[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CACT4Y+YwRE=YNQYmQ=7RWde33830YOYr5pEAoYbrofY2JG43MA@mail.gmail.com>
Date: Fri, 12 Feb 2021 13:54:47 +0100
From: Dmitry Vyukov <dvyukov@...gle.com>
To: info@...xander-lochmann.de
Cc: Andrey Konovalov <andreyknvl@...gle.com>,
Jonathan Corbet <corbet@....net>,
Andrew Morton <akpm@...ux-foundation.org>,
Wei Yongjun <weiyongjun1@...wei.com>,
Maciej Grochowski <maciej.grochowski@...me>,
kasan-dev <kasan-dev@...glegroups.com>,
"open list:DOCUMENTATION" <linux-doc@...r.kernel.org>,
LKML <linux-kernel@...r.kernel.org>,
syzkaller <syzkaller@...glegroups.com>
Subject: Re: [PATCH] KCOV: Introduced tracing unique covered PCs
On Thu, Feb 11, 2021 at 9:07 AM Alexander Lochmann
<info@...xander-lochmann.de> wrote:
>
> Introduced new tracing mode KCOV_MODE_UNIQUE.
> It simply stores the executed PCs.
> The execution order is discarded.
> Each bit in the shared buffer represents every fourth
> byte of the text segment.
> Since a call instruction on every supported
> architecture is at least four bytes, it is safe
> to just store every fourth byte of the text segment.
> In contrast to KCOV_MODE_TRACE_PC, the shared buffer
> cannot overflow. Thus, all executed PCs are recorded.
>
> Signed-off-by: Alexander Lochmann <info@...xander-lochmann.de>
> ---
> Documentation/dev-tools/kcov.rst | 80 ++++++++++++++++++++++++++++++++
> include/linux/kcov.h | 4 +-
> include/uapi/linux/kcov.h | 10 ++++
> kernel/kcov.c | 67 ++++++++++++++++++++------
> 4 files changed, 147 insertions(+), 14 deletions(-)
>
> diff --git a/Documentation/dev-tools/kcov.rst b/Documentation/dev-tools/kcov.rst
> index 8548b0b04e43..4712a730a06a 100644
> --- a/Documentation/dev-tools/kcov.rst
> +++ b/Documentation/dev-tools/kcov.rst
> @@ -127,6 +127,86 @@ That is, a parent process opens /sys/kernel/debug/kcov, enables trace mode,
> mmaps coverage buffer and then forks child processes in a loop. Child processes
> only need to enable coverage (disable happens automatically on thread end).
>
> +If someone is interested in a set of executed PCs, and does not care about
> +execution order, he or she can advise KCOV to do so:
> +
> +.. code-block:: c
> +
> + #include <stdio.h>
> + #include <stddef.h>
> + #include <stdint.h>
> + #include <stdlib.h>
> + #include <sys/types.h>
> + #include <sys/stat.h>
> + #include <sys/ioctl.h>
> + #include <sys/mman.h>
> + #include <unistd.h>
> + #include <fcntl.h>
> +
> + #define KCOV_INIT_TRACE _IOR('c', 1, unsigned long)
> + #define KCOV_INIT_UNIQUE _IOR('c', 2, unsigned long)
> + #define KCOV_ENABLE _IO('c', 100)
> + #define KCOV_DISABLE _IO('c', 101)
> +
> + #define BITS_PER_LONG 64
> + #define KCOV_TRACE_PC 0
> + #define KCOV_TRACE_CMP 1
> + #define KCOV_UNIQUE_PC 2
> + /*
> + * Determine start of text segment via 'nm vmlinux | grep _stext | cut -d " " -f1',
> + * and fill in.
> + */
> + #define STEXT_START 0xffffffff81000000
> +
> +
> +
> + int main(int argc, char **argv)
> + {
> + int fd;
> + unsigned long *cover, n, i;
> +
> + /* A single fd descriptor allows coverage collection on a single
> + * thread.
> + */
> + fd = open("/sys/kernel/debug/kcov", O_RDWR);
> + if (fd == -1)
> + perror("open"), exit(1);
> + /* Setup trace mode and trace size. */
> + if ((n = ioctl(fd, KCOV_INIT_UNIQUE, 0)) < 0)
> + perror("ioctl"), exit(1);
> + /* Mmap buffer shared between kernel- and user-space. */
> + cover = (unsigned long*)mmap(NULL, n,
> + PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
> + if ((void*)cover == MAP_FAILED)
> + perror("mmap"), exit(1);
> + /* Enable coverage collection on the current thread. */
> + if (ioctl(fd, KCOV_ENABLE, KCOV_UNIQUE_PC))
> + perror("ioctl"), exit(1);
> + /* That's the target syscal call. */
> + read(-1, NULL, 0);
> + /* Disable coverage collection for the current thread. After this call
> + * coverage can be enabled for a different thread.
> + */
> + if (ioctl(fd, KCOV_DISABLE, 0))
> + perror("ioctl"), exit(1);
> + /* Convert byte size into element size */
> + n /= sizeof(unsigned long);
> + /* Print executed PCs in sorted order */
> + for (i = 0; i < n; i++) {
> + for (int j = 0; j < BITS_PER_LONG; j++) {
> + if (cover[i] & (1L << j)) {
> + printf("0x%jx\n", (uintmax_t)(STEXT_START + (i * BITS_PER_LONG + j) * 4));
> + }
> + }
> + }
> + /* Free resources. */
> + if (munmap(cover, n * sizeof(unsigned long)))
> + perror("munmap"), exit(1);
> + if (close(fd))
> + perror("close"), exit(1);
> + return 0;
> + }
> +
> Comparison operands collection
> ------------------------------
>
> diff --git a/include/linux/kcov.h b/include/linux/kcov.h
> index a10e84707d82..aa0c8bcf8299 100644
> --- a/include/linux/kcov.h
> +++ b/include/linux/kcov.h
> @@ -19,7 +19,9 @@ enum kcov_mode {
> */
> KCOV_MODE_TRACE_PC = 2,
> /* Collecting comparison operands mode. */
> - KCOV_MODE_TRACE_CMP = 3,
> + KCOV_MODE_TRACE_CMP = 4,
> + /* Collecting unique covered PCs. Execution order is not saved. */
> + KCOV_MODE_UNIQUE_PC = 8,
> };
>
> #define KCOV_IN_CTXSW (1 << 30)
> diff --git a/include/uapi/linux/kcov.h b/include/uapi/linux/kcov.h
> index 1d0350e44ae3..5b99b6d1a1ac 100644
> --- a/include/uapi/linux/kcov.h
> +++ b/include/uapi/linux/kcov.h
> @@ -19,6 +19,7 @@ struct kcov_remote_arg {
> #define KCOV_REMOTE_MAX_HANDLES 0x100
>
> #define KCOV_INIT_TRACE _IOR('c', 1, unsigned long)
> +#define KCOV_INIT_UNIQUE _IOR('c', 2, unsigned long)
> #define KCOV_ENABLE _IO('c', 100)
> #define KCOV_DISABLE _IO('c', 101)
> #define KCOV_REMOTE_ENABLE _IOW('c', 102, struct kcov_remote_arg)
> @@ -35,6 +36,15 @@ enum {
> KCOV_TRACE_PC = 0,
> /* Collecting comparison operands mode. */
> KCOV_TRACE_CMP = 1,
> + /*
> + * Unique coverage collection mode.
> + * Unique covered PCs are collected in a per-task buffer.
> + * De-duplicates the collected PCs. Execution order is *not* saved.
> + * Each bit in the buffer represents every fourth byte of the text segment.
> + * Since a call instruction is at least four bytes on every supported
> + * architecture, storing just every fourth byte is sufficient.
> + */
> + KCOV_UNIQUE_PC = 2,
> };
>
> /*
> diff --git a/kernel/kcov.c b/kernel/kcov.c
> index 6b8368be89c8..8f00ba6e672a 100644
> --- a/kernel/kcov.c
> +++ b/kernel/kcov.c
> @@ -24,6 +24,7 @@
> #include <linux/refcount.h>
> #include <linux/log2.h>
> #include <asm/setup.h>
> +#include <asm/sections.h>
>
> #define kcov_debug(fmt, ...) pr_debug("%s: " fmt, __func__, ##__VA_ARGS__)
>
> @@ -171,7 +172,7 @@ static notrace bool check_kcov_mode(enum kcov_mode needed_mode, struct task_stru
> * kcov_start().
> */
> barrier();
> - return mode == needed_mode;
> + return (mode & needed_mode) && !(mode & KCOV_IN_CTXSW);
I see this produces an additional check and branch:
void foo1(unsigned mode) {
if ((mode & 10) && !(mode & (1<<30)))
foo();
}
0: 40 f6 c7 0a test $0xa,%dil
4: 74 0f je 15 <foo1+0x15>
6: 81 e7 00 00 00 40 and $0x40000000,%edi
c: 75 07 jne 15 <foo1+0x15>
e: 31 c0 xor %eax,%eax
10: e9 00 00 00 00 jmpq 15 <foo1+0x15>
I think we could make KCOV_IN_CTXSW sign bit and then express the check as:
void foo2(unsigned mode) {
if (((int)(mode & 0x8000000a)) > 0)
foo();
}
0000000000000020 <foo2>:
20: 81 e7 0a 00 00 80 and $0x8000000a,%edi
26: 7f 08 jg 30 <foo2+0x10>
28: c3 retq
> }
>
> static notrace unsigned long canonicalize_ip(unsigned long ip)
> @@ -191,18 +192,26 @@ void notrace __sanitizer_cov_trace_pc(void)
> struct task_struct *t;
> unsigned long *area;
> unsigned long ip = canonicalize_ip(_RET_IP_);
> - unsigned long pos;
> + unsigned long pos, idx;
>
> t = current;
> - if (!check_kcov_mode(KCOV_MODE_TRACE_PC, t))
> + if (!check_kcov_mode(KCOV_MODE_TRACE_PC | KCOV_MODE_UNIQUE_PC, t))
> return;
>
> area = t->kcov_area;
> - /* The first 64-bit word is the number of subsequent PCs. */
> - pos = READ_ONCE(area[0]) + 1;
> - if (likely(pos < t->kcov_size)) {
> - area[pos] = ip;
> - WRITE_ONCE(area[0], pos);
> + if (likely(t->kcov_mode == KCOV_MODE_TRACE_PC)) {
Does this introduce an additional real of t->kcov_mode?
If yes, please reuse the value read in check_kcov_mode.
> + /* The first 64-bit word is the number of subsequent PCs. */
> + pos = READ_ONCE(area[0]) + 1;
> + if (likely(pos < t->kcov_size)) {
> + area[pos] = ip;
> + WRITE_ONCE(area[0], pos);
> + }
> + } else {
> + idx = (ip - canonicalize_ip((unsigned long)&_stext)) / 4;
> + pos = idx % BITS_PER_LONG;
> + idx /= BITS_PER_LONG;
> + if (likely(idx < t->kcov_size))
> + WRITE_ONCE(area[idx], READ_ONCE(area[idx]) | 1L << pos);
> }
> }
> EXPORT_SYMBOL(__sanitizer_cov_trace_pc);
> @@ -474,6 +483,7 @@ static int kcov_mmap(struct file *filep, struct vm_area_struct *vma)
> goto exit;
> }
> if (!kcov->area) {
> + kcov_debug("mmap(): Allocating 0x%lx bytes\n", size);
> kcov->area = area;
> vma->vm_flags |= VM_DONTEXPAND;
> spin_unlock_irqrestore(&kcov->lock, flags);
> @@ -515,6 +525,8 @@ static int kcov_get_mode(unsigned long arg)
> {
> if (arg == KCOV_TRACE_PC)
> return KCOV_MODE_TRACE_PC;
> + else if (arg == KCOV_UNIQUE_PC)
> + return KCOV_MODE_UNIQUE_PC;
As far as I understand, users can first do KCOV_INIT_UNIQUE and then
enable KCOV_TRACE_PC, or vice versa.
It looks somewhat strange. Is it intentional? It's not possible to
specify buffer size for KCOV_INIT_UNIQUE, so most likely the buffer
will be either too large or too small for a trace.
> else if (arg == KCOV_TRACE_CMP)
> #ifdef CONFIG_KCOV_ENABLE_COMPARISONS
> return KCOV_MODE_TRACE_CMP;
> @@ -562,12 +574,13 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,
> {
> struct task_struct *t;
> unsigned long size, unused;
> - int mode, i;
> + int mode, i, text_size, ret = 0;
> struct kcov_remote_arg *remote_arg;
> struct kcov_remote *remote;
> unsigned long flags;
>
> switch (cmd) {
> + case KCOV_INIT_UNIQUE:
I think nowadays you need some annotation like fallthrough here.
> case KCOV_INIT_TRACE:
> /*
> * Enable kcov in trace mode and setup buffer size.
> @@ -581,11 +594,39 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,
> * that must not overflow.
> */
> size = arg;
> - if (size < 2 || size > INT_MAX / sizeof(unsigned long))
> - return -EINVAL;
> - kcov->size = size;
> + if (cmd == KCOV_INIT_UNIQUE) {
> + if (size != 0)
> + return -EINVAL;
> + text_size = (canonicalize_ip((unsigned long)&_etext) - canonicalize_ip((unsigned long)&_stext));
> + /**
> + * A call instr is at least four bytes on every supported architecture.
> + * Hence, just every fourth instruction can potentially be a call.
> + */
> + text_size /= 4;
Strictly saying, we need to round up text_size to 4 before dividing by
4. Otherwise we potentially don't cover up to the last 3 bytes.
> + /*
> + * Round up size of text segment to multiple of BITS_PER_LONG.
> + * Otherwise, we cannot track
> + * the last (text_size % BITS_PER_LONG) addresses.
> + */
> + text_size = roundup(text_size, BITS_PER_LONG);
> + /* Get the amount of bytes needed */
> + text_size = text_size / 8;
> + /* mmap() requires size to be a multiple of PAGE_SIZE */
> + text_size = roundup(text_size, PAGE_SIZE);
> + /* Get the cover size (= amount of longs stored) */
s/longs/bytes/
> + ret = text_size;
> + kcov->size = text_size / sizeof(unsigned long);
> + kcov_debug("text size = 0x%lx, roundup = 0x%x, kcov->size = 0x%x\n",
> + ((unsigned long)&_etext) - ((unsigned long)&_stext),
> + text_size,
> + kcov->size);
> + } else {
> + if (size < 2 || size > INT_MAX / sizeof(unsigned long))
> + return -EINVAL;
> + kcov->size = size;
> + }
> kcov->mode = KCOV_MODE_INIT;
> - return 0;
> + return ret;
> case KCOV_ENABLE:
> /*
> * Enable coverage for the current task.
> --
> 2.30.0
>
Powered by blists - more mailing lists