From: Amit Kachhap <amit.kachhap@arm.com>
To: Mark Rutland <mark.rutland@arm.com>
Cc: linux-arm-kernel@lists.infradead.org,
Steven Rostedt <rostedt@goodmis.org>,
linux-kernel@vger.kernel.org, catalin.marinas@arm.com,
deller@gmx.de, duwe@suse.de,
James.Bottomley@HansenPartnership.com, james.morse@arm.com,
jeyu@kernel.org, jpoimboe@redhat.com, jthierry@redhat.com,
linux-parisc@vger.kernel.org, mingo@redhat.com,
peterz@infradead.org, svens@stackframe.org,
takahiro.akashi@linaro.org, will@kernel.org
Subject: Re: [PATCHv2 1/8] ftrace: add ftrace_init_nop()
Date: Tue, 5 Nov 2019 12:17:26 +0530 [thread overview]
Message-ID: <8e68de1f-f961-752d-9c07-ce41ce624d35@arm.com> (raw)
In-Reply-To: <20191104133657.GE45140@lakrids.cambridge.arm.com>
Hi Mark,
On 11/4/19 7:06 PM, Mark Rutland wrote:
> On Sat, Nov 02, 2019 at 05:49:00PM +0530, Amit Daniel Kachhap wrote:
>> Hi Mark,
>
> Hi Amit,
>
> Thanks for the review!
>
>> On 10/29/19 10:28 PM, Mark Rutland wrote:
>>> Architectures may need to perform special initialization of ftrace
>>> callsites, and today they do so by special-casing ftrace_make_nop() when
>>> the expected branch address is MCOUNT_ADDR. In some cases (e.g. for
>>> patchable-function-entry), we don't have an mcount-like symbol and don't
>> s/an mcount/a mcount.
>>> want a synthetic MCOUNT_ADDR, but we may need to perform some
>>> initialization of callsites.
>>>
>>> To make it possible to separate initialization from runtime
>>> modification, and to handle cases without an mcount-like symbol, this
>> Same as above.
>
> Using 'an' in both of these cases is correct, even though 'mcount'
> starts with a consonant, since it's pronounced as-if it were 'emcount'.
> I will leave this as-is.
Ok sure. It makes sense.
>
>>> patch adds an optional ftrace_init_nop() function that architectures can
>>> implement, which does not pass a branch address.
>>>
>>> Where an architecture does not provide ftrace_init_nop(), we will fall
>>> back to the existing behaviour of calling ftrace_make_nop() with
>>> MCOUNT_ADDR.
>>>
>>> At the same time, ftrace_code_disable() is renamed to
>>> ftrace_nop_initialize() to make it clearer that it is intended to
>>> intialize a callsite into a disabled state, and is not for disabling a
>>> callsite that has been runtime enabled. The kerneldoc description of rec
>>> arguments is updated to cover non-mcount callsites.
>>>
>>> Signed-off-by: Mark Rutland <mark.rutland@arm.com>
>>> Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
>>> Cc: Ingo Molnar <mingo@redhat.com>
>>> Cc: Steven Rostedt <rostedt@goodmis.org>
>>> Cc: Torsten Duwe <duwe@suse.de>
>>> ---
>>> include/linux/ftrace.h | 35 ++++++++++++++++++++++++++++++++---
>>> kernel/trace/ftrace.c | 6 +++---
>>> 2 files changed, 35 insertions(+), 6 deletions(-)
>>>
>>> diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
>>> index 8a8cb3c401b2..9867d90d635e 100644
>>> --- a/include/linux/ftrace.h
>>> +++ b/include/linux/ftrace.h
>>> @@ -499,7 +499,7 @@ static inline int ftrace_disable_ftrace_graph_caller(void) { return 0; }
>>> /**
>>> * ftrace_make_nop - convert code into nop
>>> * @mod: module structure if called by module load initialization
>>> - * @rec: the mcount call site record
>>> + * @rec: the call site record (e.g. mcount/fentry)
>>> * @addr: the address that the call site should be calling
>>> *
>>> * This is a very sensitive operation and great care needs
>>> @@ -520,9 +520,38 @@ static inline int ftrace_disable_ftrace_graph_caller(void) { return 0; }
>>> extern int ftrace_make_nop(struct module *mod,
>>> struct dyn_ftrace *rec, unsigned long addr);
>>> +
>>> +/**
>>> + * ftrace_init_nop - initialize a nop call site
>>> + * @mod: module structure if called by module load initialization
>>> + * @rec: the call site record (e.g. mcount/fentry)
>>> + *
>>> + * This is a very sensitive operation and great care needs
>>> + * to be taken by the arch. The operation should carefully
>>> + * read the location, check to see if what is read is indeed
>>> + * what we expect it to be, and then on success of the compare,
>>> + * it should write to the location.
>>> + *
>>> + * The code segment at @rec->ip should contain the contents created by
>>> + * the compiler
>> Nit: Will it be better to write it as "@rec->ip should store the adjusted
>> ftrace entry address of the call site" or something like that.
>
> This was the specific wording requested by Steve, and it's trying to
> describe the instructions at rec->ip, rather than the value of rec->ip,
> so I think it's better to leave this as-is.
ok Its fine this way too. Actually from the comment, I could not
understand which one of the compiler contents this points to as in this
case there are 2 nops.
>
>>> + *
>>> + * Return must be:
>>> + * 0 on success
>>> + * -EFAULT on error reading the location
>>> + * -EINVAL on a failed compare of the contents
>>> + * -EPERM on error writing to the location
>>> + * Any other value will be considered a failure.
>>> + */
>>> +#ifndef ftrace_init_nop
>>> +static inline int ftrace_init_nop(struct module *mod, struct dyn_ftrace *rec)
>>> +{
>>> + return ftrace_make_nop(mod, rec, MCOUNT_ADDR);
>>> +}
>>> +#endif
>>> +
>> Now that ftrace_init_nop is also an arch implemented function so it may be
>> added in Documentation/trace/ftrace-design.rst along with ftrace_make_nop.
>> In general also, adding some description about patchable-function-entry
>> in kernel Documentation will be useful.
>
> I agree that would be a good thing, but as Steve suggests I will leave
> this for subsequent rework, as I think that also implies more
> rework/renaming in the core code (e.g. to more cleanly separate the
> notion of a callsite from mcount specifically).
ok.
Thanks,
Amit
>
> Steve, if you'd like help with that (or review), I'd be happy to help.
>
> Thanks,
> Mark.
>
next prev parent reply other threads:[~2019-11-05 20:09 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-29 16:58 [PATCHv2 0/8] arm64: ftrace cleanup + FTRACE_WITH_REGS Mark Rutland
2019-10-29 16:58 ` [PATCHv2 1/8] ftrace: add ftrace_init_nop() Mark Rutland
2019-10-30 15:00 ` Miroslav Benes
2019-11-02 12:19 ` Amit Daniel Kachhap
2019-11-04 13:11 ` Steven Rostedt
2019-11-05 6:59 ` Amit Kachhap
2019-11-04 13:36 ` Mark Rutland
2019-11-05 6:47 ` Amit Kachhap [this message]
2019-11-06 14:15 ` Mark Rutland
2019-11-07 4:40 ` Amit Kachhap
2019-11-04 13:16 ` Steven Rostedt
2019-11-04 13:38 ` Mark Rutland
2019-11-04 13:53 ` Steven Rostedt
2019-10-29 16:58 ` [PATCHv2 2/8] module/ftrace: handle patchable-function-entry Mark Rutland
2019-10-30 15:03 ` Torsten Duwe
2019-10-31 9:02 ` Mark Rutland
2019-10-31 11:42 ` Torsten Duwe
2019-10-31 13:00 ` Mark Rutland
2019-11-04 13:28 ` Steven Rostedt
2019-11-04 14:00 ` Mark Rutland
2019-11-04 13:25 ` Steven Rostedt
2019-11-04 15:51 ` Mark Rutland
2019-11-04 20:58 ` Helge Deller
2019-11-05 8:59 ` Miroslav Benes
2019-10-29 16:58 ` [PATCHv2 3/8] arm64: module: rework special section handling Mark Rutland
2019-10-30 15:25 ` Miroslav Benes
2019-10-29 16:58 ` [PATCHv2 4/8] arm64: module/ftrace: intialize PLT at load time Mark Rutland
2019-11-02 12:20 ` Amit Daniel Kachhap
2019-11-04 13:55 ` Mark Rutland
2019-10-29 16:58 ` [PATCHv2 5/8] arm64: insn: add encoder for MOV (register) Mark Rutland
2019-10-29 16:58 ` [PATCHv2 6/8] arm64: asm-offsets: add S_FP Mark Rutland
2019-10-29 16:58 ` [PATCHv2 7/8] arm64: implement ftrace with regs Mark Rutland
2019-11-02 12:21 ` Amit Daniel Kachhap
2019-11-04 13:51 ` Mark Rutland
[not found] ` <CANW9uyug8WKN2fR-FmcW-C_OO_OQ_AvukM+BR7wqiJ9eFQMO9Q@mail.gmail.com>
2019-11-15 7:45 ` Torsten Duwe
2019-11-15 13:59 ` Mark Rutland
2019-10-29 16:58 ` [PATCHv2 8/8] arm64: ftrace: minimize ifdeffery Mark Rutland
2019-10-30 17:02 ` [PATCHv2 0/8] arm64: ftrace cleanup + FTRACE_WITH_REGS Torsten Duwe
2019-10-31 17:16 ` Torsten Duwe
2019-11-01 9:08 ` Mark Rutland
2019-11-01 15:39 ` Sven Schnelle
2019-11-01 16:28 ` Mark Rutland
2019-11-02 12:12 ` Amit Daniel Kachhap
2019-11-04 12:56 ` Will Deacon
2019-11-04 13:03 ` Amit Kachhap
2019-11-04 14:04 ` Mark Rutland
2019-11-05 7:06 ` Amit Kachhap
2019-11-07 11:31 ` Catalin Marinas
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=8e68de1f-f961-752d-9c07-ce41ce624d35@arm.com \
--to=amit.kachhap@arm.com \
--cc=James.Bottomley@HansenPartnership.com \
--cc=catalin.marinas@arm.com \
--cc=deller@gmx.de \
--cc=duwe@suse.de \
--cc=james.morse@arm.com \
--cc=jeyu@kernel.org \
--cc=jpoimboe@redhat.com \
--cc=jthierry@redhat.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-parisc@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=svens@stackframe.org \
--cc=takahiro.akashi@linaro.org \
--cc=will@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).