| 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
| ||
|
Message-ID: <CAJF2gTT9gZZftoUD4eMT0Th=oYOpsdYbrFdr=5kO2-N+ki51NA@mail.gmail.com>
Date: Fri, 30 Oct 2020 11:28:01 +0800
From: Guo Ren <guoren@...nel.org>
To: Atish Patra <atishp@...shpatra.org>
Cc: Zong Li <zong.li@...ive.com>, Paul McKenney <paulmck@...nel.org>,
josh@...htriplett.org, Steven Rostedt <rostedt@...dmis.org>,
mathieu.desnoyers@...icios.com, jiangshanlai@...il.com,
joel@...lfernandes.org, vincent.whitchurch@...s.com,
Thomas Gleixner <tglx@...utronix.de>,
Paul Walmsley <paul.walmsley@...ive.com>,
Palmer Dabbelt <palmerdabbelt@...gle.com>,
Masami Hiramatsu <mhiramat@...nel.org>,
Greentime Hu <greentime.hu@...ive.com>,
Colin Ian King <colin.king@...onical.com>,
rcu@...r.kernel.org,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
linux-riscv <linux-riscv@...ts.infradead.org>
Subject: Re: [PATCH] stop_machine: Mark functions as notrace
On Fri, Oct 30, 2020 at 2:46 AM Atish Patra <atishp@...shpatra.org> wrote:
>
> On Thu, Oct 29, 2020 at 9:06 AM Guo Ren <guoren@...nel.org> wrote:
> >
> > On Thu, Oct 29, 2020 at 10:34 AM Zong Li <zong.li@...ive.com> wrote:
> > >
> > > On Thu, Oct 29, 2020 at 8:23 AM Atish Patra <atishp@...shpatra.org> wrote:
> > > >
> > > > On Wed, Oct 28, 2020 at 8:44 AM Guo Ren <guoren@...nel.org> wrote:
> > > > >
> > > > > Hi Zong & Atish,
> > > > >
> > > > > In our 2 harts c910 chip, we found:
> > > > >
> > > > > echo function > /sys/kernel/debug/tracing/current_tracer
> > > > > echo function_graph > /sys/kernel/debug/tracing/current_tracer
> > > > > echo function > /sys/kernel/debug/tracing/current_tracer
> > > > > echo function_graph > /sys/kernel/debug/tracing/current_tracer
> > > > >
> > > > > Then one core halted at stop_machine_yield:
> > > > > arch_cpu_idle () at arch/riscv/kernel/process.c:39
> > > > > 39 local_irq_enable();
> > > > > (gdb) i th
> > > > > Id Target Id Frame
> > > > > * 1 Thread 1 (CPU#0) arch_cpu_idle () at arch/riscv/kernel/process.c:39
> > > > > 2 Thread 2 (CPU#1) stop_machine_yield
> > > > > (cpumask=0xffffffe001371fa8 <__cpu_online_mask>) at
> > > > > ./arch/riscv/include/asm/vdso/processor.h:12
> > > > > (gdb) thread 2
> > > > > [Switching to thread 2 (Thread 2)]
> > > > > #0 stop_machine_yield (cpumask=0xffffffe001371fa8
> > > > > <__cpu_online_mask>) at ./arch/riscv/include/asm/vdso/processor.h:12
> > > > > 12 __asm__ __volatile__ ("div %0, %0, zero" : "=r" (dummy));
> > > > >
> > > > > With your patch, it's solved. For this patch, I'll give:
> > > > > Tested by: Guo Ren <guoren@...nel.org>
> > > > >
> > > > > But that's not enough, we still need:
> > > > >
> > > > > diff --git a/arch/riscv/kernel/sbi.c b/arch/riscv/kernel/sbi.c
> > > > > index 226ccce..12b8808 100644
> > > > > --- a/arch/riscv/kernel/sbi.c
> > > > > +++ b/arch/riscv/kernel/sbi.c
> > > > > @@ -376,7 +376,7 @@ EXPORT_SYMBOL(sbi_send_ipi);
> > > > > *
> > > > > * Return: None
> > > > > */
> > > > > -void sbi_remote_fence_i(const unsigned long *hart_mask)
> > > > > +void notrace sbi_remote_fence_i(const unsigned long *hart_mask)
> > > > > {
> > > > > __sbi_rfence(SBI_EXT_RFENCE_REMOTE_FENCE_I,
> > > > > hart_mask, 0, 0, 0, 0);
> > > > > diff --git a/arch/riscv/mm/cacheflush.c b/arch/riscv/mm/cacheflush.c
> > > > > index 400b945d..9467d987 100644
> > > > > --- a/arch/riscv/mm/cacheflush.c
> > > > > +++ b/arch/riscv/mm/cacheflush.c
> > > > > @@ -9,12 +9,12 @@
> > > > >
> > > > > #include <asm/sbi.h>
> > > > >
> > > > > -static void ipi_remote_fence_i(void *info)
> > > > > +static void notrace ipi_remote_fence_i(void *info)
> > > > > {
> > > > > return local_flush_icache_all();
> > > > > }
> > > > >
> > > > > -void flush_icache_all(void)
> > > > > +void notrace flush_icache_all(void)
> > > > > {
> > > > > if (IS_ENABLED(CONFIG_RISCV_SBI))
> > > > > sbi_remote_fence_i(NULL);
> > > > >
> > > >
> > > > Did you see any issue if these functions are not marked as notrace ?
> > > >
> > > > As per Zong's explanation, the issue was that the other harts already
> > > > fetched the next 2 nops and
> > > > executed 1 while kernel patching replaced other with one of the auipc
> > > > + jalr pair.
> > > >
> > > > @Zong can correct me if I am wrong.
> > > >
> > > > These functions are too far ahead. Can it cause such issues ? If yes,
> > > > then we need to mark each and every function
> > > > that can be invoked from patch_text_nosync and are not inlined.
> > > >
> > > > That includes copy_to_kernel_nofault, __sbi_rfence_v02,
> > > > __sbi_rfence_v02_call, sbi_ecall.
> > > >
> > > > Few of these functions may be inlined by compiler. Can we depend on that ?
> > > >
> > > > > Because:
> > > > > (gdb) bt
> > > > > #0 flush_icache_all () at arch/riscv/mm/cacheflush.c:20
> > > > > #1 0xffffffe00020473a in patch_text_nosync (addr=<optimized out>, insns=
> > > > > <optimized out>, len=<optimized out>) at arch/riscv/kernel/patch.c:96
> > > > > #2 0xffffffe000206792 in ftrace_make_call (rec=<optimized out>,
> > > > > addr=<optimized out>) at arch/riscv/kernel/ftrace.c:109
> > > > > #3 0xffffffe0002c9be4 in __ftrace_replace_code (rec=0xffffffe01ae40020, e
> > > > > nable=true) at kernel/trace/ftrace.c:2503
> > > > > #4 0xffffffe0002ca092 in ftrace_replace_code (mod_flags=<optimized
> > > > > out>) at kernel/trace/ftrace.c:2530
> > > > > #5 0xffffffe0002ca24a in ftrace_modify_all_code (command=9) at kernel
> > > > > /trace/ftrace.c:2677
> > > > > #6 0xffffffe0002ca2ee in __ftrace_modify_code (data=<optimized out>) at
> > > > > kernel/trace/ftrace.c:2703
> > > > > #7 0xffffffe0002c1390 in multi_cpu_stop (data=0x0) at kernel/stop_machin
> > > > > e.c:224
> > > > > #8 0xffffffe0002c0fbe in cpu_stopper_thread (cpu=<optimized out>) at kern
> > > > > el/stop_machine.c:491
> > > > > #9 0xffffffe0002343be in smpboot_thread_fn (data=0x0) at kernel/smpboot.
> > > > > c:165
> > > > > #10 0xffffffe00022f894 in kthread (_create=0xffffffe01af13040) at kern
> > > > > el/kthread.c:292
> > > > > #11 0xffffffe000201fac in handle_exception () at arch/riscv/kernel/entry.S:236
> > > > >
> > >
> > > It seems to me that the problem happens on the waiting threads, it
> > No, that is the call trace to show ftrace_make_call ->
> > flush_icache_all and we should give notrace on the whole path.
> >
>
> Hmm. I am curious to understand how other architectures avoid this problem.
for arm64
static int ftrace_modify_code(unsigned long pc, u32 old, u32 new,
bool validate)
{
u32 replaced;
...
if (aarch64_insn_patch_text_nosync((void *)pc, new))
return -EPERM;
int __kprobes aarch64_insn_patch_text_nosync(void *addr, u32 insn)
{
u32 *tp = addr;
int ret;
/* A64 instructions must be word aligned */
if ((uintptr_t)tp & 0x3)
return -EINVAL;
ret = aarch64_insn_write(tp, insn);
if (ret == 0)
__flush_icache_range((uintptr_t)tp,
(uintptr_t)tp + AARCH64_INSN_SIZE);
Look at arm64, they __kprobes flag and I guess it would also prevent
ftrace call site.
__flush_icache_range is written in asm and no possible ftrace call site.
> Is it a bigger issue in RISC-V because we have to switch privilege
> mode to sync I/D cache ?
We should sync I/D cache at s-mode because we need virtual address.
For c910 we've added icache broadcast invalid instructions by physical
address and virtual address.
Current linux/arch/riscv I/D cache sync is so expensive.
--
Best Regards
Guo Ren
ML: https://lore.kernel.org/linux-csky/
Powered by blists - more mailing lists