public inbox for opensbi@lists.infradead.org
 help / color / mirror / Atom feed
From: cp0613@linux.alibaba.com
To: opensbi@lists.infradead.org
Cc: anup@brainfault.org, samuel.holland@sifive.com,
	guoren@kernel.org, Chen Pei <cp0613@linux.alibaba.com>
Subject: [PATCH v2 2/2] lib: sbi: Flush cache entries after writing PMP CSRs
Date: Thu, 26 Feb 2026 20:34:08 +0800	[thread overview]
Message-ID: <20260226123408.1597-3-cp0613@linux.alibaba.com> (raw)
In-Reply-To: <20260226123408.1597-1-cp0613@linux.alibaba.com>

From: Chen Pei <cp0613@linux.alibaba.com>

As the privileged specification states, after writing to the PMP CSRs,
a SFENCE.VMA or HFENCE.GVMA instruction should be executed with rs1=x0
and rs2=x0 to flush all address translation cache entries.

Considering the performance issues caused by flushing all address
translation cache entries, sbi_hart_pmp_fence_vma is used.

Signed-off-by: Chen Pei <cp0613@linux.alibaba.com>
---
 lib/sbi/sbi_hart_pmp.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/lib/sbi/sbi_hart_pmp.c b/lib/sbi/sbi_hart_pmp.c
index 80407110..dc7f61fb 100644
--- a/lib/sbi/sbi_hart_pmp.c
+++ b/lib/sbi/sbi_hart_pmp.c
@@ -270,14 +270,21 @@ static int sbi_hart_smepmp_map_range(struct sbi_scratch *scratch,
 			     pmp_flags, base, order);
 	pmp_set(SBI_SMEPMP_RESV_ENTRY, pmp_flags, base, order);
 
+	sbi_hart_pmp_fence_vma(addr, size);
+
 	return SBI_OK;
 }
 
 static int sbi_hart_smepmp_unmap_range(struct sbi_scratch *scratch,
 				       unsigned long addr, unsigned long size)
 {
+	int ret;
+
 	sbi_platform_pmp_disable(sbi_platform_ptr(scratch), SBI_SMEPMP_RESV_ENTRY);
-	return pmp_disable(SBI_SMEPMP_RESV_ENTRY);
+	ret = pmp_disable(SBI_SMEPMP_RESV_ENTRY);
+	sbi_hart_pmp_fence_vma(addr, size);
+
+	return ret;
 }
 
 static int sbi_hart_oldpmp_configure(struct sbi_scratch *scratch)
@@ -333,6 +340,8 @@ static void sbi_hart_pmp_unconfigure(struct sbi_scratch *scratch)
 		sbi_platform_pmp_disable(sbi_platform_ptr(scratch), i);
 		pmp_disable(i);
 	}
+
+	sbi_hart_pmp_fence_all();
 }
 
 static struct sbi_hart_protection pmp_protection = {
-- 
2.50.1


-- 
opensbi mailing list
opensbi@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/opensbi

  parent reply	other threads:[~2026-02-26 12:35 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-26 12:34 [PATCH v2 0/2] lib: sbi: Flush cache entries after writing PMP CSRs cp0613
2026-02-26 12:34 ` [PATCH v2 1/2] lib: sbi: Introduce sbi_hart_pmp_fence_vma cp0613
2026-02-26 12:34 ` cp0613 [this message]
2026-02-27  1:19   ` [PATCH v2 2/2] lib: sbi: Flush cache entries after writing PMP CSRs Guo Ren
2026-02-27  1:21   ` Guo Ren

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=20260226123408.1597-3-cp0613@linux.alibaba.com \
    --to=cp0613@linux.alibaba.com \
    --cc=anup@brainfault.org \
    --cc=guoren@kernel.org \
    --cc=opensbi@lists.infradead.org \
    --cc=samuel.holland@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