From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 27919 invoked by alias); 10 Sep 2004 00:48:04 -0000 Mailing-List: contact glibc-bugs-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Post: List-Help: , Sender: glibc-bugs-owner@sources.redhat.com Received: (qmail 27874 invoked by uid 48); 10 Sep 2004 00:48:02 -0000 Date: Fri, 10 Sep 2004 00:48:00 -0000 From: "dennis at mds dot rmit dot edu dot au" To: glibc-bugs@sources.redhat.com Message-ID: <20040910004801.378.dennis@mds.rmit.edu.au> Reply-To: sourceware-bugzilla@sources.redhat.com Subject: [Bug nptl/378] New: posix_spawn implementation, use vfork/execve rather than fork/execve for NPTL Linux. X-Bugzilla-Reason: CC X-SW-Source: 2004-09/txt/msg00045.txt.bz2 List-Id: The pairing of fork/execve can still be extremely slow even on modern Unix variants that use copy-on-write semantics. If one manages to create a large process with an extremely large number of memory mapped regions then fork/execve performance degrades badly (many tens-of-times slower). Copying the page table entries from parent to child can be an extremely expensive operation. That performance penalty does not occur with vfork/execve. The current generic glibc version of posix_spawn is based around fork/execve. Hence, it will suffer from the performance problem listed above. Solaris 10 now comes with a posix_spawn implementation. Using truss and speaking to Sun engineers it has been confirmed that the new Solaris 10 posix_spawn uses vfork/execve. Also the implementation they have carried out means their new posix_spawn is thread safe (where their vfork is not). When I raised this question at Red Hat bugzilla a friendly Red Hat engineer responded with the following. > It is impossible to use vfork in the linuxthreads libc, because > sigaction modifies global state, but most probably it should be > doable for NPTL libc, provided a few calls (e.g. *gid) are changed > into inline syscalls and adding code to run atfork registered > handlers before/after vfork in spawni.c. So this enhancement request is for the glibc engineers to investigate whether NPTL glibc posix_spawn can use vfork/execve rather than fork/execve. If there are no obstacles then that enhancement should be pushed through. This will result in huge performance benefits for application developers like us. The other benefit is that Solaris10 and NPTL glibc posix_spawn would have the same performance characteristics. Again that will be extremely. Dennis. -- Summary: posix_spawn implementation, use vfork/execve rather than fork/execve for NPTL Linux. Product: glibc Version: unspecified Status: NEW Severity: enhancement Priority: P2 Component: nptl AssignedTo: drepper at redhat dot com ReportedBy: dennis at mds dot rmit dot edu dot au CC: glibc-bugs at sources dot redhat dot com http://sources.redhat.com/bugzilla/show_bug.cgi?id=378 ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.