* 9FRONT "EMAILSCHADEN" RELEASED
@ 2020-10-19 4:03 sl
0 siblings, 0 replies; 2+ messages in thread
From: sl @ 2020-10-19 4:03 UTC (permalink / raw)
To: 9front
9FRONT "EMAILSCHADEN" RELEASED
==============================
![front 9FRONT "EMAILSCHADEN" front](http://9front.org/img/emailschaden.front.png)
DOWNLOAD
--------
Multiple installation media are provided for PC and Raspberry Pi. For
PC, burn an .iso file to CD, or dd it directly to USB media. For
Raspberry Pi, dd an .img file directly to sdcard.
The pi.img file can be used for Raspberry Pi 1, 2, and 3. The pi3.img
file can be used for Raspberry Pi 3 and 4.
http://9front.org/iso/9front-8013.d9e940a768d1.386.iso.gz.torrent
http://9front.org/iso/9front-8013.d9e940a768d1.amd64.iso.gz.torrent
http://9front.org/iso/9front-8013.d9e940a768d1.pi.img.gz.torrent
http://9front.org/iso/9front-8013.d9e940a768d1.pi3.img.gz.torrent
cpu% sha1sum -2 256 9front-8013*.gz
e8a224a8220e472593dd17a2479ba5a9497f6790cac23de2522400930a4dd626 9front-8013.d9e940a768d1.386.iso.gz
7771be88148560a8e59283a93c98b9ecc37172e7a2acff356f12e64aac9e5c77 9front-8013.d9e940a768d1.amd64.iso.gz
a63a87272258b17fdd24f2807252134aee311af243ec8274f715f8dcf1b86f1f 9front-8013.d9e940a768d1.pi.img.gz
5b5baa5c3fd0b69a71aaa7bca867508210fac25d8ad17d76b97d8aece1c9f3dd 9front-8013.d9e940a768d1.pi3.img.gz
MIRRORS
-------
<http://9front.org/iso/>
<http://r-36.net/9front/>
<http://felloff.net/usr/cinap_lenrek/9front.torrent>
MERCURIAL REPOSITORY
--------------------
<https://code.9front.org/hg/plan9front>
DASH 1 MANUAL
-------------
<http://fqa.9front.org/dash1.emailschaden.pdf>
dash 1 and man pages in print: <http://9front.org/propaganda/books>
(EMAILSCHADEN print edition coming soon)
THANKS
------
presotto
NEW IN THIS RELEASE
-------------------
New date/time formaring/parsing API in libc
Kernel updates for Raspberry PI4 8GB model
Intel wireless driver now supports 8000 and 9000 family cards
New Thinkpads now fully supported: T480, T495
Big kernel PCI code overhaul
KERNEL AND DRIVERS
------------------
archacpi: use 64-bit uvlong for physical addresses
audioac97: use 64-bit physical addresses
audiohda: use 64-bit physical addresses, check pci membar types
audiohda: use PCIWADDR() instead of PADDR(), handle 64-bit dma addresses
audiosb16: cleanup audioprobe(), cast ISAConf.port to ulong
aux/acpi: connect EC region handler, fixes battery status on eeepc 1000px (thanks Peter Kosyh)
aux/acpi: forgot to commit -p option
aux/acpi: write a man page, fix usage, call threadexits
aux/vga: eepc igfx support (thanks p.kosyh)
aux/vga: use 64-bit physical addresses for pci membar
bcm64: bring pci code in line with the pc/pc64
bcm64: fix usb xhci controller on pi4 8GB variant (thanks richard miller)
bcm64: handle 8GB of physical memory for raspberry pi4
bcm64: increase maximum segment size to 8GB
bcm64: use 64-bit uvlong physical address in vmap()
bcm64: use uvlong for ISAConf.port
bootrc: allow kbmap to be set via plan9.ini (thanks Aaron Bieber)
devether: print Ether.port as 64-bit uvlong
devip: fix parseipmask() prototype in ip.h
devip: implement ipv6 support in ipmux packet filter
devip: pick less surprising interface address in header for incoming UDP packets
devlml: use 64-bit physical addresses
devmnt: print chanpath for unexpected reply tag
devpccard: use 64-bit physical addresses
devpnp, devether, devusb: cast ISAConf.port to uvlong (for 32-bit implementations)
devpnp: print pci membar as 64-bit uvlong
devusb: keep isochronous ep->hz consistent with maxpkt, ntds and pollival
devusb: print Hci.port as 64-bit uvlong
devvga: use 64-bit physical addresses for framebuffer
etherbcm: handle 64-bit host addresses, use PCIWADDR() instead of PADDR()
etheriwl: add Wireless 8260 card
etheriwl: add for Intel Wireless-AC 9260
etheriwl: add support for Intel Wireless AC 8265
etheriwl: delay before crystal calibration
etheriwl: don't break controller on command flush timeout
etheriwl: remove stations *after* disabling binding quotas, drain all queues in rxoff7000()
kbdfs: add a compose sequence to type ⑨
kbdfs: allow >16 bit runes in /lib/keyboard
kernel: don't strip binaries in bootfs.paq
kernel: fix Abind cyclic reference and mounthead leaks (thanks Alex Musolino)
kernel: get rid of unused ucallocb
kernel: make segments non-executable when icache is not maintained
kernel: massive pci code rewrite
kernel: remove unused segment argument in freepte()
kernel: segflush() needs to flush tlb of other processes
nusb/lib: use fmtprint for the entire dump to be printed out
nusb/serial: add ids for FT230X (thanks mischief)
pc, pc64: implement 64-bit pci membar support
pc, pc64: remove mystery "type" bits in pcicfgrw*raw() (fixes qemu, thanks mischief)
pc, pc64: use 64-bit physical addresses for ISAConf.port
pc, pc64: use 64-bit physical addresses for vmap() and upaalloc()
pc, xen: make PAT support optional (for xen)
pc64: disable interrupts in mmuwalk() for checkmmu()
pc/ether*: use 64-bit physical addresses and check pci membar types and sizes
pc/pcmciamodem: cast i/o port to ulong
pc/pmmc: check pci membar type
pc/vga*: use 64-bit physical addresses and check pci membar types and sizes
pc/wavelan: print Ether.port as 64-bit value
sd53c8xx: use 64-bit physical addresses
sdiahci: use 64-bit physical addresses
sdiahci: accept AHCI controllers from ASMedia vendor id (thanks mischief)
sdmv50xx: use 64-bit physical addresses and check pci membar type
sdnvme: handle machines with more cpu's than submit queues (thanks mischief)
sdnvme: make sure pci membar type is memory
sdnvme: use PCIWADDR() instead of PADDR()
sdodin: use 64-bit physical addresses and check pci membar type
uartaxp: use 64-bit physical addresses and check pci membar types
usbehcipc: use 64-bit physical addresses and check pci membar type
usbuhci: make sure pci membar type is i/o
usbxhci: implement isochronous in transfers (for webcam, audio recording)
usbxhci: use 64-bit physical addresses
wifi: add packet timestamping support
COMPILERS AND DEBUGGERS
-----------------------
7c: fix wrong type on OASxxx operations
cc: promote integer constants according to c99 spec.
cpp: fix mutually recursive macros
LIBRARIES
---------
ape/ctype.h: add isblank, fix functions (thanks staalmannen)
ape/libdraw: add missing eenter (thanks phil9)
ape/limits.h: fix typo in (U)LLONG_MAX
ape: add missing rerrstr.c
ape: make libplumb available under ape.
ape: simplify mkfile (thanks amavect)
ape: stdio.h: correct return type of putc
ape: stdio: fix warnings, make code more standard
libaml: fix fault when the second operand of comparison cannot be converted
libc: introduce new time and date apis.
libc: ignore '?' in date format strings
libc: make yday 0-based, as docs suggest
libc: tmparse: should ignore leading whitespace
libc: tmparse: put in local timezone hack
libc: tm2sec: clear new fields in tm
libndb: order subnets by prefix length for ndbipinfo() lookups
libplumb: add missing mkfile
libttf: check directory() result
libttf: fix cmap subtable offset type and rank UCS-4 higher (more codepoints). fixes runes > 0xffff
PROGRAMS
--------
abaco: add fonts.h to HFILES in mkfile (thanks eekee)
acme/win: add trailing space to window tag
acme/win: fix window recreation command
acme: add missing logf.c
acme: implement 'scratch' ctl command (thanks Drew DeVault)
acme: implement position-dependent scroll-wheel scrolling
acme: import changes from plan9port (thanks jxy)
acme: import event log from plan9port (thanks fshahriar)
audio/flacdec: add -s SECONDS option to seek before decoding
audio/mp3dec: add -s SECONDS option
audio/mp3dec: mad timer duration is all wrong, use samples instead
audio/oggdec: add -s SECONDS option to seek before decoding
audio/oggdec: allow -s 0
audio/oggdec: give it enough chance to seek from the start, where first seek might be still too close to the beginning of the file
aux/cpuid: flush instruction cache of after patching trampoline
aux/getflags: fix named rune arguments.
aux/ms2, aux/na: fix warnings
aux/trampoline: Implement inactivity timeout (-t option)
cwfs: update mtime and qid.vers for directory on rename
date: use new libc date apis.
dc: increase exponent limit (thanks unboe, lyndon)
deroff: fix out-of-bounds access if runes above 0X80 are inside EQ clauses (thanks mmnmnnmnmm, via plan9port)
diff: support unified diff via -u
faces: remove log support
g: add '.hs' and 'mkfile' to the walk. (thanks joe9)
games/4s: increase thread size (thanks majiru)
games/nes: workaround for truncated chr
hjfs: update mtime and qid.vers for directory on rename
htmlroff: fix out of bounds access (thanks Rei-sen, via plan9port)
imap4d: fix missing return in %δ format of Dfmt()
imap4d: fix mkfile to have the correct default target
imap4d: get rid of unixdate/unixfrom handling
imap4d: reject invalid month in date2tm()
imap4d: respect errors from read(), remove debug prints
ip/6in4: request ipv4 packets only in ipmux filter
ip/cifsd: add FIND_FIRST2 SMB_FIND_FILE_FULL_DIRECTORY_INFO info level
ip/ftpd: handle "allo" command by treating it as a no-op
ip/httpd: fix 'mk nuke'
ip/tinc: run script sub-shell in its own environment
ip/torrent: fix size check in 64-bit "v" unpack (thanks pr)
ip/torrent: try harder allocating ports
ndb/dns: allow multiple txt, nullrr, cert, key and sig records (thanks kvik)
ndb/dns: mark ns record authoritative when in our area for delegation
ndb/dnsdebug: add -c flag to debug caching dns server behaviour
nedmail: don't try to shoot down subcommand on interrupt
png: remove bogus chanlen warning
pre-lib9p servers: fix incorrect Tversion handling
ptrap: fix /env/status always being set on main process exit
qball: remove unused #include <stdio.h>
ramfs: make Elocked[] more consistent (thanks fazlul)
rc-httpd: fix invalid test(1) invocation in dir-index
rc: avoid forking for final command that has variable assignments (to get $apid right)
rc: avoid stat calls for directory globbing
replica: fix mkfile script installation (thanks Amavect)
rio: move the code for 'send' into a function
seconds: use new libc date api
seconds: tolerate trailing whitespace in dates
snoopy: add ipmux pseudo protocol
sshfs: update usage text to match man page
sshnet: prevent ssh process from keeping the mount alive
syscall: utility overhaul
syscall: don't append '\n' to the output
syscall: fix build problem caused by stale tab.h
upas/fs: extract proper date from unix header
upas/fs: fix handing of multi-line header fields (thanks theinicke)
upas/fs: fix memory leak in ref822()
upas/fs: fix truncation of plumb date
upas/fs: fix wrong nparts field index (changed in previous commit)
upas/fs: move unixheader handling to parseheaders()
upas/fs: port date parsing to libc apis
upas/fs: put date822 into the index, fix from and replyto handling
upas/fs: remove now unused Mtrunc mimeflags constant
upas/fs: remove unused function date822tounix
upas/fs: use memchr() instead of strchr() in hdrlen()
upas/fs: wait until the index becomes unlocked
upas/fs: make imap debug logging less noisy
upas/marshal: make attachment failure fatal
upas: appendfolder(): don't be picky about the timezone of ctime()
upas: appendfolder(): skip the address from unix header before date
upas: convert to tmdate, change timezone format
upas: fix appendfolder timestamps (thanks umbraticus)
window: fix error exit with the -m flag
vmx: add Kmod4 key
vmx: does not do well with certain fb widths, so restrict it
vmx: set xstart to either nsec or cycles depending on what is available
vmx: use _actual_ system kbmap
vmx: use _tos->cyclefreq (thanks cinap)
vmx: use cycles() instead of nsec() when possible
vncv: implement rfb 3.8 protocol (thanks Iruatã)
DOCUMENTATION
-------------
9p(2): fix typo
acme(1): document log file in manpage
audio(1): add new -s SECONDS option
ctime(2), tmdate(2): document ctime/tmparse incompatibility
chdir(2): add reference to getwd(2) in SEE ALSO section
cifsd(8): fix typo (thanks senthil)
faces(1): remove bespoke date parser
getwd(2): add reference to chdir(2) in SEE ALSO section
graphics(2): fix typo
ip(3): fix typo in manpage (thanks jstsmthrgk)
mk9660(8): fix reference to proto file format
tmdate(2): document tz field
tmdate(2): fix missing arg, fix formatting (thanks joe9)
tmdate(2): fix examples, stale references (thanks deuteron)
tmdate(2): fix mistakes in prototype (thanks moody)
tmdate(2): fix name of tm struct
tmdate(2): fix typo (thanks mveety)
torrent(1): update url in examples section
walk(1): clarify how -n argument works (thanks umbraticus)
walk(1): improve wording, formatting
walk(1): fix typos
OTHER
-----
/lib/rob, /lib/rsc, /lib/theo: So it sounds like everyone is in favor of the entire generics proposal and all the semantics, and all we have left to hammer out is the bracket characters? Do I have that right?
bullshit: decentralized ActivityPub
![back](http://9front.org/img/emailschaden.back.png)
^ permalink raw reply [flat|nested] 2+ messages in thread
[parent not found: <0230698894072CDFA179F53B8EE1DA03@ewsd.inri.net>]
* Re: 9FRONT "EMAILSCHADEN" RELEASED
[not found] <0230698894072CDFA179F53B8EE1DA03@ewsd.inri.net>
@ 2020-10-19 22:38 ` Anthony Martin
0 siblings, 0 replies; 2+ messages in thread
From: Anthony Martin @ 2020-10-19 22:38 UTC (permalink / raw)
To: 9front
sl@stanleylieber.com once said:
> pc, pc64: remove mystery "type" bits in pcicfgrw*raw() (fixes qemu, thanks mischief)
In changeset 78891f472fbf, cinap wrote:
the access functions for pci config space in config mode #1
used to set bit 0 in the register offset if the access was
to a device on any bus different from 0.
it is completely unclear why this was done and i can't find
any documentation on this.
but for sure, this breaks all pci config spacess access to
pci devices behind a bridge on qemu. with -trace pci* it
was discovered that all config space register offsets on
devies behind pci brige where off by one.
Our code looks like it was a slight misreading of the specification
and the change looks good. However, the original behavior should have
been innocuous for any conforming host bridge:
For Type 1 translations, the host bridge directly copies the
contents of the CONFIG_ADDRESS register (excluding bits 31 and
0) onto the PCI AD lines during the address phase of a
configuration transaction making sure that AD[1::0] is "01".
(from PCI Local Bus Specification, Revision 3.0, §3.2.2.3.2)
Note the phrase "excluding bits 31 and 0". Also, the PCI-PCI bridge
in QEMU should be doing a Type 1 to Type 0 translation before the
device behind the bridge sees the address.
Cheers,
Anthony
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2020-10-19 22:39 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-10-19 4:03 9FRONT "EMAILSCHADEN" RELEASED sl
[not found] <0230698894072CDFA179F53B8EE1DA03@ewsd.inri.net>
2020-10-19 22:38 ` Anthony Martin
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).