linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Paul Burton <paul.burton@mips.com>
To: "linux-mips@vger.kernel.org" <linux-mips@vger.kernel.org>
Cc: Huacai Chen <chenhc@lemote.com>,
	Jiaxun Yang <jiaxun.yang@flygoat.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Paul Burton <pburton@wavecomp.com>
Subject: [PATCH 36/37] MIPS: genex: Don't reload address unnecessarily
Date: Mon, 30 Sep 2019 23:08:45 +0000	[thread overview]
Message-ID: <20190930230806.2940505-37-paul.burton@mips.com> (raw)
In-Reply-To: <20190930230806.2940505-1-paul.burton@mips.com>

In ejtag_debug_handler() we must reload the address of
ejtag_debug_buffer_spinlock if an sc fails, since the address in k0 will
have been clobbered by the result of the sc instruction. In the case
where we simply load a non-zero value (ie. there's contention for the
lock) the address will not be clobbered & we can simply branch back to
repeat the load from memory without reloading the address into k0.

The primary motivation for this change is that it moves the target of
the bnez instruction to an instruction within the LL/SC loop (the LL
itself), which we know contains no other memory accesses & therefore
isn't affected by Loongson3 LL/SC errata.

Signed-off-by: Paul Burton <paul.burton@mips.com>
---

 arch/mips/kernel/genex.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/mips/kernel/genex.S b/arch/mips/kernel/genex.S
index ac4f2b835165..60ede6b75a3b 100644
--- a/arch/mips/kernel/genex.S
+++ b/arch/mips/kernel/genex.S
@@ -355,8 +355,8 @@ NESTED(ejtag_debug_handler, PT_SIZE, sp)
 #ifdef CONFIG_SMP
 1:	PTR_LA	k0, ejtag_debug_buffer_spinlock
 	__SYNC(full, loongson3_war)
-	ll	k0, 0(k0)
-	bnez	k0, 1b
+2:	ll	k0, 0(k0)
+	bnez	k0, 2b
 	PTR_LA	k0, ejtag_debug_buffer_spinlock
 	sc	k0, 0(k0)
 	beqz	k0, 1b
-- 
2.23.0


  parent reply	other threads:[~2019-09-30 23:09 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-30 23:08 [PATCH 00/37] MIPS: barriers & atomics cleanups Paul Burton
2019-09-30 23:08 ` [PATCH 01/37] MIPS: Unify sc beqz definition Paul Burton
2019-09-30 23:08 ` [PATCH 02/37] MIPS: Use compact branch for LL/SC loops on MIPSr6+ Paul Burton
2019-09-30 23:08 ` [PATCH 03/37] MIPS: barrier: Add __SYNC() infrastructure Paul Burton
2019-09-30 23:08 ` [PATCH 04/37] MIPS: barrier: Clean up rmb() & wmb() definitions Paul Burton
2019-09-30 23:08 ` [PATCH 05/37] MIPS: barrier: Clean up __smp_mb() definition Paul Burton
2019-09-30 23:08 ` [PATCH 06/37] MIPS: barrier: Remove fast_mb() Octeon #ifdef'ery Paul Burton
2019-09-30 23:08 ` [PATCH 07/37] MIPS: barrier: Clean up __sync() definition Paul Burton
2019-09-30 23:08 ` [PATCH 08/37] MIPS: barrier: Clean up sync_ginv() Paul Burton
2019-09-30 23:08 ` [PATCH 09/37] MIPS: atomic: Fix whitespace in ATOMIC_OP macros Paul Burton
2019-09-30 23:08 ` [PATCH 11/37] MIPS: atomic: Use one macro to generate 32b & 64b functions Paul Burton
2019-09-30 23:08 ` [PATCH 10/37] MIPS: atomic: Handle !kernel_uses_llsc first Paul Burton
2019-09-30 23:08 ` [PATCH 12/37] MIPS: atomic: Emit Loongson3 sync workarounds within asm Paul Burton
2019-09-30 23:08 ` [PATCH 13/37] MIPS: atomic: Use _atomic barriers in atomic_sub_if_positive() Paul Burton
2019-09-30 23:08 ` [PATCH 14/37] MIPS: atomic: Unify 32b & 64b sub_if_positive Paul Burton
2019-09-30 23:08 ` [PATCH 15/37] MIPS: atomic: Deduplicate 32b & 64b read, set, xchg, cmpxchg Paul Burton
2019-09-30 23:08 ` [PATCH 17/37] MIPS: bitops: Handle !kernel_uses_llsc first Paul Burton
2019-09-30 23:08 ` [PATCH 16/37] MIPS: bitops: Use generic builtin ffs/fls; drop cpu_has_clo_clz Paul Burton
2019-09-30 23:08 ` [PATCH 18/37] MIPS: bitops: Only use ins for bit 16 or higher Paul Burton
2019-09-30 23:08 ` [PATCH 19/37] MIPS: bitops: Use MIPS_ISA_REV, not #ifdefs Paul Burton
2019-09-30 23:08 ` [PATCH 20/37] MIPS: bitops: ins start position is always an immediate Paul Burton
2019-09-30 23:08 ` [PATCH 21/37] MIPS: bitops: Implement test_and_set_bit() in terms of _lock variant Paul Burton
2019-09-30 23:08 ` [PATCH 22/37] MIPS: bitops: Allow immediates in test_and_{set,clear,change}_bit Paul Burton
2019-09-30 23:08 ` [PATCH 23/37] MIPS: bitops: Use the BIT() macro Paul Burton
2019-09-30 23:08 ` [PATCH 24/37] MIPS: bitops: Avoid redundant zero-comparison for non-LLSC Paul Burton
2019-09-30 23:08 ` [PATCH 25/37] MIPS: bitops: Abstract LL/SC loops Paul Burton
2019-09-30 23:08 ` [PATCH 26/37] MIPS: bitops: Use BIT_WORD() & BITS_PER_LONG Paul Burton
2019-09-30 23:08 ` [PATCH 27/37] MIPS: bitops: Emit Loongson3 sync workarounds within asm Paul Burton
2019-09-30 23:08 ` [PATCH 28/37] MIPS: bitops: Use smp_mb__before_atomic in test_* ops Paul Burton
2019-09-30 23:08 ` [PATCH 29/37] MIPS: cmpxchg: Emit Loongson3 sync workarounds within asm Paul Burton
2019-09-30 23:08 ` [PATCH 30/37] MIPS: cmpxchg: Omit redundant barriers for Loongson3 Paul Burton
2019-09-30 23:08 ` [PATCH 31/37] MIPS: futex: Emit Loongson3 sync workarounds within asm Paul Burton
2019-09-30 23:08 ` [PATCH 32/37] MIPS: syscall: " Paul Burton
2019-09-30 23:08 ` [PATCH 33/37] MIPS: barrier: Remove loongson_llsc_mb() Paul Burton
2019-09-30 23:08 ` [PATCH 34/37] MIPS: barrier: Make __smp_mb__before_atomic() a no-op for Loongson3 Paul Burton
2019-09-30 23:08 ` [PATCH 35/37] MIPS: genex: Add Loongson3 LL/SC workaround to ejtag_debug_handler Paul Burton
2019-09-30 23:08 ` Paul Burton [this message]
2019-09-30 23:08 ` [PATCH 37/37] MIPS: Check Loongson3 LL/SC errata workaround correctness Paul Burton

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=20190930230806.2940505-37-paul.burton@mips.com \
    --to=paul.burton@mips.com \
    --cc=chenhc@lemote.com \
    --cc=jiaxun.yang@flygoat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mips@vger.kernel.org \
    --cc=pburton@wavecomp.com \
    /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).