton --- fs/proc/base.c | 6 ++---- include/linux/sched/signal.h | 11 +++++++++++ 2 files changed, 13 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/fs/proc/base.c b/fs/proc/base.c index ffd54617c354..c0e971cc6d41 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -3840,10 +3840,8 @@ static struct task_struct *next_tid(struct task_struct *start) struct task_struct *pos = NULL; rcu_read_lock(); if (pid_alive(start)) { - pos = next_thread(start); - if (thread_group_leader(pos)) - pos = NULL; - else + pos = __next_thread(start); + if (pos) get_task_struct(pos); } rcu_read_unlock(); diff --git a/include/linux/sched/signal.h b/include/linux/sched/signal.h index 0014d3adaf84..7fb34b8cda54 100644 --- a/include/linux/sched/signal.h +++ b/include/linux/sched/signal.h @@ -715,6 +715,17 @@ bool same_thread_group(struct task_struct *p1, struct task_struct *p2) return p1->signal == p2->signal; } +/* + * returns NULL if p is the last thread in the thread group + */ +static inline struct task_struct *__next_thread(struct task_struct *p) +{ + return list_next_or_null_rcu(&p->signal->thread_head, + &p->thread_node, + struct task_struct, + thread_node); +} + static inline struct task_struct *next_thread(const struct task_struct *p) { return list_entry_rcu(p->thread_group.next, -- cgit v1.2.3 From d639cf4abb4d171ab2456904da5668c42b5c1937 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 24 Aug 2023 16:32:01 +0200 Subject: change next_thread() to use __next_thread() ?: group_leader This relies on fact that group leader is always the 1st entry in the signal->thread_head list. With or without this change, if the lockless next_thread(last_thread) races with exec it can return the old or the new leader. We are almost ready to kill task->thread_group, after this change its only user is thread_group_empty(). Link: https://lkml.kernel.org/r/20230824143201.GB31222@redhat.com Signed-off-by: Oleg Nesterov Cc: Eric W. Biederman Cc: Peter Zijlstra Signed-off-by: Andrew Morton --- include/linux/sched/signal.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched/signal.h b/include/linux/sched/signal.h index 7fb34b8cda54..cffc882d367f 100644 --- a/include/linux/sched/signal.h +++ b/include/linux/sched/signal.h @@ -726,10 +726,9 @@ static inline struct task_struct *__next_thread(struct task_struct *p) thread_node); } -static inline struct task_struct *next_thread(const struct task_struct *p) +static inline struct task_struct *next_thread(struct task_struct *p) { - return list_entry_rcu(p->thread_group.next, - struct task_struct, thread_group); + return __next_thread(p) ?: p->group_leader; } static inline int thread_group_empty(struct task_struct *p) -- cgit v1.2.3 From e34a35ee1f52312af130b5ebd42fa28313fc6149 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 26 Aug 2023 13:14:06 +0200 Subject: change thread_group_empty() to use task_struct->thread_node Patch series "kill task_struct->thread_group". This patch (of 2): It could use list_is_singular() but this way it is cheaper. Plus the thread_group_leader() check makes it clear that thread_group_empty() can only return true if p is a group leader. This was not immediately obvious before this patch. task_struct->thread_group no longer has users, it can die. Link: https://lkml.kernel.org/r/20230826111200.GA22982@redhat.com Link: https://lkml.kernel.org/r/20230826111406.GA23238@redhat.com Signed-off-by: Oleg Nesterov Cc: Eric W. Biederman Cc: Peter Zijlstra Signed-off-by: Andrew Morton --- include/linux/sched/signal.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sched/signal.h b/include/linux/sched/signal.h index cffc882d367f..d7fa3ca2fa53 100644 --- a/include/linux/sched/signal.h +++ b/include/linux/sched/signal.h @@ -733,7 +733,8 @@ static inline struct task_struct *next_thread(struct task_struct *p) static inline int thread_group_empty(struct task_struct *p) { - return list_empty(&p->thread_group); + return thread_group_leader(p) && + list_is_last(&p->thread_node, &p->signal->thread_head); } #define delay_group_leader(p) \ -- cgit v1.2.3 From 8e1f385104ac044f1552686ad6e1cbc71cc05a30 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sat, 26 Aug 2023 13:14:09 +0200 Subject: kill task_struct->thread_group The last user was removed by the previous patch. Link: https://lkml.kernel.org/r/20230826111409.GA23243@redhat.com Signed-off-by: Oleg Nesterov Cc: Eric W. Biederman Cc: Peter Zijlstra Signed-off-by: Andrew Morton --- include/linux/sched.h | 1 - init/init_task.c | 1 - kernel/exit.c | 1 - kernel/fork.c | 3 --- 4 files changed, 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 77f01ac385f7..6d1341b1673f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1002,7 +1002,6 @@ struct task_struct { /* PID/PID hash table linkage. */ struct pid *thread_pid; struct hlist_node pid_links[PIDTYPE_MAX]; - struct list_head thread_group; struct list_head thread_node; struct completion *vfork_done; diff --git a/init/init_task.c b/init/init_task.c index ff6c4b9bfe6b..c0de0200fd56 100644 --- a/init/init_task.c +++ b/init/init_task.c @@ -132,7 +132,6 @@ struct task_struct init_task .pi_lock = __RAW_SPIN_LOCK_UNLOCKED(init_task.pi_lock), .timer_slack_ns = 50000, /* 50 usec default slack */ .thread_pid = &init_struct_pid, - .thread_group = LIST_HEAD_INIT(init_task.thread_group), .thread_node = LIST_HEAD_INIT(init_signals.thread_head), #ifdef CONFIG_AUDIT .loginuid = INVALID_UID, diff --git a/kernel/exit.c b/kernel/exit.c index edb50b4c9972..f3ba4b97a7d9 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -133,7 +133,6 @@ static void __unhash_process(struct task_struct *p, bool group_dead) list_del_init(&p->sibling); __this_cpu_dec(process_counts); } - list_del_rcu(&p->thread_group); list_del_rcu(&p->thread_node); } diff --git a/kernel/fork.c b/kernel/fork.c index 3b6d20dfb9a8..b9d3aa493bbd 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -2576,7 +2576,6 @@ __latent_entropy struct task_struct *copy_process( p->dirty_paused_when = 0; p->pdeath_signal = 0; - INIT_LIST_HEAD(&p->thread_group); p->task_works = NULL; clear_posix_cputimers_work(p); @@ -2704,8 +2703,6 @@ __latent_entropy struct task_struct *copy_process( atomic_inc(¤t->signal->live); refcount_inc(¤t->signal->sigcnt); task_join_group_stop(p); - list_add_tail_rcu(&p->thread_group, - &p->group_leader->thread_group); list_add_tail_rcu(&p->thread_node, &p->signal->thread_head); } -- cgit v1.2.3 From 9cba82bba500e3ce875381350f289cfb3aa633ba Mon Sep 17 00:00:00 2001 From: Xingui Yang Date: Tue, 5 Sep 2023 02:48:33 +0000 Subject: seq_file: add helper macro to define attribute for rw file Patch series "Add helper macro DEFINE_SHOW_STORE_ATTRIBUTE() at seq_file.c", v6. We already own DEFINE_SHOW_ATTRIBUTE() helper macro for defining attribute for read-only file, but we found many of drivers also want a helper macro for read-write file too. So we add this helper macro to reduce duplicated code. This patch (of 3): We already own DEFINE_SHOW_ATTRIBUTE() helper macro for defining attribute for read-only file, but many of drivers want a helper macro for read-write file too. So we add DEFINE_SHOW_STORE_ATTRIBUTE() helper to reduce duplicated code. Link: https://lkml.kernel.org/r/20230905024835.43219-1-yangxingui@huawei.com Link: https://lkml.kernel.org/r/20230905024835.43219-2-yangxingui@huawei.com Signed-off-by: Luo Jiaxing Co-developed-by: Xingui Yang Signed-off-by: Xingui Yang Reviewed-by: Andy Shevchenko Cc: Al Viro Cc: Animesh Manna Cc: Anshuman Gupta Cc: Damien Le Moal Cc: Felipe Balbi Cc: Greg Kroah-Hartman Cc: Himanshu Madhani Cc: James Bottomley Cc: John Garry Cc: Martin K. Petersen Cc: Uma Shankar Cc: Xiang Chen Cc: Zeng Tao Signed-off-by: Andrew Morton --- include/linux/seq_file.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'include/linux') diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index 386ab580b839..234bcdb1fba4 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -207,6 +207,21 @@ static const struct file_operations __name ## _fops = { \ .release = single_release, \ } +#define DEFINE_SHOW_STORE_ATTRIBUTE(__name) \ +static int __name ## _open(struct inode *inode, struct file *file) \ +{ \ + return single_open(file, __name ## _show, inode->i_private); \ +} \ + \ +static const struct file_operations __name ## _fops = { \ + .owner = THIS_MODULE, \ + .open = __name ## _open, \ + .read = seq_read, \ + .write = __name ## _write, \ + .llseek = seq_lseek, \ + .release = single_release, \ +} + #define DEFINE_PROC_SHOW_ATTRIBUTE(__name) \ static int __name ## _open(struct inode *inode, struct file *file) \ { \ -- cgit v1.2.3 From 6309727ef27162deabd5c095c11af24970fba5a2 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Fri, 8 Sep 2023 01:40:48 +0200 Subject: kthread: add kthread_stop_put Add a kthread_stop_put() helper that stops a thread and puts its task struct. Use it to replace the various instances of kthread_stop() followed by put_task_struct(). Remove the kthread_stop_put() macro in usbip that is similar but doesn't return the result of kthread_stop(). [agruenba@redhat.com: fix kerneldoc comment] Link: https://lkml.kernel.org/r/20230911111730.2565537-1-agruenba@redhat.com [akpm@linux-foundation.org: document kthread_stop_put()'s argument] Link: https://lkml.kernel.org/r/20230907234048.2499820-1-agruenba@redhat.com Signed-off-by: Andreas Gruenbacher Signed-off-by: Andrew Morton --- drivers/accel/ivpu/ivpu_job.c | 3 +-- drivers/dma-buf/st-dma-fence-chain.c | 12 ++++-------- drivers/dma-buf/st-dma-fence.c | 4 +--- drivers/gpu/drm/i915/gt/selftest_migrate.c | 4 +--- drivers/net/xen-netback/interface.c | 3 +-- drivers/usb/usbip/usbip_common.h | 6 ------ fs/gfs2/ops_fstype.c | 9 +++------ include/linux/kthread.h | 1 + kernel/irq/manage.c | 15 +++++---------- kernel/kthread.c | 18 ++++++++++++++++++ kernel/smpboot.c | 3 +-- mm/damon/core.c | 3 +-- net/core/pktgen.c | 3 +-- 13 files changed, 38 insertions(+), 46 deletions(-) (limited to 'include/linux') diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index de9e69f70af7..76f468c9f761 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -618,6 +618,5 @@ int ivpu_job_done_thread_init(struct ivpu_device *vdev) void ivpu_job_done_thread_fini(struct ivpu_device *vdev) { - kthread_stop(vdev->job_done_thread); - put_task_struct(vdev->job_done_thread); + kthread_stop_put(vdev->job_done_thread); } diff --git a/drivers/dma-buf/st-dma-fence-chain.c b/drivers/dma-buf/st-dma-fence-chain.c index c0979c8049b5..9c2a0c082a76 100644 --- a/drivers/dma-buf/st-dma-fence-chain.c +++ b/drivers/dma-buf/st-dma-fence-chain.c @@ -476,10 +476,9 @@ static int find_race(void *arg) for (i = 0; i < ncpus; i++) { int ret; - ret = kthread_stop(threads[i]); + ret = kthread_stop_put(threads[i]); if (ret && !err) err = ret; - put_task_struct(threads[i]); } kfree(threads); @@ -591,8 +590,7 @@ static int wait_forward(void *arg) for (i = 0; i < fc.chain_length; i++) dma_fence_signal(fc.fences[i]); - err = kthread_stop(tsk); - put_task_struct(tsk); + err = kthread_stop_put(tsk); err: fence_chains_fini(&fc); @@ -621,8 +619,7 @@ static int wait_backward(void *arg) for (i = fc.chain_length; i--; ) dma_fence_signal(fc.fences[i]); - err = kthread_stop(tsk); - put_task_struct(tsk); + err = kthread_stop_put(tsk); err: fence_chains_fini(&fc); @@ -669,8 +666,7 @@ static int wait_random(void *arg) for (i = 0; i < fc.chain_length; i++) dma_fence_signal(fc.fences[i]); - err = kthread_stop(tsk); - put_task_struct(tsk); + err = kthread_stop_put(tsk); err: fence_chains_fini(&fc); diff --git a/drivers/dma-buf/st-dma-fence.c b/drivers/dma-buf/st-dma-fence.c index fb6e0a6ae2c9..b7c6f7ea9e0c 100644 --- a/drivers/dma-buf/st-dma-fence.c +++ b/drivers/dma-buf/st-dma-fence.c @@ -548,11 +548,9 @@ static int race_signal_callback(void *arg) for (i = 0; i < ARRAY_SIZE(t); i++) { int err; - err = kthread_stop(t[i].task); + err = kthread_stop_put(t[i].task); if (err && !ret) ret = err; - - put_task_struct(t[i].task); } } diff --git a/drivers/gpu/drm/i915/gt/selftest_migrate.c b/drivers/gpu/drm/i915/gt/selftest_migrate.c index 3def5ca72dec..0fb07f073baa 100644 --- a/drivers/gpu/drm/i915/gt/selftest_migrate.c +++ b/drivers/gpu/drm/i915/gt/selftest_migrate.c @@ -719,11 +719,9 @@ static int threaded_migrate(struct intel_migrate *migrate, if (IS_ERR_OR_NULL(tsk)) continue; - status = kthread_stop(tsk); + status = kthread_stop_put(tsk); if (status && !err) err = status; - - put_task_struct(tsk); } kfree(thread); diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c index f3f2c07423a6..33c8143619f0 100644 --- a/drivers/net/xen-netback/interface.c +++ b/drivers/net/xen-netback/interface.c @@ -672,8 +672,7 @@ err: static void xenvif_disconnect_queue(struct xenvif_queue *queue) { if (queue->task) { - kthread_stop(queue->task); - put_task_struct(queue->task); + kthread_stop_put(queue->task); queue->task = NULL; } diff --git a/drivers/usb/usbip/usbip_common.h b/drivers/usb/usbip/usbip_common.h index d8cbd2dfc2c2..282efca64a01 100644 --- a/drivers/usb/usbip/usbip_common.h +++ b/drivers/usb/usbip/usbip_common.h @@ -298,12 +298,6 @@ struct usbip_device { __k; \ }) -#define kthread_stop_put(k) \ - do { \ - kthread_stop(k); \ - put_task_struct(k); \ - } while (0) - /* usbip_common.c */ void usbip_dump_urb(struct urb *purb); void usbip_dump_header(struct usbip_header *pdu); diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 33ca04733e93..ecf789b7168c 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1126,8 +1126,7 @@ static int init_threads(struct gfs2_sbd *sdp) return 0; fail: - kthread_stop(sdp->sd_logd_process); - put_task_struct(sdp->sd_logd_process); + kthread_stop_put(sdp->sd_logd_process); sdp->sd_logd_process = NULL; return error; } @@ -1135,13 +1134,11 @@ fail: void gfs2_destroy_threads(struct gfs2_sbd *sdp) { if (sdp->sd_logd_process) { - kthread_stop(sdp->sd_logd_process); - put_task_struct(sdp->sd_logd_process); + kthread_stop_put(sdp->sd_logd_process); sdp->sd_logd_process = NULL; } if (sdp->sd_quotad_process) { - kthread_stop(sdp->sd_quotad_process); - put_task_struct(sdp->sd_quotad_process); + kthread_stop_put(sdp->sd_quotad_process); sdp->sd_quotad_process = NULL; } } diff --git a/include/linux/kthread.h b/include/linux/kthread.h index 2c30ade43bc8..b11f53c1ba2e 100644 --- a/include/linux/kthread.h +++ b/include/linux/kthread.h @@ -86,6 +86,7 @@ void free_kthread_struct(struct task_struct *k); void kthread_bind(struct task_struct *k, unsigned int cpu); void kthread_bind_mask(struct task_struct *k, const struct cpumask *mask); int kthread_stop(struct task_struct *k); +int kthread_stop_put(struct task_struct *k); bool kthread_should_stop(void); bool kthread_should_park(void); bool kthread_should_stop_or_park(void); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index d309ba84e08a..1782f90cd8c6 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -1852,15 +1852,13 @@ out_thread: struct task_struct *t = new->thread; new->thread = NULL; - kthread_stop(t); - put_task_struct(t); + kthread_stop_put(t); } if (new->secondary && new->secondary->thread) { struct task_struct *t = new->secondary->thread; new->secondary->thread = NULL; - kthread_stop(t); - put_task_struct(t); + kthread_stop_put(t); } out_mput: module_put(desc->owner); @@ -1971,12 +1969,9 @@ static struct irqaction *__free_irq(struct irq_desc *desc, void *dev_id) * the same bit to a newly requested action. */ if (action->thread) { - kthread_stop(action->thread); - put_task_struct(action->thread); - if (action->secondary && action->secondary->thread) { - kthread_stop(action->secondary->thread); - put_task_struct(action->secondary->thread); - } + kthread_stop_put(action->thread); + if (action->secondary && action->secondary->thread) + kthread_stop_put(action->secondary->thread); } /* Last action releases resources */ diff --git a/kernel/kthread.c b/kernel/kthread.c index 1eea53050bab..290cbc845225 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -715,6 +715,24 @@ int kthread_stop(struct task_struct *k) } EXPORT_SYMBOL(kthread_stop); +/** + * kthread_stop_put - stop a thread and put its task struct + * @k: thread created by kthread_create(). + * + * Stops a thread created by kthread_create() and put its task_struct. + * Only use when holding an extra task struct reference obtained by + * calling get_task_struct(). + */ +int kthread_stop_put(struct task_struct *k) +{ + int ret; + + ret = kthread_stop(k); + put_task_struct(k); + return ret; +} +EXPORT_SYMBOL(kthread_stop_put); + int kthreadd(void *unused) { struct task_struct *tsk = current; diff --git a/kernel/smpboot.c b/kernel/smpboot.c index f47d8f375946..1992b62e980b 100644 --- a/kernel/smpboot.c +++ b/kernel/smpboot.c @@ -272,8 +272,7 @@ static void smpboot_destroy_threads(struct smp_hotplug_thread *ht) struct task_struct *tsk = *per_cpu_ptr(ht->store, cpu); if (tsk) { - kthread_stop(tsk); - put_task_struct(tsk); + kthread_stop_put(tsk); *per_cpu_ptr(ht->store, cpu) = NULL; } } diff --git a/mm/damon/core.c b/mm/damon/core.c index bcd2bd9d6c10..2f54f153d7f5 100644 --- a/mm/damon/core.c +++ b/mm/damon/core.c @@ -699,8 +699,7 @@ static int __damon_stop(struct damon_ctx *ctx) if (tsk) { get_task_struct(tsk); mutex_unlock(&ctx->kdamond_lock); - kthread_stop(tsk); - put_task_struct(tsk); + kthread_stop_put(tsk); return 0; } mutex_unlock(&ctx->kdamond_lock); diff --git a/net/core/pktgen.c b/net/core/pktgen.c index f56b8d697014..826250a0f5b1 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -3982,8 +3982,7 @@ static void __net_exit pg_net_exit(struct net *net) list_for_each_safe(q, n, &list) { t = list_entry(q, struct pktgen_thread, th_list); list_del(&t->th_list); - kthread_stop(t->tsk); - put_task_struct(t->tsk); + kthread_stop_put(t->tsk); kfree(t); } -- cgit v1.2.3 From 5e57418a2031cd5e1863efdf3d7447a16a368172 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 11 Sep 2023 18:49:13 +0300 Subject: minmax: deduplicate __unconst_integer_typeof() It appears that compiler_types.h already have an implementation of the __unconst_integer_typeof() called __unqual_scalar_typeof(). Use it instead of the copy. Link: https://lkml.kernel.org/r/20230911154913.4176033-1-andriy.shevchenko@linux.intel.com Signed-off-by: Andy Shevchenko Acked-by: Herve Codina Signed-off-by: Andrew Morton --- include/linux/minmax.h | 26 +++----------------------- 1 file changed, 3 insertions(+), 23 deletions(-) (limited to 'include/linux') diff --git a/include/linux/minmax.h b/include/linux/minmax.h index 83aebc244cba..69bbe987fa87 100644 --- a/include/linux/minmax.h +++ b/include/linux/minmax.h @@ -2,6 +2,7 @@ #ifndef _LINUX_MINMAX_H #define _LINUX_MINMAX_H +#include #include #include @@ -134,27 +135,6 @@ */ #define max_t(type, x, y) __careful_cmp((type)(x), (type)(y), >) -/* - * Remove a const qualifier from integer types - * _Generic(foo, type-name: association, ..., default: association) performs a - * comparison against the foo type (not the qualified type). - * Do not use the const keyword in the type-name as it will not match the - * unqualified type of foo. - */ -#define __unconst_integer_type_cases(type) \ - unsigned type: (unsigned type)0, \ - signed type: (signed type)0 - -#define __unconst_integer_typeof(x) typeof( \ - _Generic((x), \ - char: (char)0, \ - __unconst_integer_type_cases(char), \ - __unconst_integer_type_cases(short), \ - __unconst_integer_type_cases(int), \ - __unconst_integer_type_cases(long), \ - __unconst_integer_type_cases(long long), \ - default: (x))) - /* * Do not check the array parameter using __must_be_array(). * In the following legit use-case where the "array" passed is a simple pointer, @@ -169,13 +149,13 @@ * 'int *buff' and 'int buff[N]' types. * * The array can be an array of const items. - * typeof() keeps the const qualifier. Use __unconst_integer_typeof() in order + * typeof() keeps the const qualifier. Use __unqual_scalar_typeof() in order * to discard the const qualifier for the __element variable. */ #define __minmax_array(op, array, len) ({ \ typeof(&(array)[0]) __array = (array); \ typeof(len) __len = (len); \ - __unconst_integer_typeof(__array[0]) __element = __array[--__len]; \ + __unqual_scalar_typeof(__array[0]) __element = __array[--__len];\ while (__len--) \ __element = op(__element, __array[__len]); \ __element; }) -- cgit v1.2.3 From f6e9d38f8eb00ac8b52e6d15f6aa9bcecacb081b Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Tue, 12 Sep 2023 12:23:55 +0300 Subject: minmax: fix header inclusions BUILD_BUG_ON*() macros are defined in build_bug.h. Include it. Replace compiler_types.h by compiler.h, which provides the former, to have a definition of the __UNIQUE_ID(). Link: https://lkml.kernel.org/r/20230912092355.79280-1-andriy.shevchenko@linux.intel.com Signed-off-by: Andy Shevchenko Reviewed-by: Herve Codina Cc: Rasmus Villemoes Signed-off-by: Andrew Morton --- include/linux/minmax.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/minmax.h b/include/linux/minmax.h index 69bbe987fa87..ca69abd6151e 100644 --- a/include/linux/minmax.h +++ b/include/linux/minmax.h @@ -2,7 +2,8 @@ #ifndef _LINUX_MINMAX_H #define _LINUX_MINMAX_H -#include +#include +#include #include #include -- cgit v1.2.3 From a9e1a3d84e4a0ea560ed4d84c28d06dbfdffed22 Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Thu, 14 Sep 2023 11:31:35 +0800 Subject: crash_core: change the prototype of function parse_crashkernel() Add two parameters 'low_size' and 'high' to function parse_crashkernel(), later crashkernel=,high|low parsing will be added. Make adjustments in all call sites of parse_crashkernel() in arch. Link: https://lkml.kernel.org/r/20230914033142.676708-3-bhe@redhat.com Signed-off-by: Baoquan He Reviewed-by: Zhen Lei Cc: Catalin Marinas Cc: Chen Jiahao Signed-off-by: Andrew Morton --- arch/arm/kernel/setup.c | 3 ++- arch/arm64/mm/init.c | 2 +- arch/ia64/kernel/setup.c | 2 +- arch/loongarch/kernel/setup.c | 4 +++- arch/mips/kernel/setup.c | 3 ++- arch/powerpc/kernel/fadump.c | 2 +- arch/powerpc/kexec/core.c | 2 +- arch/powerpc/mm/nohash/kaslr_booke.c | 2 +- arch/riscv/mm/init.c | 2 +- arch/s390/kernel/setup.c | 4 ++-- arch/sh/kernel/machine_kexec.c | 2 +- arch/x86/kernel/setup.c | 3 ++- include/linux/crash_core.h | 3 ++- kernel/crash_core.c | 15 ++++++++++++--- 14 files changed, 32 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index c66b560562b3..e2bb7afd0683 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -1010,7 +1010,8 @@ static void __init reserve_crashkernel(void) total_mem = get_total_mem(); ret = parse_crashkernel(boot_command_line, total_mem, - &crash_size, &crash_base); + &crash_size, &crash_base, + NULL, NULL); /* invalid value specified or crashkernel=0 */ if (ret || !crash_size) return; diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index 8a0f8604348b..801c59c39a8f 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -142,7 +142,7 @@ static void __init reserve_crashkernel(void) /* crashkernel=X[@offset] */ ret = parse_crashkernel(cmdline, memblock_phys_mem_size(), - &crash_size, &crash_base); + &crash_size, &crash_base, NULL, NULL); if (ret == -ENOENT) { ret = parse_crashkernel_high(cmdline, 0, &crash_size, &crash_base); if (ret || !crash_size) diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 5a55ac82c13a..4faea2d2cf07 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -277,7 +277,7 @@ static void __init setup_crashkernel(unsigned long total, int *n) int ret; ret = parse_crashkernel(boot_command_line, total, - &size, &base); + &size, &base, NULL, NULL); if (ret == 0 && size > 0) { if (!base) { sort_regions(rsvd_region, *n); diff --git a/arch/loongarch/kernel/setup.c b/arch/loongarch/kernel/setup.c index 7783f0a3d742..4de32b07c0dc 100644 --- a/arch/loongarch/kernel/setup.c +++ b/arch/loongarch/kernel/setup.c @@ -267,7 +267,9 @@ static void __init arch_parse_crashkernel(void) unsigned long long crash_base, crash_size; total_mem = memblock_phys_mem_size(); - ret = parse_crashkernel(boot_command_line, total_mem, &crash_size, &crash_base); + ret = parse_crashkernel(boot_command_line, total_mem, + &crash_size, &crash_base, + NULL, NULL); if (ret < 0 || crash_size <= 0) return; diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c index cb871eb784a7..08321c945ac4 100644 --- a/arch/mips/kernel/setup.c +++ b/arch/mips/kernel/setup.c @@ -460,7 +460,8 @@ static void __init mips_parse_crashkernel(void) total_mem = memblock_phys_mem_size(); ret = parse_crashkernel(boot_command_line, total_mem, - &crash_size, &crash_base); + &crash_size, &crash_base, + NULL, NULL); if (ret != 0 || crash_size <= 0) return; diff --git a/arch/powerpc/kernel/fadump.c b/arch/powerpc/kernel/fadump.c index 3ff2da7b120b..d14eda1e8589 100644 --- a/arch/powerpc/kernel/fadump.c +++ b/arch/powerpc/kernel/fadump.c @@ -313,7 +313,7 @@ static __init u64 fadump_calculate_reserve_size(void) * memory at a predefined offset. */ ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(), - &size, &base); + &size, &base, NULL, NULL); if (ret == 0 && size > 0) { unsigned long max_size; diff --git a/arch/powerpc/kexec/core.c b/arch/powerpc/kexec/core.c index de64c7962991..9346c960b296 100644 --- a/arch/powerpc/kexec/core.c +++ b/arch/powerpc/kexec/core.c @@ -109,7 +109,7 @@ void __init reserve_crashkernel(void) total_mem_sz = memory_limit ? memory_limit : memblock_phys_mem_size(); /* use common parsing */ ret = parse_crashkernel(boot_command_line, total_mem_sz, - &crash_size, &crash_base); + &crash_size, &crash_base, NULL, NULL); if (ret == 0 && crash_size > 0) { crashk_res.start = crash_base; crashk_res.end = crash_base + crash_size - 1; diff --git a/arch/powerpc/mm/nohash/kaslr_booke.c b/arch/powerpc/mm/nohash/kaslr_booke.c index 2fb3edafe9ab..b4f2786a7d2b 100644 --- a/arch/powerpc/mm/nohash/kaslr_booke.c +++ b/arch/powerpc/mm/nohash/kaslr_booke.c @@ -178,7 +178,7 @@ static void __init get_crash_kernel(void *fdt, unsigned long size) int ret; ret = parse_crashkernel(boot_command_line, size, &crash_size, - &crash_base); + &crash_base, NULL, NULL); if (ret != 0 || crash_size == 0) return; if (crash_base == 0) diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c index 0798bd861dcb..9fe448900059 100644 --- a/arch/riscv/mm/init.c +++ b/arch/riscv/mm/init.c @@ -1388,7 +1388,7 @@ static void __init reserve_crashkernel(void) } ret = parse_crashkernel(cmdline, memblock_phys_mem_size(), - &crash_size, &crash_base); + &crash_size, &crash_base, NULL, NULL); if (ret == -ENOENT) { /* Fallback to crashkernel=X,[high,low] */ ret = parse_crashkernel_high(cmdline, 0, &crash_size, &crash_base); diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index de6ad0fb2328..e555b576d3c8 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -625,8 +625,8 @@ static void __init reserve_crashkernel(void) phys_addr_t low, high; int rc; - rc = parse_crashkernel(boot_command_line, ident_map_size, &crash_size, - &crash_base); + rc = parse_crashkernel(boot_command_line, ident_map_size, + &crash_size, &crash_base, NULL, NULL); crash_base = ALIGN(crash_base, KEXEC_CRASH_MEM_ALIGN); crash_size = ALIGN(crash_size, KEXEC_CRASH_MEM_ALIGN); diff --git a/arch/sh/kernel/machine_kexec.c b/arch/sh/kernel/machine_kexec.c index 223c14f44af7..fa3a7b36190a 100644 --- a/arch/sh/kernel/machine_kexec.c +++ b/arch/sh/kernel/machine_kexec.c @@ -154,7 +154,7 @@ void __init reserve_crashkernel(void) int ret; ret = parse_crashkernel(boot_command_line, memblock_phys_mem_size(), - &crash_size, &crash_base); + &crash_size, &crash_base, NULL, NULL); if (ret == 0 && crash_size > 0) { crashk_res.start = crash_base; crashk_res.end = crash_base + crash_size - 1; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index b098b1fa2470..655c04812905 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -553,7 +553,8 @@ static void __init reserve_crashkernel(void) total_mem = memblock_phys_mem_size(); /* crashkernel=XM */ - ret = parse_crashkernel(boot_command_line, total_mem, &crash_size, &crash_base); + ret = parse_crashkernel(boot_command_line, total_mem, + &crash_size, &crash_base, NULL, NULL); if (ret != 0 || crash_size <= 0) { /* crashkernel=X,high */ ret = parse_crashkernel_high(boot_command_line, total_mem, diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 0c06561bf5ff..6156355ef831 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -80,7 +80,8 @@ Elf_Word *append_elf_note(Elf_Word *buf, char *name, unsigned int type, void final_note(Elf_Word *buf); int __init parse_crashkernel(char *cmdline, unsigned long long system_ram, - unsigned long long *crash_size, unsigned long long *crash_base); + unsigned long long *crash_size, unsigned long long *crash_base, + unsigned long long *low_size, bool *high); int parse_crashkernel_high(char *cmdline, unsigned long long system_ram, unsigned long long *crash_size, unsigned long long *crash_base); int parse_crashkernel_low(char *cmdline, unsigned long long system_ram, diff --git a/kernel/crash_core.c b/kernel/crash_core.c index 507113932aa9..33ced5b5ed4e 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -287,10 +287,19 @@ static int __init __parse_crashkernel(char *cmdline, int __init parse_crashkernel(char *cmdline, unsigned long long system_ram, unsigned long long *crash_size, - unsigned long long *crash_base) + unsigned long long *crash_base, + unsigned long long *low_size, + bool *high) { - return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, - NULL); + int ret; + + /* crashkernel=X[@offset] */ + ret = __parse_crashkernel(cmdline, system_ram, crash_size, + crash_base, NULL); + if (!high) + return ret; + + return 0; } int __init parse_crashkernel_high(char *cmdline, -- cgit v1.2.3 From 70916e9c8d9f1a286c99727072b22e395097909f Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Thu, 14 Sep 2023 11:31:36 +0800 Subject: crash_core: change parse_crashkernel() to support crashkernel=,high|low parsing Now parse_crashkernel() is a real entry point for all kinds of crahskernel parsing on any architecture. And wrap the crahskernel=,high|low handling inside CONFIG_ARCH_HAS_GENERIC_CRASHKERNEL_RESERVATION ifdeffery scope. Link: https://lkml.kernel.org/r/20230914033142.676708-4-bhe@redhat.com Signed-off-by: Baoquan He Reviewed-by: Zhen Lei Cc: Catalin Marinas Cc: Chen Jiahao Signed-off-by: Andrew Morton --- include/linux/crash_core.h | 6 ++++++ kernel/crash_core.c | 36 +++++++++++++++++++++++++++++++++--- 2 files changed, 39 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 6156355ef831..d8050a7eab01 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -79,6 +79,12 @@ Elf_Word *append_elf_note(Elf_Word *buf, char *name, unsigned int type, void *data, size_t data_len); void final_note(Elf_Word *buf); +#ifdef CONFIG_ARCH_HAS_GENERIC_CRASHKERNEL_RESERVATION +#ifndef DEFAULT_CRASH_KERNEL_LOW_SIZE +#define DEFAULT_CRASH_KERNEL_LOW_SIZE (128UL << 20) +#endif +#endif + int __init parse_crashkernel(char *cmdline, unsigned long long system_ram, unsigned long long *crash_size, unsigned long long *crash_base, unsigned long long *low_size, bool *high); diff --git a/kernel/crash_core.c b/kernel/crash_core.c index 33ced5b5ed4e..99a243540a35 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -283,6 +283,9 @@ static int __init __parse_crashkernel(char *cmdline, /* * That function is the entry point for command line parsing and should be * called from the arch-specific code. + * + * If crashkernel=,high|low is supported on architecture, non-NULL values + * should be passed to parameters 'low_size' and 'high'. */ int __init parse_crashkernel(char *cmdline, unsigned long long system_ram, @@ -296,10 +299,37 @@ int __init parse_crashkernel(char *cmdline, /* crashkernel=X[@offset] */ ret = __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, NULL); - if (!high) - return ret; +#ifdef CONFIG_ARCH_HAS_GENERIC_CRASHKERNEL_RESERVATION + /* + * If non-NULL 'high' passed in and no normal crashkernel + * setting detected, try parsing crashkernel=,high|low. + */ + if (high && ret == -ENOENT) { + ret = __parse_crashkernel(cmdline, 0, crash_size, + crash_base, suffix_tbl[SUFFIX_HIGH]); + if (ret || !*crash_size) + return -EINVAL; - return 0; + /* + * crashkernel=Y,low can be specified or not, but invalid value + * is not allowed. + */ + ret = __parse_crashkernel(cmdline, 0, low_size, + crash_base, suffix_tbl[SUFFIX_LOW]); + if (ret == -ENOENT) { + *low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE; + ret = 0; + } else if (ret) { + return ret; + } + + *high = true; + } +#endif + if (!*crash_size) + ret = -EINVAL; + + return ret; } int __init parse_crashkernel_high(char *cmdline, -- cgit v1.2.3 From 0ab97169aa0517079b22c2e64192906caa5dc6d5 Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Thu, 14 Sep 2023 11:31:37 +0800 Subject: crash_core: add generic function to do reservation In architecture like x86_64, arm64 and riscv, they have vast virtual address space and usually have huge physical memory RAM. Their crashkernel reservation doesn't have to be limited under 4G RAM, but can be extended to the whole physical memory via crashkernel=,high support. Now add function reserve_crashkernel_generic() to reserve crashkernel memory if users specify any case of kernel pamameters, like crashkernel=xM[@offset] or crashkernel=,high|low. This is preparation to simplify code of crashkernel=,high support in architecutures. Link: https://lkml.kernel.org/r/20230914033142.676708-5-bhe@redhat.com Signed-off-by: Baoquan He Reviewed-by: Zhen Lei Cc: Catalin Marinas Cc: Chen Jiahao Signed-off-by: Andrew Morton --- include/linux/crash_core.h | 28 ++++++++++++ kernel/crash_core.c | 107 ++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 134 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index d8050a7eab01..4dbd6565e0ff 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -93,6 +93,34 @@ int parse_crashkernel_high(char *cmdline, unsigned long long system_ram, int parse_crashkernel_low(char *cmdline, unsigned long long system_ram, unsigned long long *crash_size, unsigned long long *crash_base); +#ifdef CONFIG_ARCH_HAS_GENERIC_CRASHKERNEL_RESERVATION +#ifndef DEFAULT_CRASH_KERNEL_LOW_SIZE +#define DEFAULT_CRASH_KERNEL_LOW_SIZE (128UL << 20) +#endif +#ifndef CRASH_ALIGN +#define CRASH_ALIGN SZ_2M +#endif +#ifndef CRASH_ADDR_LOW_MAX +#define CRASH_ADDR_LOW_MAX SZ_4G +#endif +#ifndef CRASH_ADDR_HIGH_MAX +#define CRASH_ADDR_HIGH_MAX memblock_end_of_DRAM() +#endif + +void __init reserve_crashkernel_generic(char *cmdline, + unsigned long long crash_size, + unsigned long long crash_base, + unsigned long long crash_low_size, + bool high); +#else +static inline void __init reserve_crashkernel_generic(char *cmdline, + unsigned long long crash_size, + unsigned long long crash_base, + unsigned long long crash_low_size, + bool high) +{} +#endif + /* Alignment required for elf header segment */ #define ELF_CORE_HEADER_ALIGN 4096 diff --git a/kernel/crash_core.c b/kernel/crash_core.c index 99a243540a35..72e358197b52 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -5,7 +5,6 @@ */ #include -#include #include #include #include @@ -13,6 +12,9 @@ #include #include #include +#include +#include +#include #include #include @@ -360,6 +362,109 @@ static int __init parse_crashkernel_dummy(char *arg) } early_param("crashkernel", parse_crashkernel_dummy); +#ifdef CONFIG_ARCH_HAS_GENERIC_CRASHKERNEL_RESERVATION +static int __init reserve_crashkernel_low(unsigned long long low_size) +{ +#ifdef CONFIG_64BIT + unsigned long long low_base; + + low_base = memblock_phys_alloc_range(low_size, CRASH_ALIGN, 0, CRASH_ADDR_LOW_MAX); + if (!low_base) { + pr_err("cannot allocate crashkernel low memory (size:0x%llx).\n", low_size); + return -ENOMEM; + } + + pr_info("crashkernel low memory reserved: 0x%08llx - 0x%08llx (%lld MB)\n", + low_base, low_base + low_size, low_size >> 20); + + crashk_low_res.start = low_base; + crashk_low_res.end = low_base + low_size - 1; + insert_resource(&iomem_resource, &crashk_low_res); +#endif + return 0; +} + +void __init reserve_crashkernel_generic(char *cmdline, + unsigned long long crash_size, + unsigned long long crash_base, + unsigned long long crash_low_size, + bool high) +{ + unsigned long long search_end = CRASH_ADDR_LOW_MAX, search_base = 0; + bool fixed_base = false; + + /* User specifies base address explicitly. */ + if (crash_base) { + fixed_base = true; + search_base = crash_base; + search_end = crash_base + crash_size; + } else if (high) { + search_base = CRASH_ADDR_LOW_MAX; + search_end = CRASH_ADDR_HIGH_MAX; + } + +retry: + crash_base = memblock_phys_alloc_range(crash_size, CRASH_ALIGN, + search_base, search_end); + if (!crash_base) { + /* + * For crashkernel=size[KMG]@offset[KMG], print out failure + * message if can't reserve the specified region. + */ + if (fixed_base) { + pr_warn("crashkernel reservation failed - memory is in use.\n"); + return; + } + + /* + * For crashkernel=size[KMG], if the first attempt was for + * low memory, fall back to high memory, the minimum required + * low memory will be reserved later. + */ + if (!high && search_end == CRASH_ADDR_LOW_MAX) { + search_end = CRASH_ADDR_HIGH_MAX; + search_base = CRASH_ADDR_LOW_MAX; + crash_low_size = DEFAULT_CRASH_KERNEL_LOW_SIZE; + goto retry; + } + + /* + * For crashkernel=size[KMG],high, if the first attempt was + * for high memory, fall back to low memory. + */ + if (high && search_end == CRASH_ADDR_HIGH_MAX) { + search_end = CRASH_ADDR_LOW_MAX; + search_base = 0; + goto retry; + } + pr_warn("cannot allocate crashkernel (size:0x%llx)\n", + crash_size); + return; + } + + if ((crash_base > CRASH_ADDR_LOW_MAX) && + crash_low_size && reserve_crashkernel_low(crash_low_size)) { + memblock_phys_free(crash_base, crash_size); + return; + } + + pr_info("crashkernel reserved: 0x%016llx - 0x%016llx (%lld MB)\n", + crash_base, crash_base + crash_size, crash_size >> 20); + + /* + * The crashkernel memory will be removed from the kernel linear + * map. Inform kmemleak so that it won't try to access it. + */ + kmemleak_ignore_phys(crash_base); + if (crashk_low_res.end) + kmemleak_ignore_phys(crashk_low_res.start); + + crashk_res.start = crash_base; + crashk_res.end = crash_base + crash_size - 1; + insert_resource(&iomem_resource, &crashk_res); +} +#endif + int crash_prepare_elf64_headers(struct crash_mem *mem, int need_kernel_map, void **addr, unsigned long *sz) { -- cgit v1.2.3 From b631b95dded5e7f007a3a79cbaf82ef50c1e2cf7 Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Thu, 14 Sep 2023 11:31:38 +0800 Subject: crash_core: move crashk_*res definition into crash_core.c Both crashk_res and crashk_low_res are used to mark the reserved crashkernel regions in iomem_resource tree. And later the generic crashkernel resrvation will be added into crash_core.c. So move crashk_res and crashk_low_res definition into crash_core.c to avoid compiling error if CONFIG_CRASH_CORE=on while CONFIG_KEXEC_CORE is unset. Meanwhile include in if generic reservation is needed. In that case, need be added by ARCH. In asm/crash_core.h, ARCH can provide its own macro definitions to override macros in if needed. Wrap the including into CONFIG_ARCH_HAS_GENERIC_CRASHKERNEL_RESERVATION ifdeffery scope to avoid compiling error in other ARCH-es which don't take the generic reservation way yet. Link: https://lkml.kernel.org/r/20230914033142.676708-6-bhe@redhat.com Signed-off-by: Baoquan He Reviewed-by: Zhen Lei Cc: Catalin Marinas Cc: Chen Jiahao Signed-off-by: Andrew Morton --- include/linux/crash_core.h | 8 ++++++++ include/linux/kexec.h | 4 ---- kernel/crash_core.c | 16 ++++++++++++++++ kernel/kexec_core.c | 17 ----------------- 4 files changed, 24 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 4dbd6565e0ff..3c735a7e33fb 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -5,6 +5,14 @@ #include #include #include +#ifdef CONFIG_ARCH_HAS_GENERIC_CRASHKERNEL_RESERVATION +#include +#endif + +/* Location of a reserved region to hold the crash kernel. + */ +extern struct resource crashk_res; +extern struct resource crashk_low_res; #define CRASH_CORE_NOTE_NAME "CORE" #define CRASH_CORE_NOTE_HEAD_BYTES ALIGN(sizeof(struct elf_note), 4) diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 32c78078552c..8227455192b7 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -22,10 +22,6 @@ #include #include -/* Location of a reserved region to hold the crash kernel. - */ -extern struct resource crashk_res; -extern struct resource crashk_low_res; extern note_buf_t __percpu *crash_notes; #ifdef CONFIG_KEXEC_CORE diff --git a/kernel/crash_core.c b/kernel/crash_core.c index 72e358197b52..fa8808c4f00e 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -35,6 +35,22 @@ u32 *vmcoreinfo_note; /* trusted vmcoreinfo, e.g. we can make a copy in the crash memory */ static unsigned char *vmcoreinfo_data_safecopy; +/* Location of the reserved area for the crash kernel */ +struct resource crashk_res = { + .name = "Crash kernel", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM, + .desc = IORES_DESC_CRASH_KERNEL +}; +struct resource crashk_low_res = { + .name = "Crash kernel", + .start = 0, + .end = 0, + .flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM, + .desc = IORES_DESC_CRASH_KERNEL +}; + /* * parsing the "crashkernel" commandline * diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c index 9dc728982d79..be5642a4ec49 100644 --- a/kernel/kexec_core.c +++ b/kernel/kexec_core.c @@ -52,23 +52,6 @@ atomic_t __kexec_lock = ATOMIC_INIT(0); /* Flag to indicate we are going to kexec a new kernel */ bool kexec_in_progress = false; - -/* Location of the reserved area for the crash kernel */ -struct resource crashk_res = { - .name = "Crash kernel", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM, - .desc = IORES_DESC_CRASH_KERNEL -}; -struct resource crashk_low_res = { - .name = "Crash kernel", - .start = 0, - .end = 0, - .flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM, - .desc = IORES_DESC_CRASH_KERNEL -}; - int kexec_should_crash(struct task_struct *p) { /* -- cgit v1.2.3 From c37e56cac3d62c69f093904afbc58fc428484d14 Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Thu, 14 Sep 2023 11:31:42 +0800 Subject: crash_core.c: remove unneeded functions So far, nobody calls functions parse_crashkernel_high() and parse_crashkernel_low(), remove both of them. Link: https://lkml.kernel.org/r/20230914033142.676708-10-bhe@redhat.com Signed-off-by: Baoquan He Reviewed-by: Zhen Lei Cc: Catalin Marinas Cc: Chen Jiahao Signed-off-by: Andrew Morton --- include/linux/crash_core.h | 4 ---- kernel/crash_core.c | 18 ------------------ 2 files changed, 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 3c735a7e33fb..3426f6eef60b 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -96,10 +96,6 @@ void final_note(Elf_Word *buf); int __init parse_crashkernel(char *cmdline, unsigned long long system_ram, unsigned long long *crash_size, unsigned long long *crash_base, unsigned long long *low_size, bool *high); -int parse_crashkernel_high(char *cmdline, unsigned long long system_ram, - unsigned long long *crash_size, unsigned long long *crash_base); -int parse_crashkernel_low(char *cmdline, unsigned long long system_ram, - unsigned long long *crash_size, unsigned long long *crash_base); #ifdef CONFIG_ARCH_HAS_GENERIC_CRASHKERNEL_RESERVATION #ifndef DEFAULT_CRASH_KERNEL_LOW_SIZE diff --git a/kernel/crash_core.c b/kernel/crash_core.c index fa8808c4f00e..efe87d501c8c 100644 --- a/kernel/crash_core.c +++ b/kernel/crash_core.c @@ -350,24 +350,6 @@ int __init parse_crashkernel(char *cmdline, return ret; } -int __init parse_crashkernel_high(char *cmdline, - unsigned long long system_ram, - unsigned long long *crash_size, - unsigned long long *crash_base) -{ - return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, - suffix_tbl[SUFFIX_HIGH]); -} - -int __init parse_crashkernel_low(char *cmdline, - unsigned long long system_ram, - unsigned long long *crash_size, - unsigned long long *crash_base) -{ - return __parse_crashkernel(cmdline, system_ram, crash_size, crash_base, - suffix_tbl[SUFFIX_LOW]); -} - /* * Add a dummy early_param handler to mark crashkernel= as a known command line * parameter and suppress incorrect warnings in init/main.c. -- cgit v1.2.3 From 5097a69d676f7e562240dbe81c21b1c62aaf5950 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Sat, 16 Sep 2023 21:21:31 +0300 Subject: extract and use FILE_LINE macro Extract nifty FILE_LINE useful for printk style debugging: printk("%s\n", FILE_LINE); It should not be used en mass probably because __FILE__ string literals can be merged while FILE_LINE's won't. But for debugging it is what the doctor ordered. Don't add leading and trailing underscores, they're painful to type. Trust me, I've tried both versions. Link: https://lkml.kernel.org/r/ebf12ac4-5a61-4b12-b8b0-1253eb371332@p183 Signed-off-by: Alexey Dobriyan Cc: Kees Cook Cc: Takashi Iwai Cc: Thomas Gleixner Signed-off-by: Andrew Morton --- include/linux/fortify-string.h | 2 +- include/linux/stringify.h | 2 ++ include/linux/timer.h | 3 +-- sound/pci/asihpi/hpidebug.h | 9 ++++----- 4 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index da51a83b2829..442ee9170259 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -643,7 +643,7 @@ __FORTIFY_INLINE bool fortify_memcpy_chk(__kernel_size_t size, __q_size_field, #op), \ #op ": detected field-spanning write (size %zu) of single %s (size %zu)\n", \ __fortify_size, \ - "field \"" #p "\" at " __FILE__ ":" __stringify(__LINE__), \ + "field \"" #p "\" at " FILE_LINE, \ __p_size_field); \ __underlying_##op(p, q, __fortify_size); \ }) diff --git a/include/linux/stringify.h b/include/linux/stringify.h index 841cec8ed525..0e84cbe65270 100644 --- a/include/linux/stringify.h +++ b/include/linux/stringify.h @@ -9,4 +9,6 @@ #define __stringify_1(x...) #x #define __stringify(x...) __stringify_1(x) +#define FILE_LINE __FILE__ ":" __stringify(__LINE__) + #endif /* !__LINUX_STRINGIFY_H */ diff --git a/include/linux/timer.h b/include/linux/timer.h index 9162f275819a..26a545bb0153 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -77,8 +77,7 @@ struct timer_list { .entry = { .next = TIMER_ENTRY_STATIC }, \ .function = (_function), \ .flags = (_flags), \ - __TIMER_LOCKDEP_MAP_INITIALIZER( \ - __FILE__ ":" __stringify(__LINE__)) \ + __TIMER_LOCKDEP_MAP_INITIALIZER(FILE_LINE) \ } #define DEFINE_TIMER(_name, _function) \ diff --git a/sound/pci/asihpi/hpidebug.h b/sound/pci/asihpi/hpidebug.h index c24ed69eb743..c6dfc229213d 100644 --- a/sound/pci/asihpi/hpidebug.h +++ b/sound/pci/asihpi/hpidebug.h @@ -29,16 +29,15 @@ enum { HPI_DEBUG_LEVEL_ERROR = 0, /* always log errors */ the start of each message, eg see linux kernel hpios.h */ #ifdef SOURCEFILE_NAME +#undef FILE_LINE #define FILE_LINE SOURCEFILE_NAME ":" __stringify(__LINE__) " " -#else -#define FILE_LINE __FILE__ ":" __stringify(__LINE__) " " #endif #define HPI_DEBUG_ASSERT(expression) \ do { \ if (!(expression)) { \ printk(KERN_ERR FILE_LINE \ - "ASSERT " __stringify(expression)); \ + " ASSERT " __stringify(expression)); \ } \ } while (0) @@ -46,7 +45,7 @@ enum { HPI_DEBUG_LEVEL_ERROR = 0, /* always log errors */ do { \ if (hpi_debug_level >= HPI_DEBUG_LEVEL_##level) { \ printk(HPI_DEBUG_FLAG_##level \ - FILE_LINE __VA_ARGS__); \ + FILE_LINE " " __VA_ARGS__); \ } \ } while (0) @@ -70,7 +69,7 @@ void hpi_debug_data(u16 *pdata, u32 len); do { \ if (hpi_debug_level >= HPI_DEBUG_LEVEL_##level) { \ hpi_debug_message(phm, HPI_DEBUG_FLAG_##level \ - FILE_LINE __stringify(level)); \ + FILE_LINE " " __stringify(level)); \ } \ } while (0) -- cgit v1.2.3 From 9bf2850c9170b52a6ab052085feced55effa78ef Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sat, 16 Sep 2023 17:50:11 +0200 Subject: kstrtox: remove strtobool() The conversion from strtobool() to kstrtobool() is completed. So strtobool() can now be removed. Link: https://lkml.kernel.org/r/87e3cc2547df174cd5af1fadbf866be4ef9e8e45.1694878151.git.christophe.jaillet@wanadoo.fr Signed-off-by: Christophe JAILLET Reviewed-by: Andy Shevchenko Signed-off-by: Andrew Morton --- include/linux/kstrtox.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kstrtox.h b/include/linux/kstrtox.h index 529974e22ea7..7fcf29a4e0de 100644 --- a/include/linux/kstrtox.h +++ b/include/linux/kstrtox.h @@ -147,9 +147,4 @@ extern long simple_strtol(const char *,char **,unsigned int); extern unsigned long long simple_strtoull(const char *,char **,unsigned int); extern long long simple_strtoll(const char *,char **,unsigned int); -static inline int strtobool(const char *s, bool *res) -{ - return kstrtobool(s, res); -} - #endif /* _LINUX_KSTRTOX_H */ -- cgit v1.2.3 From 80fcac55385ccb710d33a20dc1caaef29bd5a921 Mon Sep 17 00:00:00 2001 From: David Laight Date: Mon, 18 Sep 2023 08:16:30 +0000 Subject: minmax: add umin(a, b) and umax(a, b) Patch series "minmax: Relax type checks in min() and max()", v4. The min() (etc) functions in minmax.h require that the arguments have exactly the same types. However when the type check fails, rather than look at the types and fix the type of a variable/constant, everyone seems to jump on min_t(). In reality min_t() ought to be rare - when something unusual is being done, not normality. The orginal min() (added in 2.4.9) replaced several inline functions and included the type - so matched the implicit casting of the function call. This was renamed min_t() in 2.4.10 and the current min() added. There is no actual indication that the conversion of negatve values to large unsigned values has ever been an actual problem. A quick grep shows 5734 min() and 4597 min_t(). Having the casts on almost half of the calls shows that something is clearly wrong. If the wrong type is picked (and it is far too easy to pick the type of the result instead of the larger input) then significant bits can get discarded. Pretty much the worst example is in the derived clamp_val(), consider: unsigned char x = 200u; y = clamp_val(x, 10u, 300u); I also suspect that many of the min_t(u16, ...) are actually wrong. For example copy_data() in printk_ringbuffer.c contains: data_size = min_t(u16, buf_size, len); Here buf_size is 'unsigned int' and len 'u16', pass a 64k buffer (can you prove that doesn't happen?) and no data is returned. Apparantly it did - and has since been fixed. The only reason that most of the min_t() are 'fine' is that pretty much all the values in the kernel are between 0 and INT_MAX. Patch 1 adds umin(), this uses integer promotions to convert both arguments to 'unsigned long long'. It can be used to compare a signed type that is known to contain a non-negative value with an unsigned type. The compiler typically optimises it all away. Added first so that it can be referred to in patch 2. Patch 2 replaces the 'same type' check with a 'same signedness' one. This makes min(unsigned_int_var, sizeof()) be ok. The error message is also improved and will contain the expanded form of both arguments (useful for seeing how constants are defined). Patch 3 just fixes some whitespace. Patch 4 allows comparisons of 'unsigned char' and 'unsigned short' to signed types. The integer promotion rules convert them both to 'signed int' prior to the comparison so they can never cause a negative value be converted to a large positive one. Patch 5 (rewritted for v4) allows comparisons of unsigned values against non-negative constant integer expressions. This makes min(unsigned_int_var, 4) be ok. The only common case that is still errored is the comparison of signed values against unsigned constant integer expressions below __INT_MAX__. Typcally min(int_val, sizeof (foo)), the real fix for this is casting the constant: min(int_var, (int)sizeof (foo)). With all the patches applied pretty much all the min_t() could be replaced by min(), and most of the rest by umin(). However they all need careful inspection due to code like: sz = min_t(unsigned char, sz - 1, LIM - 1) + 1; which converts 0 to LIM. This patch (of 6): umin() and umax() can be used when min()/max() errors a signed v unsigned compare when the signed value is known to be non-negative. Unlike min_t(some_unsigned_type, a, b) umin() will never mask off high bits if an inappropriate type is selected. The '+ 0u + 0ul + 0ull' may look strange. The '+ 0u' is needed for 'signed int' on 64bit systems. The '+ 0ul' is needed for 'signed long' on 32bit systems. The '+ 0ull' is needed for 'signed long long'. Link: https://lkml.kernel.org/r/b97faef60ad24922b530241c5d7c933c@AcuMS.aculab.com Link: https://lkml.kernel.org/r/41d93ca827a248698ec64bf57e0c05a5@AcuMS.aculab.com Signed-off-by: David Laight Cc: Andy Shevchenko Cc: Christoph Hellwig Cc: Jason A. Donenfeld Cc: Linus Torvalds Cc: Matthew Wilcox (Oracle) Signed-off-by: Andrew Morton --- include/linux/minmax.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'include/linux') diff --git a/include/linux/minmax.h b/include/linux/minmax.h index ca69abd6151e..2a197f54fe05 100644 --- a/include/linux/minmax.h +++ b/include/linux/minmax.h @@ -76,6 +76,23 @@ */ #define max(x, y) __careful_cmp(x, y, >) +/** + * umin - return minimum of two non-negative values + * Signed types are zero extended to match a larger unsigned type. + * @x: first value + * @y: second value + */ +#define umin(x, y) \ + __careful_cmp((x) + 0u + 0ul + 0ull, (y) + 0u + 0ul + 0ull, <) + +/** + * umax - return maximum of two non-negative values + * @x: first value + * @y: second value + */ +#define umax(x, y) \ + __careful_cmp((x) + 0u + 0ul + 0ull, (y) + 0u + 0ul + 0ull, >) + /** * min3 - return minimum of three values * @x: first value -- cgit v1.2.3 From d03eba99f5bf7cbc6e2fdde3b6fa36954ad58e09 Mon Sep 17 00:00:00 2001 From: David Laight Date: Mon, 18 Sep 2023 08:17:15 +0000 Subject: minmax: allow min()/max()/clamp() if the arguments have the same signedness. The type-check in min()/max() is there to stop unexpected results if a negative value gets converted to a large unsigned value. However it also rejects 'unsigned int' v 'unsigned long' compares which are common and never problematc. Replace the 'same type' check with a 'same signedness' check. The new test isn't itself a compile time error, so use static_assert() to report the error and give a meaningful error message. Due to the way builtin_choose_expr() works detecting the error in the 'non-constant' side (where static_assert() can be used) also detects errors when the arguments are constant. Link: https://lkml.kernel.org/r/fe7e6c542e094bfca655abcd323c1c98@AcuMS.aculab.com Signed-off-by: David Laight Cc: Andy Shevchenko Cc: Christoph Hellwig Cc: Jason A. Donenfeld Cc: Linus Torvalds Cc: Matthew Wilcox (Oracle) Signed-off-by: Andrew Morton --- include/linux/minmax.h | 60 +++++++++++++++++++++++++++----------------------- 1 file changed, 32 insertions(+), 28 deletions(-) (limited to 'include/linux') diff --git a/include/linux/minmax.h b/include/linux/minmax.h index 2a197f54fe05..8718fd71a793 100644 --- a/include/linux/minmax.h +++ b/include/linux/minmax.h @@ -12,9 +12,8 @@ * * - avoid multiple evaluations of the arguments (so side-effects like * "x++" happen only once) when non-constant. - * - perform strict type-checking (to generate warnings instead of - * nasty runtime surprises). See the "unnecessary" pointer comparison - * in __typecheck(). + * - perform signed v unsigned type-checking (to generate compile + * errors instead of nasty runtime surprises). * - retain result as a constant expressions when called with only * constant expressions (to avoid tripping VLA warnings in stack * allocation usage). @@ -22,23 +21,30 @@ #define __typecheck(x, y) \ (!!(sizeof((typeof(x) *)1 == (typeof(y) *)1))) -#define __no_side_effects(x, y) \ - (__is_constexpr(x) && __is_constexpr(y)) +/* is_signed_type() isn't a constexpr for pointer types */ +#define __is_signed(x) \ + __builtin_choose_expr(__is_constexpr(is_signed_type(typeof(x))), \ + is_signed_type(typeof(x)), 0) -#define __safe_cmp(x, y) \ - (__typecheck(x, y) && __no_side_effects(x, y)) +#define __types_ok(x, y) \ + (__is_signed(x) == __is_signed(y)) -#define __cmp(x, y, op) ((x) op (y) ? (x) : (y)) +#define __cmp_op_min < +#define __cmp_op_max > -#define __cmp_once(x, y, unique_x, unique_y, op) ({ \ +#define __cmp(op, x, y) ((x) __cmp_op_##op (y) ? (x) : (y)) + +#define __cmp_once(op, x, y, unique_x, unique_y) ({ \ typeof(x) unique_x = (x); \ typeof(y) unique_y = (y); \ - __cmp(unique_x, unique_y, op); }) + static_assert(__types_ok(x, y), \ + #op "(" #x ", " #y ") signedness error, fix types or consider u" #op "() before " #op "_t()"); \ + __cmp(op, unique_x, unique_y); }) -#define __careful_cmp(x, y, op) \ - __builtin_choose_expr(__safe_cmp(x, y), \ - __cmp(x, y, op), \ - __cmp_once(x, y, __UNIQUE_ID(__x), __UNIQUE_ID(__y), op)) +#define __careful_cmp(op, x, y) \ + __builtin_choose_expr(__is_constexpr((x) - (y)), \ + __cmp(op, x, y), \ + __cmp_once(op, x, y, __UNIQUE_ID(__x), __UNIQUE_ID(__y))) #define __clamp(val, lo, hi) \ ((val) >= (hi) ? (hi) : ((val) <= (lo) ? (lo) : (val))) @@ -47,17 +53,15 @@ typeof(val) unique_val = (val); \ typeof(lo) unique_lo = (lo); \ typeof(hi) unique_hi = (hi); \ + static_assert(__builtin_choose_expr(__is_constexpr((lo) > (hi)), \ + (lo) <= (hi), true), \ + "clamp() low limit " #lo " greater than high limit " #hi); \ + static_assert(__types_ok(val, lo), "clamp() 'lo' signedness error"); \ + static_assert(__types_ok(val, hi), "clamp() 'hi' signedness error"); \ __clamp(unique_val, unique_lo, unique_hi); }) -#define __clamp_input_check(lo, hi) \ - (BUILD_BUG_ON_ZERO(__builtin_choose_expr( \ - __is_constexpr((lo) > (hi)), (lo) > (hi), false))) - #define __careful_clamp(val, lo, hi) ({ \ - __clamp_input_check(lo, hi) + \ - __builtin_choose_expr(__typecheck(val, lo) && __typecheck(val, hi) && \ - __typecheck(hi, lo) && __is_constexpr(val) && \ - __is_constexpr(lo) && __is_constexpr(hi), \ + __builtin_choose_expr(__is_constexpr((val) - (lo) + (hi)), \ __clamp(val, lo, hi), \ __clamp_once(val, lo, hi, __UNIQUE_ID(__val), \ __UNIQUE_ID(__lo), __UNIQUE_ID(__hi))); }) @@ -67,14 +71,14 @@ * @x: first value * @y: second value */ -#define min(x, y) __careful_cmp(x, y, <) +#define min(x, y) __careful_cmp(min, x, y) /** * max - return maximum of two values of the same or compatible types * @x: first value * @y: second value */ -#define max(x, y) __careful_cmp(x, y, >) +#define max(x, y) __careful_cmp(max, x, y) /** * umin - return minimum of two non-negative values @@ -83,7 +87,7 @@ * @y: second value */ #define umin(x, y) \ - __careful_cmp((x) + 0u + 0ul + 0ull, (y) + 0u + 0ul + 0ull, <) + __careful_cmp(min, (x) + 0u + 0ul + 0ull, (y) + 0u + 0ul + 0ull) /** * umax - return maximum of two non-negative values @@ -91,7 +95,7 @@ * @y: second value */ #define umax(x, y) \ - __careful_cmp((x) + 0u + 0ul + 0ull, (y) + 0u + 0ul + 0ull, >) + __careful_cmp(max, (x) + 0u + 0ul + 0ull, (y) + 0u + 0ul + 0ull) /** * min3 - return minimum of three values @@ -143,7 +147,7 @@ * @x: first value * @y: second value */ -#define min_t(type, x, y) __careful_cmp((type)(x), (type)(y), <) +#define min_t(type, x, y) __careful_cmp(min, (type)(x), (type)(y)) /** * max_t - return maximum of two values, using the specified type @@ -151,7 +155,7 @@ * @x: first value * @y: second value */ -#define max_t(type, x, y) __careful_cmp((type)(x), (type)(y), >) +#define max_t(type, x, y) __careful_cmp(max, (type)(x), (type)(y)) /* * Do not check the array parameter using __must_be_array(). -- cgit v1.2.3 From f4b84b2ff851f01d0fac619eadef47eb41648534 Mon Sep 17 00:00:00 2001 From: David Laight Date: Mon, 18 Sep 2023 08:17:57 +0000 Subject: minmax: fix indentation of __cmp_once() and __clamp_once() Remove the extra indentation and align continuation markers. Link: https://lkml.kernel.org/r/bed41317a05c498ea0209eafbcab45a5@AcuMS.aculab.com Signed-off-by: David Laight Cc: Andy Shevchenko Cc: Christoph Hellwig Cc: Jason A. Donenfeld Cc: Linus Torvalds Cc: Matthew Wilcox (Oracle) Signed-off-by: Andrew Morton --- include/linux/minmax.h | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) (limited to 'include/linux') diff --git a/include/linux/minmax.h b/include/linux/minmax.h index 8718fd71a793..c0e738eacefa 100644 --- a/include/linux/minmax.h +++ b/include/linux/minmax.h @@ -35,11 +35,11 @@ #define __cmp(op, x, y) ((x) __cmp_op_##op (y) ? (x) : (y)) #define __cmp_once(op, x, y, unique_x, unique_y) ({ \ - typeof(x) unique_x = (x); \ - typeof(y) unique_y = (y); \ - static_assert(__types_ok(x, y), \ - #op "(" #x ", " #y ") signedness error, fix types or consider u" #op "() before " #op "_t()"); \ - __cmp(op, unique_x, unique_y); }) + typeof(x) unique_x = (x); \ + typeof(y) unique_y = (y); \ + static_assert(__types_ok(x, y), \ + #op "(" #x ", " #y ") signedness error, fix types or consider u" #op "() before " #op "_t()"); \ + __cmp(op, unique_x, unique_y); }) #define __careful_cmp(op, x, y) \ __builtin_choose_expr(__is_constexpr((x) - (y)), \ @@ -49,16 +49,16 @@ #define __clamp(val, lo, hi) \ ((val) >= (hi) ? (hi) : ((val) <= (lo) ? (lo) : (val))) -#define __clamp_once(val, lo, hi, unique_val, unique_lo, unique_hi) ({ \ - typeof(val) unique_val = (val); \ - typeof(lo) unique_lo = (lo); \ - typeof(hi) unique_hi = (hi); \ - static_assert(__builtin_choose_expr(__is_constexpr((lo) > (hi)), \ - (lo) <= (hi), true), \ - "clamp() low limit " #lo " greater than high limit " #hi); \ - static_assert(__types_ok(val, lo), "clamp() 'lo' signedness error"); \ - static_assert(__types_ok(val, hi), "clamp() 'hi' signedness error"); \ - __clamp(unique_val, unique_lo, unique_hi); }) +#define __clamp_once(val, lo, hi, unique_val, unique_lo, unique_hi) ({ \ + typeof(val) unique_val = (val); \ + typeof(lo) unique_lo = (lo); \ + typeof(hi) unique_hi = (hi); \ + static_assert(__builtin_choose_expr(__is_constexpr((lo) > (hi)), \ + (lo) <= (hi), true), \ + "clamp() low limit " #lo " greater than high limit " #hi); \ + static_assert(__types_ok(val, lo), "clamp() 'lo' signedness error"); \ + static_assert(__types_ok(val, hi), "clamp() 'hi' signedness error"); \ + __clamp(unique_val, unique_lo, unique_hi); }) #define __careful_clamp(val, lo, hi) ({ \ __builtin_choose_expr(__is_constexpr((val) - (lo) + (hi)), \ -- cgit v1.2.3 From 4ead534fba42fc4fd41163297528d2aa731cd121 Mon Sep 17 00:00:00 2001 From: David Laight Date: Mon, 18 Sep 2023 08:18:40 +0000 Subject: minmax: allow comparisons of 'int' against 'unsigned char/short' Since 'unsigned char/short' get promoted to 'signed int' it is safe to compare them against an 'int' value. Link: https://lkml.kernel.org/r/8732ef5f809c47c28a7be47c938b28d4@AcuMS.aculab.com Signed-off-by: David Laight Cc: Andy Shevchenko Cc: Christoph Hellwig Cc: Jason A. Donenfeld Cc: Linus Torvalds Cc: Matthew Wilcox (Oracle) Signed-off-by: Andrew Morton --- include/linux/minmax.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/minmax.h b/include/linux/minmax.h index c0e738eacefa..842c1db62ffe 100644 --- a/include/linux/minmax.h +++ b/include/linux/minmax.h @@ -26,8 +26,9 @@ __builtin_choose_expr(__is_constexpr(is_signed_type(typeof(x))), \ is_signed_type(typeof(x)), 0) -#define __types_ok(x, y) \ - (__is_signed(x) == __is_signed(y)) +#define __types_ok(x, y) \ + (__is_signed(x) == __is_signed(y) || \ + __is_signed((x) + 0) == __is_signed((y) + 0)) #define __cmp_op_min < #define __cmp_op_max > -- cgit v1.2.3 From 867046cc7027703f60a46339ffde91a1970f2901 Mon Sep 17 00:00:00 2001 From: David Laight Date: Mon, 18 Sep 2023 08:19:25 +0000 Subject: minmax: relax check to allow comparison between unsigned arguments and signed constants Allow (for example) min(unsigned_var, 20). The opposite min(signed_var, 20u) is still errored. Since a comparison between signed and unsigned never makes the unsigned value negative it is only necessary to adjust the __types_ok() test. Link: https://lkml.kernel.org/r/633b64e2f39e46bb8234809c5595b8c7@AcuMS.aculab.com Signed-off-by: David Laight Cc: Andy Shevchenko Cc: Christoph Hellwig Cc: Jason A. Donenfeld Cc: Linus Torvalds Cc: Matthew Wilcox (Oracle) Signed-off-by: Andrew Morton --- include/linux/minmax.h | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/minmax.h b/include/linux/minmax.h index 842c1db62ffe..2ec559284a9f 100644 --- a/include/linux/minmax.h +++ b/include/linux/minmax.h @@ -10,13 +10,18 @@ /* * min()/max()/clamp() macros must accomplish three things: * - * - avoid multiple evaluations of the arguments (so side-effects like + * - Avoid multiple evaluations of the arguments (so side-effects like * "x++" happen only once) when non-constant. - * - perform signed v unsigned type-checking (to generate compile - * errors instead of nasty runtime surprises). - * - retain result as a constant expressions when called with only + * - Retain result as a constant expressions when called with only * constant expressions (to avoid tripping VLA warnings in stack * allocation usage). + * - Perform signed v unsigned type-checking (to generate compile + * errors instead of nasty runtime surprises). + * - Unsigned char/short are always promoted to signed int and can be + * compared against signed or unsigned arguments. + * - Unsigned arguments can be compared against non-negative signed constants. + * - Comparison of a signed argument against an unsigned constant fails + * even if the constant is below __INT_MAX__ and could be cast to int. */ #define __typecheck(x, y) \ (!!(sizeof((typeof(x) *)1 == (typeof(y) *)1))) @@ -26,9 +31,14 @@ __builtin_choose_expr(__is_constexpr(is_signed_type(typeof(x))), \ is_signed_type(typeof(x)), 0) -#define __types_ok(x, y) \ - (__is_signed(x) == __is_signed(y) || \ - __is_signed((x) + 0) == __is_signed((y) + 0)) +/* True for a non-negative signed int constant */ +#define __is_noneg_int(x) \ + (__builtin_choose_expr(__is_constexpr(x) && __is_signed(x), x, -1) >= 0) + +#define __types_ok(x, y) \ + (__is_signed(x) == __is_signed(y) || \ + __is_signed((x) + 0) == __is_signed((y) + 0) || \ + __is_noneg_int(x) || __is_noneg_int(y)) #define __cmp_op_min < #define __cmp_op_max > -- cgit v1.2.3 From 598f0ac1500d7145c696de4d38797b1dd2c651de Mon Sep 17 00:00:00 2001 From: David Laight Date: Thu, 5 Oct 2023 11:39:54 +0000 Subject: compiler.h: move __is_constexpr() to compiler.h Prior to f747e6667ebb2 __is_constexpr() was in its only user minmax.h. That commit moved it to const.h - but that file just defines ULL(x) and UL(x) so that constants can be defined for .S and .c files. So apart from the word 'const' it wasn't really a good location. Instead move the definition to compiler.h just before the similar is_signed_type() and is_unsigned_type(). This may not be a good long-term home, but the three definitions belong together. Link: https://lkml.kernel.org/r/2a6680bbe2e84459816a113730426782@AcuMS.aculab.com Signed-off-by: David Laight Reviewed-by: Kees Cook Cc: Arnd Bergmann Cc: Bart Van Assche Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Rasmus Villemoes Cc: Steven Rostedt (Google) Signed-off-by: Andrew Morton --- include/linux/compiler.h | 8 ++++++++ include/linux/const.h | 8 -------- tools/include/linux/compiler.h | 8 ++++++++ tools/include/linux/const.h | 8 -------- 4 files changed, 16 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 174099fdc485..bb1339c7057b 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -227,6 +227,14 @@ static inline void *offset_to_ptr(const int *off) /* &a[0] degrades to a pointer: a different type from an array */ #define __must_be_array(a) BUILD_BUG_ON_ZERO(__same_type((a), &(a)[0])) +/* + * This returns a constant expression while determining if an argument is + * a constant expression, most importantly without evaluating the argument. + * Glory to Martin Uecker + */ +#define __is_constexpr(x) \ + (sizeof(int) == sizeof(*(8 ? ((void *)((long)(x) * 0l)) : (int *)8))) + /* * Whether 'type' is a signed type or an unsigned type. Supports scalar types, * bool and also pointer types. diff --git a/include/linux/const.h b/include/linux/const.h index 435ddd72d2c4..81b8aae5a855 100644 --- a/include/linux/const.h +++ b/include/linux/const.h @@ -3,12 +3,4 @@ #include -/* - * This returns a constant expression while determining if an argument is - * a constant expression, most importantly without evaluating the argument. - * Glory to Martin Uecker - */ -#define __is_constexpr(x) \ - (sizeof(int) == sizeof(*(8 ? ((void *)((long)(x) * 0l)) : (int *)8))) - #endif /* _LINUX_CONST_H */ diff --git a/tools/include/linux/compiler.h b/tools/include/linux/compiler.h index 1684216e826a..7b65566f3e42 100644 --- a/tools/include/linux/compiler.h +++ b/tools/include/linux/compiler.h @@ -63,6 +63,14 @@ # define __same_type(a, b) __builtin_types_compatible_p(typeof(a), typeof(b)) #endif +/* + * This returns a constant expression while determining if an argument is + * a constant expression, most importantly without evaluating the argument. + * Glory to Martin Uecker + */ +#define __is_constexpr(x) \ + (sizeof(int) == sizeof(*(8 ? ((void *)((long)(x) * 0l)) : (int *)8))) + #ifdef __ANDROID__ /* * FIXME: Big hammer to get rid of tons of: diff --git a/tools/include/linux/const.h b/tools/include/linux/const.h index 435ddd72d2c4..81b8aae5a855 100644 --- a/tools/include/linux/const.h +++ b/tools/include/linux/const.h @@ -3,12 +3,4 @@ #include -/* - * This returns a constant expression while determining if an argument is - * a constant expression, most importantly without evaluating the argument. - * Glory to Martin Uecker - */ -#define __is_constexpr(x) \ - (sizeof(int) == sizeof(*(8 ? ((void *)((long)(x) * 0l)) : (int *)8))) - #endif /* _LINUX_CONST_H */ -- cgit v1.2.3