linux-kernel - [PATCH v2 0/8] x86/irq: trap and interrupt cleanups

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20210515014400.2999028-1-hpa@zytor.com>
Date:   Fri, 14 May 2021 18:43:52 -0700
From:   "H. Peter Anvin" <hpa@...or.com>
To:     Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        Andy Lutomirski <luto@...nel.org>,
        Borislav Petkov <bp@...en8.de>,
        "H. Peter Anvin" <hpa@...or.com>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: [PATCH v2 0/8] x86/irq: trap and interrupt cleanups

From: "H. Peter Anvin (Intel)" <hpa@...or.com>

A collection of trap/interrupt-related patches, almost all
cleanups. It does remove a modest amount of code (39 lines.) The only
patches that should have any possible effect at all are:

7/8 - x86/irq: WARN_ONCE() if irq_move_cleanup is called on a pending interrupt

	This condition is believed to be impossible after many
	improvements to the IRQ vector allocation code since this
	function was written. Per discussion with tglx, add a
	WARN_ONCE() if this happens as a first step towards excising
	this hack.

8/8 - x86/irq: merge and functonalize common code in DECLARE/DEFINE_IDTENTRY_*

	This patch reverses kvm_set_cpu_l1tf_flush_l1d() and
	__irq_enter_raw() in DEFINE_IDTENTRY_SYSVEC_SIMPLE() in order
	to be able to unify the code with DEFINE_IDTENTRY_SYSVEC().

	This replaces a lot of macros with inline functions, which
	required some amount of adjusting types in various places,
	they should have no effect.
	
	The reason for unification is mainly to avoid the possibility
	of inadvertent divergence rather than the rather modest amount
	of code, but it does remove 25 lines of code.

--- 
 1/8 x86/irq: merge and functionalize common code in DECLARE/DEFINE_IDTENTRY_*
 2/8 x86/irq: WARN_ONCE() if irq_move_cleanup is called on a pending interrupt
 3/8 x86/kexec: set_[gi]dt() -> native_[gi]dt_invalidate() in machine_kexec_*.c
 4/8 x86/desc: add native_[ig]dt_invalidate() to <asm/desc.h>
 5/8 x86/idt: remove address argument to idt_invalidate()
 6/8 x86/irq: remove unused vectors from <asm/irq_vectors.h>
 7/8 x86/irqvector: add NR_EXTERNAL_VECTORS and NR_SYSTEM_VECTORS
 8/8 x86/traps: add X86_NR_HW_TRAPS to <asm/trapnr.h>

 arch/x86/entry/common.c                  |   5 +-
 arch/x86/include/asm/desc.h              |  17 ++-
 arch/x86/include/asm/idtentry.h          | 174 +++++++++++++++----------------
 arch/x86/include/asm/irq_stack.h         |  73 +++++--------
 arch/x86/include/asm/irq_vectors.h       |   7 +-
 arch/x86/include/asm/trapnr.h            |   1 +
 arch/x86/kernel/apic/apic.c              |   2 +-
 arch/x86/kernel/apic/vector.c            |   5 +
 arch/x86/kernel/idt.c                    |   5 +-
 arch/x86/kernel/irq.c                    |   1 +
 arch/x86/kernel/machine_kexec_32.c       |  15 +--
 arch/x86/kernel/machine_kexec_64.c       |  33 +-----
 arch/x86/kernel/reboot.c                 |   2 +-
 arch/x86/kernel/sev-es.c                 |   6 +-
 arch/x86/kernel/traps.c                  |   2 +-
 tools/arch/x86/include/asm/irq_vectors.h |   7 +-
 16 files changed, 158 insertions(+), 197 deletions(-)