BSD Commit Log Search

   Fix reversed condition in new and not yet used writefulllabel().

   vmd: fix pty and file leaks in vm creation path.

   1. During vm creation, vmd duplicates the pty fd to let the privileged
   process handle cleanup after vm termination. This dup'd fd should
   be closed in the error path otherwise it leaks.

   2. The combination of `vmctl reload` with a vm using a bios or
   kernel boot image will leak a file descriptor to the image file.
   This was a regression caused by fixing a related problem. Mischa
   Peters reported vmd reporting EMFILE errors and it turns out the
   priv process was keeping the file open, but reload was overwriting
   the variable with the fd number so teardown wasn't closing it.

   There's no need to keep it open, so rewrite to not dup(2) and don't
   have the priv process keep a copy of the fd number (the source of
   the previous issue related to a race between close(2) on cleanup
   and the imsg sending the fd to the vmm process.)

   ok hshoexer@

   rpki-client: remove lower size bound of 1000 from manifestInstance

   This is no longer specified after it resulted in various complaints
   and problems:

   https://github.com/job/draft-sidrops-rpki-ccr/commit/ebf45bfefde0b983b3c6349a4cdcf4c92d005b66

   ok job

   Prefer previous cell not next when tiling/floating, from Michael Grant.

   rpki-client: remove the BN_CTX from the parser

   Before threads were added to the parser, having a global BN_CTX was easy
   enough. While it saves many small allocations, I don't think it's worth
   the complexity.

   A bignum is a struct containing a pointer to an array of words and then
   four ints. Sequence numbers are at most 20 octets (fitting into 3 8-bit
   words on 64 bits), so each seqnum requires at most 48 bytes. We can check
   with two bignums instead of three without loss of clarity, so ultimately
   we have all this complexity to avoid allocations of at most 96 bytes.

   ok claudio job

   revert: sys/vfs: stop checkalias replacing vnodes

   Unfortently this diff uncovers for everyone a bug in swaplist_find()
   which was reported by miod@.

   To fix it with peace reverting
    - cd9660_vfsops.c,v 1.100
    - udf_vfsops.c,v 1.74
    - vfs_default.c,v 1.53
    - vfs_subr.c,v 1.338
    - nfs_subs.c,v 1.155
    - vnode.h,v 1.180
    - tmpfs_subr.c,v 1.29
    - ext2fs_subr.c,v 1.39
    - ffs_subr.c,v 1.36
    - mfs_vfsops.c,v 1.64

   repair previous commit, pointed out by Mark Patruck

   Report our short OS name as "other" or "other-64" rather than "FreeBSD".
   This makes recent versions of vcenter display the full OS name and version,
   and older versions at least indicate whether the guest is amd64 or i386.

   also tested by Ian McWilliam on fusion player and Kenneth Gober on esx 5.5
   ok dlg@

   fold bpfdesc.h into bpf.c.

   bpf.c is the only user of bpfdesc.h. or was.

   ok claudio@ jmatthew@

   drm/amdkfd: evict the queues only once for CRIU PROCESS_INFO

   From Jesse Zhang
   7f695fb47b4073d6bbd5bb9d5c4f969327641eab in linux-6.18.y/6.18.63
   8f3ecda0ab42fb5ba7310fe848013d3eff31b416 in mainline linux

   drm/amdkfd: validate SRAT subtables before parsing

   From Jiale Yao
   7f1ed723c19846fd176243acb68ecdccb22bfa72 in linux-6.18.y/6.18.63
   9866cb13ba4d39aa8fed34b765d4aecabe04f02e in mainline linux

   drm/amd/display: Initialize missing TD-LUT programming inputs

   From Avi Weiss
   48f413a3999ac9421f97c2330242dc72f72ae24f in linux-6.18.y/6.18.63
   167df3b94680b8670c4cacc08af232e3bfcab5ff in mainline linux

   drm/amdkfd: split vram_pages of SVM ranges prefetched without a GPU mapping

   From Jesse Zhang
   08a3f27f91a91fe105012ce8d910540d6a54b37d in linux-6.18.y/6.18.63
   52957e813f2f025c86e4aa6e86cb4963ef94fd00 in mainline linux

   drm/amdgpu: reject CS chunks too large for kvmalloc()

   From Jesse Zhang
   cfa7fd53ed06c8cce47390a86781316bbb9d1e78 in linux-6.18.y/6.18.63
   ba4c3c202919405f461a393e823e4b72c8518494 in mainline linux

   drm/amd/display: take a runtime PM reference only when a CRTC turns on

   From Luigi Lauro
   ec01397e6c4082f78e1a40e0baff7c8979653dda in linux-6.18.y/6.18.63
   b456f591f925fbd6b33770d0eec1b51c2f9d41c1 in mainline linux

   drm/amdkfd: stop restore retries on device unplug

   From Olivier Kaloudoff
   03bb7b94d6e408bb52c5c25d4fd7bca2c794be3c in linux-6.18.y/6.18.63
   3254ed0b15f2fe656c9cb36ccb59b92422440bf8 in mainline linux

   drm/amdkfd: correct vram_available calculation

   From Zhen Han
   58d07ddaa5f33052a29cb26ca226da281e5f3475 in linux-6.18.y/6.18.63
   9131cc85d0dedd8ba4cae60dbe5985d081551f30 in mainline linux

   amdkfd: Fix memory availability double accounting of kfd pinned objects

   From Daniel Phillips
   c2a3bef88fe10cdb286ad8414b44c7a2935017ff in linux-6.18.y/6.18.63
   16df12d7158fb55f20d6e985abd1996cbabab9f8 in mainline linux

   drm/amdgpu: wait for pending page table updates before switching a VM to CPU updates

   From John Fisher
   2368f9aed09cbabb5a8980c2fbc78bde98e170e9 in linux-6.18.y/6.18.63
   53d9ef2f518add783bb48a69d77bbc8aea99e911 in mainline linux

   drm/amdkfd: wait for a GPU reset before freeing a BO

   From William Palacek
   2cbdce3fdf929ec5c13007611a37672720bcfb8e in linux-6.18.y/6.18.63
   47d799a718439764432d8ef3a0952af45f025f76 in mainline linux

   drm/amd/display: Fix swap in median3() for DSC rate control

   From Keita Morisaki
   38bb73854a4a0d38162c1776fa1a469de897bc96 in linux-6.18.y/6.18.63
   7b23bf82ccbdb4cf0752f2b8686e7319a7ce497f in mainline linux

   drm/amdgpu/gmc9: add extra type-2 TLB flush for gfx942/gfx950

   From Elena Sakhnovitch
   b260effe83a6566fd1d4038fa5ca9bc8a43c7208 in linux-6.18.y/6.18.63
   232213c9bd7f703836164f23160b76657094a7db in mainline linux

   drm/amdkfd: don't drop an SVM CWSR reference that was never taken

   From Jesse Zhang
   1436e441ad19a475ce3fdaacee7efe670405cae9 in linux-6.18.y/6.18.63
   030d47cc6720e0f07b502876cdbd266e309f9461 in mainline linux

   drm/amd/pm: reapply umd profile state after recovery

   From Kevin Wang
   01d08b4de4b1fbac7f980643ed9b4d7460ff1412 in linux-6.18.y/6.18.63
   ad6eb088a508d927641c8a76d1ca7911dfb7c80b in mainline linux

   drm/amdgpu: reject ANONONLY userptrs that span multiple VMAs

   From Jesse Zhang
   b900cb2316ff78943f6b19360d931e312b751011 in linux-6.18.y/6.18.63
   d577b68a854d0f590f480777249b041507d0708c in mainline linux

   drm/amd/display: don't WARN when removing an unallocated MST payload

   From Mario Limonciello
   87a531b56f2b8992e6082c2e57ec034548bd11d4 in linux-6.18.y/6.18.63
   b18a7c6d75bcc883b39497863cc88c9ef4c6c85f in mainline linux

   drm/amdkfd: fix doorbell allocation race creating duplicate allocations

   From Henry Martin
   2081dc291247066cf174f9ad4b20eaffc5663dc0 in linux-6.18.y/6.18.63
   bebad45050945a3d668d6701fb77ae2571c5d44b in mainline linux

   drm/amdgpu: reset VI ASIC on Apple boards with VRAM at 0

   From Maciej Kucharek
   7c43b9c2e9f803f078e22b25fb4f0250988fa17d in linux-6.18.y/6.18.63
   92a5184787b81e67f7b831ad860ca9b053505cf8 in mainline linux

   drm/amdgpu: reset CIK ASIC on Apple boards with VRAM at 0

   From Maciej Kucharek
   6130c38c3e735215694b579927e6406bc00ad784 in linux-6.18.y/6.18.63
   095f634ea0aa5b9da19495520b5fc1a155b2c137 in mainline linux

   drm/amdgpu: reset SI ASIC on MacBookPros with VRAM at 0

   From Maciej Kucharek
   94b703ea1438cff1de0f3036a0800bbb90f66256 in linux-6.18.y/6.18.63
   5b56ede880317bd912f914d5cd38ba1d357677fa in mainline linux