From: Nadav Amit <namit@vmware.com>
To: Joe Perches <joe@perches.com>
Cc: Dave Hansen <dave.hansen@intel.com>,
Andy Lutomirski <luto@kernel.org>,
Dave Hansen <dave.hansen@linux.intel.com>,
"x86@kernel.org" <x86@kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, Rik van Riel <riel@surriel.com>,
Josh Poimboeuf <jpoimboe@redhat.com>
Subject: Re: [PATCH v3 3/9] x86/mm/tlb: Open-code on_each_cpu_cond_mask() for tlb_is_not_lazy()
Date: Fri, 19 Jul 2019 23:02:47 +0000 [thread overview]
Message-ID: <D13BA044-AC02-46A7-9C5C-48FE8B6E8F63@vmware.com> (raw)
In-Reply-To: <511eb129bc0a6c92e3547c69a6e55695241dfe4a.camel@perches.com>
> On Jul 19, 2019, at 3:44 PM, Joe Perches <joe@perches.com> wrote:
>
> On Fri, 2019-07-19 at 18:41 +0000, Nadav Amit wrote:
>>> On Jul 19, 2019, at 11:36 AM, Dave Hansen <dave.hansen@intel.com> wrote:
>>>
>>> On 7/18/19 5:58 PM, Nadav Amit wrote:
>>>> @@ -865,7 +893,7 @@ void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch)
>>>> if (cpumask_test_cpu(cpu, &batch->cpumask)) {
>>>> lockdep_assert_irqs_enabled();
>>>> local_irq_disable();
>>>> - flush_tlb_func_local(&full_flush_tlb_info);
>>>> + flush_tlb_func_local((void *)&full_flush_tlb_info);
>>>> local_irq_enable();
>>>> }
>>>
>>> This looks like superfluous churn. Is it?
>>
>> Unfortunately not, since full_flush_tlb_info is defined as const. Without it
>> you would get:
>>
>> warning: passing argument 1 of ‘flush_tlb_func_local’ discards ‘const’ qualifier from pointer target type [-Wdiscarded-qualifiers]
>>
>> And flush_tlb_func_local() should get (void *) argument since it is also
>> used by the SMP infrastructure.
>
> huh?
>
> commit 3db6d5a5ecaf0a778d721ccf9809248350d4bfaf
> Author: Nadav Amit <namit@vmware.com>
> Date: Thu Apr 25 16:01:43 2019 -0700
>
> []
>
> -static void flush_tlb_func_local(void *info, enum tlb_flush_reason reason)
> +static void flush_tlb_func_local(const void *info, enum tlb_flush_reason reason)
Hopefully I decipher the “huh?” correctly...
When we moved the flush_tlb_info off the stack, in the patch that you
reference, we created a global full_flush_tlb_info variable, which was set
as const and delivered to flush_tlb_func_local. I changed the signature of
the function to avoid casting. IIRC, setting the variable as constant
slightly improved the generated assembly, and anyhow made sense.
In this patch-set I need to change the first argument of
flush_tlb_func_local to be non-const to match the SMP function signature
which takes a single non-const void * argument. Yes, there is what seems to
be non-safe casting, but flush_tlb_func_common() casts it back into a const
variable.
I know that I also added the infrastructure to run a function locally in the
SMP infrastructure, but it made sense to have the same signature for local
function and remote ones.
Does it make more sense now?
next prev parent reply other threads:[~2019-07-19 23:03 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-19 0:58 [PATCH v3 0/9] x86: Concurrent TLB flushes Nadav Amit
2019-07-19 0:58 ` [PATCH v3 1/9] smp: Run functions concurrently in smp_call_function_many() Nadav Amit
2019-07-19 18:23 ` Dave Hansen
2019-07-22 18:16 ` Peter Zijlstra
2019-07-22 18:41 ` Nadav Amit
2019-07-22 19:34 ` Peter Zijlstra
2019-07-22 18:21 ` Peter Zijlstra
2019-07-22 18:34 ` Nadav Amit
2019-07-22 19:19 ` Peter Zijlstra
2019-07-22 18:37 ` Thomas Gleixner
2019-07-22 18:40 ` Nadav Amit
2019-07-22 18:51 ` Thomas Gleixner
2019-07-22 19:02 ` Nadav Amit
2019-07-25 12:36 ` Thomas Gleixner
2019-07-25 19:10 ` Nadav Amit
2019-07-19 0:58 ` [PATCH v3 2/9] x86/mm/tlb: Remove reason as argument for flush_tlb_func_local() Nadav Amit
2019-07-19 0:58 ` [PATCH v3 3/9] x86/mm/tlb: Open-code on_each_cpu_cond_mask() for tlb_is_not_lazy() Nadav Amit
2019-07-19 18:36 ` Dave Hansen
2019-07-19 18:41 ` Nadav Amit
2019-07-19 22:44 ` Joe Perches
2019-07-19 23:02 ` Nadav Amit [this message]
2019-07-22 18:27 ` Peter Zijlstra
2019-07-22 19:47 ` Rasmus Villemoes
2019-07-22 19:51 ` Nadav Amit
2019-07-19 0:58 ` [PATCH v3 4/9] x86/mm/tlb: Flush remote and local TLBs concurrently Nadav Amit
2019-07-22 19:14 ` Peter Zijlstra
2019-07-22 19:27 ` Nadav Amit
2019-07-22 19:32 ` Peter Zijlstra
2019-07-26 7:28 ` Juergen Gross
2019-07-31 0:13 ` Michael Kelley
2019-07-19 0:58 ` [PATCH v3 5/9] x86/mm/tlb: Privatize cpu_tlbstate Nadav Amit
2019-07-19 18:38 ` Dave Hansen
2019-07-19 18:43 ` Nadav Amit
2019-07-19 18:48 ` Dave Hansen
2019-07-19 18:54 ` Nadav Amit
2019-07-20 13:58 ` Andy Lutomirski
2019-07-21 20:21 ` Nadav Amit
2019-07-19 0:58 ` [PATCH v3 6/9] x86/mm/tlb: Do not make is_lazy dirty for no reason Nadav Amit
2019-07-19 0:58 ` [PATCH v3 7/9] cpumask: Mark functions as pure Nadav Amit
2019-07-19 0:58 ` [PATCH v3 8/9] x86/mm/tlb: Remove UV special case Nadav Amit
2019-07-19 2:25 ` Mike Travis
2019-07-19 4:58 ` Nadav Amit
2019-07-31 3:11 ` Nadav Amit
2019-07-19 0:58 ` [PATCH v3 9/9] x86/mm/tlb: Remove unnecessary uses of the inline keyword Nadav Amit
2019-07-19 21:36 ` [PATCH v3 0/9] x86: Concurrent TLB flushes Dave Hansen
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=D13BA044-AC02-46A7-9C5C-48FE8B6E8F63@vmware.com \
--to=namit@vmware.com \
--cc=dave.hansen@intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=joe@perches.com \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=riel@surriel.com \
--cc=tglx@linutronix.de \
--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 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).