ok, thanks. ________________________________ 发件人: Steven Rostedt <[email protected]> 发送时间: 2025年12月17日 23:52:39 收件人: Xiang Gao 抄送: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; 高翔 主题: [External Mail]Re: [PATCH v7] dma-buf: add some tracepoints to debug.
[外部邮件] 此邮件来源于小米公司外部，请谨慎处理。若对邮件安全性存疑，请将邮件转发给[email protected]进行反馈 On Wed, 17 Dec 2025 18:51:32 +0800 Xiang Gao <[email protected]> wrote: > From: gaoxiang17 <[email protected]> > > Since we can only inspect dmabuf by iterating over process FDs or the > dmabuf_list, we need to add our own tracepoints to track its status in > real time in production. > > For example: > binder:3016_1-3102 [006] ...1. 255.126521: dma_buf_export: > exp_name=qcom,system size=12685312 ino=2738 > binder:3016_1-3102 [006] ...1. 255.126528: dma_buf_fd: > exp_name=qcom,system size=12685312 ino=2738 fd=8 > binder:3016_1-3102 [006] ...1. 255.126642: dma_buf_mmap_internal: > exp_name=qcom,system size=28672 ino=2739 > kworker/6:1-86 [006] ...1. 255.127194: dma_buf_put: > exp_name=qcom,system size=12685312 ino=2738 > RenderThread-9293 [006] ...1. 316.618179: dma_buf_get: > exp_name=qcom,system size=12771328 ino=2762 fd=176 > RenderThread-9293 [006] ...1. 316.618195: dma_buf_dynamic_attach: > exp_name=qcom,system size=12771328 ino=2762 attachment:ffffff880a18dd00 > is_dynamic=0 dev_name=kgsl-3d0 > RenderThread-9293 [006] ...1. 318.878220: dma_buf_detach: > exp_name=qcom,system size=12771328 ino=2762 attachment:ffffff880a18dd00 > is_dynamic=0 dev_name=kgsl-3d0 > > Signed-off-by: Xiang Gao <[email protected]> > --- > drivers/dma-buf/dma-buf.c | 42 ++++++++- > include/trace/events/dma_buf.h | 157 +++++++++++++++++++++++++++++++++ > 2 files changed, 198 insertions(+), 1 deletion(-) > create mode 100644 include/trace/events/dma_buf.h > > diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c > index 2bcf9ceca997..ce39bc19e13f 100644 > --- a/drivers/dma-buf/dma-buf.c > +++ b/drivers/dma-buf/dma-buf.c > @@ -35,6 +35,25 @@ > > #include "dma-buf-sysfs-stats.h" > > +#define CREATE_TRACE_POINTS > +#include <trace/events/dma_buf.h> > + > +/* > + * dmabuf->name must be accessed with holding dmabuf->name_lock. > + * we need to take the lock around the tracepoint call itself where > + * it is called in the code. > + * > + * Note: FUNC##_enabled() is a static branch that will only > + * be set when the trace event is enabled. > + */ Much better. > +#define DMA_BUF_TRACE(FUNC, ...) \ > + do { > \ > + if (FUNC##_enabled()) { > \ > + guard(spinlock)(&dmabuf->name_lock); \ > + FUNC(__VA_ARGS__); > \ > + } > \ Hmm, I wonder if we should also add: } else if (IS_ENABLED(CONFIG_LOCKDEP)) { \ /* Expose this lock when lockdep is enabled */ \ guard(spinlock)(&dmabuf->name_lock); \ } This way, if there is any issue taking the lock, lockdep will flag it without having to enable the tracepoint. When LOCKDEP is not configured, the compiler should remove that block. > + } while (0) > + > static inline int is_dma_buf_file(struct file *); > > > --- /dev/null > +++ b/include/trace/events/dma_buf.h > @@ -0,0 +1,157 @@ > +/* SPDX-License-Identifier: GPL-2.0 */ > +#undef TRACE_SYSTEM > +#define TRACE_SYSTEM dma_buf > + > +#if !defined(_TRACE_DMA_BUF_H) || defined(TRACE_HEADER_MULTI_READ) > +#define _TRACE_DMA_BUF_H > + > +#include <linux/dma-buf.h> > +#include <linux/tracepoint.h> > + > +DECLARE_EVENT_CLASS(dma_buf, > + > + TP_PROTO(struct dma_buf *dmabuf), > + > + TP_ARGS(dmabuf), > + > + TP_STRUCT__entry( > + __string(exp_name, dmabuf->exp_name) > + __field(size_t, size) > + __field(ino_t, ino) > + ), > + > + TP_fast_assign( > + __assign_str(exp_name); > + __entry->size = dmabuf->size; > + __entry->ino = dmabuf->file->f_inode->i_ino; > + ), > + > + TP_printk("exp_name=%s size=%zu ino=%lu", > + __get_str(exp_name), > + __entry->size, > + __entry->ino) > +); For the rest of the patch, from a tracing point of view: Reviewed-by: Steven Rostedt (Google) <[email protected]> -- Steve
