Apply changes between 2026b and 2026c
Release 2026c - 2026-07-08 10:23:58 -0700
Briefly:
Alberta moved to permanent -06 on 2026-06-18.
Morocco moves to permanent +00 on 2026-09-20.
More integer overflow bugs have been fixed in zic.
Changes to future timestamps
Alberta's 2026-03-08 spring forward was its last foreseeable clock
change, as it moved to permanent -06 thereafter. (Thanks to Roozbeh
Pournader and others.) Model this with its traditional abbreviation
CST. Although the change to permanent -06 legally took place on
2026-06-18, temporarily model the change to occur on 2026-11-01 at
02:00 instead, for the same reason we introduced a similarly
temporary hack for British Columbia in 2026b.
[45 lines not shown]
sh - read - handle read errors in read builtin
It used to be (incorrectly) that a read error was treated
the same as EOF in the read builtin utility. They should
be errors, with a message, and exit status > 1. Now they are.
Ooops, previous commit log should have been:
Fix conditional avaiability of code to convert between partitioning
schemes. It is needed if extended partitioning is available
(where we have a menu entry to invoke it manually) or we
compile in the "this disk is too big for disklabel" checks.
Define a "BI_MACH_FDT" machine type. The value for this record type
is 0x46445400 ('FDT\0') and unlikely to collide with any other new
machine types that Linux/m68k might decide to allocate.
This machine type has 3 machine-type-specific bootinfo record types:
- BI_FDT_PLATFORM -- a C string providing more detail about the specific
machine / platform.
- BI_FDT_BLOB -- a data record containing a Flattened Device Tree blob.
- BI_FDT_ELF_SYMS -- a memory record pointing to the location of any
loaded ELF symbols, assumed to be contiguous with the kernel +
bootinfo data. This record is needed early, possibly before virtual
memory has been bootstrapped, so that the symbols can be preserved.
Support 8-bit PIO if the front-end indicates that it's necessary:
- In wdc_drvprobe(), if neither ATAC_CAP_DATA16 nor ATAC_CAP_DATA32
is set in the controller capabilities, then attempt to enable 8-bit
PIO on the drive. If that command fails, ignore the drive.
- In wdc_datain_pio(), if we're doing 8-bit PIO, check for ATA_DRIVE_NOSTREAM
and simulate a NOSTREAM 16-bit PIO. This is necessary because we
treat the IDENTIFY command as "special" for historical (and almost
certainly bad) reasons.
- wdc_dataout_pio() gets the same NOSTREAM shenanigans for 8-bit PIO,
although I would be surprised if it is ever requested.
PR 60412: provide defines for the message files to adapt kernel names
to the various evbmips variants, so the texts can match what the
conditionalized code does.
With lots of help from gson@, thanks!
Enable radeonfb and veritefb by default on EFIKA.
The board can now boot to multi-user with wscons on graphics card.
While here, disable audio, since codec detection is flaky and tends to
cause problems.
Several fixes that help with bringup of non-POSTed Radeon cards.
- Stop preserving sync-disable bits in CRTC_EXT_CNTL
- Unblank path also clears HSYNC_DIS/VSYNC_DIS (for CRTC0)
- Reordered the TMDS transmitter enable after BIOS init
- Skip the FP_SEL_CRTC2 "iBook/XVR-100" restore hack when a BIOS was found
- Program TMDS_PLL_CNTL for internal-TMDS ports
- Skip phantom ports when building the CRTC list (for single-port RV100)
- Pixel clock ungating where appropriate
Tested on RV100 and RV280 with RADEONFB_BIOS_INIT.
Ok macallan@.
various updates for new versions.
define NO_I18N for xclock, we don't have uselocale(3) that it demands.
include xorg-pkg-ver.mk in a few places.
xdpyinfo has present support.