linux-riscv.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Palmer Dabbelt <palmer@sifive.com>,
	Paul Walmsley <paul.walmsley@sifive.com>
Cc: linux-mm@kvack.org, Damien Le Moal <damien.lemoal@wdc.com>,
	linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: [PATCH 12/17] riscv: implement remote sfence.i natively for M-mode
Date: Mon, 24 Jun 2019 07:43:06 +0200	[thread overview]
Message-ID: <20190624054311.30256-13-hch@lst.de> (raw)
In-Reply-To: <20190624054311.30256-1-hch@lst.de>

The RISC-V ISA only supports flushing the instruction cache for the local
CPU core.  For normal S-mode Linux remote flushing is offloaded to
machine mode using ecalls, but for M-mode Linux we'll have to do it
ourselves.  Use the same implementation as all the existing open source
SBI implementations by just doing an IPI to all remote cores to execute
th sfence.i instruction on every live core.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 arch/riscv/mm/cacheflush.c | 31 +++++++++++++++++++++++++++----
 1 file changed, 27 insertions(+), 4 deletions(-)

diff --git a/arch/riscv/mm/cacheflush.c b/arch/riscv/mm/cacheflush.c
index 9ebcff8ba263..10875ea1065e 100644
--- a/arch/riscv/mm/cacheflush.c
+++ b/arch/riscv/mm/cacheflush.c
@@ -10,10 +10,35 @@
 
 #include <asm/sbi.h>
 
+#ifdef CONFIG_M_MODE
+static void ipi_remote_fence_i(void *info)
+{
+	return local_flush_icache_all();
+}
+
+void flush_icache_all(void)
+{
+	on_each_cpu(ipi_remote_fence_i, NULL, 1);
+}
+
+static void flush_icache_cpumask(const cpumask_t *mask)
+{
+	on_each_cpu_mask(mask, ipi_remote_fence_i, NULL, 1);
+}
+#else /* CONFIG_M_MODE */
 void flush_icache_all(void)
 {
 	sbi_remote_fence_i(NULL);
 }
+static void flush_icache_cpumask(const cpumask_t *mask)
+{
+	cpumask_t hmask;
+
+	cpumask_clear(&hmask);
+	riscv_cpuid_to_hartid_mask(mask, &hmask);
+	sbi_remote_fence_i(hmask.bits);
+}
+#endif /* CONFIG_M_MODE */
 
 /*
  * Performs an icache flush for the given MM context.  RISC-V has no direct
@@ -28,7 +53,7 @@ void flush_icache_all(void)
 void flush_icache_mm(struct mm_struct *mm, bool local)
 {
 	unsigned int cpu;
-	cpumask_t others, hmask, *mask;
+	cpumask_t others, *mask;
 
 	preempt_disable();
 
@@ -47,9 +72,7 @@ void flush_icache_mm(struct mm_struct *mm, bool local)
 	cpumask_andnot(&others, mm_cpumask(mm), cpumask_of(cpu));
 	local |= cpumask_empty(&others);
 	if (mm != current->active_mm || !local) {
-		cpumask_clear(&hmask);
-		riscv_cpuid_to_hartid_mask(&others, &hmask);
-		sbi_remote_fence_i(hmask.bits);
+		flush_icache_cpumask(&others);
 	} else {
 		/*
 		 * It's assumed that at least one strongly ordered operation is
-- 
2.20.1


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

  parent reply	other threads:[~2019-06-24  5:44 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-24  5:42 RISC-V nommu support v2 Christoph Hellwig
2019-06-24  5:42 ` [PATCH 01/17] mm: provide a print_vma_addr stub for !CONFIG_MMU Christoph Hellwig
2019-06-24  5:42 ` [PATCH 02/17] mm: stub out all of swapops.h " Christoph Hellwig
2019-06-24  5:42 ` [PATCH 03/17] mm/nommu: fix the MAP_UNINITIALIZED flag Christoph Hellwig
2019-06-24  5:42 ` [PATCH 04/17] irqchip/sifive-plic: set max threshold for ignored handlers Christoph Hellwig
2019-06-24  5:42 ` [PATCH 05/17] riscv: use CSR_SATP instead of the legacy sptbr name in switch_mm Christoph Hellwig
2019-07-01 18:53   ` Atish Patra
2019-06-24  5:43 ` [PATCH 06/17] riscv: refactor the IPI code Christoph Hellwig
2019-06-24  5:43 ` [PATCH 07/17] riscv: abstract out CSR names for supervisor vs machine mode Christoph Hellwig
2019-07-01 18:37   ` Atish Patra
2019-06-24  5:43 ` [PATCH 08/17] riscv: improve the default power off implementation Christoph Hellwig
2019-07-01 21:07   ` Atish Patra
2019-06-24  5:43 ` [PATCH 09/17] riscv: provide a flat entry loader Christoph Hellwig
2019-06-24  5:43 ` [PATCH 10/17] riscv: read the hart ID from mhartid on boot Christoph Hellwig
2019-07-01 21:15   ` Atish Patra
2019-06-24  5:43 ` [PATCH 11/17] riscv: provide native clint access for M-mode Christoph Hellwig
2019-06-24  5:43 ` Christoph Hellwig [this message]
2019-06-24  5:43 ` [PATCH 13/17] riscv: poison SBI calls " Christoph Hellwig
2019-06-24  5:43 ` [PATCH 14/17] riscv: don't allow selecting SBI-based drivers " Christoph Hellwig
2019-06-24  5:43 ` [PATCH 15/17] riscv: use the correct interrupt levels " Christoph Hellwig
2019-06-24  5:43 ` [PATCH 16/17] riscv: clear the instruction cache and all registers when booting Christoph Hellwig
2019-07-01 21:26   ` Atish Patra
2019-07-08  8:26     ` Palmer Dabbelt
2019-08-13 15:40       ` Christoph Hellwig
2019-08-13 15:37     ` hch
2019-06-24  5:43 ` [PATCH 17/17] riscv: add nommu support Christoph Hellwig
2019-07-12 14:52   ` Vladimir Murzin
2019-06-24 11:47 ` RISC-V nommu support v2 Vladimir Murzin
2019-06-24 11:54   ` Christoph Hellwig
2019-06-24 13:08     ` Vladimir Murzin
2019-06-24 13:16       ` Christoph Hellwig
2019-06-25  7:31       ` Palmer Dabbelt
2019-06-25 12:37         ` Vladimir Murzin
2019-07-01  6:56 ` Christoph Hellwig
2019-07-01 16:06   ` Paul Walmsley

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=20190624054311.30256-13-hch@lst.de \
    --to=hch@lst.de \
    --cc=damien.lemoal@wdc.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=palmer@sifive.com \
    --cc=paul.walmsley@sifive.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).