Changes in Kconfig

d1d7f01f - um: mark rodata read-only and implement _nofault accesses

Mon, 10 Feb 2025 16:09:25 +0000

um: mark rodata read-only and implement _nofault accessesMark read-only data actually read-only (simple mprotect), andto be able to test it also implement _nofault accesses. Thisworks by setting up a new "segv_continue" pointer in current,and then when we hit a segfault we change the signal returncontext so that we continue at that address. The code usingthis sets it up so that it jumps to a label and then abortsthe access that way, returning -EFAULT.It's possible to optimize the ___backtrack_faulted() thing byusing asm goto (compiler version dependent) and/or gcc's (notsure if clang has it) &&label extension, but at least in oneattempt I made the && caused the compiler to not load -EFAULTinto the register in case of jumping to the &&label from thefault handler. So leave it like this for now.Signed-off-by: Johannes Berg Co-developed-by: Benjamin Berg Signed-off-by: Benjamin Berg Link: https://patch.msgid.link/20250210160926.420133-2-benjamin@sipsolutions.netSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

2f681ba4 - um: move thread info into task

Mon, 11 Nov 2024 10:29:10 +0000

um: move thread info into taskThis selects the THREAD_INFO_IN_TASK option for UM and changes the waythat the current task is discovered. This is trivial though, as UMLalready tracks the current task in cpu_tasks[] and this can be used toretrieve it.Also remove the signal handler code that copies the thread informationinto the IRQ stack. It is obsolete now, which also means that thementioned race condition cannot happen anymore.Signed-off-by: Benjamin Berg Reviewed-by: Hajime Tazaki Link: https://patch.msgid.link/20241111102910.46512-1-benjamin@sipsolutions.netSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

3f17fed2 - um: switch to regset API and depend on XSTATE

Wed, 23 Oct 2024 09:41:20 +0000

um: switch to regset API and depend on XSTATEThe PTRACE_GETREGSET API has now existed since Linux 2.6.33. The XSAVECPU feature should also be sufficiently common to be able to rely on it.With this, define our internal FP state to be the hosts XSAVE data. Adddiscovery for the hosts XSAVE size and place the FP registers at the endof task_struct so that we can adjust the size at runtime.Next we can implement the regset API on top and update the signalhandling as well as ptrace APIs to use them. Also switch coredumpcreation to use the regset API and finally set HAVE_ARCH_TRACEHOOK.This considerably improves the signal frames. Previously they might nothave contained all the registers (i386) and also did not have thesizes and magic values set to the correct values to permit userspace todecode the frame.As a side effect, this will permit UML to run on hosts with newer CPUextensions (such as AMX) that need even more register state.Signed-off-by: Benjamin Berg Link: https://patch.msgid.link/20241023094120.4083426-1-benjamin@sipsolutions.netSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

0b8b2668 - um: insert scheduler ticks when userspace does not yield

Thu, 10 Oct 2024 14:25:37 +0000

um: insert scheduler ticks when userspace does not yieldIn time-travel mode userspace can do a lot of work without any timepassing. Unfortunately, this can result in OOM situations as the RCUcore code will never be run.Work around this by keeping track of userspace processes that do notyield for a lot of operations. When this happens, insert a jiffie intothe sched_clock clock to account time against the process and cause thebookkeeping to run.As sched_clock is used for tracing, it is useful to keep it in syncbetween the different VMs. As such, try to remove added ticks again whenthe actual clock ticks.Signed-off-by: Benjamin Berg Link: https://patch.msgid.link/20241010142537.1134685-1-benjamin@sipsolutions.netSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

41ab5fe7 - um: Switch to 4 level page tables on 64 bit

Thu, 19 Sep 2024 12:45:11 +0000

um: Switch to 4 level page tables on 64 bitThe larger memory space is useful to support more applications insideUML. One example for this is ASAN instrumentation of userspaceapplications which requires addresses that would otherwise not beavailable.Signed-off-by: Benjamin Berg Link: https://patch.msgid.link/20240919124511.282088-11-benjamin@sipsolutions.netSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

48a858e0 - um: remove dependency on undefined CC_CAN_LINK_STATIC_NO_RUNTIME_DEPS

Tue, 24 Sep 2024 11:33:40 +0000

um: remove dependency on undefined CC_CAN_LINK_STATIC_NO_RUNTIME_DEPSCC_CAN_LINK_STATIC_NO_RUNTIME_DEPS is not defined anywhere.In the submitted patch set [1], the first patch "um/kconfig: introduceCC_CAN_LINK_STATIC_NO_RUNTIME_DEPS" was not applied.Only 2/3 and 3/3 were applied, which are now: - 730586ff7fad ("um: Allow static linking for non-glibc implementations") - 5e1121cd43d4 ("um: Some fixes to build UML with musl")Given that nobody has noticed the missing first patch for many years,it seems it was unnecessary.[1]: https://lore.kernel.org/lkml/20200719210222.2811-1-ignat@cloudflare.com/Signed-off-by: Masahiro Yamada Link: https://patch.msgid.link/20240924113342.32530-1-masahiroy@kernel.orgSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

64dcf0b8 - um: remove ARCH_NO_PREEMPT_DYNAMIC

Tue, 23 Jul 2024 12:24:56 +0000

um: remove ARCH_NO_PREEMPT_DYNAMICThere's no such symbol and we currently don't have any of themechanisms to make boot-time selection cheap enough, so we can'thave HAVE_PREEMPT_DYNAMIC_CALL or HAVE_PREEMPT_DYNAMIC_KEY.Remove the select statement.Reported-by: Lukas Bulwahn Fixes: cd01672d64a3 ("um: Enable preemption in UML")Signed-off-by: Johannes Berg Signed-off-by: Richard Weinberger List of files: /linux-6.15/arch/um/Kconfig

cd01672d - um: Enable preemption in UML

Tue, 02 Jul 2024 08:25:50 +0000

um: Enable preemption in UMLSince userspace state is saved in the MM process, kernel usingFPU still doesn't really need to do anything, so this reallyis as simple as enabling preemption. The irq critical sectionin sigio_handler() needs preempt_disable()/preempt_enable().Signed-off-by: Anton Ivanov Link: https://patch.msgid.link/20240702102549.d2fcea450854.I12f5a53d80ec1e425e66ef272b1e95cb523b608e@changeid[rebase, remove FPU save/restore, fix x86/um Makefile, rewrite commit message]Signed-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

53585f9e - um: enable UBSAN

Mon, 01 Jul 2024 20:00:35 +0000

um: enable UBSANWe can select ARCH_HAS_UBSAN, it works just fine. It had beenenabled and we even used it, but then commit 890a64810d59("ubsan: Restore dependency on ARCH_HAS_UBSAN") (correctly)disabled it again, enable ARCH_HAS_UBSAN to get it.Signed-off-by: Johannes Berg Link: https://patch.msgid.link/20240701220034.995eb04d656d.Ia29fe091b207fe66b5e26298c1e427ebcf131642@changeidSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

ab0f4ced - arch: um: rust: Add i386 support for Rust

Tue, 04 Jun 2024 22:40:50 +0000

arch: um: rust: Add i386 support for RustAt present, Rust in the kernel only supports 64-bit x86, so UML hasfollowed suit. However, it's significantly easier to support 32-bit i386on UML than on bare metal, as UML does not use the -mregparm option(which alters the ABI), which is not yet supported by rustc[1].Add support for CONFIG_RUST on um/i386, by adding a new target config togenerate_rust_target, and replacing various checks on CONFIG_X86_64 toalso support CONFIG_X86_32.We still use generate_rust_target, rather than a built-in rustc target,in order to match x86_64, provide a future place for -mregparm, and moreeasily disable floating point instructions.With these changes, the KUnit tests pass with:kunit.py run --make_options LLVM=1 --kconfig_add CONFIG_RUST=y--kconfig_add CONFIG_64BIT=n --kconfig_add CONFIG_FORTIFY_SOURCE=nAn earlier version of these changes was proposed on the Rust-for-Linuxgithub[2].[1]: https://github.com/rust-lang/rust/issues/116972[2]: https://github.com/Rust-for-Linux/linux/pull/966Signed-off-by: David Gow Link: https://patch.msgid.link/20240604224052.3138504-1-davidgow@google.comSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

ddd268c4 - um: Select HAS_IOREMAP for UML_IOMEM_EMULATION

Wed, 03 Apr 2024 12:43:00 +0000

um: Select HAS_IOREMAP for UML_IOMEM_EMULATIONIn a future patch HAS_IOPORT=n will disable inb()/outb() and friends atcompile time. UML supports these via its UML_IOMEM_EMULATION so let thatselect HAS_IOPORT and also reflect this in NO_IOPORT_MAP.Co-developed-by: Arnd Bergmann Signed-off-by: Arnd Bergmann Signed-off-by: Niklas Schnelle Link: https://patch.msgid.link/20240403124300.65379-2-schnelle@linux.ibm.comSigned-off-by: Johannes Berg List of files: /linux-6.15/arch/um/Kconfig

5394f1e9 - arch: define CONFIG_PAGE_SIZE_*KB on all architectures

Mon, 26 Feb 2024 16:14:13 +0000

arch: define CONFIG_PAGE_SIZE_*KB on all architecturesMost architectures only support a single hardcoded page size. In orderto ensure that each one of these sets the corresponding Kconfig symbols,change over the PAGE_SHIFT definition to the common one and allowonly the hardware page size to be selected.Acked-by: Guo Ren Acked-by: Heiko Carstens Acked-by: Stafford Horne Acked-by: Johannes Berg Acked-by: Geert Uytterhoeven Reviewed-by: Thomas Gleixner Signed-off-by: Arnd Bergmann List of files: /linux-6.15/arch/um/Kconfig

9349b5cd - um/cpu: Switch to arch_cpu_finalize_init()

Tue, 13 Jun 2023 23:39:36 +0000

um/cpu: Switch to arch_cpu_finalize_init()check_bugs() is about to be phased out. Switch over to the newarch_cpu_finalize_init() implementation.No functional change.Signed-off-by: Thomas Gleixner Acked-by: Richard Weinberger Link: https://lore.kernel.org/r/20230613224545.493148694@linutronix.de List of files: /linux-6.15/arch/um/Kconfig

74ce793b - hostfs: Fix ephemeral inodes

Mon, 12 Jun 2023 19:14:25 +0000

hostfs: Fix ephemeral inodeshostfs creates a new inode for each opened or created file, whichcreated useless inode allocations and forbade identifying a host filewith a kernel inode.Fix this uncommon filesystem behavior by tying kernel inodes to hostfile's inode and device IDs. Even if the host filesystem inodes may berecycled, this cannot happen while a file referencing it is opened,which is the case with hostfs. It should be noted that hostfs inode IDsmay not be unique for the same hostfs superblock because multiple host's(backed) superblocks may be used.Delete inodes when dropping them to force backed host's file descriptorsclosing.This enables to entirely remove ARCH_EPHEMERAL_INODES, and then makesLandlock fully supported by UML. This is very useful for testingchanges.These changes also factor out and simplify some helpers thanks to thenew hostfs_inode_update() and the hostfs_iget() revamp: read_name(),hostfs_create(), hostfs_lookup(), hostfs_mknod(), andhostfs_fill_sb_common().A following commit with new Landlock tests check this new hostfs inodeconsistency.Cc: Anton Ivanov Cc: Johannes Berg Acked-by: Richard Weinberger Link: https://lore.kernel.org/r/20230612191430.339153-2-mic@digikod.netSigned-off-by: Mickaël Salaün List of files: /linux-6.15/arch/um/Kconfig

83e913f5 - um: Support LTO

Fri, 13 Jan 2023 17:03:00 +0000

um: Support LTOOnly a handful of changes are necessary to get it to work.Signed-off-by: Peter Foley Signed-off-by: Richard Weinberger List of files: /linux-6.15/arch/um/Kconfig

f09c3fcf - um: put power options in a menu

Fri, 13 Jan 2023 16:56:11 +0000

um: put power options in a menuBecause having them all dumped at top-level is a bit messy.Signed-off-by: Peter Foley Signed-off-by: Richard Weinberger List of files: /linux-6.15/arch/um/Kconfig

0438aadf - rust: arch/um: Add support for CONFIG_RUST under x86_64 UML

Sat, 17 Dec 2022 04:44:36 +0000

rust: arch/um: Add support for CONFIG_RUST under x86_64 UMLCONFIG_RUST currently supports x86_64, but does not support it underUML. With the previous patches applied, adding support is trivial:add CONFIG_HAVE_RUST to UML if X86_64 is set.The scripts/generate_rust_target.rs file already checks forCONFIG_X86_64, not CONFIG_X86, so is prepared for UML support.The Rust support does not currently support X86_32.Also, update the Rust architecture support documentation to not thatthis is being maintained: I intend to look after this as best I can.Signed-off-by: David Gow Reviewed-by: Sergio González Collado Tested-by: Sergio González Collado Signed-off-by: Richard Weinberger List of files: /linux-6.15/arch/um/Kconfig

ba38961a - um: Enable FORTIFY_SOURCE

Thu, 10 Feb 2022 00:32:24 +0000

um: Enable FORTIFY_SOURCEEnable FORTIFY_SOURCE so running Kunit tests can test fortifiedfunctions.Signed-off-by: Kees Cook Tested-by: David Gow Link: https://lore.kernel.org/r/20220210003224.773957-1-keescook@chromium.org List of files: /linux-6.15/arch/um/Kconfig

3d923c5f - mm/mmap: drop ARCH_HAS_VM_GET_PAGE_PROT

Mon, 11 Jul 2022 07:06:00 +0000

mm/mmap: drop ARCH_HAS_VM_GET_PAGE_PROTNow all the platforms enable ARCH_HAS_GET_PAGE_PROT. They define andexport own vm_get_page_prot() whether custom or standardDECLARE_VM_GET_PAGE_PROT. Hence there is no need for default genericfallback for vm_get_page_prot(). Just drop this fallback and alsoARCH_HAS_GET_PAGE_PROT mechanism.Link: https://lkml.kernel.org/r/20220711070600.2378316-27-anshuman.khandual@arm.comSigned-off-by: Anshuman Khandual Reviewed-by: Geert Uytterhoeven Reviewed-by: Christoph Hellwig Reviewed-by: Christophe Leroy Acked-by: Geert Uytterhoeven Cc: Arnd Bergmann Cc: Brian Cain Cc: Catalin Marinas Cc: Christoph Hellwig Cc: Chris Zankel Cc: "David S. Miller" Cc: Dinh Nguyen Cc: Guo Ren Cc: Heiko Carstens Cc: Huacai Chen Cc: Ingo Molnar Cc: "James E.J. Bottomley" Cc: Jeff Dike Cc: Jonas Bonn Cc: Michael Ellerman Cc: Michal Simek Cc: Nicholas Piggin Cc: Palmer Dabbelt Cc: Paul Mackerras Cc: Paul Walmsley Cc: Richard Henderson Cc: Rich Felker Cc: Russell King Cc: Sam Ravnborg Cc: Stafford Horne Cc: Thomas Bogendoerfer Cc: Thomas Gleixner Cc: Vasily Gorbik Cc: Vineet Gupta Cc: WANG Xuerui Cc: Will Deacon Cc: Yoshinori Sato Signed-off-by: Andrew Morton List of files: /linux-6.15/arch/um/Kconfig

91a8da02 - um/mm: enable ARCH_HAS_VM_GET_PAGE_PROT

Mon, 11 Jul 2022 07:05:58 +0000

um/mm: enable ARCH_HAS_VM_GET_PAGE_PROTThis enables ARCH_HAS_VM_GET_PAGE_PROT on the platform and exportsstandard vm_get_page_prot() implementation via DECLARE_VM_GET_PAGE_PROT,which looks up a private and static protection_map[] array. Subsequentlyall __SXXX and __PXXX macros can be dropped which are no longer needed.Link: https://lkml.kernel.org/r/20220711070600.2378316-25-anshuman.khandual@arm.comSigned-off-by: Anshuman Khandual Cc: Jeff Dike Cc: Arnd Bergmann Cc: Brian Cain Cc: Catalin Marinas Cc: Christophe Leroy Cc: Christoph Hellwig Cc: Christoph Hellwig Cc: Chris Zankel Cc: "David S. Miller" Cc: Dinh Nguyen Cc: Geert Uytterhoeven Cc: Guo Ren Cc: Heiko Carstens Cc: Huacai Chen Cc: Ingo Molnar Cc: "James E.J. Bottomley" Cc: Jonas Bonn Cc: Michael Ellerman Cc: Michal Simek Cc: Nicholas Piggin Cc: Palmer Dabbelt Cc: Paul Mackerras Cc: Paul Walmsley Cc: Richard Henderson Cc: Rich Felker Cc: Russell King Cc: Sam Ravnborg Cc: Stafford Horne Cc: Thomas Bogendoerfer Cc: Thomas Gleixner Cc: Vasily Gorbik Cc: Vineet Gupta Cc: WANG Xuerui Cc: Will Deacon Cc: Yoshinori Sato Signed-off-by: Andrew Morton List of files: /linux-6.15/arch/um/Kconfig