105
edits
Changes
Done upstream: Continue removing TIF_IA32 from uprobes & Oprofile
== Compatible applications ==
On x86_64 there are two types of compatible applications:
On x86_64 there are two types of compatible applications:
The following text uses ''compatible'' and ''32-bit'' in the meaning of ia32 applications unless otherwise specified.
The following text uses ''compatible'' and ''32-bit'' in the meaning of ia32 applications unless otherwise specified.
== Difference between native and compat applications ==
From the CPU's point of view, 32-bit compatibility mode applications differ to 64-bit application by current CS (code segment selector): if corresponding value of L-bit from flags of entry in descriptors table is set the CPU will be in 64-bit mode when this segment descriptor is being used. There are some other differences between 32 and 64-bit selectors, one can read about them [https://www.malwaretech.com/2014/02/the-0x33-segment-selector-heavens-gate.html in the article "The 0x33 Segment Selector (Heavens Gate)"]. Code selectors for both bits are defined in kernel headers as <code>__USER32_CS</code> and <code>__USER_CS</code> and corresponds to descriptors in GDT (Global Descriptors Table). One can change 64-bit mode to compatibility mode by swapping CS value (e.g., with longjump).
From the CPU's point of view, 32-bit compatibility mode applications differ to 64-bit application by current CS (code segment selector): if corresponding value of L-bit from flags of entry in descriptors table is set the CPU will be in 64-bit mode when this segment descriptor is being used. There are some other differences between 32 and 64-bit selectors, one can read about them [https://www.malwaretech.com/2014/02/the-0x33-segment-selector-heavens-gate.html in the article "The 0x33 Segment Selector (Heavens Gate)"]. Code selectors for both bits are defined in kernel headers as <code>__USER32_CS</code> and <code>__USER_CS</code> and corresponds to descriptors in GDT (Global Descriptors Table). One can change 64-bit mode to compatibility mode by swapping CS value (e.g., with longjump).
Both native and compat applications can do 32 or 64-bit syscalls.
Both native and compat applications can do 32 or 64-bit syscalls.
=== Approaches to C/R compatible applications ===
== Mixed-bitness applications ==
That's entirely possible with current kernel ABI to create mixed-bitness applications, which may be ''very'' entangled.
For example, one could set ''both'' 32-bit and 64-bit robust futex list pointers.
Or one can create multi-threaded application where some threads are executing 32-bit code, some 64-bit code.
If we ever meet application of such mixed-bitness kind, the support may be added to CRIU quite easily, but it should be done under some compile-time config as it'll add more syscalls to usual C/R where they aren't needed.
At this moment there is no plans to add such support and it's quite unlikely that we'll find such application in real world (non-syntetic test).
== Approaches to C/R compatible applications ==
C/R of compatible applications can be done differently, this section describes cons/pros of each, to address decision why C/R of 32-bit tasks done ''that'' way and not some other.
C/R of compatible applications can be done differently, this section describes cons/pros of each, to address decision why C/R of 32-bit tasks done ''that'' way and not some other.
=== Restore with exec() of 32-bit dummy binary vs from 64-bit CRIU ===
Restore of 32-bit application can be done with some daemon that runs in 32-bit mode and communicates with CRIU binary (or 32-bit CRIU subprocess).
'''Pros''':
'''Pros''':
* no kernel patches expected (not quite true: vDSO mremap() still needed support)
* no kernel patches expected (not quite true: vDSO mremap() still needed support)
* restoring becomes more complicated, and if looking forward to restoring user/pid sub-namespaces, it will be too entangled
* restoring becomes more complicated, and if looking forward to restoring user/pid sub-namespaces, it will be too entangled
* no optimized inheritance for task's properties those erase with exec()
* no optimized inheritance for task's properties those erase with exec()
* will need also another daemon for x32
=== Restore with a flag to sigreturn() or arch_prctl() ===
The initial attempt to do 32-bit C/R, was rejected by lkml community by many reasons. It should have swapped thread info flags (such as <code>TIF_ADDR32</code>/<code>TIF_IA32</code>/<code>TIF_X32</code>), unmap native 64-bit vDSO blob from process's address space and map compatible 32-bit vDSO - all according to some bit in sigframe in <code>rt_sigreturn()</code> call or some dedicated for it <code>arch_prctl()</code> call.
'''Pros''':
* Simple from the point of CRIU: just do sigreturn with a new bit set or call arch_prctl() and do sigreturn
'''Cons''':
* If 32-bit vDSO image on restored host differ from dumped (in image), need to catch task after sigreturn and make jump trampolines separately - in case of arch_prctl() simpler ([https://lkml.org/lkml/2016/6/1/425 that's why arch_prctl was in initial RFC])
* Too many points of failure for one syscall, too complicated
* Just adding a way to swap those thread info flags from userspace would result in a new races/bugs (as e.g., TASK_SIZE macro depends on TIF_ADDR32, the mmap code may do unexpected things)
After discussion in lkml, conclusion was: separate changing personality (like thread info flags) from API to map vDSO blobs, remove TIF_IA32 flag that differs 32 from 64-bit tasks and look on syscall's nature: compat syscall, x32 syscall or native syscall.
=== Seizing with two 32-bit and 64-bit parasites ===
'''Pros''':
* no 32-bit calls in 64-bit parasite and vice-versa
* no need in exit in parasite: ptrace code doesn't allow to set 32-bit regset to 64-bit task and the reverse, running parasite the same nature as task bereaves us from those limits
'''Cons''':
* need to have two/three (for x32 also) blobs for seizing
* macros in makefiles to build two parasites
* serialization of parasite's answers: arguments to parasite differ in size - serialize them, which added not nice-looking and less readable C macros
=== Current approach ===
FIXME
==== List of failed tests ====
== Needs to be done (TODO) ==
==== Bug with mmaping over 4Gb ====
=== Kernel patch for vsyscall page ===
That's emulated page, not a vma - affects only in /proc/<pid>/maps for restored process. Depends on !TIF_IA32 && !TIF_X32 - Andy got patches for disabling the emulation on per-pid basics, for now I ran tests with <code>vsyscall=none</code> boot parameter because zdtm.py checks maps before/after C/R.
=== Error dump on x32-bit app dumping ===
At this moment we'll support only compat ia32 applications, attempt to dump x32 compat binary should result in error.
=== Continue removing TIF_IA32 from uprobes & Oprofile ===
This flag should be gone as it's suggested by Andy & Oleg.
There is quite lot of work to make kernel work without it, but small gain:
the restored ia32 process will be traced by uprobes/oprofile and stuff like that.
'''Updated''': Done by [https://lore.kernel.org/all/20201004032536.1229030-1-krisman@collabora.com/T/#u patches] - that merged to v5.11
== External links ==
* [https://github.com/checkpoint-restore/criu/issues/43 github issue]
[[Category: Under the hood]]