Michal, I think this is getting promoted the wrong way, to the kernel developers.
A patch to implement I_PEEK on pipes, is a generic 'shell accelerator' because shells no longer have to read byte by byte from pipes where the dominant locale is the multibyte en_US.UTF-8. In that light, any shell, including bash and dash, would benefit from using I_PEEK. Just no one can use it because it does not work with the default pipes. Olga On Mon, May 27, 2013 at 4:18 PM, Michal Hlavinka <[email protected]> wrote: > Hi Olga, > > I'm still trying to push it through kernel team todo list. I talked with a > few people without success yet. I'm waiting on status information from > kernel team leader, but he's unreachable now (I guess he's on vacation now). > > Josh: > No, it's in our bugzilla our kernel developers use, bug itself is not > publicly visible. I was discouraged by kernel developers from filing this to > upstream bugzilla, because it's extremely difficult to get any attention to > bug reports and almost impossible in case of feature requests. The usual > process is to bribe^Wconvince any kernel developer you know to do the work. > > Michal > > > On 12.5.2013 23:47, ольга крыжановская wrote: >> >> Michal, has been there any progress? >> >> Olga >> >> On Thu, Dec 13, 2012 at 12:22 PM, Michal Hlavinka <[email protected]> >> wrote: >>> >>> On 12/12/2012 01:35 PM, ольга крыжановская wrote: >>>> >>>> >>>> Michal, has been any one working on the Linux kernel fifo/pipe >>>> implementation to implement I_PEEK? Is there a Bugzilla entry for >>>> this? >>>> >>>> Olga >>> >>> >>> >>> Hi Olga, >>> >>> unfortunately, it's still in kernel team's todo list. >>> There is bugzilla for this, but it is internal only, so not public. >>> >>> Michal >>> >>> >>>> On Wed, Jun 13, 2012 at 5:54 PM, Michal Hlavinka <[email protected]> >>>> wrote: >>>>> >>>>> >>>>> On 06/05/2012 03:03 PM, Cedric Blancher wrote: >>>>>> >>>>>> >>>>>> >>>>>> On 5 June 2012 09:11, Michal Hlavinka<[email protected]> wrote: >>>>>>> >>>>>>> >>>>>>> >>>>>>> On 06/01/2012 02:32 PM, Irek Szczesniak wrote: >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> Michal, did anyone ever filed a bug against Linux's FIFO/PIPE >>>>>>>> implementations to support I_PEEK? As far as I can check all SystemV >>>>>>>> derivatives (including Solaris), AIX and HP/UX support I_PEEK on >>>>>>>> pipes >>>>>>>> and fifos. >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> I don't know if there was any official attempt. I asked a few kernel >>>>>>> developers in person and they told me that no one will bother with >>>>>>> this. >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> <rant> >>>>>> Michal, I recall that Redhat staff once ridiculed and mocked a patch >>>>>> (...why add extra performance support for a dead OS [Solaris] ...) >>>>>> which added support for I_PEEK to bash2 and finally convinced the >>>>>> maintainers NOT to take it. So basically this issue is blocked from >>>>>> both sides, kernel and bash, by Redhat. >>>>>> Why? >>>>>> </rant> >>>>> >>>>> >>>>> >>>>> >>>>> I'm not aware about anything related, so I can't answer your >>>>> ran^W^W^Wquestion. >>>>> >>>>> >>>>>> As for an implementation in the Linux fifo kernel module, the I_PEEK >>>>>> ioctl() can be implemented along the lines of a read() syscall but >>>>>> without disposing the data which have been read. An implementation >>>>>> should therefore be very easy and should give the shells in Linux a >>>>>> SERIOUS performance advantage. I'm wondering why Redhat isn't >>>>>> interested in performance. Oh yes, see<rant /> >>>>> >>>>> >>>>> >>>>> >>>>> Cedric was somewhat right in his reply. There are a lot of processes >>>>> which >>>>> are quite resource demanding even for one line change. Also, there is >>>>> always >>>>> more work than what can be done (especially on kernel side. afaik, >>>>> we're >>>>> still hiring ;-) ). If someone does X, it means he won't be able to do >>>>> Y. >>>>> So, we do what product management tell us to do. >>>>> >>>>> Also, please don't forget that Red Hat is only Linux, but Linux is not >>>>> only >>>>> Red Hat. (Well, nowadays even the first part is no longer completely >>>>> true.) >>>>> I wanted to add I_PEEK feature request to upstream bugzilla, but I've >>>>> been >>>>> told that it does not work that way. It'd rot there for ages until >>>>> someone >>>>> would close that a few years later during some bugzilla clean up. >>>>> >>>>> Anyway, I found up a few colleagues willing to code it. I need a >>>>> feature >>>>> request to be filed now. I'm going to ask a few customers who >>>>> complained >>>>> about 'connection reset by peer' messages if they are willing to do it. >>>>> If I >>>>> fail, I'll try to convince them to code it in their spare time. >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> ast-developers mailing list >>>>> [email protected] >>>>> https://mailman.research.att.com/mailman/listinfo/ast-developers >>>> >>>> >>>> >>>> >>>> >>> >> >> >> > -- , _ _ , { \/`o;====- Olga Kryzhanovska -====;o`\/ } .----'-/`-/ [email protected] \-`\-'----. `'-..-| / http://twitter.com/fleyta \ |-..-'` /\/\ Solaris/BSD//C/C++ programmer /\/\ `--` `--` _______________________________________________ ast-developers mailing list [email protected] http://lists.research.att.com/mailman/listinfo/ast-developers
