Skip to content

x86/pv: Multiple fixes to SYSCALL/SYSENTER handling (XSA-339 followup)

Andrew Cooper xen-devel 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest3 patches 5 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/3x86/pv: Don't deliver #GP for a SYSENTER with NT setxen/arch/x86/x86_64/entry.S+0 -11 reply
2/3x86/pv: Don't clobber NT on return-to-guestxen/arch/x86/x86_64/compat/entry.S xen/arch/x86/x86_64/entry.S+3 -31 reply
3/3x86/pv: Inject #UD for missing SYSCALL callbacksxen/arch/x86/x86_64/entry.S+19 -712 replies

Reviews trailers per patch

All 3 patches have a Reviewed-by or Acked-by.

PatchJan BeulichRoger Pau Monné
1/3R-b
2/3R-b
3/3A-b

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [PATCH 0/3] x86/pv: Multiple fixes to SYSCALL/SYSENTER handling (XSA-339 followup)

Patches 1 and 2 are a consequence of trying to get the Linux x86 selftests to pass even when running under Xen.

Patches 3 and XSA-339 were further fallout from trying to put in place testing to cover all aspects of the PV fast system call entrypoints.

Patch 3 was almost an XSA itself, but was ultimately argued as not affecting any known PV guest. It turns out that this is only true because of c/s dba899de14 in 2018, which did fix a real userspace => VM DoS on NetBSD.

All fixes need backporting.

Andrew Cooper (3):
x86/pv: Don't deliver #GP for a SYSENTER with NT set x86/pv: Don't clobber NT on return-to-guest x86/pv: Inject #UD for missing SYSCALL callbacks

 xen/arch/x86/x86_64/compat/entry.S |  2 +-
 xen/arch/x86/x86_64/entry.S        | 31 +++++++++++++++++++++----------
 2 files changed, 22 insertions(+), 11 deletions(-)
Signature · 1 line
2.11.0