Re: [PATCH v2 0/2] sparc64: D-cache alias flushing fixes

[John Paul Adrian Glaubitz]

Hi Imre,

On Mon, 2026-09-28 at 15:21 +0200, Imre Kaloz wrote:
> Two fixes to D-cache alias flushing on sparc64 CPUs with a virtually
> indexed L1 D-cache.
> 
> Patch 1 makes move_pte() and tlb_batch_add() flush only the page a PTE
> maps, instead of every page of its folio once per PTE, so the number
> of cross-calls for an mremap() of a large folio no longer grows with
> the square of its size.
> 
> Patch 2 implements flush_cache_vmap() and flush_cache_vunmap(), which
> have been no-ops on sparc64.  Stale lines left by a vmalloc mapping
> corrupt BPF programs and module data on UltraSPARC III.  It uses
> flush_dcache_page_all() as patch 1 restores it.
> 
> Both patches are tagged for stable.  Tested on a Sun Ultra 45.
> 
> Changes in v2:
> - Patch 2: flush_cache_vmap() also runs on the vmemmap, from
>   sparse_init() before init_IRQ() sets up the cross-call mondo
>   blocks; skip addresses outside vmalloc and module space, since the
>   vmemmap has no aliases to flush.  Picked up Stian Halseth's
>   Tested-by.
> 
> Imre Kaloz (2):
>   sparc64: flush only the aliased page, not its whole folio
>   sparc64: flush vmalloc ranges from the D-cache on map and unmap
> 
>  arch/sparc/include/asm/cacheflush_64.h |  9 ++--
>  arch/sparc/include/asm/pgtable_64.h    |  7 ++-
>  arch/sparc/kernel/smp_64.c             | 42 ++++++++++++-----
>  arch/sparc/mm/init_64.c                | 65 ++++++++++++++++++++++++++
>  arch/sparc/mm/tlb.c                    |  2 +-
>  arch/sparc/mm/ultra.S                  | 25 ++++++++++
>  6 files changed, 132 insertions(+), 18 deletions(-)

These two patches were actually necessary to allow me to boot Andreas' for-next
branch [1] on the V240 I'm maintaining for Debian. Without your two patches, the
mount command would fail with:

[   42.327479] hid-generic 0003:413C:2106.0001: input,hidraw0: USB HID v1.10 Keyboard [Dell Dell QuietKey Keyboard] on usb-0001:00:0a.0-1/input0
Begin: Loading essential drivers ... done.
Begin: Running /scripts/init-premount ... done.
Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done.
Begin: Running /scripts/local-premount ... done.
Begin: Will now check root file system ... fsck from util-linux 2.42.4
[/sbin/fsck.ext4 (1) -- /dev/sda2] fsck.ext4 -a -C0 /dev/sda2 
/dev/sda2: clean, 2028114/8871936 files, 27597228/35467[   43.259718] ------------[ ftrace bug ]------------
312 blocks
done[   43.334910] ftrace faulted on unknown error 
.
[   43.334915] [<00000000102ee004>] 0x102ee004
[   43.465739] Initializing ftrace call sites
[   43.519487] ftrace record flags: 2000000
[   43.570955]  (0)    
[   43.570955]  expected tramp: f854d0
[   43.645302] ------------[ cut here ]------------
[   43.705920] WARNING: kernel/trace/ftrace.c:2278 at ftrace_bug+0x2f4/0x328, CPU#0: modprobe/131
[   43.819259] Modules linked in: crc16 mbcache jbd2 hid_generic usbhid hid sr_mod sd_mod cdrom ata_generic ohci_pci ehci_pci ehci_hcd ohci_hcd pata_ali libata sym53c8xx scsi_transport_spi usbcore tg3
scsi_mod libphy scsi_common usb_common mdio_bus
[   44.105327] CPU: 0 UID: 0 PID: 131 Comm: modprobe Not tainted 7.3.0-rc1+ #1 VOLUNTARY 
[   44.209473] Call Trace:
[   44.241493] [<00000000004356ec>] dump_stack+0x8/0x18
[   44.306688] [<0000000000477448>] __warn+0xe8/0x160
[   44.369595] [<00000000004294d0>] warn_slowpath_fmt+0x34/0x74
[   44.443937] [<000000000042b0dc>] ftrace_bug+0x2f4/0x328
[   44.512563] [<0000000000578220>] ftrace_process_locs+0x420/0x620
[   44.591482] [<000000000057c87c>] ftrace_module_init+0x3c/0x80
[   44.666969] [<000000000051a544>] load_module+0x1bc4/0x23c0
[   44.739026] [<000000000051af9c>] init_module_from_file+0x9c/0x120
[   44.819089] [<000000000051b24c>] sys_finit_module+0x1cc/0x320
[   44.894577] [<0000000000406194>] linux_sparc_syscall+0x34/0x44
[   44.971222] ---[ end trace 0000000000000000 ]---
[   45.073228]               \|/ ____ \|/
[   45.073228]               "@'/ .. \`@"
[   45.073228]               /_| \__/ |_\
[   45.073228]                  \__U_/
[   45.266666] mount(130): Kernel illegal instruction [#1]
[   45.335350] CPU: 0 UID: 0 PID: 130 Comm: mount Tainted: G        W           7.3.0-rc1+ #1 VOLUNTARY 
[   45.456591] Tainted: [W]=WARN
[   45.495469] TSTATE: 0000009911001606 TPC: 00000000102ee384 TNPC: 00000000102ee388 Y: 00000000    Tainted: G        W          
[   45.645312] TPC: <ext4_get_group_desc+0x4/0x100 [ext4]>
[   45.714138] g0: 0000000000000001 g1: 000000000000043b g2: fff0000008d44000 g3: 00000000c2020000
[   45.828514] g4: fff0000007bcd800 g5: 0000000000bc5c34 g6: fff0000007d60000 g7: 0000000000020000
[   45.942891] o0: fff0000008d45800 o1: 0000000000000000 o2: 0000000000000009 o3: 0000000000001000
[   46.057267] o4: fff0000008d45800 o5: fff0000008d44000 sp: fff0000007d67011 ret_pc: 00000000103615fc
[   46.176223] RPC: <ext4_fill_super+0x1b9c/0x3000 [ext4]>
[   46.244941] l0: 0000000000000000 l1: 0000000000000011 l2: fff0000008d443f8 l3: 0000000000000000
[   46.359317] l4: fff0000008d44000 l5: 00000000103acc00 l6: fff0000008808400 l7: 0000000000000000
[   46.473689] i0: fff0000008d44000 i1: fff00000001a8e40 i2: 0000000000000000 i3: fff0000008d45800
[   46.588067] i4: 0000000000000001 i5: 0000000000020000 i6: fff0000007d67181 i7: 00000000007b8e6c
[   46.702443] I7: <get_tree_bdev_flags+0x10c/0x1c0>
[   46.764200] Call Trace:
[   46.796220] [<00000000007b8e6c>] get_tree_bdev_flags+0x10c/0x1c0
[   46.875144] [<00000000007b8f34>] get_tree_bdev+0x14/0x40
[   46.944915] [<000000001034e254>] ext4_get_tree+0x14/0x40 [ext4]
[   47.022794] [<00000000007b5840>] vfs_get_tree+0x20/0x120
[   47.092568] [<00000000007e292c>] fc_mount+0xc/0xa0
[   47.155468] [<00000000007e8c54>] path_mount+0x4b4/0xac0
[   47.224089] [<00000000007e9814>] sys_mount+0xf4/0x1a0
[   47.290430] [<0000000000406194>] linux_sparc_syscall+0x34/0x44
[   47.367059] Disabling lock debugging due to kernel taint
[   47.436829] Caller[00000000007b8e6c]: get_tree_bdev_flags+0x10c/0x1c0
[   47.521465] Caller[00000000007b8f34]: get_tree_bdev+0x14/0x40
[   47.596955] Caller[000000001034e254]: ext4_get_tree+0x14/0x40 [ext4]
[   47.680552] Caller[00000000007b5840]: vfs_get_tree+0x20/0x120
[   47.756040] Caller[00000000007e292c]: fc_mount+0xc/0xa0
[   47.824662] Caller[00000000007e8c54]: path_mount+0x4b4/0xac0
[   47.899010] Caller[00000000007e9814]: sys_mount+0xf4/0x1a0
[   47.971063] Caller[0000000000406194]: linux_sparc_syscall+0x34/0x44
[   48.053415] Caller[000001000008fea8]: 0x1000008fea8
[   48.117461] Instruction DUMP:
[   48.117465]  c4286105 
[   48.156348]  96102000 
[   48.187229]  94102009 
[   48.218112] <bc325c4f>
[   48.248992]  7c02dc45 
[   48.279874]  901221d8 
[   48.310764]  106fffaa 
[   48.341636]  ba103fea 
[   48.372517]  03040e21 
[   48.403398] 
Killed
Failed to mount /dev/sda2 as root file system.


BusyBox v1.38.0 (Debian 1:1.38.0-3) built-in shell (ash)
Enter 'help' for a list of built-in commands.

(initramfs) 

With the patches applied, I'm also no longer seeing the crash with CONFIG_CGROUP_BPF
on the V240 as reported in [1].

Tested-by: John Paul Adrian Glaubitz <glaubitz@xxxxxxxxxxxxxxxxxxx>

Adrian

> [1] https://github.com/sparclinux/issues/issues/29

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-    GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913