Re: [PATCH, RFC] warn about raw pointers that can be unique_ptr

2017-05-08 Thread Joseph Myers
Note that the new option will need documenting in invoke.texi for any patch version actually proposed for inclusion. -- Joseph S. Myers jos...@codesourcery.com

Re: [PATCH, RFC] warn about raw pointers that can be unique_ptr

2017-05-08 Thread Trevor Saunders
On Mon, May 08, 2017 at 10:28:13AM +0200, Marc Glisse wrote: > On Sun, 7 May 2017, Trevor Saunders wrote: > > > On Sun, May 07, 2017 at 07:32:48PM +0200, Marc Glisse wrote: > > > On Sun, 7 May 2017, tbsaunde+...@tbsaunde.org wrote: > > > > > > > This is a start at warning about various resource a

Re: [PATCH, RFC] warn about raw pointers that can be unique_ptr

2017-05-08 Thread Marc Glisse
On Sun, 7 May 2017, Trevor Saunders wrote: On Sun, May 07, 2017 at 07:32:48PM +0200, Marc Glisse wrote: On Sun, 7 May 2017, tbsaunde+...@tbsaunde.org wrote: This is a start at warning about various resource allocation issues that can be improved. Currently it only warns about functions that

Re: [PATCH, RFC] warn about raw pointers that can be unique_ptr

2017-05-07 Thread Trevor Saunders
On Sun, May 07, 2017 at 07:32:48PM +0200, Marc Glisse wrote: > On Sun, 7 May 2017, tbsaunde+...@tbsaunde.org wrote: > > > This is a start at warning about various resource allocation issues that > > can be > > improved. Currently it only warns about functions that call malloc and then > > always

Re: [PATCH, RFC] warn about raw pointers that can be unique_ptr

2017-05-07 Thread Marc Glisse
On Sun, 7 May 2017, tbsaunde+...@tbsaunde.org wrote: This is a start at warning about various resource allocation issues that can be improved. Currently it only warns about functions that call malloc and then always pass the resulting pointer to free(). It should be pretty simple to extend thi

[PATCH, RFC] warn about raw pointers that can be unique_ptr

2017-05-07 Thread tbsaunde+gcc
From: Trevor Saunders Hi, This is a start at warning about various resource allocation issues that can be improved. Currently it only warns about functions that call malloc and then always pass the resulting pointer to free(). It should be pretty simple to extend this to new/delete and new[]/d