[Qemu-devel] [PATCH v2] linux-user: Pass through nanosecond timestamp components for stat syscalls

Chen-Yu Tsai posted 1 patch 4 years, 11 months ago
Test FreeBSD passed
Test docker-clang@ubuntu passed
Test s390x passed
Test docker-mingw@fedora passed
Test checkpatch failed
Test asan passed
Patches applied successfully (tree, apply log)
git fetch https://github.com/patchew-project/qemu tags/patchew/20190522153631.17152-1-wens@kernel.org
Maintainers: Laurent Vivier <laurent@vivier.eu>, Riku Voipio <riku.voipio@iki.fi>
There is a newer version of this series
linux-user/syscall.c      | 19 +++++++++++++++
linux-user/syscall_defs.h | 49 +++++++++++++++++++++++++--------------
2 files changed, 50 insertions(+), 18 deletions(-)
[Qemu-devel] [PATCH v2] linux-user: Pass through nanosecond timestamp components for stat syscalls
Posted by Chen-Yu Tsai 4 years, 11 months ago
From: Chen-Yu Tsai <wens@csie.org>

Since Linux 2.6 the stat syscalls have mostly supported nanosecond
components for each of the file-related timestamps.

QEMU user mode emulation currently does not pass through the nanosecond
portion of the timestamp, even when the host system fills in the value.
This results in a mismatch when run on subsecond resolution filesystems
such as ext4 or XFS.

An example of this leading to inconsistency is cross-debootstraping a
full desktop root filesystem of Debian Buster. Recent versions of
fontconfig store the full timestamp (instead of just the second portion)
of the directory in its per-directory cache file, and checks this against
the directory to see if the cache is up-to-date. With QEMU user mode
emulation, the timestamp stored is incorrect, and upon booting the rootfs
natively, fontconfig discovers the mismatch, and proceeds to rebuild the
cache on the comparatively slow machine (low-power ARM vs x86). This
stalls the first attempt to open whatever application that incorporates
fontconfig.

This patch renames the "unused" padding trailing each timestamp element
to its nanosecond counterpart name if such an element exists in the
kernel sources for the given platform. Not all do. Then have the syscall
wrapper fill in the nanosecond portion if the host supports it, as
specified by the _POSIX_C_SOURCE and _XOPEN_SOURCE feature macros.

Recent versions of glibc only use stat64 and newfstatat syscalls on
32-bit and 64-bit platforms respectively. The changes in this patch
were tested by directly calling the stat, stat64 and newfstatat syscalls
directly, in addition to the glibc wrapper, on arm and aarch64 little
endian targets.

Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Chen-Yu Tsai <wens@csie.org>

---

Changes since v1:

  - Define TARGET_STAT_HAS_NSEC for targets that have the *_nsec fields
    in their struct stat

  - Only copy the *_nsec fields in the stat (not stat64) code path if
    TARGET_STAT_HAS_NSEC is defined

  - Picked up Reviewed-by from Laurent

As Laurent mentioned, linux-user/syscall_defs.h is indented with tabs,
so the checkpatch.pl errors can be ignored.

Original cover letter:

This issue was found while integrating some software that uses newer
versions of fontconfig into Raspbian images. We found that the first
launch of said software always stalls with fontconfig regenerating its
font cache files. Upon closer examination I found the timestamps were
not matching. The rest is explained above. Currently we're just working
around the problem by patching the correct timestamps into the cache
files after the fact.

Please consider this a drive-by scratch-my-own-itch contribution, but I
will stick around to deal with any comments raised during review. I'm
not on the mailing lists either, so please keep me in CC.

checkpatch returns "ERROR: code indent should never use tabs" for
linux-user/syscall_defs.h, however as far as I can tell the whole file
is indented with tabs. I'm not sure what to make of this.

Finally, I think this could be worth backporting to older versions.
---
 linux-user/syscall.c      | 19 +++++++++++++++
 linux-user/syscall_defs.h | 49 +++++++++++++++++++++++++--------------
 2 files changed, 50 insertions(+), 18 deletions(-)

diff --git a/linux-user/syscall.c b/linux-user/syscall.c
index efa3ec2837..6d0f13c155 100644
--- a/linux-user/syscall.c
+++ b/linux-user/syscall.c
@@ -6408,6 +6408,11 @@ static inline abi_long host_to_target_stat64(void *cpu_env,
         __put_user(host_st->st_atime, &target_st->target_st_atime);
         __put_user(host_st->st_mtime, &target_st->target_st_mtime);
         __put_user(host_st->st_ctime, &target_st->target_st_ctime);
+#if _POSIX_C_SOURCE >= 200809L || _XOPEN_SOURCE >= 700
+        __put_user(host_st->st_atim.tv_nsec, &target_st->target_st_atime_nsec);
+        __put_user(host_st->st_mtim.tv_nsec, &target_st->target_st_mtime_nsec);
+        __put_user(host_st->st_ctim.tv_nsec, &target_st->target_st_ctime_nsec);
+#endif
         unlock_user_struct(target_st, target_addr, 1);
     } else
 #endif
@@ -6438,6 +6443,11 @@ static inline abi_long host_to_target_stat64(void *cpu_env,
         __put_user(host_st->st_atime, &target_st->target_st_atime);
         __put_user(host_st->st_mtime, &target_st->target_st_mtime);
         __put_user(host_st->st_ctime, &target_st->target_st_ctime);
+#if _POSIX_C_SOURCE >= 200809L || _XOPEN_SOURCE >= 700
+        __put_user(host_st->st_atim.tv_nsec, &target_st->target_st_atime_nsec);
+        __put_user(host_st->st_mtim.tv_nsec, &target_st->target_st_mtime_nsec);
+        __put_user(host_st->st_ctim.tv_nsec, &target_st->target_st_ctime_nsec);
+#endif
         unlock_user_struct(target_st, target_addr, 1);
     }
 
@@ -8866,6 +8876,15 @@ static abi_long do_syscall1(void *cpu_env, int num, abi_long arg1,
                 __put_user(st.st_atime, &target_st->target_st_atime);
                 __put_user(st.st_mtime, &target_st->target_st_mtime);
                 __put_user(st.st_ctime, &target_st->target_st_ctime);
+#if (_POSIX_C_SOURCE >= 200809L || _XOPEN_SOURCE >= 700) && \
+    defined(TARGET_STAT_HAS_NSEC)
+                __put_user(st.st_atim.tv_nsec,
+                           &target_st->target_st_atime_nsec);
+                __put_user(st.st_mtim.tv_nsec,
+                           &target_st->target_st_mtime_nsec);
+                __put_user(st.st_ctim.tv_nsec,
+                           &target_st->target_st_ctime_nsec);
+#endif
                 unlock_user_struct(target_st, arg2, 1);
             }
         }
diff --git a/linux-user/syscall_defs.h b/linux-user/syscall_defs.h
index 1f5b2d18db..f02f3deb6a 100644
--- a/linux-user/syscall_defs.h
+++ b/linux-user/syscall_defs.h
@@ -1179,6 +1179,7 @@ struct target_winsize {
 #if (defined(TARGET_I386) && defined(TARGET_ABI32)) \
     || (defined(TARGET_ARM) && defined(TARGET_ABI32)) \
     || defined(TARGET_CRIS)
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
 	unsigned short st_dev;
 	unsigned short __pad1;
@@ -1193,11 +1194,11 @@ struct target_stat {
 	abi_ulong  st_blksize;
 	abi_ulong  st_blocks;
 	abi_ulong  target_st_atime;
-	abi_ulong  __unused1;
+	abi_ulong  target_st_atime_nsec;
 	abi_ulong  target_st_mtime;
-	abi_ulong  __unused2;
+	abi_ulong  target_st_mtime_nsec;
 	abi_ulong  target_st_ctime;
-	abi_ulong  __unused3;
+	abi_ulong  target_st_ctime_nsec;
 	abi_ulong  __unused4;
 	abi_ulong  __unused5;
 };
@@ -1229,13 +1230,13 @@ struct target_stat64 {
 	abi_ulong	__pad4;		/* future possible st_blocks high bits */
 
 	abi_ulong	target_st_atime;
-	abi_ulong	__pad5;
+	abi_ulong	target_st_atime_nsec;
 
 	abi_ulong	target_st_mtime;
-	abi_ulong	__pad6;
+	abi_ulong	target_st_mtime_nsec;
 
 	abi_ulong	target_st_ctime;
-	abi_ulong	__pad7;		/* will be high 32 bits of ctime someday */
+	abi_ulong	target_st_ctime_nsec;
 
 	unsigned long long	st_ino;
 } QEMU_PACKED;
@@ -1314,19 +1315,20 @@ struct target_stat64 {
 	unsigned int	st_blocks;
 
 	abi_ulong	target_st_atime;
-	abi_ulong	__unused1;
+	abi_ulong	target_st_atime_nsec;
 
 	abi_ulong	target_st_mtime;
-	abi_ulong	__unused2;
+	abi_ulong	target_st_mtime_nsec;
 
 	abi_ulong	target_st_ctime;
-	abi_ulong	__unused3;
+	abi_ulong	target_st_ctime_nsec;
 
 	abi_ulong	__unused4[3];
 };
 
 #elif defined(TARGET_SPARC)
 
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
 	unsigned short	st_dev;
 	abi_ulong	st_ino;
@@ -1337,14 +1339,14 @@ struct target_stat {
 	unsigned short	st_rdev;
 	abi_long	st_size;
 	abi_long	target_st_atime;
-	abi_ulong	__unused1;
+	abi_ulong	target_st_atime_nsec;
 	abi_long	target_st_mtime;
-	abi_ulong	__unused2;
+	abi_ulong	target_st_mtime_nsec;
 	abi_long	target_st_ctime;
-	abi_ulong	__unused3;
+	abi_ulong	target_st_ctime_nsec;
 	abi_long	st_blksize;
 	abi_long	st_blocks;
-	abi_ulong	__unused4[2];
+	abi_ulong	__unused1[2];
 };
 
 #define TARGET_HAS_STRUCT_STAT64
@@ -1372,20 +1374,21 @@ struct target_stat64 {
 	unsigned int	st_blocks;
 
 	unsigned int	target_st_atime;
-	unsigned int	__unused1;
+	unsigned int	target_st_atime_nsec;
 
 	unsigned int	target_st_mtime;
-	unsigned int	__unused2;
+	unsigned int	target_st_mtime_nsec;
 
 	unsigned int	target_st_ctime;
-	unsigned int	__unused3;
+	unsigned int	target_st_ctime_nsec;
 
-	unsigned int	__unused4;
-	unsigned int	__unused5;
+	unsigned int	__unused1;
+	unsigned int	__unused2;
 };
 
 #elif defined(TARGET_PPC)
 
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
 	abi_ulong st_dev;
 	abi_ulong st_ino;
@@ -1443,6 +1446,7 @@ struct QEMU_PACKED target_stat64 {
 
 #elif defined(TARGET_MICROBLAZE)
 
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
 	abi_ulong st_dev;
 	abi_ulong st_ino;
@@ -1558,6 +1562,7 @@ struct target_stat64 {
 
 #elif defined(TARGET_ABI_MIPSN64)
 
+#define TARGET_STAT_HAS_NSEC
 /* The memory layout is the same as of struct stat64 of the 32-bit kernel.  */
 struct target_stat {
 	unsigned int		st_dev;
@@ -1597,6 +1602,7 @@ struct target_stat {
 
 #elif defined(TARGET_ABI_MIPSN32)
 
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
         abi_ulong    st_dev;
         abi_ulong    st_pad0[3]; /* Reserved for st_dev expansion */
@@ -1621,6 +1627,7 @@ struct target_stat {
 
 #elif defined(TARGET_ABI_MIPSO32)
 
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
 	unsigned	st_dev;
 	abi_long	st_pad1[3];		/* Reserved for network id */
@@ -1737,6 +1744,7 @@ struct target_stat64 {
 
 #elif defined(TARGET_SH4)
 
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
 	abi_ulong  st_dev;
 	abi_ulong  st_ino;
@@ -1796,6 +1804,7 @@ struct QEMU_PACKED target_stat64 {
 };
 
 #elif defined(TARGET_I386) && !defined(TARGET_ABI32)
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
 	abi_ulong	st_dev;
 	abi_ulong	st_ino;
@@ -1841,6 +1850,7 @@ struct target_stat {
     abi_ulong  __unused[3];
 };
 #elif defined(TARGET_AARCH64)
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
     abi_ulong  st_dev;
     abi_ulong  st_ino;
@@ -1863,6 +1873,7 @@ struct target_stat {
     unsigned int __unused[2];
 };
 #elif defined(TARGET_XTENSA)
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
     abi_ulong       st_dev;
     abi_ulong       st_ino;
@@ -1912,6 +1923,7 @@ struct target_stat64  {
 
 /* These are the asm-generic versions of the stat and stat64 structures */
 
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
     abi_ulong st_dev;
     abi_ulong st_ino;
@@ -1963,6 +1975,7 @@ struct target_stat64 {
 
 #elif defined(TARGET_HPPA)
 
+#define TARGET_STAT_HAS_NSEC
 struct target_stat {
     abi_uint   st_dev;
     abi_uint   st_ino;
-- 
2.20.1


Re: [Qemu-devel] [PATCH v2] linux-user: Pass through nanosecond timestamp components for stat syscalls
Posted by Laurent Vivier 4 years, 11 months ago
On 22/05/2019 17:36, Chen-Yu Tsai wrote:
> From: Chen-Yu Tsai <wens@csie.org>
> 
> Since Linux 2.6 the stat syscalls have mostly supported nanosecond
> components for each of the file-related timestamps.
> 
> QEMU user mode emulation currently does not pass through the nanosecond
> portion of the timestamp, even when the host system fills in the value.
> This results in a mismatch when run on subsecond resolution filesystems
> such as ext4 or XFS.
> 
> An example of this leading to inconsistency is cross-debootstraping a
> full desktop root filesystem of Debian Buster. Recent versions of
> fontconfig store the full timestamp (instead of just the second portion)
> of the directory in its per-directory cache file, and checks this against
> the directory to see if the cache is up-to-date. With QEMU user mode
> emulation, the timestamp stored is incorrect, and upon booting the rootfs
> natively, fontconfig discovers the mismatch, and proceeds to rebuild the
> cache on the comparatively slow machine (low-power ARM vs x86). This
> stalls the first attempt to open whatever application that incorporates
> fontconfig.
> 
> This patch renames the "unused" padding trailing each timestamp element
> to its nanosecond counterpart name if such an element exists in the
> kernel sources for the given platform. Not all do. Then have the syscall
> wrapper fill in the nanosecond portion if the host supports it, as
> specified by the _POSIX_C_SOURCE and _XOPEN_SOURCE feature macros.
> 
> Recent versions of glibc only use stat64 and newfstatat syscalls on
> 32-bit and 64-bit platforms respectively. The changes in this patch
> were tested by directly calling the stat, stat64 and newfstatat syscalls
> directly, in addition to the glibc wrapper, on arm and aarch64 little
> endian targets.
> 
> Reviewed-by: Laurent Vivier <laurent@vivier.eu>
> Signed-off-by: Chen-Yu Tsai <wens@csie.org>
> 
> ---
> 
> Changes since v1:
> 
>    - Define TARGET_STAT_HAS_NSEC for targets that have the *_nsec fields
>      in their struct stat
> 
>    - Only copy the *_nsec fields in the stat (not stat64) code path if
>      TARGET_STAT_HAS_NSEC is defined


Please use the same syntax as the kernel does, it helps for code review:

   TARGET_STAT_HAVE_NSEC

Thanks,
Laurent

Re: [Qemu-devel] [PATCH v2] linux-user: Pass through nanosecond timestamp components for stat syscalls
Posted by no-reply@patchew.org 4 years, 11 months ago
Patchew URL: https://patchew.org/QEMU/20190522153631.17152-1-wens@kernel.org/



Hi,

This series seems to have some coding style problems. See output below for
more information:

Message-id: 20190522153631.17152-1-wens@kernel.org
Type: series
Subject: [Qemu-devel] [PATCH v2] linux-user: Pass through nanosecond timestamp components for stat syscalls

=== TEST SCRIPT BEGIN ===
#!/bin/bash
git rev-parse base > /dev/null || exit 0
git config --local diff.renamelimit 0
git config --local diff.renames True
git config --local diff.algorithm histogram
./scripts/checkpatch.pl --mailback base..
=== TEST SCRIPT END ===

From https://github.com/patchew-project/qemu
 * [new tag]               patchew/20190522153631.17152-1-wens@kernel.org -> patchew/20190522153631.17152-1-wens@kernel.org
Switched to a new branch 'test'
0ae03af3f9 linux-user: Pass through nanosecond timestamp components for stat syscalls

=== OUTPUT BEGIN ===
ERROR: code indent should never use tabs
#104: FILE: linux-user/syscall_defs.h:1197:
+^Iabi_ulong  target_st_atime_nsec;$

ERROR: code indent should never use tabs
#107: FILE: linux-user/syscall_defs.h:1199:
+^Iabi_ulong  target_st_mtime_nsec;$

ERROR: code indent should never use tabs
#110: FILE: linux-user/syscall_defs.h:1201:
+^Iabi_ulong  target_st_ctime_nsec;$

ERROR: code indent should never use tabs
#119: FILE: linux-user/syscall_defs.h:1233:
+^Iabi_ulong^Itarget_st_atime_nsec;$

ERROR: code indent should never use tabs
#123: FILE: linux-user/syscall_defs.h:1236:
+^Iabi_ulong^Itarget_st_mtime_nsec;$

ERROR: code indent should never use tabs
#127: FILE: linux-user/syscall_defs.h:1239:
+^Iabi_ulong^Itarget_st_ctime_nsec;$

ERROR: code indent should never use tabs
#136: FILE: linux-user/syscall_defs.h:1318:
+^Iabi_ulong^Itarget_st_atime_nsec;$

ERROR: code indent should never use tabs
#140: FILE: linux-user/syscall_defs.h:1321:
+^Iabi_ulong^Itarget_st_mtime_nsec;$

ERROR: code indent should never use tabs
#144: FILE: linux-user/syscall_defs.h:1324:
+^Iabi_ulong^Itarget_st_ctime_nsec;$

ERROR: code indent should never use tabs
#160: FILE: linux-user/syscall_defs.h:1342:
+^Iabi_ulong^Itarget_st_atime_nsec;$

ERROR: code indent should never use tabs
#163: FILE: linux-user/syscall_defs.h:1344:
+^Iabi_ulong^Itarget_st_mtime_nsec;$

ERROR: code indent should never use tabs
#166: FILE: linux-user/syscall_defs.h:1346:
+^Iabi_ulong^Itarget_st_ctime_nsec;$

ERROR: code indent should never use tabs
#170: FILE: linux-user/syscall_defs.h:1349:
+^Iabi_ulong^I__unused1[2];$

ERROR: code indent should never use tabs
#179: FILE: linux-user/syscall_defs.h:1377:
+^Iunsigned int^Itarget_st_atime_nsec;$

ERROR: code indent should never use tabs
#183: FILE: linux-user/syscall_defs.h:1380:
+^Iunsigned int^Itarget_st_mtime_nsec;$

ERROR: code indent should never use tabs
#187: FILE: linux-user/syscall_defs.h:1383:
+^Iunsigned int^Itarget_st_ctime_nsec;$

ERROR: code indent should never use tabs
#191: FILE: linux-user/syscall_defs.h:1385:
+^Iunsigned int^I__unused1;$

ERROR: code indent should never use tabs
#192: FILE: linux-user/syscall_defs.h:1386:
+^Iunsigned int^I__unused2;$

total: 18 errors, 0 warnings, 211 lines checked

Commit 0ae03af3f97f (linux-user: Pass through nanosecond timestamp components for stat syscalls) has style problems, please review.  If any of these errors
are false positives report them to the maintainer, see
CHECKPATCH in MAINTAINERS.
=== OUTPUT END ===

Test command exited with code: 1


The full log is available at
http://patchew.org/logs/20190522153631.17152-1-wens@kernel.org/testing.checkpatch/?type=message.
---
Email generated automatically by Patchew [https://patchew.org/].
Please send your feedback to patchew-devel@redhat.com