Implemented -e trace=%clock option

2017-03-10 Thread Rishi Bhatt
Hey, So for extending a trace class option syntax,i have implemented clock_* related syscall. The idea came in my mind after the scheduler option was committed recently. I thought clock related syscalls would be good. So i have attached the patch,please review it and give me the feedback. Cheers

Re: Implemented -e trace=%clock option

2017-03-10 Thread Dmitry V. Levin
On Fri, Mar 10, 2017 at 08:30:55PM +0530, Rishi Bhatt wrote: > Hey, > So for extending a trace class option syntax,i have implemented clock_* > related syscall. > > The idea came in my mind after the scheduler option was committed recently. > I thought clock related syscalls would be good. Yes, w

Re: GSOC: Introduction

2017-03-10 Thread Abhishek Tiwari
On Thu, Mar 9, 2017 at 5:39 AM, Dmitry V. Levin wrote: > On Wed, Mar 08, 2017 at 07:51:49AM +0530, Abhishek Tiwari wrote: >> Made changes in description of commit message and INSTALL-git.md. >> From bfb8f2576e947b8b1abd7709f1ae8277d77e3c2a Mon Sep 17 00:00:00 2001 >> From: Abhishek Tiwari >> Date

Re: GSOC: Introduction

2017-03-10 Thread Abhishek Tiwari
> Thank you for the patch acceptance. > Is the following list of open system calls complete ? It is based on > reply that function calls that returns new file descriptor > > 1) mq_open > 2) openat > 3) perf_event_open > 4) open_by_handle_at > 5) memfd_create > 6) epoll_create1 > 7) dup > 8) dup3 >

Re: GSoC 2017 introduction

2017-03-10 Thread Victor Krapivensky
On 03/10/2017 04:36 AM, Dmitry V. Levin wrote: This cannot be correct. OK, removed it for now, and fixed all the issues you mentioned. I've also added a test, so please review. I've also stumbled upon a wrong comment in util.c. Attaching a separate patch for it. This is wrong, the type of n