From: Kees Cook <keescook@chromium.org>
To: linux-kernel@vger.kernel.org
Cc: Kees Cook <keescook@chromium.org>,
Nobuhiro Iwamatsu <nobuhiro.iwamatsu.kw@hitachi.com>,
Qiuxu Zhuo <qiuxu.zhuo@intel.com>,
Ard Biesheuvel <ard.biesheuvel@linaro.org>,
Anton Vorontsov <anton@enomsg.org>,
Colin Cross <ccross@android.com>, Tony Luck <tony.luck@intel.com>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Paul Mackerras <paulus@samba.org>,
Michael Ellerman <mpe@ellerman.id.au>,
"Rafael J. Wysocki" <rjw@rjwysocki.net>,
Len Brown <lenb@kernel.org>,
Matt Fleming <matt@codeblueprint.co.uk>,
Nathan Fontenot <nfont@linux.vnet.ibm.com>,
Pan Xinhui <xinhui.pan@linux.vnet.ibm.com>,
Daniel Axtens <dja@axtens.net>,
Paul Gortmaker <paul.gortmaker@windriver.com>,
Geliang Tang <geliangtang@163.com>,
linuxppc-dev@lists.ozlabs.org, linux-acpi@vger.kernel.org,
linux-efi@vger.kernel.org, linux-doc@vger.kernel.org
Subject: [PATCH 12/18] pstore: Pass record contents instead of copying
Date: Mon, 6 Mar 2017 13:55:26 -0800 [thread overview]
Message-ID: <1488837332-71582-13-git-send-email-keescook@chromium.org> (raw)
In-Reply-To: <1488837332-71582-1-git-send-email-keescook@chromium.org>
pstore_mkfile() shouldn't have to memcpy the record contents. It can use
the existing copy instead. This adjusts the allocation lifetime management
and renames the contents variable from "data" to "buf" to assist moving to
struct pstore_record in the future.
Signed-off-by: Kees Cook <keescook@chromium.org>
---
fs/pstore/inode.c | 22 +++++++++++++++-------
fs/pstore/platform.c | 16 ++++++++++++----
2 files changed, 27 insertions(+), 11 deletions(-)
diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c
index a98787bab3e6..3d1f047e4f41 100644
--- a/fs/pstore/inode.c
+++ b/fs/pstore/inode.c
@@ -52,7 +52,7 @@ struct pstore_private {
u64 id;
int count;
ssize_t size;
- char data[];
+ char *buf;
};
struct pstore_ftrace_seq_data {
@@ -63,6 +63,14 @@ struct pstore_ftrace_seq_data {
#define REC_SIZE sizeof(struct pstore_ftrace_record)
+static void free_pstore_private(struct pstore_private *private)
+{
+ if (!private)
+ return;
+ kfree(private->buf);
+ kfree(private);
+}
+
static void *pstore_ftrace_seq_start(struct seq_file *s, loff_t *pos)
{
struct pstore_private *ps = s->private;
@@ -105,7 +113,7 @@ static int pstore_ftrace_seq_show(struct seq_file *s, void *v)
{
struct pstore_private *ps = s->private;
struct pstore_ftrace_seq_data *data = v;
- struct pstore_ftrace_record *rec = (void *)(ps->data + data->off);
+ struct pstore_ftrace_record *rec = (void *)(ps->buf + data->off);
seq_printf(s, "CPU:%d ts:%llu %08lx %08lx %pf <- %pF\n",
pstore_ftrace_decode_cpu(rec),
@@ -143,7 +151,7 @@ static ssize_t pstore_file_read(struct file *file, char __user *userbuf,
if (ps->type == PSTORE_TYPE_FTRACE)
return seq_read(file, userbuf, count, ppos);
- return simple_read_from_buffer(userbuf, count, ppos, ps->data, ps->size);
+ return simple_read_from_buffer(userbuf, count, ppos, ps->buf, ps->size);
}
static int pstore_file_open(struct inode *inode, struct file *file)
@@ -221,7 +229,7 @@ static void pstore_evict_inode(struct inode *inode)
spin_lock_irqsave(&allpstore_lock, flags);
list_del(&p->list);
spin_unlock_irqrestore(&allpstore_lock, flags);
- kfree(p);
+ free_pstore_private(p);
}
}
@@ -332,7 +340,7 @@ int pstore_mkfile(struct pstore_record *record)
goto fail;
inode->i_mode = S_IFREG | 0444;
inode->i_fop = &pstore_file_operations;
- private = kmalloc(sizeof *private + size, GFP_KERNEL);
+ private = kzalloc(sizeof(*private), GFP_KERNEL);
if (!private)
goto fail_alloc;
private->type = record->type;
@@ -394,7 +402,7 @@ int pstore_mkfile(struct pstore_record *record)
if (!dentry)
goto fail_lockedalloc;
- memcpy(private->data, record->buf, size);
+ private->buf = record->buf;
inode->i_size = private->size = size;
inode->i_private = private;
@@ -414,7 +422,7 @@ int pstore_mkfile(struct pstore_record *record)
fail_lockedalloc:
inode_unlock(d_inode(root));
- kfree(private);
+ free_pstore_private(private);
fail_alloc:
iput(inode);
diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c
index c0d401e732e6..d897e2f11b6a 100644
--- a/fs/pstore/platform.c
+++ b/fs/pstore/platform.c
@@ -828,14 +828,22 @@ void pstore_get_records(int quiet)
if (psi->open && psi->open(psi))
goto out;
+ /*
+ * Backend callback read() allocates record.buf. decompress_record()
+ * may reallocate record.buf. On success, pstore_mkfile() will keep
+ * the record.buf, so free it only on failure.
+ */
while ((record.size = psi->read(&record)) > 0) {
decompress_record(&record);
rc = pstore_mkfile(&record);
+ if (rc) {
+ /* pstore_mkfile() did not take buf, so free it. */
+ kfree(record.buf);
+ if (rc != -EEXIST || !quiet)
+ failed++;
+ }
- if (rc && (rc != -EEXIST || !quiet))
- failed++;
-
- kfree(record.buf);
+ /* Reset for next record. */
memset(&record, 0, sizeof(record));
record.psi = psi;
}
--
2.7.4
next prev parent reply other threads:[~2017-03-06 21:55 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-06 21:55 [PATCH 00/18] pstore: refactor internal APIs Kees Cook
2017-03-06 21:55 ` [PATCH 01/18] pstore: Use dynamic spinlock initializer Kees Cook
2017-03-06 21:55 ` [PATCH 02/18] pstore: Shut down worker when unregistering Kees Cook
2017-03-06 21:55 ` [PATCH 03/18] pstore: Avoid race in module unloading Kees Cook
2017-03-07 16:16 ` Namhyung Kim
2017-03-07 16:18 ` Kees Cook
2017-03-06 21:55 ` [PATCH 04/18] pstore: Improve register_pstore() error reporting Kees Cook
2017-03-06 21:55 ` [PATCH 05/18] pstore: Add kernel-doc for struct pstore_info Kees Cook
2017-03-06 21:55 ` [PATCH 06/18] pstore: Extract common arguments into structure Kees Cook
2017-03-07 16:22 ` Namhyung Kim
2017-03-07 19:31 ` Kees Cook
2017-03-06 21:55 ` [PATCH 07/18] pstore: Move record decompression to function Kees Cook
2017-03-06 21:55 ` [PATCH 08/18] pstore: Switch pstore_mkfile to pass record Kees Cook
2017-03-06 21:55 ` [PATCH 09/18] pstore: Replace arguments for read() API Kees Cook
2017-03-06 21:55 ` [PATCH 10/18] pstore: Replace arguments for write() API Kees Cook
2017-03-06 21:55 ` [PATCH 11/18] pstore: Always allocate buffer for decompression Kees Cook
2017-03-06 21:55 ` Kees Cook [this message]
2017-03-06 21:55 ` [PATCH 13/18] pstore: Allocate records on heap instead of stack Kees Cook
2017-03-06 21:55 ` [PATCH 14/18] pstore: Do not duplicate record metadata Kees Cook
2017-03-06 21:55 ` [PATCH 15/18] pstore: Replace arguments for erase() API Kees Cook
2017-03-06 21:55 ` [PATCH 16/18] pstore: Replace arguments for write_buf() API Kees Cook
2017-03-06 21:55 ` [PATCH 17/18] pstore: Replace arguments for write_buf_user() API Kees Cook
2017-03-06 21:55 ` [PATCH 18/18] pstore: Remove write_buf() callback Kees Cook
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=1488837332-71582-13-git-send-email-keescook@chromium.org \
--to=keescook@chromium.org \
--cc=anton@enomsg.org \
--cc=ard.biesheuvel@linaro.org \
--cc=benh@kernel.crashing.org \
--cc=ccross@android.com \
--cc=dja@axtens.net \
--cc=geliangtang@163.com \
--cc=lenb@kernel.org \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-efi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=matt@codeblueprint.co.uk \
--cc=mpe@ellerman.id.au \
--cc=nfont@linux.vnet.ibm.com \
--cc=nobuhiro.iwamatsu.kw@hitachi.com \
--cc=paul.gortmaker@windriver.com \
--cc=paulus@samba.org \
--cc=qiuxu.zhuo@intel.com \
--cc=rjw@rjwysocki.net \
--cc=tony.luck@intel.com \
--cc=xinhui.pan@linux.vnet.ibm.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).