Re: AT_EXECPATH aux_info vector contains path of interpreter when directly exec'ing rtld

2019-12-20 Thread Konstantin Belousov
On Fri, Dec 20, 2019 at 04:26:41PM -0500, Ryan Stone wrote: > I've noticed that on head, if I directly execute rtld to run an > executable, AT_EXECPATH contains the path to rtld on head (on > 12.0-RELEASE it will contain nothing). This is causing me a problem > because clang uses AT_EXECPATH to

Re: AT_EXECPATH aux_info vector contains path of interpreter when directly exec'ing rtld

2019-12-20 Thread Ed Maste
On Fri, 20 Dec 2019 at 16:26, Ryan Stone wrote: > > I've noticed that on head, if I directly execute rtld to run an > executable, AT_EXECPATH contains the path to rtld on head (on > 12.0-RELEASE it will contain nothing). This is causing me a problem > because clang uses AT_EXECPATH to

Re: AT_EXECPATH aux_info vector contains path of interpreter when directly exec'ing rtld

2019-12-20 Thread Conrad Meyer
Isn’t rtld’s behavior here correct? It’s really Clang which is doing something quite odd. On Fri, Dec 20, 2019 at 13:27 Ryan Stone wrote: > I've noticed that on head, if I directly execute rtld to run an > executable, AT_EXECPATH contains the path to rtld on head (on > 12.0-RELEASE it will

AT_EXECPATH aux_info vector contains path of interpreter when directly exec'ing rtld

2019-12-20 Thread Ryan Stone
I've noticed that on head, if I directly execute rtld to run an executable, AT_EXECPATH contains the path to rtld on head (on 12.0-RELEASE it will contain nothing). This is causing me a problem because clang uses AT_EXECPATH to preferentially locate where it's installed, which it uses to locate