Patch "x86/microcode: Dont abuse the TLB-flush interface" has been added to the 4.14-stable tree
Subsystems (3)
This is a note to let you know that I've just added the patch titled
x86/microcode: Dont abuse the TLB-flush interface
to the 4.14-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
x86-microcode-dont-abuse-the-tlb-flush-interface.patch
and it can be found in the queue-4.14 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.
Quoted 1 line
From
23cb7d46f371844c004784ad9552a57446f73e5aMon Sep 17 00:00:00 2001
From: Peter Zijlstra <[email protected]>
Date: Tue, 5 Dec 2017 13:34:51 +0100
Subject: x86/microcode: Dont abuse the TLB-flush interface
From: Peter Zijlstra <[email protected]>
commit 23cb7d46f371844c004784ad9552a57446f73e5a upstream.
Commit:
ec400ddeff20 ("x86/microcode_intel_early.c: Early update ucode on Intel's CPU")
... grubbed into tlbflush internals without coherent explanation.
Since it says its a precaution and the SDM doesn't mention anything like this, take it out back.
--- arch/x86/include/asm/tlbflush.h | 19 ++++++------------- arch/x86/kernel/cpu/microcode/intel.c | 13 ------------- 2 files changed, 6 insertions(+), 26 deletions(-)
arch/x86/include/asm/tlbflush.h +6 -13
@@ -246,20 +246,9 @@ static inline void __native_flush_tlb(vopreempt_enable (); } -static inline void __native_flush_tlb_global_irq_disabled (void ) -{ -unsigned long cr4; - - cr4 =this_cpu_read (cpu_tlbstate.cr4); -/* clear PGE */ -native_write_cr4 (cr4 & ~X86_CR4_PGE ); -/* write old PGE again and flush TLBs */ -native_write_cr4 (cr4); -} -static inline void __native_flush_tlb_global (void ) { -unsigned long flags; +unsigned long cr4, flags;if (static_cpu_has (X86_FEATURE_INVPCID )) {/*
@@ -277,7 +266,11 @@ static inline void __native_flush_tlb_gl*/ raw_local_irq_save (flags); -__native_flush_tlb_global_irq_disabled (); + cr4 =this_cpu_read (cpu_tlbstate.cr4); +/* toggle PGE */ +native_write_cr4 (cr4 ^X86_CR4_PGE ); +/* write old PGE again and flush TLBs */ +native_write_cr4 (cr4);raw_local_irq_restore (flags); }
arch/x86/kernel/cpu/microcode/intel.c +0 -13
@@ -565,15 +565,6 @@ static void print_ucode(struct ucode_cpu }#else -/* -* Flush global tlb. We only do this in x86_64 where paging has been enabled -* already and PGE should be enabled as well. -*/ -static inline void flush_tlb_early (void ) -{ -__native_flush_tlb_global_irq_disabled (); -} -static inline void print_ucode (struct ucode_cpu_info *uci) {struct microcode_intel *mc;
@@ -602,10 +593,6 @@ static int apply_microcode_early(structif (rev != mc->hdr.rev)return -1 ; -#ifdef CONFIG_X86_64 -/* Flush global tlb. This is precaution. */ -flush_tlb_early (); -#endif uci->cpu_sig.rev = rev;if (early)
Patches currently in stable-queue which might be from [email protected] are
queue-4.14/x86-entry-rename-sysenter_stack-to-cpu_entry_area_entry_stack.patch queue-4.14/x86-mm-put-mmu-to-hardware-asid-translation-in-one-place.patch queue-4.14/x86-vsyscall-64-explicitly-set-_page_user-in-the-pagetable-hierarchy.patch queue-4.14/x86-uv-use-the-right-tlb-flush-api.patch queue-4.14/x86-decoder-fix-and-update-the-opcodes-map.patch queue-4.14/x86-mm-dump_pagetables-check-page_present-for-real.patch queue-4.14/x86-ldt-prevent-ldt-inheritance-on-exec.patch queue-4.14/x86-microcode-dont-abuse-the-tlb-flush-interface.patch queue-4.14/x86-doc-remove-obvious-weirdnesses-from-the-x86-mm-layout-documentation.patch queue-4.14/init-invoke-init_espfix_bsp-from-mm_init.patch queue-4.14/x86-cpu_entry_area-move-it-to-a-separate-unit.patch queue-4.14/x86-vsyscall-64-warn-and-fail-vsyscall-emulation-in-native-mode.patch queue-4.14/x86-mm-create-asm-invpcid.h.patch queue-4.14/x86-mm-remove-superfluous-barriers.patch queue-4.14/x86-ldt-rework-locking.patch queue-4.14/arch-mm-allow-arch_dup_mmap-to-fail.patch queue-4.14/x86-cpu_entry_area-move-it-out-of-the-fixmap.patch queue-4.14/tools-headers-sync-objtool-uapi-header.patch queue-4.14/x86-mm-remove-hard-coded-asid-limit-checks.patch queue-4.14/x86-kconfig-limit-nr_cpus-on-32-bit-to-a-sane-amount.patch queue-4.14/objtool-fix-64-bit-build-on-32-bit-host.patch queue-4.14/x86-mm-add-comments-to-clarify-which-tlb-flush-functions-are-supposed-to-flush-what.patch queue-4.14/x86-mm-move-the-cr3-construction-functions-to-tlbflush.h.patch queue-4.14/x86-mm-dump_pagetables-make-the-address-hints-correct-and-readable.patch queue-4.14/x86-insn-eval-add-utility-functions-to-get-segment-selector.patch queue-4.14/objtool-move-synced-files-to-their-original-relative-locations.patch queue-4.14/x86-mm-use-__flush_tlb_one-for-kernel-memory.patch queue-4.14/objtool-move-kernel-headers-code-sync-check-to-a-script.patch queue-4.14/x86-mm-64-improve-the-memory-map-documentation.patch queue-4.14/objtool-fix-cross-build.patch