Hi Lianbo,
>I would recommend packing them into two patches as below:
>BTW: If there is a better way, you could rearrange them when merging. Thanks.
>And with the warning fix, otherwise:
Thank you for your review. I will send a new patch with these fixes.
Thanks,
Shogo Matsumoto
Thank you for the patch, Shogo.
Date: Thu, 10 Feb 2022 06:38:32 +0000
From: HAGIO KAZUHITO(?????) <k-hagio-ab@xxxxxxx>
To: "shogo.matsumoto@xxxxxxxxxxx" <shogo.matsumoto@xxxxxxxxxxx>
Cc: "Discussion list for crash utility usage, maintenance and
development" <crash-utility@xxxxxxxxxx>
Subject: Re: [Crash-utility] [PATCH v3 0/7] log: output logs of printk
safe buffers
Message-ID:
<TYYPR01MB6777FE775F5CD6438FC2E448DD2F9@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset="iso-2022-jp"
-----Original Message-----
> This patch set introduces -s option for log builtin command to display
> printk safe buffers (safe_print_seq/nmi_print_seq) as follows:
>
> ===
> crash> log -s
> PRINTK_SAFE_SEQ_BUF: nmi_print_seq
> CPU: 0 ADDR: ffff969d7bc19ce0 LEN: 150 MESSAGE_LOST: 0
> Uhhuh. NMI received for unknown reason 20 on CPU 0.
> Do you have a strange power saving mode enabled?
> Dazed and confused, but trying to continue
> ...
> ===
>
> The printk safe buffers are also displayed at the bottom of
> 'log' output so as not to overlook them.
>
> ===
> crash> log
> ...
> [nmi_print_seq] Uhhuh. NMI received for unknown reason 20 on CPU 0.
> [nmi_print_seq] Do you have a strange power saving mode enabled?
> [nmi_print_seq] Dazed and confused, but trying to continue
> ===
>
> -m and -t options are also supported.
>
> Note that the safe buffer (struct printk_safe_seq_buf) was introduced
> in kernel-4.11 (Merge commit 7d91de74436a69c2b78a7a72f1e7f97f8b4396fa)
> and removed in kernel-5.15 (93d102f094be9beab28e5afb656c188b16a3793b).
>
> Changes since v2:
> - Add support new options -s, -t, -m (Kazu)
> - Add help text (Kazu)
Thank you for the update.
Maybe I will join the patches into two or three and the following warning
is emitted, so I will adjust a little when merging, but otherwise the
I would recommend packing them into two patches as below:
[PATCH v3 1/7] log: introduce -s option
[PATCH v3 2/7] log: adjust indent and line breaks for log -s
[PATCH v3 3/7] log: append printk safe buffer output to 'log'
[PATCH v3 6/7] symbols: add support 'help -o' for printk safe buffers
[PATCH v3 7/7] log: add help text for printk safe buffers
[PATCH v3 4/7] log: add support -t option for output of printk safe buffers
[PATCH v3 5/7] log: add support -m for output of printk safe buffers
BTW: If there is a better way, you could rearrange them when merging. Thanks.
And with the warning fix, otherwise:
patchset and the output of the commands look nice to me!
Acked-by: Kazuhito Hagio <k-hagio-ab@xxxxxxx>
$ make clean ; make warn
...
cc -c -g -DX86_64 -DLZO -DSNAPPY -DGDB_10_2 kernel.c -Wall -O2 -Wstrict-prototypes -Wmissing-prototypes -fstack-protector -Wformat-security
kernel.c: In function ?__dump_printk_safe_seq_buf?:
kernel.c:11623:7: warning: format not a string literal and no format arguments [-Wformat-security]
fprintf(fp, space(PRINTK_SAFE_SEQ_BUF_INDENT));
^~~~~~~
Will add "%s".
Thanks,
Kazu
>
> [v1]:
https://listman.redhat.com/archives/crash-utility/2021-December/msg00031.html
> [v2]:
https://listman.redhat.com/archives/crash-utility/2022-January/msg00004.html
>
> Test program is attached in the above v2 patch e-mail.
>
> Shogo Matsumoto (7):
> log: introduce -s option
> log: adjust indent and line breaks for log -s
> log: append printk safe buffer output to 'log'
> log: add support -t option for output of printk safe buffers
> log: add support -m for output of printk safe buffers
> symbols: add support 'help -o' for printk safe buffers
> log: add help text for printk safe buffers
>
> defs.h | 5 ++
> help.c | 25 ++++++++-
> kernel.c | 159 +++++++++++++++++++++++++++++++++++++++++++++++++++++-
> symbols.c | 5 ++
> 4 files changed, 192 insertions(+), 2 deletions(-)
>
> --
> 2.29.2
>
>
> --
> Crash-utility mailing list
> Crash-utility@xxxxxxxxxx
>
https://listman.redhat.com/mailman/listinfo/crash-utility
------------------------------
Message: 2
Date: Thu, 10 Feb 2022 08:21:52 +0000
From: HAGIO KAZUHITO(?????) <k-hagio-ab@xxxxxxx>
To: "Discussion list for crash utility usage, maintenance and
development" <crash-utility@xxxxxxxxxx>
Cc: "zwang@xxxxxxxxxxxxxxxxxxx" <zwang@xxxxxxxxxxxxxxxxxxx>,
"patches@xxxxxxxxxxxxxxxxxxx" <patches@xxxxxxxxxxxxxxxxxxx>,
"lijiang@xxxxxxxxxx" <lijiang@xxxxxxxxxx>,
"darren@xxxxxxxxxxxxxxxxxxxxxx" <darren@xxxxxxxxxxxxxxxxxxxxxx>
Subject: Re: [Crash-utility] [PATCH] arm64: Use CONFIG_ARM64_VA_BITS
to initialize VA_BITS_ACTUAL
Message-ID:
<TYYPR01MB6777479D629CD1F2A18A7A6DDD2F9@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset="iso-2022-jp"
Hi Huang,
thanks for the patch.
-----Original Message-----
> For DISKDUMP case, we can get VA_BITS_ACTUAL from CONFIG_ARM64_VA_BITS.
I could not understand this, there is a case where CONFIG_ARM64_VA_BITS
is different from VA_BITS_ACTUAL and why is this only for DISKDUMP case?
If the patch intends to guess the value of VA_BITS_ACTUAL to be the same as
CONFIG_ARM64_VA_BITS when no NUMBER(TCR_EL1_T1SZ), I think that DISKDUMP
check is not needed and it would be better to write such a commit log and
a comment e.g. "/* guess */" on the else if block.
Thanks,
Kazu
> Without this patch, we may have to use "--machdep vabits_actual=48" to
> set the VA_BITS_ACTUAL.
>
> Signed-off-by: Huang Shijie <shijie@xxxxxxxxxxxxxxxxxxxxxx>
> ---
> arm64.c | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/arm64.c b/arm64.c
> index 4f2c2b5..2b3ec02 100644
> --- a/arm64.c
> +++ b/arm64.c
> @@ -4170,6 +4170,12 @@ arm64_calc_VA_BITS(void)
> } else if (machdep->machspec->VA_BITS_ACTUAL) {
> machdep->machspec->VA_BITS = machdep->machspec->VA_BITS_ACTUAL;
> machdep->machspec->VA_START = _VA_START(machdep->machspec->VA_BITS_ACTUAL);
> + } else if (pc->flags & DISKDUMP) {
> + if (machdep->machspec->CONFIG_ARM64_VA_BITS) {
> + machdep->machspec->VA_BITS_ACTUAL =
> machdep->machspec->CONFIG_ARM64_VA_BITS;
> + machdep->machspec->VA_BITS =
> machdep->machspec->CONFIG_ARM64_VA_BITS;
> + machdep->machspec->VA_START =
> _VA_START(machdep->machspec->VA_BITS_ACTUAL);
> + }
> } else
> error(FATAL, "cannot determine VA_BITS_ACTUAL\n");
> }
> --
> 2.30.2
>
>
> --
> Crash-utility mailing list
> Crash-utility@xxxxxxxxxx
>
https://listman.redhat.com/mailman/listinfo/crash-utility
------------------------------
--
Crash-utility mailing list
Crash-utility@xxxxxxxxxx
https://listman.redhat.com/mailman/listinfo/crash-utility
End of Crash-utility Digest, Vol 197, Issue 7
*********************************************
|
