From nobody Sun Feb 8 12:36:37 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 5F75E7FBC2 for ; Wed, 21 Feb 2024 14:06:41 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708524401; cv=none; b=UZ7AOeLEW5zKY4mbAFZvMDUrow0MprilU0PrkekZp8TnSBIzNXF83rZa8A0IrMP2hu2p6BK+0FJcWss7ppsl7VMHXdvF0MtlkWRKU0PU66x8zwsZAA+jQ0gl1FNdhj83UuWfuVltX2gQjRBFN/plHFNReOo7kwAKMyHvnLysGeA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1708524401; c=relaxed/simple; bh=aZBsiUdC/sqkVAN/zoqT47h9j1DEO0rYRQQLWdJ5vGU=; h=Message-ID:Date:From:To:Cc:Subject:References:MIME-Version: Content-Type; b=Dd9CD3ILdAyLDVys5uH1Ahf5fVRcSDQcfkhEzOT9WocONISA5IDSQZDEfYEVOH/2yaOrLZcUKsvB2a4A7TahTxV4ZPST/6pxveJjWj1eQPCkOeRTDSJWeV0pGwXbV3y/s1F7wr1ow6wnye0OC7dY0yEVhtkHcZqm1bx9uWplFOM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 Received: by smtp.kernel.org (Postfix) with ESMTPSA id 396FAC43399; Wed, 21 Feb 2024 14:06:41 +0000 (UTC) Received: from rostedt by gandalf with local (Exim 4.97) (envelope-from ) id 1rcnGv-00000002iBK-1qTr; Wed, 21 Feb 2024 09:08:29 -0500 Message-ID: <20240221140829.305327406@goodmis.org> User-Agent: quilt/0.67 Date: Wed, 21 Feb 2024 09:08:05 -0500 From: Steven Rostedt To: linux-kernel@vger.kernel.org Cc: Masami Hiramatsu , Mark Rutland , Mathieu Desnoyers , Andrew Morton , Vincent Donnefort Subject: [for-next][PATCH 09/11] tracing: Allow user-space mapping of the ring-buffer References: <20240221140756.797572998@goodmis.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" From: Vincent Donnefort Currently, user-space extracts data from the ring-buffer via splice, which is handy for storage or network sharing. However, due to splice limitations, it is imposible to do real-time analysis without a copy. A solution for that problem is to let the user-space map the ring-buffer directly. The mapping is exposed via the per-CPU file trace_pipe_raw. The first element of the mapping is the meta-page. It is followed by each subbuffer constituting the ring-buffer, ordered by their unique page ID: * Meta-page -- include/uapi/linux/trace_mmap.h for a description * Subbuf ID 0 * Subbuf ID 1 ... It is therefore easy to translate a subbuf ID into an offset in the mapping: reader_id =3D meta->reader->id; reader_offset =3D meta->meta_page_size + reader_id * meta->subbuf_size; When new data is available, the mapper must call a newly introduced ioctl: TRACE_MMAP_IOCTL_GET_READER. This will update the Meta-page reader ID to point to the next reader containing unread data. Mapping will prevent snapshot and buffer size modifications. Link: https://lore.kernel.org/linux-trace-kernel/20240220202310.2489614-5-v= donnefort@google.com Signed-off-by: Vincent Donnefort Signed-off-by: Steven Rostedt (Google) --- include/uapi/linux/trace_mmap.h | 2 + kernel/trace/trace.c | 137 ++++++++++++++++++++++++++++++-- kernel/trace/trace.h | 1 + 3 files changed, 135 insertions(+), 5 deletions(-) diff --git a/include/uapi/linux/trace_mmap.h b/include/uapi/linux/trace_mma= p.h index ffcd8dfcaa4f..d25b9d504a7c 100644 --- a/include/uapi/linux/trace_mmap.h +++ b/include/uapi/linux/trace_mmap.h @@ -43,4 +43,6 @@ struct trace_buffer_meta { __u64 Reserved2; }; =20 +#define TRACE_MMAP_IOCTL_GET_READER _IO('T', 0x1) + #endif /* _TRACE_MMAP_H_ */ diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 2b922a79c553..8fdd68dbcf6d 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1175,6 +1175,12 @@ static void tracing_snapshot_instance_cond(struct tr= ace_array *tr, return; } =20 + if (tr->mapped) { + trace_array_puts(tr, "*** BUFFER MEMORY MAPPED ***\n"); + trace_array_puts(tr, "*** Can not use snapshot (sorry) ***\n"); + return; + } + local_irq_save(flags); update_max_tr(tr, current, smp_processor_id(), cond_data); local_irq_restore(flags); @@ -1307,7 +1313,7 @@ static int tracing_arm_snapshot_locked(struct trace_a= rray *tr) lockdep_assert_held(&trace_types_lock); =20 spin_lock(&tr->snapshot_trigger_lock); - if (tr->snapshot =3D=3D UINT_MAX) { + if (tr->snapshot =3D=3D UINT_MAX || tr->mapped) { spin_unlock(&tr->snapshot_trigger_lock); return -EBUSY; } @@ -6032,7 +6038,7 @@ static void tracing_set_nop(struct trace_array *tr) { if (tr->current_trace =3D=3D &nop_trace) return; -=09 + tr->current_trace->enabled--; =20 if (tr->current_trace->reset) @@ -8151,15 +8157,31 @@ tracing_buffers_splice_read(struct file *file, loff= _t *ppos, return ret; } =20 -/* An ioctl call with cmd 0 to the ring buffer file will wake up all waite= rs */ static long tracing_buffers_ioctl(struct file *file, unsigned int cmd, uns= igned long arg) { struct ftrace_buffer_info *info =3D file->private_data; struct trace_iterator *iter =3D &info->iter; + int err; + + if (cmd =3D=3D TRACE_MMAP_IOCTL_GET_READER) { + if (!(file->f_flags & O_NONBLOCK)) { + err =3D ring_buffer_wait(iter->array_buffer->buffer, + iter->cpu_file, + iter->tr->buffer_percent); + if (err) + return err; + } =20 - if (cmd) - return -ENOIOCTLCMD; + return ring_buffer_map_get_reader(iter->array_buffer->buffer, + iter->cpu_file); + } else if (cmd) { + return -ENOTTY; + } =20 + /* + * An ioctl call with cmd 0 to the ring buffer file will wake up all + * waiters + */ mutex_lock(&trace_types_lock); =20 iter->wait_index++; @@ -8172,6 +8194,110 @@ static long tracing_buffers_ioctl(struct file *file= , unsigned int cmd, unsigned return 0; } =20 +static vm_fault_t tracing_buffers_mmap_fault(struct vm_fault *vmf) +{ + struct ftrace_buffer_info *info =3D vmf->vma->vm_file->private_data; + struct trace_iterator *iter =3D &info->iter; + vm_fault_t ret =3D VM_FAULT_SIGBUS; + struct page *page; + + page =3D ring_buffer_map_fault(iter->array_buffer->buffer, iter->cpu_file, + vmf->pgoff); + if (!page) + return ret; + + get_page(page); + vmf->page =3D page; + vmf->page->mapping =3D vmf->vma->vm_file->f_mapping; + vmf->page->index =3D vmf->pgoff; + + return 0; +} + +#ifdef CONFIG_TRACER_MAX_TRACE +static int get_snapshot_map(struct trace_array *tr) +{ + int err =3D 0; + + /* + * Called with mmap_lock held. lockdep would be unhappy if we would now + * take trace_types_lock. Instead use the specific + * snapshot_trigger_lock. + */ + spin_lock(&tr->snapshot_trigger_lock); + + if (tr->snapshot || tr->mapped =3D=3D UINT_MAX) + err =3D -EBUSY; + else + tr->mapped++; + + spin_unlock(&tr->snapshot_trigger_lock); + + /* Wait for update_max_tr() to observe iter->tr->mapped */ + if (tr->mapped =3D=3D 1) + synchronize_rcu(); + + return err; + +} +static void put_snapshot_map(struct trace_array *tr) +{ + spin_lock(&tr->snapshot_trigger_lock); + if (!WARN_ON(!tr->mapped)) + tr->mapped--; + spin_unlock(&tr->snapshot_trigger_lock); +} +#else +static inline int get_snapshot_map(struct trace_array *tr) { return 0; } +static inline void put_snapshot_map(struct trace_array *tr) { } +#endif + +static void tracing_buffers_mmap_close(struct vm_area_struct *vma) +{ + struct ftrace_buffer_info *info =3D vma->vm_file->private_data; + struct trace_iterator *iter =3D &info->iter; + + ring_buffer_unmap(iter->array_buffer->buffer, iter->cpu_file); + put_snapshot_map(iter->tr); +} + +static void tracing_buffers_mmap_open(struct vm_area_struct *vma) +{ + struct ftrace_buffer_info *info =3D vma->vm_file->private_data; + struct trace_iterator *iter =3D &info->iter; + + WARN_ON(ring_buffer_map(iter->array_buffer->buffer, iter->cpu_file)); +} + +static const struct vm_operations_struct tracing_buffers_vmops =3D { + .open =3D tracing_buffers_mmap_open, + .close =3D tracing_buffers_mmap_close, + .fault =3D tracing_buffers_mmap_fault, +}; + +static int tracing_buffers_mmap(struct file *filp, struct vm_area_struct *= vma) +{ + struct ftrace_buffer_info *info =3D filp->private_data; + struct trace_iterator *iter =3D &info->iter; + int ret =3D 0; + + if (vma->vm_flags & VM_WRITE || vma->vm_flags & VM_EXEC) + return -EPERM; + + vm_flags_mod(vma, VM_DONTCOPY | VM_DONTDUMP, VM_MAYWRITE); + vma->vm_ops =3D &tracing_buffers_vmops; + + ret =3D get_snapshot_map(iter->tr); + if (ret) + return ret; + + ret =3D ring_buffer_map(iter->array_buffer->buffer, iter->cpu_file); + if (ret) + put_snapshot_map(iter->tr); + + return ret; +} + static const struct file_operations tracing_buffers_fops =3D { .open =3D tracing_buffers_open, .read =3D tracing_buffers_read, @@ -8180,6 +8306,7 @@ static const struct file_operations tracing_buffers_f= ops =3D { .splice_read =3D tracing_buffers_splice_read, .unlocked_ioctl =3D tracing_buffers_ioctl, .llseek =3D no_llseek, + .mmap =3D tracing_buffers_mmap, }; =20 static ssize_t diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 64450615ca0c..749a182dab48 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -336,6 +336,7 @@ struct trace_array { bool allocated_snapshot; spinlock_t snapshot_trigger_lock; unsigned int snapshot; + unsigned int mapped; unsigned long max_latency; #ifdef CONFIG_FSNOTIFY struct dentry *d_max_latency; --=20 2.43.0