Revert "x86/mm/ident_map: Use gbpages only where full GB page should be mapped."
authorIngo Molnar <mingo@kernel.org>
Mon, 25 Mar 2024 10:47:51 +0000 (11:47 +0100)
committerIngo Molnar <mingo@kernel.org>
Mon, 25 Mar 2024 10:54:35 +0000 (11:54 +0100)
commitc567f2948f57bdc03ed03403ae0234085f376b7d
treec77676f421fbfe2cbb946bdb3a439f51202214a8
parent91491e5fb09624116950f9f2e1767a42e1da7868
Revert "x86/mm/ident_map: Use gbpages only where full GB page should be mapped."

This reverts commit d794734c9bbfe22f86686dc2909c25f5ffe1a572.

While the original change tries to fix a bug, it also unintentionally broke
existing systems, see the regressions reported at:

  https://lore.kernel.org/all/3a1b9909-45ac-4f97-ad68-d16ef1ce99db@pavinjoseph.com/

Since d794734c9bbf was also marked for -stable, let's back it out before
causing more damage.

Note that due to another upstream change the revert was not 100% automatic:

  0a845e0f6348 mm/treewide: replace pud_large() with pud_leaf()

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: <stable@vger.kernel.org>
Cc: Russ Anderson <rja@hpe.com>
Cc: Steve Wahl <steve.wahl@hpe.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Link: https://lore.kernel.org/all/3a1b9909-45ac-4f97-ad68-d16ef1ce99db@pavinjoseph.com/
Fixes: d794734c9bbf ("x86/mm/ident_map: Use gbpages only where full GB page should be mapped.")
arch/x86/mm/ident_map.c