linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: David Howells <dhowells@redhat.com>
To: Al Viro <viro@zeniv.linux.org.uk>,
	Linus Torvalds <torvalds@linux-foundation.org>
Cc: David Howells <dhowells@redhat.com>, Jens Axboe <axboe@kernel.dk>,
	Christoph Hellwig <hch@lst.de>,
	Christian Brauner <christian@brauner.io>,
	David Laight <David.Laight@ACULAB.COM>,
	Matthew Wilcox <willy@infradead.org>,
	Jeff Layton <jlayton@kernel.org>,
	linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org,
	linux-mm@kvack.org, netdev@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: [PATCH v4 07/13] iov_iter: Make copy_from_iter() always handle MCE
Date: Wed, 13 Sep 2023 17:56:42 +0100	[thread overview]
Message-ID: <20230913165648.2570623-8-dhowells@redhat.com> (raw)
In-Reply-To: <20230913165648.2570623-1-dhowells@redhat.com>

Make copy_from_iter() always catch an MCE and return a short copy and make
the coredump code rely on that.  This requires arch support in the form of
a memcpy_mc() function that returns the length copied.

[?] Is it better to kill the thread in the event of an MCE occurring?

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Alexander Viro <viro@zeniv.linux.org.uk>
cc: Jens Axboe <axboe@kernel.dk>
cc: Christoph Hellwig <hch@lst.de>
cc: Christian Brauner <christian@brauner.io>
cc: Matthew Wilcox <willy@infradead.org>
cc: Linus Torvalds <torvalds@linux-foundation.org>
cc: David Laight <David.Laight@ACULAB.COM>
cc: linux-block@vger.kernel.org
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
---
 arch/x86/include/asm/mce.h | 23 +++++++++++++++++++++++
 fs/coredump.c              |  1 -
 lib/iov_iter.c             | 12 +++++-------
 3 files changed, 28 insertions(+), 8 deletions(-)

diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index 180b1cbfcc4e..77ce2044536c 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -353,4 +353,27 @@ static inline void mce_hygon_feature_init(struct cpuinfo_x86 *c)	{ return mce_am
 
 unsigned long copy_mc_fragile_handle_tail(char *to, char *from, unsigned len);
 
+static __always_inline __must_check
+size_t memcpy_mc(void *to, const void *from, size_t len)
+{
+#ifdef CONFIG_ARCH_HAS_COPY_MC
+	/*
+	 * If CPU has FSRM feature, use 'rep movs'.
+	 * Otherwise, use rep_movs_alternative.
+	 */
+	asm volatile(
+		"1:\n\t"
+		ALTERNATIVE("rep movsb",
+			    "call rep_movs_alternative", ALT_NOT(X86_FEATURE_FSRM))
+		"2:\n"
+		_ASM_EXTABLE_TYPE(1b, 2b, EX_TYPE_DEFAULT_MCE_SAFE)
+		:"+c" (len), "+D" (to), "+S" (from), ASM_CALL_CONSTRAINT
+		: : "memory", "rax", "r8", "r9", "r10", "r11");
+#else
+	memcpy(to, from, len);
+	return 0;
+#endif
+	return len;
+}
+
 #endif /* _ASM_X86_MCE_H */
diff --git a/fs/coredump.c b/fs/coredump.c
index 9d235fa14ab9..ad54102a5e14 100644
--- a/fs/coredump.c
+++ b/fs/coredump.c
@@ -884,7 +884,6 @@ static int dump_emit_page(struct coredump_params *cprm, struct page *page)
 	pos = file->f_pos;
 	bvec_set_page(&bvec, page, PAGE_SIZE, 0);
 	iov_iter_bvec(&iter, ITER_SOURCE, &bvec, 1, PAGE_SIZE);
-	iov_iter_set_copy_mc(&iter);
 	n = __kernel_write_iter(cprm->file, &iter, &pos);
 	if (n != PAGE_SIZE)
 		return 0;
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index 65374ee91ecd..b574601783bc 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -14,6 +14,7 @@
 #include <linux/scatterlist.h>
 #include <linux/instrumented.h>
 #include <linux/iov_iter.h>
+#include <asm/mce.h>
 
 static __always_inline
 size_t copy_to_user_iter(void __user *iter_to, size_t progress,
@@ -253,14 +254,11 @@ size_t _copy_mc_to_iter(const void *addr, size_t bytes, struct iov_iter *i)
 EXPORT_SYMBOL_GPL(_copy_mc_to_iter);
 #endif /* CONFIG_ARCH_HAS_COPY_MC */
 
-static size_t memcpy_from_iter_mc(void *iter_from, size_t progress,
-				  size_t len, void *to, void *priv2)
+static __always_inline
+size_t memcpy_from_iter_mc(void *iter_from, size_t progress,
+			   size_t len, void *to, void *priv2)
 {
-	struct iov_iter *iter = priv2;
-
-	if (iov_iter_is_copy_mc(iter))
-		return copy_mc_to_kernel(to + progress, iter_from, len);
-	return memcpy_from_iter(iter_from, progress, len, to, priv2);
+	return memcpy_mc(to + progress, iter_from, len);
 }
 
 size_t _copy_from_iter(void *addr, size_t bytes, struct iov_iter *i)


  parent reply	other threads:[~2023-09-13 16:58 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-13 16:56 [PATCH v4 00/13] iov_iter: Convert the iterator macros into inline funcs David Howells
2023-09-13 16:56 ` [PATCH v4 01/13] iov_iter: Add a benchmarking kunit test David Howells
2023-09-14  6:04   ` Johannes Thumshirn
2023-09-13 16:56 ` [PATCH v4 02/13] iov_iter: Renumber ITER_* constants David Howells
2023-09-13 16:56 ` [PATCH v4 03/13] iov_iter: Derive user-backedness from the iterator type David Howells
2023-09-13 16:56 ` [PATCH v4 04/13] iov_iter: Convert iterate*() to inline funcs David Howells
2023-09-13 16:56 ` [PATCH v4 05/13] iov: Move iterator functions to a header file David Howells
2023-09-14  9:06   ` David Laight
2023-09-15  9:38   ` David Howells
2023-09-13 16:56 ` [PATCH v4 06/13] iov_iter: Add a kernel-type iterator-only iteration function David Howells
2023-09-13 16:56 ` David Howells [this message]
2023-09-13 19:43   ` [PATCH v4 07/13] iov_iter: Make copy_from_iter() always handle MCE Linus Torvalds
2023-09-13 16:56 ` [PATCH v4 08/13] iov_iter: Remove the copy_mc flag and associated functions David Howells
2023-09-13 16:56 ` [PATCH v4 09/13] iov_iter, net: Move csum_and_copy_to/from_iter() to net/ David Howells
2023-09-13 16:56 ` [PATCH v4 10/13] iov_iter, net: Fold in csum_and_memcpy() David Howells
2023-09-13 16:56 ` [PATCH v4 11/13] iov_iter, net: Merge csum_and_copy_from_iter{,_full}() together David Howells
2023-09-13 16:56 ` [PATCH v4 12/13] iov_iter, net: Move hash_and_copy_to_iter() to net/ David Howells
2023-09-13 16:56 ` [PATCH v4 13/13] iov_iter: Create a fake device to allow iov_iter testing/benchmarking David Howells

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=20230913165648.2570623-8-dhowells@redhat.com \
    --to=dhowells@redhat.com \
    --cc=David.Laight@ACULAB.COM \
    --cc=axboe@kernel.dk \
    --cc=christian@brauner.io \
    --cc=hch@lst.de \
    --cc=jlayton@kernel.org \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=netdev@vger.kernel.org \
    --cc=torvalds@linux-foundation.org \
    --cc=viro@zeniv.linux.org.uk \
    --cc=willy@infradead.org \
    /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).