Script 'mail_helper' called by obssrc
Hello community,

here is the log from the commit of package limine for openSUSE:Factory checked 
in at 2026-10-01 16:48:30
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/limine (Old)
 and      /work/SRC/openSUSE:Factory/.limine.new.1253 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Package is "limine"

Thu Oct  1 16:48:30 2026 rev:50 rq:1381838 version:12.9.1

Changes:
--------
--- /work/SRC/openSUSE:Factory/limine/limine.changes    2026-09-18 
22:08:45.984195615 +0200
+++ /work/SRC/openSUSE:Factory/.limine.new.1253/limine.changes  2026-10-01 
16:50:04.825669026 +0200
@@ -1,0 +2,45 @@
+Thu Oct  1 07:38:29 UTC 2026 - Marvin Friedrich <[email protected]>
+
+- Update to 12.9.1:
+  * Reveal the menu on any key pressed during a `quiet` countdown, as
+    CONFIG.md describes, without acting on the key. A key the menu does
+    not handle previously left a blank screen with autoboot cancelled.
+  * Show the menu in `quiet` mode wherever automatic boot is disabled,
+    whether by `timeout: no` or by a Boot Loader Interface `menu-force`
+    or one-shot timeout of 0. It was previously never drawn, leaving the
+    firmware's screen up.
+  * ISO 9660: Refuse a multi-extent file whose continuation record is
+    missing or belongs to another file, rather than returning it
+    truncated. Fixes #665.
+  * Stretch wallpapers without losing horizontal precision. The rightmost
+    columns of an image could be cut off, and one narrower than 1/64 of
+    the screen showed only its first column. Fixes #669.
+  * Keep the addend of a 64-bit absolute relocation against an undefined
+    weak symbol, and on aarch64 that of a GLOB_DAT or JUMP_SLOT one too,
+    which AAELF64 defines as S + A. Fixes #655.
+  * aarch64, loongarch64: Stop marking the mappings handed to the
+    executable global, as x86 never has. Global TLB entries match under
+    every ASID, so they could outlive an ASID switch.
+  * loongarch64: Declare that csrwr overwrites its register operand,
+    which the compiler otherwise assumed still held the value written.
+  * loongarch64: Set the TLB refill page size to 4KiB at handoff, rather
+    than to the page size the firmware left in STLBPS.
+  * Limine protocol/loongarch64: Only set the NX bit on CPUs implementing
+    execution protection, without which the bit is undefined and `PF_X`
+    cannot be honoured.
+  * x86: Branch right after enabling paging on every switch into or out
+    of long mode, as the Intel and AMD manuals call for.
+  * x86: Load a 32-bit TSS on each AP before it activates long mode. AMD
+    documents a 16-bit TSS as the INIT state, and Intel faults the
+    activation with one loaded.
+  * x86: Write the I/O APIC index register as a single byte, its width on
+    chipsets such as Intel's ICH10.
+  * Host tool: Reject excess positional arguments to `bios-install`,
+    which installed to the last partition index given, and to
+    `enroll-config`, which ignored them. Fixes #659.
+  * Host tool: Undo a `bios-install` and fail if its uninstall data
+    cannot be saved. The install was kept and reported as a success, and
+    a rerun would have recorded the installed sectors as the originals.
+  * Miscellaneous bug fixes.
+
+-------------------------------------------------------------------

Old:
----
  limine-12.9.0.tar.gz

New:
----
  limine-12.9.1.tar.gz

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Other differences:
------------------
++++++ limine.spec ++++++
--- /var/tmp/diff_new_pack.YD7nsm/_old  2026-10-01 16:50:05.571699012 +0200
+++ /var/tmp/diff_new_pack.YD7nsm/_new  2026-10-01 16:50:05.576699213 +0200
@@ -15,7 +15,7 @@
 #
 
 Name:           limine
-Version:        12.9.0
+Version:        12.9.1
 Release:        0
 Summary:        Modern, advanced, portable, multiprotocol bootloader and boot 
manager
 License:        BSD-2-Clause

++++++ limine-12.9.0.tar.gz -> limine-12.9.1.tar.gz ++++++
++++ 3289 lines of diff (skipped)

Reply via email to