From: Allen Pais <allen.pais@oracle.com>
To: Kirill Tkhai <tkhai@yandex.ru>
Cc: linux-rt-users <linux-rt-users@vger.kernel.org>,
"sparclinux@vger.kernel.org" <sparclinux@vger.kernel.org>,
"davem@davemloft.net" <davem@davemloft.net>,
"bigeasy@linutronix.de" <bigeasy@linutronix.de>
Subject: Re: [PATCH 3/4] sparc64: convert spinlock_t to raw_spinlock_t in mmu_context_t
Date: Wed, 19 Feb 2014 09:23:46 +0530 [thread overview]
Message-ID: <53042ACA.6060907@oracle.com> (raw)
In-Reply-To: <268891392209126@web5h.yandex.ru>
Kirill,
>
> 12.02.2014, 16:15, "Allen Pais" <allen.pais@oracle.com>:
>> On Wednesday 12 February 2014 05:13 PM, Kirill Tkhai wrote:
>>
>>> 12.02.2014, 15:29, "Allen Pais" <allen.pais@oracle.com>:
>>>>>>>> [ 1487.027884] I7: <rt_mutex_setprio+0x3c/0x2c0>
>>>>>>>> [ 1487.027885] Call Trace:
>>>>>>>> [ 1487.027887] [00000000004967dc] rt_mutex_setprio+0x3c/0x2c0
>>>>>>>> [ 1487.027892] [00000000004afe20] task_blocks_on_rt_mutex+0x180/0x200
>>>>>>>> [ 1487.027895] [0000000000819114] rt_spin_lock_slowlock+0x94/0x300
>>>>>>>> [ 1487.027897] [0000000000817ebc] __schedule+0x39c/0x53c
>>>>>>>> [ 1487.027899] [00000000008185fc] schedule+0x1c/0xc0
>>>>>>>> [ 1487.027908] [000000000048fff4] smpboot_thread_fn+0x154/0x2e0
>>>>>>>> [ 1487.027913] [000000000048753c] kthread+0x7c/0xa0
>>>>>>>> [ 1487.027920] [00000000004060c4] ret_from_syscall+0x1c/0x2c
>>>>>>>> [ 1487.027922] [0000000000000000] (null)
>>
>> I am not convinced that I've covered all tlb/smp code. Guess I'll need to dig more.
>
> ++all above. May we have to add one more crutch... Put preempt_disable() at begining of
> __set_pte_at() and enable at end...
I realized locking in tsb is very tricky. My attempts to try and get hackbench run
without causing a stall failed. So here's what I tried to fix it, am not sure if it's
an appropriate fix, I would love to get comments. I have tested this fix for over 24 hours
with hackbench and dd, the system did not stall :)
diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c
index 9eb10b4..24dcd29 100644
--- a/arch/sparc/mm/tsb.c
+++ b/arch/sparc/mm/tsb.c
@@ -6,6 +6,7 @@
#include <linux/kernel.h>
#include <linux/preempt.h>
#include <linux/slab.h>
+#include <linux/locallock.h>
#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/mmu_context.h>
@@ -14,6 +15,7 @@
#include <asm/oplib.h>
extern struct tsb swapper_tsb[KERNEL_TSB_NENTRIES];
+static DEFINE_LOCAL_IRQ_LOCK(tsb_lock);
static inline unsigned long tsb_hash(unsigned long vaddr, unsigned long hash_sh
{
@@ -71,9 +73,9 @@ static void __flush_tsb_one(struct tlb_batch *tb, unsigned lon
void flush_tsb_user(struct tlb_batch *tb)
{
struct mm_struct *mm = tb->mm;
- unsigned long nentries, base, flags;
+ unsigned long nentries, base;
- raw_spin_lock_irqsave(&mm->context.lock, flags);
+ local_lock(tsb_lock);
base = (unsigned long) mm->context.tsb_block[MM_TSB_BASE].tsb;
nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries;
@@ -90,7 +92,7 @@ void flush_tsb_user(struct tlb_batch *tb)
__flush_tsb_one(tb, HPAGE_SHIFT, base, nentries);
}
#endif
- raw_spin_unlock_irqrestore(&mm->context.lock, flags);
+ local_unlock(tsb_lock);
}
Thanks,
- Allen
WARNING: multiple messages have this Message-ID (diff)
From: Allen Pais <allen.pais@oracle.com>
To: Kirill Tkhai <tkhai@yandex.ru>
Cc: linux-rt-users <linux-rt-users@vger.kernel.org>,
"sparclinux@vger.kernel.org" <sparclinux@vger.kernel.org>,
"davem@davemloft.net" <davem@davemloft.net>,
"bigeasy@linutronix.de" <bigeasy@linutronix.de>
Subject: Re: [PATCH 3/4] sparc64: convert spinlock_t to raw_spinlock_t in mmu_context_t
Date: Wed, 19 Feb 2014 03:54:15 +0000 [thread overview]
Message-ID: <53042ACA.6060907@oracle.com> (raw)
In-Reply-To: <268891392209126@web5h.yandex.ru>
Kirill,
>
> 12.02.2014, 16:15, "Allen Pais" <allen.pais@oracle.com>:
>> On Wednesday 12 February 2014 05:13 PM, Kirill Tkhai wrote:
>>
>>> 12.02.2014, 15:29, "Allen Pais" <allen.pais@oracle.com>:
>>>>>>>> [ 1487.027884] I7: <rt_mutex_setprio+0x3c/0x2c0>
>>>>>>>> [ 1487.027885] Call Trace:
>>>>>>>> [ 1487.027887] [00000000004967dc] rt_mutex_setprio+0x3c/0x2c0
>>>>>>>> [ 1487.027892] [00000000004afe20] task_blocks_on_rt_mutex+0x180/0x200
>>>>>>>> [ 1487.027895] [0000000000819114] rt_spin_lock_slowlock+0x94/0x300
>>>>>>>> [ 1487.027897] [0000000000817ebc] __schedule+0x39c/0x53c
>>>>>>>> [ 1487.027899] [00000000008185fc] schedule+0x1c/0xc0
>>>>>>>> [ 1487.027908] [000000000048fff4] smpboot_thread_fn+0x154/0x2e0
>>>>>>>> [ 1487.027913] [000000000048753c] kthread+0x7c/0xa0
>>>>>>>> [ 1487.027920] [00000000004060c4] ret_from_syscall+0x1c/0x2c
>>>>>>>> [ 1487.027922] [0000000000000000] (null)
>>
>> I am not convinced that I've covered all tlb/smp code. Guess I'll need to dig more.
>
> ++all above. May we have to add one more crutch... Put preempt_disable() at begining of
> __set_pte_at() and enable at end...
I realized locking in tsb is very tricky. My attempts to try and get hackbench run
without causing a stall failed. So here's what I tried to fix it, am not sure if it's
an appropriate fix, I would love to get comments. I have tested this fix for over 24 hours
with hackbench and dd, the system did not stall :)
diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c
index 9eb10b4..24dcd29 100644
--- a/arch/sparc/mm/tsb.c
+++ b/arch/sparc/mm/tsb.c
@@ -6,6 +6,7 @@
#include <linux/kernel.h>
#include <linux/preempt.h>
#include <linux/slab.h>
+#include <linux/locallock.h>
#include <asm/page.h>
#include <asm/pgtable.h>
#include <asm/mmu_context.h>
@@ -14,6 +15,7 @@
#include <asm/oplib.h>
extern struct tsb swapper_tsb[KERNEL_TSB_NENTRIES];
+static DEFINE_LOCAL_IRQ_LOCK(tsb_lock);
static inline unsigned long tsb_hash(unsigned long vaddr, unsigned long hash_sh
{
@@ -71,9 +73,9 @@ static void __flush_tsb_one(struct tlb_batch *tb, unsigned lon
void flush_tsb_user(struct tlb_batch *tb)
{
struct mm_struct *mm = tb->mm;
- unsigned long nentries, base, flags;
+ unsigned long nentries, base;
- raw_spin_lock_irqsave(&mm->context.lock, flags);
+ local_lock(tsb_lock);
base = (unsigned long) mm->context.tsb_block[MM_TSB_BASE].tsb;
nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries;
@@ -90,7 +92,7 @@ void flush_tsb_user(struct tlb_batch *tb)
__flush_tsb_one(tb, HPAGE_SHIFT, base, nentries);
}
#endif
- raw_spin_unlock_irqrestore(&mm->context.lock, flags);
+ local_unlock(tsb_lock);
}
Thanks,
- Allen
next prev parent reply other threads:[~2014-02-19 3:53 UTC|newest]
Thread overview: 85+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-01-06 3:55 [PATCH 0/4] PREEMPT_RT support for sparc64 Allen Pais
2014-01-06 3:55 ` Allen Pais
2014-01-06 3:55 ` [PATCH 1/4] sparc64: use generic rwsem spinlocks rt Allen Pais
2014-01-06 3:56 ` Allen Pais
2014-01-06 3:55 ` [PATCH 2/4] sparc64: allow forced irq threading Allen Pais
2014-01-06 3:56 ` Allen Pais
2014-01-06 3:55 ` [PATCH 3/4] sparc64: convert spinlock_t to raw_spinlock_t in mmu_context_t Allen Pais
2014-01-06 3:56 ` Allen Pais
2014-02-11 21:13 ` Kirill Tkhai
2014-02-11 21:13 ` Kirill Tkhai
2014-02-12 7:31 ` Allen Pais
2014-02-12 7:43 ` Allen Pais
2014-02-12 7:48 ` Allen Pais
2014-02-12 7:48 ` Allen Pais
2014-02-12 8:33 ` Kirill Tkhai
2014-02-12 8:33 ` Kirill Tkhai
2014-02-12 11:28 ` Allen Pais
2014-02-12 11:40 ` Allen Pais
2014-02-12 11:43 ` Kirill Tkhai
2014-02-12 11:43 ` Kirill Tkhai
2014-02-12 12:14 ` Allen Pais
2014-02-12 12:26 ` Allen Pais
2014-02-12 12:45 ` Kirill Tkhai
2014-02-12 12:45 ` Kirill Tkhai
2014-02-12 13:05 ` Allen Pais
2014-02-12 13:17 ` Allen Pais
2014-02-19 3:53 ` Allen Pais [this message]
2014-02-19 3:54 ` Allen Pais
2014-02-19 8:09 ` Kirill Tkhai
2014-02-19 8:09 ` Kirill Tkhai
2014-02-19 8:12 ` Allen Pais
2014-02-19 8:24 ` Allen Pais
2014-02-19 8:57 ` Kirill Tkhai
2014-02-19 8:57 ` Kirill Tkhai
2014-02-19 8:59 ` Allen Pais
2014-02-19 8:59 ` Allen Pais
2014-02-19 9:13 ` Allen Pais
2014-02-19 9:25 ` Allen Pais
2014-02-19 9:25 ` Kirill Tkhai
2014-02-19 9:25 ` Kirill Tkhai
2014-02-19 9:31 ` Allen Pais
2014-02-19 9:43 ` Allen Pais
2014-02-26 7:51 ` Allen Pais
2014-02-26 7:52 ` Allen Pais
2014-02-28 14:51 ` Kirill Tkhai
2014-02-28 14:51 ` Kirill Tkhai
2014-03-04 19:10 ` David Miller
2014-03-04 19:10 ` David Miller
2014-03-04 20:28 ` David Miller
2014-03-04 20:28 ` David Miller
2014-03-05 4:30 ` Allen Pais
2014-03-05 4:42 ` Allen Pais
2014-03-06 21:36 ` David Miller
2014-03-06 21:36 ` David Miller
2014-03-07 14:05 ` Sebastian Andrzej Siewior
2014-03-07 14:05 ` Sebastian Andrzej Siewior
2014-03-04 20:39 ` Kirill Tkhai
2014-03-04 20:39 ` Kirill Tkhai
2014-03-07 13:41 ` Sebastian Andrzej Siewior
2014-03-07 13:41 ` Sebastian Andrzej Siewior
2014-03-04 20:03 ` David Miller
2014-03-04 20:03 ` David Miller
2014-03-04 21:26 ` Kirill Tkhai
2014-03-04 21:26 ` Kirill Tkhai
2014-03-04 20:01 ` David Miller
2014-03-04 20:01 ` David Miller
2014-03-05 4:34 ` Allen Pais
2014-03-05 4:46 ` Allen Pais
2014-03-05 4:52 ` David Miller
2014-03-05 4:52 ` David Miller
2014-03-04 19:59 ` David Miller
2014-03-04 19:59 ` David Miller
2014-03-04 19:55 ` David Miller
2014-03-04 19:55 ` David Miller
2014-03-04 20:44 ` Kirill Tkhai
2014-03-04 20:44 ` Kirill Tkhai
2014-03-07 14:29 ` Sebastian Andrzej Siewior
2014-03-07 14:29 ` Sebastian Andrzej Siewior
2014-01-06 3:55 ` [PATCH 4/4] sparc64: convert ctx_alloc_lock raw_spinlock_t Allen Pais
2014-01-06 3:56 ` Allen Pais
2014-02-05 3:31 ` [PATCH 0/4] PREEMPT_RT support for sparc64 Allen Pais
2014-02-05 8:28 ` Sebastian Andrzej Siewior
2014-02-05 10:38 ` Allen Pais
2014-02-05 10:43 ` Sebastian Andrzej Siewior
2014-02-05 10:51 ` Allen Pais
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=53042ACA.6060907@oracle.com \
--to=allen.pais@oracle.com \
--cc=bigeasy@linutronix.de \
--cc=davem@davemloft.net \
--cc=linux-rt-users@vger.kernel.org \
--cc=sparclinux@vger.kernel.org \
--cc=tkhai@yandex.ru \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.