Re: Adding shared memory implementation

2009-01-27 Thread Bernhard Reutner-Fischer
On Tue, Jan 27, 2009 at 10:47:35AM +0100, Mikael Lund Jepsen, ICCC wrote: > Bernhard Reutner-Fischer wrote: >> This can simply be dropped into the proper tests/ subdir, no? >> > I'm not yet familiar with the test suite, so I had not considered > putting it in there, but I think it is definitel

Re: Adding shared memory implementation

2009-01-27 Thread Mikael Lund Jepsen, ICCC
Bernhard Reutner-Fischer wrote: > This can simply be dropped into the proper tests/ subdir, no? > I'm not yet familiar with the test suite, so I had not considered putting it in there, but I think it is definitely a good idea. I just looked briefly at the (quite elaborate) test skeleton and a f

Re: Adding shared memory implementation

2009-01-23 Thread Bernhard Reutner-Fischer
On Fri, Jan 23, 2009 at 03:39:33PM +0100, Mikael Lund Jepsen, ICCC wrote: > Mikael Lund Jepsen, ICCC wrote: >> >> Joakim Tjernlund wrote: + /* skip opening slash */ + if (*name == '/') +++name; >>> >>> I think skipping all leading '/' is better: >>>for (; *name ==

Re: Adding shared memory implementation

2009-01-23 Thread Mikael Lund Jepsen, ICCC
Mikael Lund Jepsen, ICCC wrote: Joakim Tjernlund wrote: + /* skip opening slash */ + if (*name == '/') +++name; I think skipping all leading '/' is better: for (; *name == '/'; ++name); I took the newlib code verbatim to keep the number of different implementations down, but

Re: Adding shared memory implementation

2009-01-16 Thread Joakim Tjernlund
> > > Joakim Tjernlund wrote: > + /* skip opening slash */ > + if (*name == '/') > +++name; > > > I think skipping all leading '/' is better: >for (; *name == '/'; ++name); > > I took the newlib code verbatim to keep the number of different > implementations down, > but if we are m

Re: Adding shared memory implementation

2009-01-16 Thread Mikael Lund Jepsen, ICCC
Joakim Tjernlund wrote: + /* skip opening slash */ + if (*name == '/') +++name; I think skipping all leading '/' is better: for (; *name == '/'; ++name); I took the newlib code verbatim to keep the number of different implementations down, but if we are making changes I think

Re: Adding shared memory implementation

2009-01-16 Thread Joakim Tjernlund
> + /* skip opening slash */ > + if (*name == '/') > +++name; I think skipping all leading '/' is better: for (; *name == '/'; ++name); Jocke ___ uClibc mailing list uClibc@uclibc.org http://lists.busybox.net/mailman/listinfo/uclibc

Re: Adding shared memory implementation

2009-01-16 Thread Mikael Lund Jepsen, ICCC
Hi all, This patch adds shm_open() and shm_unlink() to librt. For my own purposes, I have git'ed uclibc and made the patch from there, so let me know if you need it in another format. The included testapp simply creates shared memory area in a parent process, writes test data to it and opens

Re: Adding shared memory implementation

2009-01-14 Thread Carmelo AMOROSO
Mikael Lund Jepsen, ICCC wrote: > Hi all, > I cannot see the shm_open / shm_unlink functions implemented in librt. > Is there any reason (bloat, ?) for not including these? > as simply it is reported in uClibc_vs_SuSv3.txt, they are optional and simply not implemented until now. > I've tried tak

Adding shared memory implementation

2009-01-14 Thread Mikael Lund Jepsen, ICCC
Hi all, I cannot see the shm_open / shm_unlink functions implemented in librt. Is there any reason (bloat, ?) for not including these? I've tried taking shm_open.c and shm_unlink.c from newlib and dropped them in librt, and so far it seems to be working as expected. I have not done any extensive