Home | History | Annotate | Line # | Download | only in roadmaps
mobile revision 1.2 
      1 $NetBSD: mobile,v 1.2 2025/05/19 18:02:53 nia Exp $
      2 
      3 NetBSD Mobile Roadmap
      4 =====================
      5 
      6 This roadmap is meant to cover issues specifically pertaining to
      7 mobile use, that is, devices that run on batteries and get carried
      8 around. This includes:
      9    - phones
     10    - tablets
     11    - tablet PCs
     12    - laptops
     13 The typical assumption right now is that phones and tablets have one
     14 software stack (iOS, Android) and work one way, and laptops, including
     15 tablet PCs, have another software stack (Windows, MacOS, Linux) and
     16 work another way. The "laptop" software stack is more or less the same
     17 as the "desktop" software stack, modulo some laptop-specific issues.
     18 Those laptop-specific issues are covered in this file; the rest of
     19 that software stack is discussed in the "desktop" roadmap. This file
     20 also covers the phone/tablet software stack.
     21 
     22 The following elements, projects, and goals are considered strategic
     23 priorities for the project:
     24 
     25  1. Tickless timers/scheduling
     26 
     27 These elements, projects, and goals are more or less long-term goals:
     28 
     29  2. Power management concerns
     30  3. Suspending
     31  4. atrun considered harmful
     32  5. (Wireless config issues are in the "desktop" roadmap)
     33 
     34 These elements, projects, and goals are for the time being pretty much
     35 blue sky:
     36 
     37  6. Touchscreen (phone/tablet) UI
     38  7. Support for phone hardware
     39 
     40 
     41 Explanations
     42 ============
     43 
     44 1. Tickless timers/scheduling
     45 
     46 The basic premise with a tickless system is that instead of generating
     47 a timer interrupt HZ times a second, one programs a high-resolution
     48 timer on the fly to interrupt the next time something needs to happen.
     49 This can substantially reduce the number of timer interrupts taken,
     50 and also importantly it avoids waking the system up regularly when
     51 otherwise idle and reduces power consumption and heating.
     52 
     53 There has been a fair amount of talk about this but so far no real
     54 action.
     55 
     56   - As of January 2017 nobody is known to be working on this.
     57   - There is currently no clear timeframe or release target.
     58   - Contact: ? (XXX)
     59 
     60 
     61 2. Power management concerns
     62 
     63 NetBSD's power management infrastructure is fairly lacking. We don't
     64 have good CPU clock rate throttling, we mostly don't have the ability
     65 to power down idle devices, and we don't have a configuration and
     66 control setup to manage it either. On x86 we also don't support a
     67 number of important ACPI sleep/idle states.
     68 
     69 At the moment there isn't even a good inventory of what needs to be
     70 done in this department. Someone please write it and put it here.
     71 
     72   - As of January 2017 nobody is known to be working on this.
     73   - There is currently no clear timeframe or release target.
     74   - Contact: ? (XXX)
     75 
     76 
     77 3. Suspending
     78 
     79 Currently suspending mostly doesn't work, and the chances of being
     80 able to suspend any given laptop model successfully are low until
     81 someone using it gets annoyed enough to sit down and make it behave.
     82 
     83 We need to fix this, both by adding suspend hooks to drivers that are
     84 missing them and also (ideally) by coming up with a better way to cope
     85 with drivers that don't know how to suspend.
     86 
     87   - As of January 2017 nobody is known to be specifically working on
     88     this, although work on individual drivers occurs sporadically.
     89   - There is currently no clear timeframe or release target.
     90   - Contact: ? (XXX)
     91 
     92 
     93 4. atrun considered harmful
     94 
     95 There are a number of things on the system that unnecessarily wake up
     96 and take cpu time and power on a regular basis. One of the big
     97 offenders is atrun -- it should be changed either to be a daemon that
     98 wakes up only when it has a job to run, integrated into cron to the
     99 same end, or changed around in some other similar fashion.
    100 
    101 One can always turn atrun off, but there's no particular reason that
    102 at(1) functionality should be unavailable on laptops.
    103 
    104   - As of January 2017 nobody is known to be specifically working on
    105     this, although work on individual drivers occurs sporadically.
    106   - There is currently no clear timeframe or release target.
    107   - Contact: ? (XXX)
    108 
    109 
    110 6. Touchscreen (phone/tablet) UI
    111 
    112 We'd rather like to be able to run on phones, and that means having a
    113 UI suitable for a phone -- a shell isn't going to cut it, and even a
    114 shell coupled with a keyboard app isn't really the ticket.
    115 
    116 This has many of the same kinds of issues as desktop software. Some of
    117 the specific issues are different; e.g. location handling is a lot
    118 more critical for phones than for desktops and even laptops.
    119 
    120 While we don't currently run on any phone platforms (see below)
    121 there's nothing stopping working on this using older PDA/palmtop
    122 hardware like hpcarm.
    123 
    124 
    125 7. Support for phone hardware
    126 
    127 As of 2025, much of the underlying work to support various Arm
    128 SoCs is done, but this point is heavily dependent on item 6 (UI).
    129 
    130