RFE: Sync to Linux 6.12 syscall definitions #435
Conversation
xen0n
force-pushed
the
linux-6.11-syscalls
branch
from
aecd5e9
to
f36e149
Compare
xen0n
changed the title
xen0n
force-pushed
the
linux-6.11-syscalls
branch
from
f36e149
to
f4afd16
Compare
|
After the change, |
this is really helpful for LoongArch, thanks. |
|
Hi maintainers,
Please review. |
|
Could you review this PR? Thanks. |
https://github.com/seccomp/libseccomp?tab=readme-ov-file#release-process TL;DR: When it's ready. |
| #else | ||
| #define __SNR_map_shadow_stack __PNR_map_shadow_stack | ||
| #endif | ||
|
|
There was a problem hiding this comment.
Isn't map_shadow_stack() defined for all the arches?
There was a problem hiding this comment.
Yes it is, but in this commit the PNR was allocated anyway and the change is already shipped in v2.5.5. So if we're to consider PNR values as part of the public API we would have to keep __PNR_map_shadow_stack defined, and for stylistic consistency we would have to keep this #ifdef-fery as well.
|
Just one small comment/clarification/question above, but otherwise I think this looks good. Thanks for all your work on this @xen0n! |
include/seccomp-syscalls.h
Outdated
| @@ -280,6 +280,9 @@ | |||
| #define __PNR_atomic_barrier -10246 | |||
| #define __PNR_atomic_cmpxchg_32 -10247 | |||
| #define __PNR_getpagesize -10248 | |||
| #define __PNR_map_shadow_stack -10249 | |||
There was a problem hiding this comment.
FTR: The PNR numbers below -10244 (__PNR_memfd_secret) have already diverged between libseccomp:main and libseccomp:release-2.5.
There was a problem hiding this comment.
As a FYI, I'm okay with fixing that sort of thing while merging the PR.
There was a problem hiding this comment.
I have preserved the __PNR_map_shadow_stack position while rebasing.
Apart from de-duplication of logic, this refactor is also going to help syncing to the Linux 6.11+ definitions, where all architectures are converted to source their syscall definitions from syscall.tbl files. The change is tested on Linux 6.2 sources to not affect the generated syscalls.csv apart from timestamp changes. Signed-off-by: WANG Xuerui <git@xen0n.name>
The aarch64, loongarch64 and riscv64 architectures have their syscall table sources changed to scripts/syscall.tbl, from the original inclusion of asm-generic/unistd.h. Make the script recognize the new format for these architectures. Signed-off-by: WANG Xuerui <git@xen0n.name>
Due to the addition of fstat & newfstatat to the LoongArch syscall ABI, tests 38 and 55 have to be updated for the changed syscall numbers. As for the PNR additions, normally they are allocated alphabetically for the syscalls introduced between updates of the table, but in the v2.5 release branch -10245 is already assigned to map_shadow_stack in commit 53267af ("all: update the syscall table for Linux v6.7-rc3"). While the map_shadow_stack syscall is in fact available across all architectures, for consistency with v2.5.5 and later it is kept in the same position in this update. Signed-off-by: WANG Xuerui <git@xen0n.name>
xen0n
force-pushed
the
linux-6.11-syscalls
branch
from
f4afd16
to
d909801
Compare
xen0n
changed the title
|
This is now rebased and updated to reflect Linux v6.12-rc5 syscall info (no new syscalls since 6.11). I have also incorporated @rusty-snake's review comment of keeping PNR allocations consistent with v2.5.x. Tests still pass on aarch64, loongarch64, riscv64 and x86_64. |
The
arch-syscall-validatescript updates will need to be backported to v2.5.x later, for continuing to sync aarch64 and riscv64 that got refactored to source the syscall definitions from asyscall.tblinstead ofasm-generic/unistd.h.