mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Denys Vlasenko <dvlasenk@redhat.com>
To: Tom Lendacky <thomas.lendacky@amd.com>, Joerg Roedel <jroedel@suse.de>
Cc: Denys Vlasenko <dvlasenk@redhat.com>, linux-kernel@vger.kernel.org
Subject: [PATCH] iommu/amd: Do not re-fetch iommu->cmd_buf_tail
Date: Thu, 24 Oct 2019 14:54:10 +0200	[thread overview]
Message-ID: <20191024125410.19224-1-dvlasenk@redhat.com> (raw)

The compiler is not smart enough to realize that iommu->cmd_buf_tail
can't be modified across memcpy:

41 8b 45 74          mov    0x74(%r13),%eax   # iommu->cmd_buf_tail
44 8d 78 10          lea    0x10(%rax),%r15d  # += sizeof(*cmd)
41 81 e7 ff 1f 00 00 and    $0x1fff,%r15d     # %= CMD_BUFFER_SIZE
49 03 45 68          add    0x68(%r13),%rax   # target = iommu->cmd_buf + iommu->cmd_buf_tail
45 89 7d 74          mov    %r15d,0x74(%r13)  # store to iommu->cmd_buf_tail
49 8b 34 24          mov    (%r12),%rsi       # memcpy
49 8b 7c 24 08       mov    0x8(%r12),%rdi    # memcpy
48 89 30             mov    %rsi,(%rax)       # memcpy
48 89 78 08          mov    %rdi,0x8(%rax)    # memcpy
49 8b 55 38          mov    0x38(%r13),%rdx   # iommu->mmio_base
41 8b 45 74          mov    0x74(%r13),%eax   # redundant load of iommu->cmd_buf_tail
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
89 82 08 20 00 00    mov    %eax,0x2008(%rdx) # writel

CC: Tom Lendacky <thomas.lendacky@amd.com>
CC: Joerg Roedel <jroedel@suse.de>
CC: linux-kernel@vger.kernel.org
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
---
 drivers/iommu/amd_iommu.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index dd555078258c..34c497c4b0a7 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -876,17 +876,18 @@ static void copy_cmd_to_buffer(struct amd_iommu *iommu,
 			       struct iommu_cmd *cmd)
 {
 	u8 *target;
-
-	target = iommu->cmd_buf + iommu->cmd_buf_tail;
-
-	iommu->cmd_buf_tail += sizeof(*cmd);
-	iommu->cmd_buf_tail %= CMD_BUFFER_SIZE;
+	u32 tail;
 
 	/* Copy command to buffer */
+	tail = iommu->cmd_buf_tail;
+	target = iommu->cmd_buf + tail;
 	memcpy(target, cmd, sizeof(*cmd));
 
+	tail = (tail + sizeof(*cmd)) % CMD_BUFFER_SIZE;
+	iommu->cmd_buf_tail = tail;
+
 	/* Tell the IOMMU about it */
-	writel(iommu->cmd_buf_tail, iommu->mmio_base + MMIO_CMD_TAIL_OFFSET);
+	writel(tail, iommu->mmio_base + MMIO_CMD_TAIL_OFFSET);
 }
 
 static void build_completion_wait(struct iommu_cmd *cmd, u64 address)
-- 
2.21.0


             reply	other threads:[~2019-10-24 12:54 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-24 12:54 Denys Vlasenko [this message]
2019-10-30  9:20 ` Joerg Roedel

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=20191024125410.19224-1-dvlasenk@redhat.com \
    --to=dvlasenk@redhat.com \
    --cc=jroedel@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=thomas.lendacky@amd.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

all inboxes | Powered by JetHome®