linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeremy Kerr <jeremy.kerr@canonical.com>
To: linux-efi@vger.kernel.org, linux-kernel@vger.kernel.org
Cc: <glin@suse.com>,
	Lee@grenadilla.canonical.com, Chun-Yi <jlee@suse.com>,
	Matthew Garrett <mjg@redhat.com>, Peter Jones <pjones@redhat.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Matt Fleming <matt.fleming@intel.com>
Subject: [PATCH 3/3] efi: Handle deletions and size changes in efivarfs_write_file
Date: Fri, 05 Oct 2012 13:54:56 +0800	[thread overview]
Message-ID: <1349416496.810982.659496647274.3.gpush@pecola> (raw)
In-Reply-To: <1349416496.810727.310563927016.1.gpush@pecola>

A write to an efivarfs file will not always result in a variable of
'count' size after the EFI SetVariable() call. We may have appended to
the existing data (ie, with the EFI_VARIABLE_APPEND_WRITE attribute), or
even have deleted the variable (with an authenticated variable update,
with a zero datasize).

This change re-reads the updated variable from firmware, to check for
size changes and deletions. In the latter case, we need to drop the
dentry.

Signed-off-by: Jeremy Kerr <jeremy.kerr@canonical.com>

---
 drivers/firmware/efivars.c |   49 +++++++++++++++++++++++++++++--------
 1 file changed, 39 insertions(+), 10 deletions(-)

diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c
index e1253d6..a422de3 100644
--- a/drivers/firmware/efivars.c
+++ b/drivers/firmware/efivars.c
@@ -647,6 +647,7 @@ static ssize_t efivarfs_file_write(struct file *file,
 	u32 attributes;
 	struct inode *inode = file->f_mapping->host;
 	int datasize = count - sizeof(attributes);
+	unsigned long newdatasize;
 
 	if (count < sizeof(attributes))
 		return -EINVAL;
@@ -685,32 +686,60 @@ static ssize_t efivarfs_file_write(struct file *file,
 
 	switch (status) {
 	case EFI_SUCCESS:
-		mutex_lock(&inode->i_mutex);
-		i_size_write(inode, count);
-		mutex_unlock(&inode->i_mutex);
 		break;
 	case EFI_INVALID_PARAMETER:
 		count = -EINVAL;
-		break;
+		goto out;
 	case EFI_OUT_OF_RESOURCES:
 		count = -ENOSPC;
-		break;
+		goto out;
 	case EFI_DEVICE_ERROR:
 		count = -EIO;
-		break;
+		goto out;
 	case EFI_WRITE_PROTECTED:
 		count = -EROFS;
-		break;
+		goto out;
 	case EFI_SECURITY_VIOLATION:
 		count = -EACCES;
-		break;
+		goto out;
 	case EFI_NOT_FOUND:
 		count = -ENOENT;
-		break;
+		goto out;
 	default:
 		count = -EINVAL;
-		break;
+		goto out;
 	}
+
+	/*
+	 * Writing to the variable may have caused a change in size (which
+	 * could either be an append or an overwrite), or the variable to be
+	 * deleted. Perform a GetVariable() so we can tell what actually
+	 * happened.
+	 */
+	newdatasize = 0;
+	status = efivars->ops->get_variable(var->var.VariableName,
+					    &var->var.VendorGuid,
+					    NULL, &newdatasize,
+					    NULL);
+
+	if (status == EFI_BUFFER_TOO_SMALL) {
+		mutex_lock(&inode->i_mutex);
+		i_size_write(inode, newdatasize + sizeof(attributes));
+		mutex_unlock(&inode->i_mutex);
+
+	} else if (status == EFI_NOT_FOUND) {
+		spin_lock(&efivars->lock);
+		list_del(&var->list);
+		spin_unlock(&efivars->lock);
+		efivar_unregister(var);
+		drop_nlink(inode);
+		dput(file->f_dentry);
+
+	} else {
+		pr_warn("efivarfs: inconsistent EFI variable implementation? "
+				"status = %lx\n", status);
+	}
+
 out:
 	kfree(data);
 

  parent reply	other threads:[~2012-10-05  6:56 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-05  5:54 [PATCH 1/3] efi: Add support for a UEFI variable filesystem Jeremy Kerr
2012-10-05  5:54 ` [PATCH 2/3] efi: add efivars kobject to efi sysfs folder Jeremy Kerr
2012-10-05  6:51   ` joeyli
2012-10-05  7:44     ` Jeremy Kerr
2012-10-05  5:54 ` Jeremy Kerr [this message]
2012-10-06 19:32 ` [PATCH 1/3] efi: Add support for a UEFI variable filesystem Matt Fleming
2012-10-11 10:32 ` [PATCH 0/5] efivarfs: fixes and cleanups Andy Whitcroft
2012-10-11 10:32   ` [PATCH 1/5] efivarfs: efivarfs_file_read ensure we free data in error paths Andy Whitcroft
2012-10-11 13:53     ` Jeremy Kerr
2012-10-11 10:32   ` [PATCH 2/5] efivarfs: efivarfs_create() ensure we drop our reference on inode on error Andy Whitcroft
2012-10-11 14:13     ` Jeremy Kerr
2012-10-12 19:03     ` Khalid Aziz
2012-10-12 19:21       ` Matt Fleming
2012-10-12 20:11         ` Khalid Aziz
2012-10-11 10:32   ` [PATCH 3/5] efivarfs: efivarfs_fill_super() fix inode reference counts Andy Whitcroft
2012-10-11 14:10     ` Jeremy Kerr
2012-10-11 10:32   ` [PATCH 4/5] efivarfs: efivarfs_fill_super() ensure we free our temporary name Andy Whitcroft
2012-10-11 13:59     ` Jeremy Kerr
2012-10-11 10:32   ` [PATCH 5/5] efivarfs: efivarfs_fill_super() ensure we clean up correctly on error Andy Whitcroft
2012-10-11 14:04     ` Jeremy Kerr
2012-10-11 16:06       ` Andy Whitcroft
2012-10-16  9:16         ` Jeremy Kerr
2012-10-11 12:40   ` [PATCH 0/5] efivarfs: fixes and cleanups Matthew Garrett
2012-10-11 12:48   ` Matt Fleming

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=1349416496.810982.659496647274.3.gpush@pecola \
    --to=jeremy.kerr@canonical.com \
    --cc=Lee@grenadilla.canonical.com \
    --cc=glin@suse.com \
    --cc=hpa@zytor.com \
    --cc=jlee@suse.com \
    --cc=linux-efi@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=matt.fleming@intel.com \
    --cc=mjg@redhat.com \
    --cc=pjones@redhat.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).