From: "Bakul Shah via 9fans" <9fans@9fans.net>
To: 9fans <9fans@9fans.net>
Subject: Re: mmaping on plan9? (was Re: [9fans] venti /plan9port mmapped
Date: Wed, 7 Jan 2026 12:35:28 -0800 [thread overview]
Message-ID: <5E011A70-314B-47DB-A5FD-03D6E54F2EFE@iitbombay.org> (raw)
In-Reply-To: <B26977CC40F5FE4C442C52B1073997B6@eigenstate.org>
> On Jan 7, 2026, at 8:41 AM, ori@eigenstate.org wrote:
>
> Quoth Bakul Shah via 9fans <9fans@9fans.net>:
>> I have this idea that will horrify most of you!
>>
>> 1. Create an mmap device driver. You ask it to a new file handle which you use to communicate about memory mapping.
>> 2. If you want to mmap some file, you open it and write its file descriptor along with other parameters (file offset, base addr, size, mode, flags) to your mmap file handle.
>> 3. The mmap driver sets up necessary page table entries but doesn't actually fetch any data before returning from the write.
>> 4. It can asynchronously kick off io requests on your behalf and fixup page table entries as needed.
>> 5. Page faults in the mmapped area are serviced by making appropriate read/write calls.
>> 6. Flags can be used to indicate read-ahead or write-behind for typical serial access.
>> 7. Similarly msync, munmap etc. can be implemented.
>>
>> In a sneaky way this avoids the need for adding any mmap specific syscalls! But the underlying work would be mostly similar in either case.
>>
>> The main benefits of mmap are reduced initial latency , "pay as you go" cost structure and ease of use. It is certainly more expensive than reading/writing the same amount of data directly from a program.
>>
>> No idea how horrible a hack is needed to implement such a thing or even if it is possible at all but I had to share this ;-)
>
> To what end? The problems with mmap have little to do with adding a syscall;
> they're about how you do things like communicating I/O errors. Especially
> when flushing the cache.
>
> Imagine the following setup -- I've imported 9p.io:
>
> 9fs 9pio
>
> and then I map a file from it:
>
> mapped = mmap("/n/9pio/plan9/lib/words", OWRITE);
>
> Now, I want to write something into the file:
>
> *mapped = 1234;
>
> The cached version of the page is dirty, so the OS will
> eventually need to flush it back with a 9p Twrite; Let's
> assume that before this happens, the network goes down.
>
> How do you communicate the error with userspace?
This was just a brainwave but...
You have a (control) connection with the mmap device to
set up mmap so might as well use it to convey errors!
This device would be strictly local to where a program
runs.
I'd even consider allowing a separate process to mmap,
by making an address space a first class object. That'd
move more stuff out of the kernel and allow for more
interesting/esoteric uses.
------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/Te8d7c6e48b5c075b-M6fb0ef830c8e525cb591fb48
Delivery options: https://9fans.topicbox.com/groups/9fans/subscription
next prev parent reply other threads:[~2026-01-07 21:09 UTC|newest]
Thread overview: 92+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-02 19:54 [9fans] venti /plan9port mmapped wb.kloke
2026-01-02 20:39 ` ori
2026-01-02 20:58 ` Bakul Shah via 9fans
2026-01-06 22:59 ` Ron Minnich
2026-01-07 4:27 ` Noam Preil
2026-01-07 6:15 ` Shawn Rutledge
2026-01-07 15:46 ` Persistent memory (was Re: [9fans] venti /plan9port mmapped) arnold
2026-01-07 16:11 ` Noam Preil
2026-01-07 17:26 ` Wes Kussmaul
2026-01-07 8:52 ` [9fans] venti /plan9port mmapped wb.kloke
2026-01-07 16:30 ` mmaping on plan9? (was " Bakul Shah via 9fans
2026-01-07 16:40 ` Noam Preil
2026-01-07 16:41 ` ori
2026-01-07 20:35 ` Bakul Shah via 9fans [this message]
2026-01-07 21:31 ` ron minnich
2026-01-08 7:56 ` arnold
2026-01-08 10:31 ` wb.kloke
2026-01-09 0:02 ` ron minnich
2026-01-09 3:57 ` Paul Lalonde
2026-01-09 5:10 ` ron minnich
2026-01-09 5:18 ` arnold
2026-01-09 6:06 ` David Leimbach via 9fans
2026-01-09 17:13 ` ron minnich
2026-01-09 17:39 ` tlaronde
2026-01-09 19:48 ` David Leimbach via 9fans
2026-02-05 21:30 ` Alyssa M via 9fans
2026-02-08 14:18 ` Ethan Azariah
2026-02-08 15:10 ` Alyssa M via 9fans
2026-02-08 20:43 ` Ethan Azariah
2026-02-09 1:35 ` ron minnich
2026-02-09 15:23 ` ron minnich
2026-02-09 17:13 ` Bakul Shah via 9fans
2026-02-09 21:38 ` ron minnich
2026-02-10 10:13 ` Alyssa M via 9fans
2026-02-11 1:43 ` Ron Minnich
2026-02-11 2:19 ` Bakul Shah via 9fans
2026-02-11 3:21 ` Ori Bernstein
2026-02-11 10:01 ` hiro
2026-02-12 1:36 ` Dan Cross
2026-02-12 5:39 ` Alyssa M via 9fans
2026-02-12 9:08 ` hiro via 9fans
2026-02-12 13:34 ` Alyssa M via 9fans
2026-02-13 13:48 ` hiro
2026-02-13 17:21 ` ron minnich
2026-02-15 16:12 ` Danny Wilkins via 9fans
2026-02-17 3:13 ` Alyssa M via 9fans
2026-02-17 13:02 ` Dan Cross
2026-02-17 16:00 ` ron minnich
2026-02-17 16:39 ` hiro
2026-02-17 16:56 ` Bakul Shah via 9fans
2026-02-17 17:54 ` hiro
2026-02-17 22:21 ` Alyssa M via 9fans
2026-02-16 2:24 ` Alyssa M via 9fans
2026-02-16 3:17 ` Ori Bernstein
2026-02-16 10:55 ` Frank D. Engel, Jr.
2026-02-16 13:49 ` Ori Bernstein
2026-02-16 19:40 ` Bakul Shah via 9fans
2026-02-16 19:43 ` Bakul Shah via 9fans
2026-02-16 9:50 ` tlaronde
2026-02-16 12:24 ` hiro via 9fans
2026-02-16 12:33 ` hiro via 9fans
2026-02-11 14:22 ` Dan Cross
2026-02-11 18:44 ` Ori Bernstein
2026-02-12 1:22 ` Dan Cross
2026-02-12 4:26 ` Ori Bernstein
2026-02-12 4:34 ` Dan Cross
2026-02-12 3:12 ` Alyssa M via 9fans
2026-02-12 4:52 ` Dan Cross
2026-02-12 8:37 ` Alyssa M via 9fans
2026-02-12 12:37 ` hiro via 9fans
2026-02-13 1:36 ` Dan Cross
2026-02-14 3:35 ` Alyssa M via 9fans
2026-02-14 14:26 ` Dan Cross
2026-02-15 4:34 ` Bakul Shah via 9fans
2026-02-15 10:19 ` hiro
2026-02-10 16:49 ` wb.kloke
2026-02-08 14:08 ` Ethan Azariah
2026-01-07 21:40 ` ori
2026-01-07 16:52 ` ori
2026-01-07 17:37 ` wb.kloke
2026-01-07 17:46 ` Noam Preil
2026-01-07 17:56 ` wb.kloke
2026-01-07 18:07 ` Noam Preil
2026-01-07 18:58 ` wb.kloke
2026-01-07 14:57 ` Thaddeus Woskowiak
2026-01-07 16:07 ` Wes Kussmaul
2026-01-07 16:22 ` Noam Preil
2026-01-07 17:31 ` Wes Kussmaul
2026-01-07 16:13 ` Noam Preil
2026-01-02 21:01 ` ori
2026-01-08 15:59 ` wb.kloke
2026-02-11 23:19 ` red
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5E011A70-314B-47DB-A5FD-03D6E54F2EFE@iitbombay.org \
--to=9fans@9fans.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).