Commit Graph

5 Commits

Author SHA1 Message Date
Linus Torvalds
2a8120d7b4 more s390 updates for 6.10 merge window
- Switch read and write software bits for PUDs
 
 - Add missing hardware bits for PUDs and PMDs
 
 - Generate unwind information for C modules to fix GDB unwind
   error for vDSO functions
 
 - Create .build-id links for unstripped vDSO files to enable
   vDSO debugging with symbols
 
 - Use standard stack frame layout for vDSO generated stack frames
   to manually walk stack frames without DWARF information
 
 - Rework perf_callchain_user() and arch_stack_walk_user() functions
   to reduce code duplication
 
 - Skip first stack frame when walking user stack
 
 - Add basic checks to identify invalid instruction pointers when
   walking stack frames
 
 - Introduce and use struct stack_frame_vdso_wrapper within vDSO user
   wrapper code to automatically generate an asm-offset define. Also
   use STACK_FRAME_USER_OVERHEAD instead of STACK_FRAME_OVERHEAD to
   document that the code works with user space stack
 
 - Clear the backchain of the extra stack frame added by the vDSO user
   wrapper code. This allows the user stack walker to detect and skip
   the non-standard stack frame. Without this an incorrect instruction
   pointer would be added to stack traces.
 
 - Rewrite psw_idle() function in C to ease maintenance and further
   enhancements
 
 - Remove get_vtimer() function and use get_cpu_timer() instead
 
 - Mark psw variable in __load_psw_mask() as __unitialized to avoid
   superfluous clearing of PSW
 
 - Remove obsolete and superfluous comment about removed TIF_FPU flag
 
 - Replace memzero_explicit() and kfree() with kfree_sensitive() to
   fix warnings reported by Coccinelle
 
 - Wipe sensitive data and all copies of protected- or secure-keys
   from stack when an IOCTL fails
 
 - Both do_airq_interrupt() and do_io_interrupt() functions set
   CIF_NOHZ_DELAY flag. Move it in do_io_irq() to simplify the code
 
 - Provide iucv_alloc_device() and iucv_release_device() helpers,
   which can be used to deduplicate more or less identical IUCV
   device allocation and release code in four different drivers
 
 - Make use of iucv_alloc_device() and iucv_release_device()
   helpers to get rid of quite some code and also remove a
   cast to an incompatible function (clang W=1)
 
 - There is no user of iucv_root outside of the core IUCV code left.
   Therefore remove the EXPORT_SYMBOL
 
 - __apply_alternatives() contains a runtime check which verifies
   that the size of the to be patched code area is even. Convert
   this to a compile time check
 
 - Increase size of buffers for sending z/VM CP DIAGNOSE X'008'
   commands from 128 to 240
 
 - Do not accept z/VM CP DIAGNOSE X'008' commands longer than
   maximally allowed
 
 - Use correct defines IPL_BP_NVME_LEN and IPL_BP0_NVME_LEN instead
   of IPL_BP_FCP_LEN and IPL_BP0_FCP_LEN ones to initialize NVMe
   reIPL block on 'scp_data' sysfs attribute update
 
 - Initialize the correct fields of the NVMe dump block, which
   were confused with FCP fields
 
 - Refactor macros for 'scp_data' (re-)IPL sysfs attribute to
   reduce code duplication
 
 - Introduce 'scp_data' sysfs attribute for dump IPL to allow tools
   such as dumpconf passing additional kernel command line parameters
   to a stand-alone dumper
 
 - Rework the CPACF query functions to use the correct RRE or RRF
   instruction formats and set instruction register fields correctly
 
 - Instead of calling BUG() at runtime force a link error during
   compile when a unsupported opcode is used with __cpacf_query()
   or __cpacf_check_opcode() functions
 
 - Fix a crash in ap_parse_bitmap_str() function on /sys/bus/ap/apmask
   or /sys/bus/ap/aqmask sysfs file update with a relative mask value
 
 - Fix "bindings complete" udev event which should be sent once all AP
   devices have been bound to device drivers and again when unbind/bind
   actions take place and all AP devices are bound again
 
 - Facility list alt_stfle_fac_list is nowhere used in the decompressor,
   therefore remove it there
 
 - Remove custom kprobes insn slot allocator in favour of the standard
   module_alloc() one, since kernel image and module areas are located
   within 4GB
 
 - Use kvcalloc() instead of kvmalloc_array() in zcrypt driver to avoid
   calling memset() with a large byte count and get rid of the sparse
   warning as result
 -----BEGIN PGP SIGNATURE-----
 
 iI0EABYIADUWIQQrtrZiYVkVzKQcYivNdxKlNrRb8AUCZkyx2BccYWdvcmRlZXZA
 bGludXguaWJtLmNvbQAKCRDNdxKlNrRb8PYZAP9KxEfTyUmIh61Gx8+m3BW5dy7p
 E2Q8yotlUpGj49ul+AD8CEAyTiWR95AlMOVZZLV/0J7XIjhALvpKAGfiJWkvXgc=
 =pife
 -----END PGP SIGNATURE-----

Merge tag 's390-6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull more s390 updates from Alexander Gordeev:

 - Switch read and write software bits for PUDs

 - Add missing hardware bits for PUDs and PMDs

 - Generate unwind information for C modules to fix GDB unwind error for
   vDSO functions

 - Create .build-id links for unstripped vDSO files to enable vDSO
   debugging with symbols

 - Use standard stack frame layout for vDSO generated stack frames to
   manually walk stack frames without DWARF information

 - Rework perf_callchain_user() and arch_stack_walk_user() functions to
   reduce code duplication

 - Skip first stack frame when walking user stack

 - Add basic checks to identify invalid instruction pointers when
   walking stack frames

 - Introduce and use struct stack_frame_vdso_wrapper within vDSO user
   wrapper code to automatically generate an asm-offset define. Also use
   STACK_FRAME_USER_OVERHEAD instead of STACK_FRAME_OVERHEAD to document
   that the code works with user space stack

 - Clear the backchain of the extra stack frame added by the vDSO user
   wrapper code. This allows the user stack walker to detect and skip
   the non-standard stack frame. Without this an incorrect instruction
   pointer would be added to stack traces.

 - Rewrite psw_idle() function in C to ease maintenance and further
   enhancements

 - Remove get_vtimer() function and use get_cpu_timer() instead

 - Mark psw variable in __load_psw_mask() as __unitialized to avoid
   superfluous clearing of PSW

 - Remove obsolete and superfluous comment about removed TIF_FPU flag

 - Replace memzero_explicit() and kfree() with kfree_sensitive() to fix
   warnings reported by Coccinelle

 - Wipe sensitive data and all copies of protected- or secure-keys from
   stack when an IOCTL fails

 - Both do_airq_interrupt() and do_io_interrupt() functions set
   CIF_NOHZ_DELAY flag. Move it in do_io_irq() to simplify the code

 - Provide iucv_alloc_device() and iucv_release_device() helpers, which
   can be used to deduplicate more or less identical IUCV device
   allocation and release code in four different drivers

 - Make use of iucv_alloc_device() and iucv_release_device() helpers to
   get rid of quite some code and also remove a cast to an incompatible
   function (clang W=1)

 - There is no user of iucv_root outside of the core IUCV code left.
   Therefore remove the EXPORT_SYMBOL

 - __apply_alternatives() contains a runtime check which verifies that
   the size of the to be patched code area is even. Convert this to a
   compile time check

 - Increase size of buffers for sending z/VM CP DIAGNOSE X'008' commands
   from 128 to 240

 - Do not accept z/VM CP DIAGNOSE X'008' commands longer than maximally
   allowed

 - Use correct defines IPL_BP_NVME_LEN and IPL_BP0_NVME_LEN instead of
   IPL_BP_FCP_LEN and IPL_BP0_FCP_LEN ones to initialize NVMe reIPL
   block on 'scp_data' sysfs attribute update

 - Initialize the correct fields of the NVMe dump block, which were
   confused with FCP fields

 - Refactor macros for 'scp_data' (re-)IPL sysfs attribute to reduce
   code duplication

 - Introduce 'scp_data' sysfs attribute for dump IPL to allow tools such
   as dumpconf passing additional kernel command line parameters to a
   stand-alone dumper

 - Rework the CPACF query functions to use the correct RRE or RRF
   instruction formats and set instruction register fields correctly

 - Instead of calling BUG() at runtime force a link error during compile
   when a unsupported opcode is used with __cpacf_query() or
   __cpacf_check_opcode() functions

 - Fix a crash in ap_parse_bitmap_str() function on /sys/bus/ap/apmask
   or /sys/bus/ap/aqmask sysfs file update with a relative mask value

 - Fix "bindings complete" udev event which should be sent once all AP
   devices have been bound to device drivers and again when unbind/bind
   actions take place and all AP devices are bound again

 - Facility list alt_stfle_fac_list is nowhere used in the decompressor,
   therefore remove it there

 - Remove custom kprobes insn slot allocator in favour of the standard
   module_alloc() one, since kernel image and module areas are located
   within 4GB

 - Use kvcalloc() instead of kvmalloc_array() in zcrypt driver to avoid
   calling memset() with a large byte count and get rid of the sparse
   warning as result

* tag 's390-6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (39 commits)
  s390/zcrypt: Use kvcalloc() instead of kvmalloc_array()
  s390/kprobes: Remove custom insn slot allocator
  s390/boot: Remove alt_stfle_fac_list from decompressor
  s390/ap: Fix bind complete udev event sent after each AP bus scan
  s390/ap: Fix crash in AP internal function modify_bitmap()
  s390/cpacf: Make use of invalid opcode produce a link error
  s390/cpacf: Split and rework cpacf query functions
  s390/ipl: Introduce sysfs attribute 'scp_data' for dump ipl
  s390/ipl: Introduce macros for (re)ipl sysfs attribute 'scp_data'
  s390/ipl: Fix incorrect initialization of nvme dump block
  s390/ipl: Fix incorrect initialization of len fields in nvme reipl block
  s390/ipl: Do not accept z/VM CP diag X'008' cmds longer than max length
  s390/ipl: Fix size of vmcmd buffers for sending z/VM CP diag X'008' cmds
  s390/alternatives: Convert runtime sanity check into compile time check
  s390/iucv: Unexport iucv_root
  tty: hvc-iucv: Make use of iucv_alloc_device()
  s390/smsgiucv_app: Make use of iucv_alloc_device()
  s390/netiucv: Make use of iucv_alloc_device()
  s390/vmlogrdr: Make use of iucv_alloc_device()
  s390/iucv: Provide iucv_alloc_device() / iucv_release_device()
  ...
2024-05-21 12:09:36 -07:00
Jens Remus
fc2f5f10f9 s390/vdso: Create .build-id links for unstripped vdso files
Citing Andy Lutomirski from commit dda1e95cee ("x86/vdso: Create
.build-id links for unstripped vdso files"):

"With this change, doing 'make vdso_install' and telling gdb:

set debug-file-directory /lib/modules/KVER/vdso

will enable vdso debugging with symbols.  This is useful for
testing, but kernel RPM builds will probably want to manually delete
these symlinks or otherwise do something sensible when they strip
the vdso/*.so files."

Fixes: 4bff8cb545 ("s390: convert to GENERIC_VDSO")
Signed-off-by: Jens Remus <jremus@linux.ibm.com>
Signed-off-by: Alexander Gordeev <agordeev@linux.ibm.com>
2024-05-14 13:37:06 +02:00
Masahiro Yamada
1c369b6c94 kbuild: simplify generic vdso installation code
With commit 4b0bf9a012 ("riscv: compat_vdso: install compat_vdso.so.dbg
to /lib/modules/*/vdso/") applied, all debug VDSO files are installed in
$(MODLIB)/vdso/.

Simplify the installation rule.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
2024-05-10 04:34:52 +09:00
Masahiro Yamada
c1a8627164 kbuild: fix build ID symlinks to installed debug VDSO files
Commit 56769ba4b2 ("kbuild: unify vdso_install rules") accidentally
dropped the '.debug' suffix from the build ID symlinks.

Fixes: 56769ba4b2 ("kbuild: unify vdso_install rules")
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
2023-12-23 23:24:03 +09:00
Masahiro Yamada
56769ba4b2 kbuild: unify vdso_install rules
Currently, there is no standard implementation for vdso_install,
leading to various issues:

 1. Code duplication

    Many architectures duplicate similar code just for copying files
    to the install destination.

    Some architectures (arm, sparc, x86) create build-id symlinks,
    introducing more code duplication.

 2. Unintended updates of in-tree build artifacts

    The vdso_install rule depends on the vdso files to install.
    It may update in-tree build artifacts. This can be problematic,
    as explained in commit 19514fc665 ("arm, kbuild: make
    "make install" not depend on vmlinux").

 3. Broken code in some architectures

    Makefile code is often copied from one architecture to another
    without proper adaptation.

    'make vdso_install' for parisc does not work.

    'make vdso_install' for s390 installs vdso64, but not vdso32.

To address these problems, this commit introduces a generic vdso_install
rule.

Architectures that support vdso_install need to define vdso-install-y
in arch/*/Makefile. vdso-install-y lists the files to install.

For example, arch/x86/Makefile looks like this:

  vdso-install-$(CONFIG_X86_64)           += arch/x86/entry/vdso/vdso64.so.dbg
  vdso-install-$(CONFIG_X86_X32_ABI)      += arch/x86/entry/vdso/vdsox32.so.dbg
  vdso-install-$(CONFIG_X86_32)           += arch/x86/entry/vdso/vdso32.so.dbg
  vdso-install-$(CONFIG_IA32_EMULATION)   += arch/x86/entry/vdso/vdso32.so.dbg

These files will be installed to $(MODLIB)/vdso/ with the .dbg suffix,
if exists, stripped away.

vdso-install-y can optionally take the second field after the colon
separator. This is needed because some architectures install a vdso
file as a different base name.

The following is a snippet from arch/arm64/Makefile.

  vdso-install-$(CONFIG_COMPAT_VDSO)      += arch/arm64/kernel/vdso32/vdso.so.dbg:vdso32.so

This will rename vdso.so.dbg to vdso32.so during installation. If such
architectures change their implementation so that the base names match,
this workaround will go away.

Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
Acked-by: Sven Schnelle <svens@linux.ibm.com> # s390
Reviewed-by: Nicolas Schier <nicolas@fjasle.eu>
Reviewed-by: Guo Ren <guoren@kernel.org>
Acked-by: Helge Deller <deller@gmx.de>  # parisc
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
2023-10-28 21:09:02 +09:00