June 2025 - Linux-stable-mirror

[PATCH RESEND 1/3] HID: wacom: fix memory leak on kobject creation failure

by Qasim Ijaz

During wacom_initialize_remotes() a fifo buffer is allocated with kfifo_alloc() and later a cleanup action is registered during devm_add_action_or_reset() to clean it up. However if the code fails to create a kobject and register it with sysfs the code simply returns -ENOMEM before the cleanup action is registered leading to a memory leak. Fix this by ensuring the fifo is freed when the kobject creation and registration process fails. Fixes: 83e6b40e2de6 ("HID: wacom: EKR: have the wacom resources dynamically allocated") Reviewed-by: Ping Cheng <ping.cheng(a)wacom.com> Cc: stable(a)vger.kernel.org Signed-off-by: Qasim Ijaz <qasdev00(a)gmail.com> --- drivers/hid/wacom_sys.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index eaf099b2efdb..ec5282bc69d6 100644 --- a/drivers/hid/wacom_sys.c +++ b/drivers/hid/wacom_sys.c @@ -2048,8 +2048,10 @@ static int wacom_initialize_remotes(struct wacom *wacom) remote->remote_dir = kobject_create_and_add("wacom_remote", &wacom->hdev->dev.kobj); - if (!remote->remote_dir) + if (!remote->remote_dir) { + kfifo_free(&remote->remote_fifo); return -ENOMEM; + } error = sysfs_create_files(remote->remote_dir, remote_unpair_attrs); -- 2.39.5

2 weeks, 5 days

1
0
0 0

[PATCH 6.15.y] pinctrl: mediatek: eint: Fix invalid pointer dereference for v1 platforms

by Uwe Kleine-König

From: Nícolas F. R. A. Prado <nfraprado(a)collabora.com> commit 1c9977b263475373b31bbf86af94a5c9ae2be42c upstream. Commit 3ef9f710efcb ("pinctrl: mediatek: Add EINT support for multiple addresses") introduced an access to the 'soc' field of struct mtk_pinctrl in mtk_eint_do_init() and for that an include of pinctrl-mtk-common-v2.h. However, pinctrl drivers relying on the v1 common driver include pinctrl-mtk-common.h instead, which provides another definition of struct mtk_pinctrl that does not contain an 'soc' field. Since mtk_eint_do_init() can be called both by v1 and v2 drivers, it will now try to dereference an invalid pointer when called on v1 platforms. This has been observed on Genio 350 EVK (MT8365), which crashes very early in boot (the kernel trace can only be seen with earlycon). In order to fix this, since 'struct mtk_pinctrl' was only needed to get a 'struct mtk_eint_pin', make 'struct mtk_eint_pin' a parameter of mtk_eint_do_init() so that callers need to supply it, removing mtk_eint_do_init()'s dependency on any particular 'struct mtk_pinctrl'. Fixes: 3ef9f710efcb ("pinctrl: mediatek: Add EINT support for multiple addresses") Suggested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno(a)collabora.com> Signed-off-by: Nícolas F. R. A. Prado <nfraprado(a)collabora.com> Link: https://lore.kernel.org/20250520-genio-350-eint-null-ptr-deref-fix-v2-1-6a3… Signed-off-by: Linus Walleij <linus.walleij(a)linaro.org> [ukleinek: backport to 6.15.y] Signed-off-by: Uwe Kleine-König <u.kleine-koenig(a)baylibre.com> --- Hello, would be great to have this in 6.15. Further backporting isn't needed as 3ef9f710efcb == v6.15-rc1~106^2 isn't in 6.14. This patch fixes booting on mt8365-evk (and probably a few more machines based on mediatek SoCs. There was an easy conflict with 86dee87f4b2e6ac119b03810e58723d0b27787a4 in drivers/pinctrl/mediatek/mtk-eint.c. Thanks Uwe drivers/pinctrl/mediatek/mtk-eint.c | 26 ++++++++----------- drivers/pinctrl/mediatek/mtk-eint.h | 5 ++-- .../pinctrl/mediatek/pinctrl-mtk-common-v2.c | 2 +- drivers/pinctrl/mediatek/pinctrl-mtk-common.c | 2 +- 4 files changed, 16 insertions(+), 19 deletions(-) diff --git a/drivers/pinctrl/mediatek/mtk-eint.c b/drivers/pinctrl/mediatek/mtk-eint.c index b4eb2beab691..c516c34aaaf6 100644 --- a/drivers/pinctrl/mediatek/mtk-eint.c +++ b/drivers/pinctrl/mediatek/mtk-eint.c @@ -22,7 +22,6 @@ #include <linux/platform_device.h> #include "mtk-eint.h" -#include "pinctrl-mtk-common-v2.h" #define MTK_EINT_EDGE_SENSITIVE 0 #define MTK_EINT_LEVEL_SENSITIVE 1 @@ -505,10 +504,9 @@ int mtk_eint_find_irq(struct mtk_eint *eint, unsigned long eint_n) } EXPORT_SYMBOL_GPL(mtk_eint_find_irq); -int mtk_eint_do_init(struct mtk_eint *eint) +int mtk_eint_do_init(struct mtk_eint *eint, struct mtk_eint_pin *eint_pin) { unsigned int size, i, port, inst = 0; - struct mtk_pinctrl *hw = (struct mtk_pinctrl *)eint->pctl; /* If clients don't assign a specific regs, let's use generic one */ if (!eint->regs) @@ -519,7 +517,15 @@ int mtk_eint_do_init(struct mtk_eint *eint) if (!eint->base_pin_num) return -ENOMEM; - if (eint->nbase == 1) { + if (eint_pin) { + eint->pins = eint_pin; + for (i = 0; i < eint->hw->ap_num; i++) { + inst = eint->pins[i].instance; + if (inst >= eint->nbase) + continue; + eint->base_pin_num[inst]++; + } + } else { size = eint->hw->ap_num * sizeof(struct mtk_eint_pin); eint->pins = devm_kmalloc(eint->dev, size, GFP_KERNEL); if (!eint->pins) @@ -533,16 +539,6 @@ int mtk_eint_do_init(struct mtk_eint *eint) } } - if (hw && hw->soc && hw->soc->eint_pin) { - eint->pins = hw->soc->eint_pin; - for (i = 0; i < eint->hw->ap_num; i++) { - inst = eint->pins[i].instance; - if (inst >= eint->nbase) - continue; - eint->base_pin_num[inst]++; - } - } - eint->pin_list = devm_kmalloc(eint->dev, eint->nbase * sizeof(u16 *), GFP_KERNEL); if (!eint->pin_list) goto err_pin_list; @@ -610,7 +606,7 @@ int mtk_eint_do_init(struct mtk_eint *eint) err_wake_mask: devm_kfree(eint->dev, eint->pin_list); err_pin_list: - if (eint->nbase == 1) + if (!eint_pin) devm_kfree(eint->dev, eint->pins); err_pins: devm_kfree(eint->dev, eint->base_pin_num); diff --git a/drivers/pinctrl/mediatek/mtk-eint.h b/drivers/pinctrl/mediatek/mtk-eint.h index f7f58cca0d5e..23801d4b636f 100644 --- a/drivers/pinctrl/mediatek/mtk-eint.h +++ b/drivers/pinctrl/mediatek/mtk-eint.h @@ -88,7 +88,7 @@ struct mtk_eint { }; #if IS_ENABLED(CONFIG_EINT_MTK) -int mtk_eint_do_init(struct mtk_eint *eint); +int mtk_eint_do_init(struct mtk_eint *eint, struct mtk_eint_pin *eint_pin); int mtk_eint_do_suspend(struct mtk_eint *eint); int mtk_eint_do_resume(struct mtk_eint *eint); int mtk_eint_set_debounce(struct mtk_eint *eint, unsigned long eint_n, @@ -96,7 +96,8 @@ int mtk_eint_set_debounce(struct mtk_eint *eint, unsigned long eint_n, int mtk_eint_find_irq(struct mtk_eint *eint, unsigned long eint_n); #else -static inline int mtk_eint_do_init(struct mtk_eint *eint) +static inline int mtk_eint_do_init(struct mtk_eint *eint, + struct mtk_eint_pin *eint_pin) { return -EOPNOTSUPP; } diff --git a/drivers/pinctrl/mediatek/pinctrl-mtk-common-v2.c b/drivers/pinctrl/mediatek/pinctrl-mtk-common-v2.c index d1556b75d9ef..ba13558bfcd7 100644 --- a/drivers/pinctrl/mediatek/pinctrl-mtk-common-v2.c +++ b/drivers/pinctrl/mediatek/pinctrl-mtk-common-v2.c @@ -416,7 +416,7 @@ int mtk_build_eint(struct mtk_pinctrl *hw, struct platform_device *pdev) hw->eint->pctl = hw; hw->eint->gpio_xlate = &mtk_eint_xt; - ret = mtk_eint_do_init(hw->eint); + ret = mtk_eint_do_init(hw->eint, hw->soc->eint_pin); if (ret) goto err_free_eint; diff --git a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c index 8596f3541265..7289648eaa02 100644 --- a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c +++ b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c @@ -1039,7 +1039,7 @@ static int mtk_eint_init(struct mtk_pinctrl *pctl, struct platform_device *pdev) pctl->eint->pctl = pctl; pctl->eint->gpio_xlate = &mtk_eint_xt; - return mtk_eint_do_init(pctl->eint); + return mtk_eint_do_init(pctl->eint, NULL); } /* This is used as a common probe function */ -- 2.47.2

2 weeks, 5 days

1
0
0 0

[for-linus][PATCH 2/3] ring-buffer: Fix buffer locking in ring_buffer_subbuf_order_set()

by Steven Rostedt

From: Dmitry Antipov <dmantipov(a)yandex.ru> Enlarge the critical section in ring_buffer_subbuf_order_set() to ensure that error handling takes place with per-buffer mutex held, thus preventing list corruption and other concurrency-related issues. Cc: stable(a)vger.kernel.org Cc: Masami Hiramatsu <mhiramat(a)kernel.org> Cc: Mathieu Desnoyers <mathieu.desnoyers(a)efficios.com> Cc: Tzvetomir Stoyanov <tz.stoyanov(a)gmail.com> Link: https://lore.kernel.org/20250606112242.1510605-1-dmantipov@yandex.ru Reported-by: syzbot+05d673e83ec640f0ced9(a)syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=05d673e83ec640f0ced9 Fixes: f9b94daa542a8 ("ring-buffer: Set new size of the ring buffer sub page") Signed-off-by: Dmitry Antipov <dmantipov(a)yandex.ru> Signed-off-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> --- kernel/trace/ring_buffer.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index e24509bd0af5..00fc38d70e86 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -6795,7 +6795,7 @@ int ring_buffer_subbuf_order_set(struct trace_buffer *buffer, int order) old_size = buffer->subbuf_size; /* prevent another thread from changing buffer sizes */ - mutex_lock(&buffer->mutex); + guard(mutex)(&buffer->mutex); atomic_inc(&buffer->record_disabled); /* Make sure all commits have finished */ @@ -6900,7 +6900,6 @@ int ring_buffer_subbuf_order_set(struct trace_buffer *buffer, int order) } atomic_dec(&buffer->record_disabled); - mutex_unlock(&buffer->mutex); return 0; @@ -6909,7 +6908,6 @@ int ring_buffer_subbuf_order_set(struct trace_buffer *buffer, int order) buffer->subbuf_size = old_size; atomic_dec(&buffer->record_disabled); - mutex_unlock(&buffer->mutex); for_each_buffer_cpu(buffer, cpu) { cpu_buffer = buffer->buffers[cpu]; -- 2.47.2

2 weeks, 5 days

1
0
0 0

[for-linus][PATCH 1/3] tracing: Fix regression of filter waiting a long time on RCU synchronization

by Steven Rostedt

From: Steven Rostedt <rostedt(a)goodmis.org> When faultable trace events were added, a trace event may no longer use normal RCU to synchronize but instead used synchronize_rcu_tasks_trace(). This synchronization takes a much longer time to synchronize. The filter logic would free the filters by calling tracepoint_synchronize_unregister() after it unhooked the filter strings and before freeing them. With this function now calling synchronize_rcu_tasks_trace() this increased the time to free a filter tremendously. On a PREEMPT_RT system, it was even more noticeable. # time trace-cmd record -p function sleep 1 [..] real 2m29.052s user 0m0.244s sys 0m20.136s As trace-cmd would clear out all the filters before recording, it could take up to 2 minutes to do a recording of "sleep 1". To find out where the issues was: ~# trace-cmd sqlhist -e -n sched_stack select start.prev_state as state, end.next_comm as comm, TIMESTAMP_DELTA_USECS as delta, start.STACKTRACE as stack from sched_switch as start join sched_switch as end on start.prev_pid = end.next_pid Which will produce the following commands (and -e will also execute them): echo 's:sched_stack s64 state; char comm[16]; u64 delta; unsigned long stack[];' >> /sys/kernel/tracing/dynamic_events echo 'hist:keys=prev_pid:__arg_18057_2=prev_state,__arg_18057_4=common_timestamp.usecs,__arg_18057_7=common_stacktrace' >> /sys/kernel/tracing/events/sched/sched_switch/trigger echo 'hist:keys=next_pid:__state_18057_1=$__arg_18057_2,__comm_18057_3=next_comm,__delta_18057_5=common_timestamp.usecs-$__arg_18057_4,__stack_18057_6=$__arg_18057_7:onmatch(sched.sched_switch).trace(sched_stack,$__state_18057_1,$__comm_18057_3,$__delta_18057_5,$__stack_18057_6)' >> /sys/kernel/tracing/events/sched/sched_switch/trigger The above creates a synthetic event that creates a stack trace when a task schedules out and records it with the time it scheduled back in. Basically the time a task is off the CPU. It also records the state of the task when it left the CPU (running, blocked, sleeping, etc). It also saves the comm of the task as "comm" (needed for the next command). ~# echo 'hist:keys=state,stack.stacktrace:vals=delta:sort=state,delta if comm == "trace-cmd" && state & 3' > /sys/kernel/tracing/events/synthetic/sched_stack/trigger The above creates a histogram with buckets per state, per stack, and the value of the total time it was off the CPU for that stack trace. It filters on tasks with "comm == trace-cmd" and only the sleeping and blocked states (1 - sleeping, 2 - blocked). ~# trace-cmd record -p function sleep 1 ~# cat /sys/kernel/tracing/events/synthetic/sched_stack/hist | tail -18 { state: 2, stack.stacktrace __schedule+0x1545/0x3700 schedule+0xe2/0x390 schedule_timeout+0x175/0x200 wait_for_completion_state+0x294/0x440 __wait_rcu_gp+0x247/0x4f0 synchronize_rcu_tasks_generic+0x151/0x230 apply_subsystem_event_filter+0xa2b/0x1300 subsystem_filter_write+0x67/0xc0 vfs_write+0x1e2/0xeb0 ksys_write+0xff/0x1d0 do_syscall_64+0x7b/0x420 entry_SYSCALL_64_after_hwframe+0x76/0x7e } hitcount: 237 delta: 99756288 <<--------------- Delta is 99 seconds! Totals: Hits: 525 Entries: 21 Dropped: 0 This shows that this particular trace waited for 99 seconds on synchronize_rcu_tasks() in apply_subsystem_event_filter(). In fact, there's a lot of places in the filter code that spends a lot of time waiting for synchronize_rcu_tasks_trace() in order to free the filters. Add helper functions that will use call_rcu*() variants to asynchronously free the filters. This brings the timings back to normal: # time trace-cmd record -p function sleep 1 [..] real 0m14.681s user 0m0.335s sys 0m28.616s And the histogram also shows this: ~# cat /sys/kernel/tracing/events/synthetic/sched_stack/hist | tail -21 { state: 2, stack.stacktrace __schedule+0x1545/0x3700 schedule+0xe2/0x390 schedule_timeout+0x175/0x200 wait_for_completion_state+0x294/0x440 __wait_rcu_gp+0x247/0x4f0 synchronize_rcu_normal+0x3db/0x5c0 tracing_reset_online_cpus+0x8f/0x1e0 tracing_open+0x335/0x440 do_dentry_open+0x4c6/0x17a0 vfs_open+0x82/0x360 path_openat+0x1a36/0x2990 do_filp_open+0x1c5/0x420 do_sys_openat2+0xed/0x180 __x64_sys_openat+0x108/0x1d0 do_syscall_64+0x7b/0x420 } hitcount: 2 delta: 77044 Totals: Hits: 55 Entries: 28 Dropped: 0 Where the total waiting time of synchronize_rcu_tasks_trace() is 77 milliseconds. Cc: stable(a)vger.kernel.org Cc: Masami Hiramatsu <mhiramat(a)kernel.org> Cc: Mathieu Desnoyers <mathieu.desnoyers(a)efficios.com> Cc: "Paul E. McKenney" <paulmck(a)kernel.org> Cc: "Kiszka, Jan" <jan.kiszka(a)siemens.com> Cc: "Ziegler, Andreas" <ziegler.andreas(a)siemens.com> Cc: "MOESSBAUER, Felix" <felix.moessbauer(a)siemens.com> Link: https://lore.kernel.org/20250605161701.35f7989a@gandalf.local.home Reported-by: "Flot, Julien" <julien.flot(a)siemens.com> Tested-by: Julien Flot <julien.flot(a)siemens.com> Fixes: a363d27cdbc2 ("tracing: Allow system call tracepoints to handle page faults") Closes: https://lore.kernel.org/all/240017f656631c7dd4017aa93d91f41f653788ea.camel@… Signed-off-by: Steven Rostedt (Google) <rostedt(a)goodmis.org> --- kernel/trace/trace_events_filter.c | 164 ++++++++++++++++++++++------- 1 file changed, 127 insertions(+), 37 deletions(-) diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 2048560264bb..3ff782d6b522 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1335,6 +1335,74 @@ static void filter_free_subsystem_preds(struct trace_subsystem_dir *dir, } } +struct filter_list { + struct list_head list; + struct event_filter *filter; +}; + +struct filter_head { + struct list_head list; + struct rcu_head rcu; +}; + + +static void free_filter_list(struct rcu_head *rhp) +{ + struct filter_head *filter_list = container_of(rhp, struct filter_head, rcu); + struct filter_list *filter_item, *tmp; + + list_for_each_entry_safe(filter_item, tmp, &filter_list->list, list) { + __free_filter(filter_item->filter); + list_del(&filter_item->list); + kfree(filter_item); + } + kfree(filter_list); +} + +static void free_filter_list_tasks(struct rcu_head *rhp) +{ + call_rcu(rhp, free_filter_list); +} + +/* + * The tracepoint_synchronize_unregister() is a double rcu call. + * It calls synchronize_rcu_tasks_trace() followed by synchronize_rcu(). + * Instead of waiting for it, simply call these via the call_rcu*() + * variants. + */ +static void delay_free_filter(struct filter_head *head) +{ + call_rcu_tasks_trace(&head->rcu, free_filter_list_tasks); +} + +static void try_delay_free_filter(struct event_filter *filter) +{ + struct filter_head *head; + struct filter_list *item; + + head = kmalloc(sizeof(*head), GFP_KERNEL); + if (!head) + goto free_now; + + INIT_LIST_HEAD(&head->list); + + item = kmalloc(sizeof(*item), GFP_KERNEL); + if (!item) { + kfree(head); + goto free_now; + } + + item->filter = filter; + list_add_tail(&item->list, &head->list); + delay_free_filter(head); + return; + + free_now: + /* Make sure the filter is not being used */ + tracepoint_synchronize_unregister(); + __free_filter(filter); +} + static inline void __free_subsystem_filter(struct trace_event_file *file) { __free_filter(file->filter); @@ -1342,15 +1410,53 @@ static inline void __free_subsystem_filter(struct trace_event_file *file) } static void filter_free_subsystem_filters(struct trace_subsystem_dir *dir, - struct trace_array *tr) + struct trace_array *tr, + struct event_filter *filter) { struct trace_event_file *file; + struct filter_head *head; + struct filter_list *item; + + head = kmalloc(sizeof(*head), GFP_KERNEL); + if (!head) + goto free_now; + + INIT_LIST_HEAD(&head->list); + + item = kmalloc(sizeof(*item), GFP_KERNEL); + if (!item) { + kfree(head); + goto free_now; + } + + item->filter = filter; + list_add_tail(&item->list, &head->list); list_for_each_entry(file, &tr->events, list) { if (file->system != dir) continue; - __free_subsystem_filter(file); + item = kmalloc(sizeof(*item), GFP_KERNEL); + if (!item) + goto free_now; + item->filter = file->filter; + list_add_tail(&item->list, &head->list); + file->filter = NULL; + } + + delay_free_filter(head); + return; + free_now: + tracepoint_synchronize_unregister(); + + if (head) + free_filter_list(&head->rcu); + + list_for_each_entry(file, &tr->events, list) { + if (file->system != dir || !file->filter) + continue; + __free_filter(file->filter); } + __free_filter(filter); } int filter_assign_type(const char *type) @@ -2131,11 +2237,6 @@ static inline void event_clear_filter(struct trace_event_file *file) RCU_INIT_POINTER(file->filter, NULL); } -struct filter_list { - struct list_head list; - struct event_filter *filter; -}; - static int process_system_preds(struct trace_subsystem_dir *dir, struct trace_array *tr, struct filter_parse_error *pe, @@ -2144,11 +2245,16 @@ static int process_system_preds(struct trace_subsystem_dir *dir, struct trace_event_file *file; struct filter_list *filter_item; struct event_filter *filter = NULL; - struct filter_list *tmp; - LIST_HEAD(filter_list); + struct filter_head *filter_list; bool fail = true; int err; + filter_list = kmalloc(sizeof(*filter_list), GFP_KERNEL); + if (!filter_list) + return -ENOMEM; + + INIT_LIST_HEAD(&filter_list->list); + list_for_each_entry(file, &tr->events, list) { if (file->system != dir) @@ -2175,7 +2281,7 @@ static int process_system_preds(struct trace_subsystem_dir *dir, if (!filter_item) goto fail_mem; - list_add_tail(&filter_item->list, &filter_list); + list_add_tail(&filter_item->list, &filter_list->list); /* * Regardless of if this returned an error, we still * replace the filter for the call. @@ -2195,31 +2301,22 @@ static int process_system_preds(struct trace_subsystem_dir *dir, * Do a synchronize_rcu() and to ensure all calls are * done with them before we free them. */ - tracepoint_synchronize_unregister(); - list_for_each_entry_safe(filter_item, tmp, &filter_list, list) { - __free_filter(filter_item->filter); - list_del(&filter_item->list); - kfree(filter_item); - } + delay_free_filter(filter_list); return 0; fail: /* No call succeeded */ - list_for_each_entry_safe(filter_item, tmp, &filter_list, list) { - list_del(&filter_item->list); - kfree(filter_item); - } + free_filter_list(&filter_list->rcu); parse_error(pe, FILT_ERR_BAD_SUBSYS_FILTER, 0); return -EINVAL; fail_mem: __free_filter(filter); + /* If any call succeeded, we still need to sync */ if (!fail) - tracepoint_synchronize_unregister(); - list_for_each_entry_safe(filter_item, tmp, &filter_list, list) { - __free_filter(filter_item->filter); - list_del(&filter_item->list); - kfree(filter_item); - } + delay_free_filter(filter_list); + else + free_filter_list(&filter_list->rcu); + return -ENOMEM; } @@ -2361,9 +2458,7 @@ int apply_event_filter(struct trace_event_file *file, char *filter_string) event_clear_filter(file); - /* Make sure the filter is not being used */ - tracepoint_synchronize_unregister(); - __free_filter(filter); + try_delay_free_filter(filter); return 0; } @@ -2387,11 +2482,8 @@ int apply_event_filter(struct trace_event_file *file, char *filter_string) event_set_filter(file, filter); - if (tmp) { - /* Make sure the call is done with the filter */ - tracepoint_synchronize_unregister(); - __free_filter(tmp); - } + if (tmp) + try_delay_free_filter(tmp); } return err; @@ -2417,9 +2509,7 @@ int apply_subsystem_event_filter(struct trace_subsystem_dir *dir, filter = system->filter; system->filter = NULL; /* Ensure all filters are no longer used */ - tracepoint_synchronize_unregister(); - filter_free_subsystem_filters(dir, tr); - __free_filter(filter); + filter_free_subsystem_filters(dir, tr, filter); return 0; } -- 2.47.2

2 weeks, 5 days

1
0
0 0

[PATCH 0/2] mm/memory: fix memory tearing on threaded fork

by Jann Horn

The first patch is a fix with an explanation of the issue, you should read that first. The second patch adds a comment to document the rules because figuring this out from scratch causes brain pain. Accidentally hitting this issue and getting negative consequences from it would require several stars to line up just right; but if someone out there is using a malloc() implementation that uses lockless data structures across threads or such, this could actually be a problem. In case someone wants a testcase, here's a very artificial one: ``` #include <pthread.h> #include <err.h> #include <stdio.h> #include <unistd.h> #include <sys/syscall.h> #include <sys/uio.h> #include <sys/mman.h> #include <sys/wait.h> #include <linux/io_uring.h> #define SYSCHK(x) ({ \ typeof(x) __res = (x); \ if (__res == (typeof(x))-1) \ err(1, "SYSCHK(" #x ")"); \ __res; \ }) #define NUM_SQ_PAGES 4 static int uring_init(struct io_uring_sqe **sqesp, void **cqesp) { struct io_uring_sqe *sqes = SYSCHK(mmap(NULL, NUM_SQ_PAGES*0x1000, PROT_READ|PROT_WRITE, MAP_SHARED|MAP_ANONYMOUS, -1, 0)); void *cqes = SYSCHK(mmap(NULL, NUM_SQ_PAGES*0x1000, PROT_READ|PROT_WRITE, MAP_SHARED|MAP_ANONYMOUS, -1, 0)); *(volatile unsigned int *)(cqes+4) = 64 * NUM_SQ_PAGES; struct io_uring_params params = { .flags = IORING_SETUP_NO_MMAP|IORING_SETUP_NO_SQARRAY, .sq_off = { .user_addr = (unsigned long)sqes }, .cq_off = { .user_addr = (unsigned long)cqes } }; int uring_fd = SYSCHK(syscall(__NR_io_uring_setup, /*entries=*/10, &params)); if (sqesp) *sqesp = sqes; if (cqesp) *cqesp = cqes; return uring_fd; } static char *bufmem[0x3000] __attribute__((aligned(0x1000))); static void *thread_fn(void *dummy) { unsigned long i = 0; while (1) { *(volatile unsigned long *)(bufmem + 0x0000) = i; *(volatile unsigned long *)(bufmem + 0x0f00) = i; *(volatile unsigned long *)(bufmem + 0x1000) = i; *(volatile unsigned long *)(bufmem + 0x1f00) = i; *(volatile unsigned long *)(bufmem + 0x2000) = i; *(volatile unsigned long *)(bufmem + 0x2f00) = i; i++; } } int main(void) { #if 1 int uring_fd = uring_init(NULL, NULL); struct iovec reg_iov = { .iov_base = bufmem, .iov_len = 0x2000 }; SYSCHK(syscall(__NR_io_uring_register, uring_fd, IORING_REGISTER_BUFFERS, &reg_iov, 1)); #endif pthread_t thread; if (pthread_create(&thread, NULL, thread_fn, NULL)) errx(1, "pthread_create"); sleep(1); int child = SYSCHK(fork()); if (child == 0) { printf("bufmem values:\n"); printf(" 0x0000: 0x%lx\n", *(volatile unsigned long *)(bufmem + 0x0000)); printf(" 0x0f00: 0x%lx\n", *(volatile unsigned long *)(bufmem + 0x0f00)); printf(" 0x1000: 0x%lx\n", *(volatile unsigned long *)(bufmem + 0x1000)); printf(" 0x1f00: 0x%lx\n", *(volatile unsigned long *)(bufmem + 0x1f00)); printf(" 0x2000: 0x%lx\n", *(volatile unsigned long *)(bufmem + 0x2000)); printf(" 0x2f00: 0x%lx\n", *(volatile unsigned long *)(bufmem + 0x2f00)); return 0; } int wstatus; SYSCHK(wait(&wstatus)); return 0; } ``` Without this series, the child will usually print results that are apart by more than 1, which is not a state that ever occurred in the parent; in my opinion, that counts as a bug. If you change the "#if 1" to "#if 0", the bug won't manifest. Signed-off-by: Jann Horn <jannh(a)google.com> --- Jann Horn (2): mm/memory: ensure fork child sees coherent memory snapshot mm/memory: Document how we make a coherent memory snapshot kernel/fork.c | 34 ++++++++++++++++++++++++++++++++++ mm/memory.c | 18 ++++++++++++++++++ 2 files changed, 52 insertions(+) --- base-commit: 8477ab143069c6b05d6da4a8184ded8b969240f5 change-id: 20250530-fork-tearing-71da211a50cf -- Jann Horn <jannh(a)google.com>

2 weeks, 5 days

5
17
0 0

[PATCH AUTOSEL 5.4 1/3] cifs: Fix cifs_query_path_info() for Windows NT servers

by Sasha Levin

From: Pali Rohár <pali(a)kernel.org> [ Upstream commit a3e771afbb3bce91c8296828304903e7348003fe ] For TRANS2 QUERY_PATH_INFO request when the path does not exist, the Windows NT SMB server returns error response STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile without the SMBFLG_RESPONSE flag set. Similarly it returns STATUS_DELETE_PENDING when the file is being deleted. And looks like that any error response from TRANS2 QUERY_PATH_INFO does not have SMBFLG_RESPONSE flag set. So relax check in check_smb_hdr() for detecting if the packet is response for this special case. This change fixes stat() operation against Windows NT SMB servers and also all operations which depends on -ENOENT result from stat like creat() or mkdir(). Signed-off-by: Pali Rohár <pali(a)kernel.org> Signed-off-by: Steve French <stfrench(a)microsoft.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- **YES** This commit should be backported to stable kernel trees. Here's my extensive analysis: ## Backport Recommendation: YES ### Nature of the Fix This is a **targeted bug fix** that addresses a specific protocol compliance issue with Windows NT SMB servers. The commit fixes a fundamental problem where basic file operations (`stat()`, `creat()`, `mkdir()`) fail against Windows NT servers due to overly strict SMB header validation. ### Code Change Analysis The fix is remarkably **simple and contained** - only 8 lines of code in a single function: ```c /bin /bin.usr-is-merged /boot /dev /etc /home /init /lib /lib.usr- is-merged /lib64 /lost+found /media /mnt /opt /proc /root /run /sbin /sbin.usr-is-merged /snap /srv /sys /tmp /usr /var * Windows NT server returns error resposne (e.g. STATUS_DELETE_PENDING * or STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile or any other) * for some TRANS2 requests without the RESPONSE flag set in header. */ if (smb->Command == SMB_COM_TRANSACTION2 && smb->Status.CifsError != 0) return 0; ``` The change is in the `check_smb_hdr()` function in `fs/smb/client/misc.c`, which validates SMB packet headers. The fix **relaxes validation** only for a very specific case: TRANS2 commands returning errors from Windows NT servers. ### Risk Assessment: LOW 1. **Surgical precision**: Only affects SMB1 TRANS2 error responses, leaving normal operation paths untouched 2. **Conservative approach**: The condition `smb->Status.CifsError != 0` ensures this only applies to actual error responses 3. **No behavioral changes** for compliant servers that properly set the RESPONSE flag 4. **Well-established code path**: The `check_smb_hdr()` function is mature and stable ### Impact Assessment: HIGH 1. **Fixes broken functionality**: Without this fix, basic file operations fail completely against Windows NT servers 2. **Backward compatibility**: Restores support for legacy but still-used server environments 3. **User-visible improvement**: Directly fixes `stat()`, `creat()`, and `mkdir()` operations 4. **No regressions**: Modern SMB servers continue to work as before ### Comparison with Similar Commits Looking at the provided similar commits: - **Similar Commit #2** (Status: YES): Also adds new status code mappings for better server compatibility - **Similar Commit #3** (Status: YES): Reduces unnecessary network roundtrips by improving error handling - **Similar Commit #4** (Status: YES): Fixes WSL reparse point querying over SMB1 - **Similar Commit #5** (Status: YES): Fixes missing resource cleanup This commit follows the **same pattern** as these approved backports: small, targeted fixes that improve compatibility and fix real-world issues without introducing new features or architectural changes. ### Technical Justification The Windows NT server behavior described in the commit is **non-compliant but real**: these servers return error responses for TRANS2 QUERY_PATH_INFO requests without setting the `SMBFLG_RESPONSE` flag. The current strict validation incorrectly treats these as invalid packets, causing the CIFS client to fail when it should handle the errors properly. The fix is **protocol-aware** and **conservative** - it only relaxes validation for the specific case where we know Windows NT behaves differently, ensuring no impact on standard-compliant servers. ### Stable Tree Suitability This commit perfectly fits stable tree criteria: - ✅ **Important bug fix** affecting real-world usage - ✅ **Minimal and contained** change with clear scope - ✅ **No new features** - purely fixes existing broken functionality - ✅ **Low regression risk** due to targeted nature - ✅ **Production-ready** code following established patterns This should be backported to all stable kernels that support SMB1 client functionality, as it fixes a fundamental compatibility issue without any meaningful risk of regression. fs/cifs/misc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index db1fcdedf289a..af9752535dbab 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -306,6 +306,14 @@ check_smb_hdr(struct smb_hdr *smb) if (smb->Command == SMB_COM_LOCKING_ANDX) return 0; + /* + * Windows NT server returns error resposne (e.g. STATUS_DELETE_PENDING + * or STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile or any other) + * for some TRANS2 requests without the RESPONSE flag set in header. + */ + if (smb->Command == SMB_COM_TRANSACTION2 && smb->Status.CifsError != 0) + return 0; + cifs_dbg(VFS, "Server sent request, not response. mid=%u\n", get_mid(smb)); return 1; -- 2.39.5

2 weeks, 5 days

1
2
0 0

[PATCH AUTOSEL 5.10 1/6] cifs: Fix cifs_query_path_info() for Windows NT servers

by Sasha Levin

From: Pali Rohár <pali(a)kernel.org> [ Upstream commit a3e771afbb3bce91c8296828304903e7348003fe ] For TRANS2 QUERY_PATH_INFO request when the path does not exist, the Windows NT SMB server returns error response STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile without the SMBFLG_RESPONSE flag set. Similarly it returns STATUS_DELETE_PENDING when the file is being deleted. And looks like that any error response from TRANS2 QUERY_PATH_INFO does not have SMBFLG_RESPONSE flag set. So relax check in check_smb_hdr() for detecting if the packet is response for this special case. This change fixes stat() operation against Windows NT SMB servers and also all operations which depends on -ENOENT result from stat like creat() or mkdir(). Signed-off-by: Pali Rohár <pali(a)kernel.org> Signed-off-by: Steve French <stfrench(a)microsoft.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- **YES** This commit should be backported to stable kernel trees. Here's my extensive analysis: ## Backport Recommendation: YES ### Nature of the Fix This is a **targeted bug fix** that addresses a specific protocol compliance issue with Windows NT SMB servers. The commit fixes a fundamental problem where basic file operations (`stat()`, `creat()`, `mkdir()`) fail against Windows NT servers due to overly strict SMB header validation. ### Code Change Analysis The fix is remarkably **simple and contained** - only 8 lines of code in a single function: ```c /bin /bin.usr-is-merged /boot /dev /etc /home /init /lib /lib.usr- is-merged /lib64 /lost+found /media /mnt /opt /proc /root /run /sbin /sbin.usr-is-merged /snap /srv /sys /tmp /usr /var * Windows NT server returns error resposne (e.g. STATUS_DELETE_PENDING * or STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile or any other) * for some TRANS2 requests without the RESPONSE flag set in header. */ if (smb->Command == SMB_COM_TRANSACTION2 && smb->Status.CifsError != 0) return 0; ``` The change is in the `check_smb_hdr()` function in `fs/smb/client/misc.c`, which validates SMB packet headers. The fix **relaxes validation** only for a very specific case: TRANS2 commands returning errors from Windows NT servers. ### Risk Assessment: LOW 1. **Surgical precision**: Only affects SMB1 TRANS2 error responses, leaving normal operation paths untouched 2. **Conservative approach**: The condition `smb->Status.CifsError != 0` ensures this only applies to actual error responses 3. **No behavioral changes** for compliant servers that properly set the RESPONSE flag 4. **Well-established code path**: The `check_smb_hdr()` function is mature and stable ### Impact Assessment: HIGH 1. **Fixes broken functionality**: Without this fix, basic file operations fail completely against Windows NT servers 2. **Backward compatibility**: Restores support for legacy but still-used server environments 3. **User-visible improvement**: Directly fixes `stat()`, `creat()`, and `mkdir()` operations 4. **No regressions**: Modern SMB servers continue to work as before ### Comparison with Similar Commits Looking at the provided similar commits: - **Similar Commit #2** (Status: YES): Also adds new status code mappings for better server compatibility - **Similar Commit #3** (Status: YES): Reduces unnecessary network roundtrips by improving error handling - **Similar Commit #4** (Status: YES): Fixes WSL reparse point querying over SMB1 - **Similar Commit #5** (Status: YES): Fixes missing resource cleanup This commit follows the **same pattern** as these approved backports: small, targeted fixes that improve compatibility and fix real-world issues without introducing new features or architectural changes. ### Technical Justification The Windows NT server behavior described in the commit is **non-compliant but real**: these servers return error responses for TRANS2 QUERY_PATH_INFO requests without setting the `SMBFLG_RESPONSE` flag. The current strict validation incorrectly treats these as invalid packets, causing the CIFS client to fail when it should handle the errors properly. The fix is **protocol-aware** and **conservative** - it only relaxes validation for the specific case where we know Windows NT behaves differently, ensuring no impact on standard-compliant servers. ### Stable Tree Suitability This commit perfectly fits stable tree criteria: - ✅ **Important bug fix** affecting real-world usage - ✅ **Minimal and contained** change with clear scope - ✅ **No new features** - purely fixes existing broken functionality - ✅ **Low regression risk** due to targeted nature - ✅ **Production-ready** code following established patterns This should be backported to all stable kernels that support SMB1 client functionality, as it fixes a fundamental compatibility issue without any meaningful risk of regression. fs/cifs/misc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 2d46018b02839..54c443686daba 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -310,6 +310,14 @@ check_smb_hdr(struct smb_hdr *smb) if (smb->Command == SMB_COM_LOCKING_ANDX) return 0; + /* + * Windows NT server returns error resposne (e.g. STATUS_DELETE_PENDING + * or STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile or any other) + * for some TRANS2 requests without the RESPONSE flag set in header. + */ + if (smb->Command == SMB_COM_TRANSACTION2 && smb->Status.CifsError != 0) + return 0; + cifs_dbg(VFS, "Server sent request, not response. mid=%u\n", get_mid(smb)); return 1; -- 2.39.5

2 weeks, 5 days

1
5
0 0

[PATCH AUTOSEL 5.15 1/7] cifs: Fix cifs_query_path_info() for Windows NT servers

by Sasha Levin

From: Pali Rohár <pali(a)kernel.org> [ Upstream commit a3e771afbb3bce91c8296828304903e7348003fe ] For TRANS2 QUERY_PATH_INFO request when the path does not exist, the Windows NT SMB server returns error response STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile without the SMBFLG_RESPONSE flag set. Similarly it returns STATUS_DELETE_PENDING when the file is being deleted. And looks like that any error response from TRANS2 QUERY_PATH_INFO does not have SMBFLG_RESPONSE flag set. So relax check in check_smb_hdr() for detecting if the packet is response for this special case. This change fixes stat() operation against Windows NT SMB servers and also all operations which depends on -ENOENT result from stat like creat() or mkdir(). Signed-off-by: Pali Rohár <pali(a)kernel.org> Signed-off-by: Steve French <stfrench(a)microsoft.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- **YES** This commit should be backported to stable kernel trees. Here's my extensive analysis: ## Backport Recommendation: YES ### Nature of the Fix This is a **targeted bug fix** that addresses a specific protocol compliance issue with Windows NT SMB servers. The commit fixes a fundamental problem where basic file operations (`stat()`, `creat()`, `mkdir()`) fail against Windows NT servers due to overly strict SMB header validation. ### Code Change Analysis The fix is remarkably **simple and contained** - only 8 lines of code in a single function: ```c /bin /bin.usr-is-merged /boot /dev /etc /home /init /lib /lib.usr- is-merged /lib64 /lost+found /media /mnt /opt /proc /root /run /sbin /sbin.usr-is-merged /snap /srv /sys /tmp /usr /var * Windows NT server returns error resposne (e.g. STATUS_DELETE_PENDING * or STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile or any other) * for some TRANS2 requests without the RESPONSE flag set in header. */ if (smb->Command == SMB_COM_TRANSACTION2 && smb->Status.CifsError != 0) return 0; ``` The change is in the `check_smb_hdr()` function in `fs/smb/client/misc.c`, which validates SMB packet headers. The fix **relaxes validation** only for a very specific case: TRANS2 commands returning errors from Windows NT servers. ### Risk Assessment: LOW 1. **Surgical precision**: Only affects SMB1 TRANS2 error responses, leaving normal operation paths untouched 2. **Conservative approach**: The condition `smb->Status.CifsError != 0` ensures this only applies to actual error responses 3. **No behavioral changes** for compliant servers that properly set the RESPONSE flag 4. **Well-established code path**: The `check_smb_hdr()` function is mature and stable ### Impact Assessment: HIGH 1. **Fixes broken functionality**: Without this fix, basic file operations fail completely against Windows NT servers 2. **Backward compatibility**: Restores support for legacy but still-used server environments 3. **User-visible improvement**: Directly fixes `stat()`, `creat()`, and `mkdir()` operations 4. **No regressions**: Modern SMB servers continue to work as before ### Comparison with Similar Commits Looking at the provided similar commits: - **Similar Commit #2** (Status: YES): Also adds new status code mappings for better server compatibility - **Similar Commit #3** (Status: YES): Reduces unnecessary network roundtrips by improving error handling - **Similar Commit #4** (Status: YES): Fixes WSL reparse point querying over SMB1 - **Similar Commit #5** (Status: YES): Fixes missing resource cleanup This commit follows the **same pattern** as these approved backports: small, targeted fixes that improve compatibility and fix real-world issues without introducing new features or architectural changes. ### Technical Justification The Windows NT server behavior described in the commit is **non-compliant but real**: these servers return error responses for TRANS2 QUERY_PATH_INFO requests without setting the `SMBFLG_RESPONSE` flag. The current strict validation incorrectly treats these as invalid packets, causing the CIFS client to fail when it should handle the errors properly. The fix is **protocol-aware** and **conservative** - it only relaxes validation for the specific case where we know Windows NT behaves differently, ensuring no impact on standard-compliant servers. ### Stable Tree Suitability This commit perfectly fits stable tree criteria: - ✅ **Important bug fix** affecting real-world usage - ✅ **Minimal and contained** change with clear scope - ✅ **No new features** - purely fixes existing broken functionality - ✅ **Low regression risk** due to targeted nature - ✅ **Production-ready** code following established patterns This should be backported to all stable kernels that support SMB1 client functionality, as it fixes a fundamental compatibility issue without any meaningful risk of regression. fs/cifs/misc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 33328eae03d7a..a3d37e7769e61 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -297,6 +297,14 @@ check_smb_hdr(struct smb_hdr *smb) if (smb->Command == SMB_COM_LOCKING_ANDX) return 0; + /* + * Windows NT server returns error resposne (e.g. STATUS_DELETE_PENDING + * or STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile or any other) + * for some TRANS2 requests without the RESPONSE flag set in header. + */ + if (smb->Command == SMB_COM_TRANSACTION2 && smb->Status.CifsError != 0) + return 0; + cifs_dbg(VFS, "Server sent request, not response. mid=%u\n", get_mid(smb)); return 1; -- 2.39.5

2 weeks, 5 days

1
6
0 0

[PATCH AUTOSEL 6.6 01/13] cifs: Correctly set SMB1 SessionKey field in Session Setup Request

by Sasha Levin

From: Pali Rohár <pali(a)kernel.org> [ Upstream commit 89381c72d52094988e11d23ef24a00066a0fa458 ] [MS-CIFS] specification in section 2.2.4.53.1 where is described SMB_COM_SESSION_SETUP_ANDX Request, for SessionKey field says: The client MUST set this field to be equal to the SessionKey field in the SMB_COM_NEGOTIATE Response for this SMB connection. Linux SMB client currently set this field to zero. This is working fine against Windows NT SMB servers thanks to [MS-CIFS] product behavior <94>: Windows NT Server ignores the client's SessionKey. For compatibility with [MS-CIFS], set this SessionKey field in Session Setup Request to value retrieved from Negotiate response. Signed-off-by: Pali Rohár <pali(a)kernel.org> Signed-off-by: Steve French <stfrench(a)microsoft.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- Based on my comprehensive analysis of both the commit message and code changes, examining the current kernel tree context: **YES** ## Extensive Analysis ### **Nature of the Fix** This is a **specification compliance fix** for SMB1 protocol handling. The commit addresses a clear violation of the MS-CIFS specification section 2.2.4.53.1, which mandates that the SessionKey field in SMB_COM_SESSION_SETUP_ANDX Request must be set to the value from the SMB_COM_NEGOTIATE Response. ### **Code Changes Analysis** **1. Structural Changes (Low Risk)** - **cifsglob.h**: Adds `__le32 session_key_id` to `TCP_Server_Info` structure - minimal ABI impact as this is kernel-internal - **cifspdu.h**: Corrects endianness annotations from `__u32` to `__le32` - pure annotation fix, no behavioral change **2. Functional Changes (Contained and Safe)** - **cifssmb.c**: Captures SessionKey from negotiate response (`server->session_key_id = pSMBr->SessionKey`) - simple field assignment - **sess.c**: Sets SessionKey in session setup request (`pSMB->req.SessionKey = server->session_key_id`) - follows specification requirement ### **Risk Assessment** **Minimal Regression Risk:** - Changes are isolated to SMB1 legacy protocol code - Current behavior (setting SessionKey to zero) works with Windows NT due to MS- CIFS product behavior <94> - New behavior follows specification exactly, improving compatibility with non-Windows SMB1 servers - No security implications - purely a protocol compliance improvement **Code Maturity:** - SMB1 session setup code is very mature and stable - Limited recent changes in this area except compatibility fixes - Well- understood code path with clear specification backing ### **Backport Suitability Criteria Met** **1. Important Bug Fix:** ✅ Fixes protocol specification violation that could cause compatibility issues with strict SMB1 implementations **2. Small and Contained:** ✅ Four minimal changes across related files, all implementing the same specification requirement **3. Clear Side Effects:** ✅ No side effects beyond fixing the specification compliance issue **4. No Architectural Changes:** ✅ Simple field additions and assignments, no structural changes **5. Non- Critical Subsystem:** ✅ Legacy SMB1 protocol code, deprecated since 2017 **6. Minimal Risk:** ✅ Maintains backward compatibility while improving forward compatibility **7. Stable Tree Rules Compliance:** ✅ Important compatibility bugfix with minimal regression risk ### **Comparison to Similar Commits** This commit closely matches the pattern of **Similar Commit #3** and **Similar Commit #5** (both marked YES for backporting): - Protocol compliance improvements - Session setup related fixes - Clear specification backing - Minimal, contained changes - Authored by maintainers (Steve French involvement) ### **Conclusion** This commit represents an ideal stable backport candidate: a clear specification compliance fix with minimal code changes, no security implications, and improvement in interoperability. The fix ensures Linux kernel SMB client properly follows MS-CIFS specification, which is valuable for enterprise environments using diverse SMB1 server implementations. fs/smb/client/cifsglob.h | 1 + fs/smb/client/cifspdu.h | 6 +++--- fs/smb/client/cifssmb.c | 1 + fs/smb/client/sess.c | 1 + 4 files changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 39117343b703f..1f3451202fe8d 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -739,6 +739,7 @@ struct TCP_Server_Info { char workstation_RFC1001_name[RFC1001_NAME_LEN_WITH_NULL]; __u32 sequence_number; /* for signing, protected by srv_mutex */ __u32 reconnect_instance; /* incremented on each reconnect */ + __le32 session_key_id; /* retrieved from negotiate response and send in session setup request */ struct session_key session_key; unsigned long lstrp; /* when we got last response from this server */ struct cifs_secmech secmech; /* crypto sec mech functs, descriptors */ diff --git a/fs/smb/client/cifspdu.h b/fs/smb/client/cifspdu.h index ca33f6cd6a800..763178b774542 100644 --- a/fs/smb/client/cifspdu.h +++ b/fs/smb/client/cifspdu.h @@ -557,7 +557,7 @@ typedef union smb_com_session_setup_andx { __le16 MaxBufferSize; __le16 MaxMpxCount; __le16 VcNumber; - __u32 SessionKey; + __le32 SessionKey; __le16 SecurityBlobLength; __u32 Reserved; __le32 Capabilities; /* see below */ @@ -576,7 +576,7 @@ typedef union smb_com_session_setup_andx { __le16 MaxBufferSize; __le16 MaxMpxCount; __le16 VcNumber; - __u32 SessionKey; + __le32 SessionKey; __le16 CaseInsensitivePasswordLength; /* ASCII password len */ __le16 CaseSensitivePasswordLength; /* Unicode password length*/ __u32 Reserved; /* see below */ @@ -614,7 +614,7 @@ typedef union smb_com_session_setup_andx { __le16 MaxBufferSize; __le16 MaxMpxCount; __le16 VcNumber; - __u32 SessionKey; + __le32 SessionKey; __le16 PasswordLength; __u32 Reserved; /* encrypt key len and offset */ __le16 ByteCount; diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index b91184ebce02c..096ea62e87ea9 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -479,6 +479,7 @@ CIFSSMBNegotiate(const unsigned int xid, server->max_rw = le32_to_cpu(pSMBr->MaxRawSize); cifs_dbg(NOISY, "Max buf = %d\n", ses->server->maxBuf); server->capabilities = le32_to_cpu(pSMBr->Capabilities); + server->session_key_id = pSMBr->SessionKey; server->timeAdj = (int)(__s16)le16_to_cpu(pSMBr->ServerTimeZone); server->timeAdj *= 60; diff --git a/fs/smb/client/sess.c b/fs/smb/client/sess.c index f04922eb45d4c..a7ece67ee9630 100644 --- a/fs/smb/client/sess.c +++ b/fs/smb/client/sess.c @@ -680,6 +680,7 @@ static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, USHRT_MAX)); pSMB->req.MaxMpxCount = cpu_to_le16(server->maxReq); pSMB->req.VcNumber = cpu_to_le16(1); + pSMB->req.SessionKey = server->session_key_id; /* Now no need to set SMBFLG_CASELESS or obsolete CANONICAL PATH */ -- 2.39.5

2 weeks, 5 days

1
12
0 0

[PATCH AUTOSEL 6.12 01/15] cifs: Correctly set SMB1 SessionKey field in Session Setup Request

by Sasha Levin

From: Pali Rohár <pali(a)kernel.org> [ Upstream commit 89381c72d52094988e11d23ef24a00066a0fa458 ] [MS-CIFS] specification in section 2.2.4.53.1 where is described SMB_COM_SESSION_SETUP_ANDX Request, for SessionKey field says: The client MUST set this field to be equal to the SessionKey field in the SMB_COM_NEGOTIATE Response for this SMB connection. Linux SMB client currently set this field to zero. This is working fine against Windows NT SMB servers thanks to [MS-CIFS] product behavior <94>: Windows NT Server ignores the client's SessionKey. For compatibility with [MS-CIFS], set this SessionKey field in Session Setup Request to value retrieved from Negotiate response. Signed-off-by: Pali Rohár <pali(a)kernel.org> Signed-off-by: Steve French <stfrench(a)microsoft.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- Based on my comprehensive analysis of both the commit message and code changes, examining the current kernel tree context: **YES** ## Extensive Analysis ### **Nature of the Fix** This is a **specification compliance fix** for SMB1 protocol handling. The commit addresses a clear violation of the MS-CIFS specification section 2.2.4.53.1, which mandates that the SessionKey field in SMB_COM_SESSION_SETUP_ANDX Request must be set to the value from the SMB_COM_NEGOTIATE Response. ### **Code Changes Analysis** **1. Structural Changes (Low Risk)** - **cifsglob.h**: Adds `__le32 session_key_id` to `TCP_Server_Info` structure - minimal ABI impact as this is kernel-internal - **cifspdu.h**: Corrects endianness annotations from `__u32` to `__le32` - pure annotation fix, no behavioral change **2. Functional Changes (Contained and Safe)** - **cifssmb.c**: Captures SessionKey from negotiate response (`server->session_key_id = pSMBr->SessionKey`) - simple field assignment - **sess.c**: Sets SessionKey in session setup request (`pSMB->req.SessionKey = server->session_key_id`) - follows specification requirement ### **Risk Assessment** **Minimal Regression Risk:** - Changes are isolated to SMB1 legacy protocol code - Current behavior (setting SessionKey to zero) works with Windows NT due to MS- CIFS product behavior <94> - New behavior follows specification exactly, improving compatibility with non-Windows SMB1 servers - No security implications - purely a protocol compliance improvement **Code Maturity:** - SMB1 session setup code is very mature and stable - Limited recent changes in this area except compatibility fixes - Well- understood code path with clear specification backing ### **Backport Suitability Criteria Met** **1. Important Bug Fix:** ✅ Fixes protocol specification violation that could cause compatibility issues with strict SMB1 implementations **2. Small and Contained:** ✅ Four minimal changes across related files, all implementing the same specification requirement **3. Clear Side Effects:** ✅ No side effects beyond fixing the specification compliance issue **4. No Architectural Changes:** ✅ Simple field additions and assignments, no structural changes **5. Non- Critical Subsystem:** ✅ Legacy SMB1 protocol code, deprecated since 2017 **6. Minimal Risk:** ✅ Maintains backward compatibility while improving forward compatibility **7. Stable Tree Rules Compliance:** ✅ Important compatibility bugfix with minimal regression risk ### **Comparison to Similar Commits** This commit closely matches the pattern of **Similar Commit #3** and **Similar Commit #5** (both marked YES for backporting): - Protocol compliance improvements - Session setup related fixes - Clear specification backing - Minimal, contained changes - Authored by maintainers (Steve French involvement) ### **Conclusion** This commit represents an ideal stable backport candidate: a clear specification compliance fix with minimal code changes, no security implications, and improvement in interoperability. The fix ensures Linux kernel SMB client properly follows MS-CIFS specification, which is valuable for enterprise environments using diverse SMB1 server implementations. fs/smb/client/cifsglob.h | 1 + fs/smb/client/cifspdu.h | 6 +++--- fs/smb/client/cifssmb.c | 1 + fs/smb/client/sess.c | 1 + 4 files changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index a38b40d68b14f..9cd39cf96b99a 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -739,6 +739,7 @@ struct TCP_Server_Info { char workstation_RFC1001_name[RFC1001_NAME_LEN_WITH_NULL]; __u32 sequence_number; /* for signing, protected by srv_mutex */ __u32 reconnect_instance; /* incremented on each reconnect */ + __le32 session_key_id; /* retrieved from negotiate response and send in session setup request */ struct session_key session_key; unsigned long lstrp; /* when we got last response from this server */ struct cifs_secmech secmech; /* crypto sec mech functs, descriptors */ diff --git a/fs/smb/client/cifspdu.h b/fs/smb/client/cifspdu.h index 28f8ca470770d..688a26aeef3b4 100644 --- a/fs/smb/client/cifspdu.h +++ b/fs/smb/client/cifspdu.h @@ -557,7 +557,7 @@ typedef union smb_com_session_setup_andx { __le16 MaxBufferSize; __le16 MaxMpxCount; __le16 VcNumber; - __u32 SessionKey; + __le32 SessionKey; __le16 SecurityBlobLength; __u32 Reserved; __le32 Capabilities; /* see below */ @@ -576,7 +576,7 @@ typedef union smb_com_session_setup_andx { __le16 MaxBufferSize; __le16 MaxMpxCount; __le16 VcNumber; - __u32 SessionKey; + __le32 SessionKey; __le16 CaseInsensitivePasswordLength; /* ASCII password len */ __le16 CaseSensitivePasswordLength; /* Unicode password length*/ __u32 Reserved; /* see below */ @@ -614,7 +614,7 @@ typedef union smb_com_session_setup_andx { __le16 MaxBufferSize; __le16 MaxMpxCount; __le16 VcNumber; - __u32 SessionKey; + __le32 SessionKey; __le16 PasswordLength; __u32 Reserved; /* encrypt key len and offset */ __le16 ByteCount; diff --git a/fs/smb/client/cifssmb.c b/fs/smb/client/cifssmb.c index 8667f403a0ab6..c83b7aba24972 100644 --- a/fs/smb/client/cifssmb.c +++ b/fs/smb/client/cifssmb.c @@ -481,6 +481,7 @@ CIFSSMBNegotiate(const unsigned int xid, server->max_rw = le32_to_cpu(pSMBr->MaxRawSize); cifs_dbg(NOISY, "Max buf = %d\n", ses->server->maxBuf); server->capabilities = le32_to_cpu(pSMBr->Capabilities); + server->session_key_id = pSMBr->SessionKey; server->timeAdj = (int)(__s16)le16_to_cpu(pSMBr->ServerTimeZone); server->timeAdj *= 60; diff --git a/fs/smb/client/sess.c b/fs/smb/client/sess.c index 9b32f7821b718..6e48e6efe656f 100644 --- a/fs/smb/client/sess.c +++ b/fs/smb/client/sess.c @@ -655,6 +655,7 @@ static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, USHRT_MAX)); pSMB->req.MaxMpxCount = cpu_to_le16(server->maxReq); pSMB->req.VcNumber = cpu_to_le16(1); + pSMB->req.SessionKey = server->session_key_id; /* Now no need to set SMBFLG_CASELESS or obsolete CANONICAL PATH */ -- 2.39.5

2 weeks, 5 days

1
14
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror June 2025