All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: x86@kernel.org, Steven Rostedt <rostedt@goodmis.org>,
	Brian Gerst <brgerst@gmail.com>, Juergen Gross <jgross@suse.com>,
	Frederic Weisbecker <frederic@kernel.org>,
	Alexandre Chartre <alexandre.chartre@oracle.com>
Subject: [patch part-II V2 06/13] x86/entry/common: Mark syscall entry points notrace and NOKPROBE
Date: Sun, 08 Mar 2020 23:24:05 +0100	[thread overview]
Message-ID: <20200308222609.522613084@linutronix.de> (raw)
In-Reply-To: 20200308222359.370649591@linutronix.de

The entry code has some limitations for instrumentation. Anything before
invoking enter_from_user_mode() cannot be probed because kprobes depend on
RCU and with NOHZ_FULL user mode can be accounted similar to idle from a
RCU point of view. enter_from_user_mode() calls into context tracking which
adjusts the RCU state.

A similar problem exists vs. function tracing. The function entry/exit
points can be used by BPF which again is not safe before CONTEXT_KERNEL has
been reached.

Mark the C-entry points for the various syscalls with notrace and
NOKPROBE_SYMBOL().

Note, that this still leaves the ASM invocations of trace_hardirqs_off()
unprotected. While this is safe vs. RCU at least from the ftrace POV, these
are trace points which can be utilized by BPF... This will be addressed in
later patches.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Alexandre Chartre <alexandre.chartre@oracle.com>
---
V2: Amend changelog
---
 arch/x86/entry/common.c |    9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

--- a/arch/x86/entry/common.c
+++ b/arch/x86/entry/common.c
@@ -315,11 +315,12 @@ void do_syscall_64_irqs_on(unsigned long
 	syscall_return_slowpath(regs);
 }
 
-__visible void do_syscall_64(unsigned long nr, struct pt_regs *regs)
+__visible notrace void do_syscall_64(unsigned long nr, struct pt_regs *regs)
 {
 	syscall_entry_apply_fixups();
 	do_syscall_64_irqs_on(nr, regs);
 }
+NOKPROBE_SYMBOL(do_syscall_64);
 #endif
 
 #if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION)
@@ -370,11 +371,12 @@ static __always_inline void do_syscall_3
 }
 
 /* Handles int $0x80 */
-__visible void do_int80_syscall_32(struct pt_regs *regs)
+__visible notrace void do_int80_syscall_32(struct pt_regs *regs)
 {
 	syscall_entry_apply_fixups();
 	do_syscall_32_irqs_on(regs);
 }
+NOKPROBE_SYMBOL(do_int80_syscall_32);
 
 /* Fast syscall 32bit variant */
 static __always_inline long do_fast_syscall_32_irqs_on(struct pt_regs *regs)
@@ -450,10 +452,11 @@ static __always_inline long do_fast_sysc
 }
 
 /* Returns 0 to return using IRET or 1 to return using SYSEXIT/SYSRETL. */
-__visible long do_fast_syscall_32(struct pt_regs *regs)
+__visible notrace long do_fast_syscall_32(struct pt_regs *regs)
 {
 	syscall_entry_apply_fixups();
 	return do_fast_syscall_32_irqs_on(regs);
 }
+NOKPROBE_SYMBOL(do_fast_syscall_32);
 
 #endif /* CONFIG_X86_32 || CONFIG_IA32_EMULATION */


  parent reply	other threads:[~2020-03-08 23:25 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-08 22:23 [patch part-II V2 00/13] x86/entry: Consolidation - Part II (syscalls) Thomas Gleixner
2020-03-08 22:24 ` [patch part-II V2 01/13] context_tracking: Ensure that the critical path cannot be instrumented Thomas Gleixner
2020-03-09 14:22   ` Frederic Weisbecker
2020-03-09 14:40     ` Thomas Gleixner
2020-03-10 10:12   ` Alexandre Chartre
2020-03-08 22:24 ` [patch part-II V2 02/13] x86/entry: Mark enter_from_user_mode() notrace and NOKPROBE Thomas Gleixner
2020-03-09 15:14   ` Frederic Weisbecker
2020-03-09 15:40     ` Thomas Gleixner
2020-03-11 22:21       ` Frederic Weisbecker
2020-03-10 10:15   ` Alexandre Chartre
2020-03-08 22:24 ` [patch part-II V2 03/13] x86/entry/32: Remove unused label restore_nocheck Thomas Gleixner
2020-03-10 10:16   ` Alexandre Chartre
2020-03-10 13:02   ` [tip: x86/entry] " tip-bot2 for Thomas Gleixner
2020-03-08 22:24 ` [patch part-II V2 04/13] x86/entry/64: Trace irqflags unconditionally as ON when returning to user space Thomas Gleixner
2020-03-10 10:25   ` Alexandre Chartre
2020-03-10 13:02   ` [tip: x86/entry] " tip-bot2 for Thomas Gleixner
2020-03-08 22:24 ` [patch part-II V2 05/13] x86/entry/common: Consolidate syscall entry code Thomas Gleixner
2020-03-08 22:24 ` Thomas Gleixner [this message]
2020-03-13 15:12   ` [patch part-II V2 06/13] x86/entry/common: Mark syscall entry points notrace and NOKPROBE Frederic Weisbecker
2020-03-08 22:24 ` [patch part-II V2 07/13] x86/entry: Move irq tracing on syscall entry to C-code Thomas Gleixner
2020-03-13 15:16   ` Frederic Weisbecker
2020-03-13 23:17     ` Thomas Gleixner
2020-03-08 22:24 ` [patch part-II V2 08/13] tracing: Provide lockdep less trace_hardirqs_on/off() variants Thomas Gleixner
2020-03-10 10:55   ` Alexandre Chartre
2020-03-10 11:08     ` Borislav Petkov
2020-03-10 11:21       ` Alexandre Chartre
2020-03-08 22:24 ` [patch part-II V2 09/13] x86/entry/common: Split hardirq tracing into lockdep and ftrace parts Thomas Gleixner
2020-03-10 11:20   ` Borislav Petkov
2020-03-10 13:40     ` Steven Rostedt
2020-03-10 13:28   ` Alexandre Chartre
2020-03-23  9:08   ` [x86/entry/common] bae397f6e7: WARNING:at_kernel/sched/cputime.c:#get_vtime_delta kernel test robot
2020-03-23  9:08     ` kernel test robot
2020-03-08 22:24 ` [patch part-II V2 10/13] x86/entry/common: Split prepare_exit_to_usermode() and syscall_return_slowpath() Thomas Gleixner
2020-03-10 13:37   ` Alexandre Chartre
2020-03-08 22:24 ` [patch part-II V2 11/13] x86/speculation/mds: Mark mds_user_clear_cpu_buffers() __always_inline Thomas Gleixner
2020-03-10 13:48   ` Alexandre Chartre
2020-03-10 16:38     ` Thomas Gleixner
2020-03-08 22:24 ` [patch part-II V2 12/13] x86/entry: Move irq flags tracing to prepare_exit_to_usermode() Thomas Gleixner
2020-03-10 14:03   ` Alexandre Chartre
2020-03-08 22:24 ` [patch part-II V2 13/13] x86/entry/common: Split irq tracing in prepare_exit_to_usermode() Thomas Gleixner
2020-03-10 14:09   ` Alexandre Chartre

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200308222609.522613084@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=alexandre.chartre@oracle.com \
    --cc=brgerst@gmail.com \
    --cc=frederic@kernel.org \
    --cc=jgross@suse.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rostedt@goodmis.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.