drm/amdgpu: free the userptr ranges when a CS fails after parser_bos
From Jesse Zhang
6ac46ae5ac561d4fc7d4246ad83d71bed0d9af9a in linux-6.18.y/6.18.59
1b4da805a0ba354a549ce07459c871c17dd94fe0 in mainline linux
drm/i915/dmc: Harden DMC firmware payload parsing
From Dibin Moolakadan Subrahmanian
dfcfe2874e7c16e8460dc86ac1eeb54ea0e7f8aa in linux-6.18.y/6.18.59
8bb1858db453cd4250e3b633f5d2d09206024ca6 in mainline linux
drm/amdgpu: keep the eviction fence until the last handle is closed
From Jesse Zhang
dce8f7f03a47b1744dccf506268f9229b67fab47 in linux-6.18.y/6.18.59
3c44cbcec5eb370713bbd2c315ed7c34d8ccec2b in mainline linux
drm/amdgpu: Fix VCE 3 ring align_mask
From David Rosca
61f376654fb9d8c35896f3d5e85b95994513f647 in linux-6.18.y/6.18.59
2ee9836545e690c9e66ec203445d7705959fd7a8 in mainline linux
drm/i915/dmc: Add sanity check for DMC load address
From Dibin Moolakadan Subrahmanian
ef82b4d5cb4a3fe77c7f99008d78b8c64709e258 in linux-6.18.y/6.18.59
7dedc5bb11f95e96fe80757c924d8d9aa7327759 in mainline linux
drm/exec: don't immediately add prelocked obj to array of locked objs
From Christian Koenig
96d9996e5dbcfabed052c0d9377a382c95fed607 in linux-6.18.y/6.18.59
26401a4999f1596a515b2e704e7066ac09306488 in mainline linux
drm/amdgpu: report the BO alignment in bytes in GEM_LIST_HANDLES
From Jesse Zhang
07002a326c7cc03fd11be10e9cbc828aa4587908 in linux-6.18.y/6.18.59
d65a459dbc9a6c3b051e22c96f73598e34b31bfc in mainline linux
drm/amdgpu: don't drop the runtime PM reference twice in amdgpu_gpr
From Jesse Zhang
652b1f26125fe48aed5f4944698a85b457eef491 in linux-6.18.y/6.18.59
de9f6bcdb44f0e410031002e2f103826ea6b6259 in mainline linux
drm/amdgpu/sdma3: Work around POLL_REGMEM / FENCE hang
From Timur Kristof
05507b9b19f95ce1ca1f5fc5d394ae1f4c5b59e2 in linux-6.18.y/6.18.59
752480ad62964edf0254e151110d3e53a7cb37de in mainline linux
drm/amd/pm: add zero fan stop temperature support on smu 14.0.2
From Kenneth Feng
6cfc4a25d4f02935977a6a45a60e7a0c966e38c6 in linux-6.18.y/6.18.59
a2eae9f4fa1b07f8e6e61b48931228da051943f0 in mainline linux
drm/ttm: fix swapped-out resources never leaving their bulk_move range
From Vadim Nikitushkin
b48ebe399b4e72baa323462580fdd83ebda561e1 in linux-6.18.y/6.18.59
3db7d7d583419f7b1f2e141e36418802dbb25cf8 in mainline linux
Use loadfirmware(9) to load the rge(4) microcode from /etc/firmware
instead of hard-coding a gigantic firmware into the kernel.
suggested by deraadt@
ok mlarkin@
Update to iodine-0.9.0
With this release I couldn't reproduce the client/server issues I had
with 0.8. Drop our patches to force the use of arc4random(3) instead of
srand(3)/rand(3) - not sure if they were still useful anyway since our
rand(3) doens't respect srand(3) since some years already. Now upstream
detects and uses arc4random etc. This release also switches to meson
instead of a Makefile.
Rev 1.153 moved the initialization of buckstring[] from lazily at first
sysctl kern.malloc.bucket retrieval, to kmeminit(), but forgot to remove the
buckstring_init multiple initialization guard. Remove it.
Extremely old PA-1.0 processors did not have integer multiply and divide
instructions, and relied upon SFU (coprocessor) instructions to do so.
The floating-point emulation code written by HP for OSF/1 had support to
emulate these instruction as well.
That code eventually made its way to Utah' HPBSD, and later to
PA-RISC MkLinux; but both code bases had the appropriate logic to check for
these SFU instructions and dispatch to the emulation code commented out,
because none of these code bases were supporting PA-1.0 (PCX or older)
processors anymore.
When this code was carried away in OpenBSD, the unneeded dispatch logic was
left out, but it was not noticed that this had caused some code in spmath/
to become dead code. Let it rot in the Attic/ now.
