Implement the mmap system call for FreeBSD.
cover Implement the mmap system call for FreeBSD.
Upstream the implementation of the following mmap system calls, from the qemu-bsd-user fork:
mmap(2), munmap(2), mprotect(2), msync(2), mlock(2), munlock(2), mlockall(2) munlockall(2), madvise(2), minherit(2), mincore(2), shm_open(2),shm_open2(2), shm_rename2(2), shm_unlink(2), shmget(2), shmctl(2), shmat(2), shmdt(2)
Karim Taha (1):
Add bsd-mem.c to meson.build
Kyle Evans (2):
Implement shm_open2(2) system call
Implement shm_rename(2) system call
Stacey Son (19):
Implement struct target_ipc_perm
Implement struct target_shmid_ds
Declarations for ipc_perm and shmid_ds conversion functions Introduce freebsd/os-misc.h to the source tree
Implement target_set_brk function in bsd-mem.c instead of os-syscall.c Implement ipc_perm conversion between host and target. Implement shmid_ds conversion between host and target. Introduce bsd-mem.h to the source tree
Implement mmap(2) and munmap(2)
Implement mprotect(2)
Implement msync(2)
Implement mlock(2), munlock(2), mlockall(2), munlockall(2),
madvise(2), minherit(2)
Implement mincore(2)
Implement do_obreak function
Implement shm_open(2)
Implement shm_unlink(2) and shmget(2)
Implement shmctl(2)
Implement shmat(2) and shmdt(2)
Add stubs for vadvise(), sbrk() and sstk()
bsd-user/bsd-mem.c | 125 ++++++++++ bsd-user/bsd-mem.h | 431 ++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-misc.h | 104 ++++++++ bsd-user/freebsd/os-syscall.c | 112 ++++++++- bsd-user/meson.build | 1 + bsd-user/qemu-bsd.h | 45 ++++ bsd-user/syscall_defs.h | 37 +++ 7 files changed, 851 insertions(+), 4 deletions(-) create mode 100644 bsd-user/bsd-mem.c create mode 100644 bsd-user/bsd-mem.h create mode 100644 bsd-user/freebsd/os-misc.h create mode 100644 bsd-user/qemu-bsd.h
Signature · 1 line
2.40.0
PATCH 1/22 Implement struct target_ipc_perm
From: Stacey Son <[email protected]>
--- bsd-user/syscall_defs.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)
bsd-user/syscall_defs.h +17 -0
@@ -55,6 +55,23 @@ struct target_iovec {abi_long iov_len;/* Number of bytes */ }; +/* +* sys/ipc.h +*/ +struct target_ipc_perm { +uint32_t cuid;/* creator user id */ +uint32_t cgid;/* creator group id */ +uint32_t uid;/* user id */ +uint32_t gid;/* group id */ +uint16_t mode;/* r/w permission */ +uint16_t seq;/* sequence # */ +abi_long key;/* user specified msg/sem/shm key */ +}; + +#define TARGET_IPC_RMID 0/* remove identifier */ +#define TARGET_IPC_SET 1/* set options */ +#define TARGET_IPC_STAT 2/* get options */ +/* * sys/mman.h */
Signature · 1 line
2.40.0
Re: [PATCH 01/22] Implement struct target_ipc_perm
Quoted 7 lines · Karim Taha
From: Stacey Son<[email protected]>
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/syscall_defs.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)
r~
Re: [PATCH 01/22] Implement struct target_ipc_perm
In reply to Karim TahaOne general thing about all the patches in this series. The first line in the
commit message should start with 'bsd-user: ' which I'll only say once. You can fix it in v2 easily enough.
Quoted 8 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/syscall_defs.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)
Quoted 32 lines
bsd-user/syscall_defs.h +17
-1@@ -55,6 +55,23 @@ struct target_iovec {abi_long iov_len;/* Number of bytes */ }; +/* +* sys/ipc.h +*/ +struct target_ipc_perm { +uint32_t cuid;/* creator user id */ +uint32_t cgid;/* creator group id */ +uint32_t uid;/* user id */ +uint32_t gid;/* group id */ +uint16_t mode;/* r/w permission */ +uint16_t seq;/* sequence # */ +abi_long key;/* user specified msg/sem/shm key */ +}; + +#define TARGET_IPC_RMID 0/* remove identifier */ +#define TARGET_IPC_SET 1/* set options */ +#define TARGET_IPC_STAT 2/* get options */ +/* * sys/mman.h */ --2.40.0
PATCH 2/22 Implement struct target_shmid_ds
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/syscall_defs.h | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+)
bsd-user/syscall_defs.h +20 -0
@@ -72,6 +72,26 @@ struct target_ipc_perm {#define TARGET_IPC_SET 1/* set options */ #define TARGET_IPC_STAT 2/* get options */ +/* +* sys/shm.h +*/ +struct target_shmid_ds { +struct target_ipc_perm shm_perm;/* peration permission structure */ +abi_ulong shm_segsz;/* size of segment in bytes */ +int32_t shm_lpid;/* process ID of last shared memory op */ +int32_t shm_cpid;/* process ID of creator */ +int32_t shm_nattch;/* number of current attaches */ +target_time_t shm_atime;/* time of last shmat() */ +target_time_t shm_dtime;/* time of last shmdt() */ +target_time_t shm_ctime;/* time of last change by shmctl() */ +}; + +#define N_BSD_SHM_REGIONS 32 +struct bsd_shm_regions { +abi_long start; +abi_long size; +}; +/* * sys/mman.h */
Signature · 1 line
2.40.0
Re: [PATCH 02/22] Implement struct target_shmid_ds
Quoted 7 lines · Karim Taha
From: Stacey Son<[email protected]>
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/syscall_defs.h | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+)
r~
Re: [PATCH 02/22] Implement struct target_shmid_ds
In reply to Karim TahaQuoted 8 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/syscall_defs.h | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+)
Quoted 35 lines
bsd-user/syscall_defs.h +20
-1@@ -72,6 +72,26 @@ struct target_ipc_perm {#define TARGET_IPC_SET 1/* set options */ #define TARGET_IPC_STAT 2/* get options */ +/* +* sys/shm.h +*/ +struct target_shmid_ds { +struct target_ipc_perm shm_perm;/* peration permission structure */ +abi_ulong shm_segsz;/* size of segment in bytes */ +int32_t shm_lpid;/* process ID of last shared memory op */ +int32_t shm_cpid;/* process ID of creator */ +int32_t shm_nattch;/* number of current attaches */ +target_time_t shm_atime;/* time of last shmat() */ +target_time_t shm_dtime;/* time of last shmdt() */ +target_time_t shm_ctime;/* time of last change by shmctl() */ +}; + +#define N_BSD_SHM_REGIONS 32 +struct bsd_shm_regions { +abi_long start; +abi_long size; +}; +/* * sys/mman.h */ --2.40.0
PATCH 3/22 Declarations for ipc_perm and shmid_ds conversion functions
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/qemu-bsd.h | 45 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 45 insertions(+) create mode 100644 bsd-user/qemu-bsd.h
bsd-user/qemu-bsd.h new file +45 -0
@@ -0,0 +1,45 @@ +/* +* BSD conversion extern declarations +* +* Copyright (c) 2013 Stacey D. Son +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>. +*/ + +#ifndef QEMU_BSD_H +#define QEMU_BSD_H + +#include <sys/types.h> +#include <sys/ipc.h> +#include <sys/msg.h> +#include <sys/resource.h> +#include <sys/sem.h> +#include <sys/shm.h> +#include <sys/socket.h> +#include <sys/un.h> +#include <sys/uuid.h> +#include <sys/wait.h> +#include <netinet/in.h> + +/* bsd-mem.c */ +abi_long target_to_host_ipc_perm (struct ipc_perm *host_ip, +abi_ulong target_addr); +abi_long host_to_target_ipc_perm (abi_ulong target_addr, +struct ipc_perm *host_ip); +abi_long target_to_host_shmid_ds (struct shmid_ds *host_sd, +abi_ulong target_addr); +abi_long host_to_target_shmid_ds (abi_ulong target_addr, +struct shmid_ds *host_sd); + +#endif /* QEMU_BSD_H */
Signature · 1 line
2.40.0
Re: [PATCH 03/22] Declarations for ipc_perm and shmid_ds conversion functions
Quoted 8 lines · Karim Taha
From: Stacey Son<[email protected]>
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/qemu-bsd.h | 45 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 45 insertions(+)create mode 100644
bsd-user/qemu-bsd.h
r~
Re: [PATCH 03/22] Declarations for ipc_perm and shmid_ds conversion functions
In reply to Karim TahaQuoted 9 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/qemu-bsd.h | 45 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 45 insertions(+) create mode 100644 bsd-user/qemu-bsd.h
Quoted 55 lines
bsd-user/qemu-bsd.h new file +45
-1@@ -0,0 +1,45 @@ +/* +* BSD conversion extern declarations +* +* Copyright (c) 2013 Stacey D. Son +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>. +*/ + +#ifndef QEMU_BSD_H +#define QEMU_BSD_H + +#include <sys/types.h> +#include <sys/ipc.h> +#include <sys/msg.h> +#include <sys/resource.h> +#include <sys/sem.h> +#include <sys/shm.h> +#include <sys/socket.h> +#include <sys/un.h> +#include <sys/uuid.h> +#include <sys/wait.h> +#include <netinet/in.h> + +/* bsd-mem.c */ +abi_long target_to_host_ipc_perm (struct ipc_perm *host_ip, +abi_ulong target_addr); +abi_long host_to_target_ipc_perm (abi_ulong target_addr, +struct ipc_perm *host_ip); +abi_long target_to_host_shmid_ds (struct shmid_ds *host_sd, +abi_ulong target_addr); +abi_long host_to_target_shmid_ds (abi_ulong target_addr, +struct shmid_ds *host_sd); + +#endif /* QEMU_BSD_H */ --2.40.0
PATCH 4/22 Introduce freebsd/os-misc.h to the source tree
In reply to Karim TahaFrom: Stacey Son <[email protected]>
To preserve the copyright notice and help with the 'Author' info for subsequent changes to the file.
--- bsd-user/freebsd/os-misc.h | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) create mode 100644 bsd-user/freebsd/os-misc.h
bsd-user/freebsd/os-misc.h new file +28 -0
@@ -0,0 +1,28 @@ +/* +* miscellaneous FreeBSD system call shims +* +* Copyright (c) 2013-14 Stacey D. Son +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>. +*/ + +#ifndef OS_MISC_H +#define OS_MISC_H + +#include <sys/cpuset.h> +#include <sys/random.h> +#include <sched.h> + + +#endif /* OS_MISC_H */
Signature · 1 line
2.40.0
Re: [PATCH 04/22] Introduce freebsd/os-misc.h to the source tree
Quoted 11 lines · Karim Taha
From: Stacey Son<[email protected]>
To preserve the copyright notice and help with the 'Author' info for subsequent changes to the file.
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/freebsd/os-misc.h | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+)create mode 100644
bsd-user/freebsd/os-misc.h
r~
Re: [PATCH 04/22] Introduce freebsd/os-misc.h to the source tree
In reply to Karim TahaQuoted 5 lines · Karim Taha
From: Stacey Son <[email protected]>
To preserve the copyright notice and help with the 'Author' info for subsequent changes to the file.
Quoted 7 lines
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/freebsd/os-misc.h | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) create mode 100644 bsd-user/freebsd/os-misc.h
PATCH 5/22 Implement shm_open2(2) system call
In reply to Karim TahaFrom: Kyle Evans <[email protected]>
--- bsd-user/freebsd/os-misc.h | 52 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 13 +++++++++ 2 files changed, 65 insertions(+)
bsd-user/freebsd/os-misc.h +52 -0
@@ -24,5 +24,57 @@#include <sys/random.h> #include <sched.h> +int shm_open2 (const char *path,int flags,mode_t mode,int shmflags, +const char *); + +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 +/* shm_open2(2) */ +static inline abi_long do_freebsd_shm_open2 (abi_ulong pathptr,abi_ulong flags, +abi_long mode,abi_ulong shmflags,abi_ulong nameptr) +{ +int ret; +void *uname, *upath; + +#ifdef SHM_ANON +#define SHM_PATH (p) (p) == SHM_ANON ? (p) : path(p) +if (pathptr == (uintptr_t )SHM_ANON ) { + upath =SHM_ANON ; + } else +#else +#define SHM_PATH (p) path(p) +#endif + { + upath =lock_user_string (pathptr); +if (upath ==NULL ) { +return -TARGET_EFAULT ; + } + } + + uname =NULL ; +if (nameptr !=0 ) { + uname =lock_user_string (nameptr); +if (uname ==NULL ) { +unlock_user (upath, pathptr,0 ); +return -TARGET_EFAULT ; + } + } + ret =get_errno (shm_open2 (SHM_PATH (upath), +target_to_host_bitmask (flags, fcntl_flags_tbl), mode, +target_to_host_bitmask (shmflags, shmflag_flags_tbl), uname)); + +#ifdef SHM_ANON + if (upath != SHM_ANON) +#endif + { +unlock_user (upath, pathptr,0 ); + } +if (uname !=NULL ) { +unlock_user (uname, nameptr,0 ); + } +return ret; +} +#undef SHM_PATH +#endif /* __FreeBSD_version >= 1300048 */ +#endif /* OS_MISC_H */
bsd-user/freebsd/os-syscall.c +13 -0
@@ -33,9 +33,13 @@#include "signal-common.h" #include "user/syscall-trace.h" +/* BSD independent syscall shims */ #include "bsd-file.h" #include "bsd-proc.h" +/* *BSD dependent syscall shims */ +#include "os-misc.h" +/* I/O */ safe_syscall3(int, open, const char *, path, int, flags, mode_t, mode); safe_syscall4(int, openat, int, fd, const char *, path, int, flags, mode_t,
@@ -482,6 +486,15 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_undelete (arg1);break ; +/* +* Memory management system calls. +*/ +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 +case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ + ret =do_freebsd_shm_open2 (arg1, arg2, arg3, arg4, arg5); +break ; +#endif +/* * sys{ctl, arch, call} */
Signature · 1 line
2.40.0
Re: [PATCH 05/22] Implement shm_open2(2) system call
Quoted 19 lines · Karim Taha
From: Kyle Evans <[email protected]>
Signed-off-by Kyle Evans <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/freebsd/os-misc.h | 52 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 13 +++++++++ 2 files changed, 65 insertions(+)bsd-user/freebsd/os-misc.h +2
-0@@ -24,5 +24,57 @@#include <sys/random.h> #include <sched.h> +int shm_open2 (const char *path,int flags,mode_t mode,int shmflags, +const char *);
bsd-user/freebsd/os-misc.h line 38
Quoted 10 lines
bsd-user/freebsd/os-misc.h +10
-0+ +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 +/* shm_open2(2) */ +static inline abi_long do_freebsd_shm_open2 (abi_ulong pathptr,abi_ulong flags, +abi_long mode,abi_ulong shmflags,abi_ulong nameptr) +{ +int ret; +void *uname, *upath; + +#ifdef SHM_ANON
Why would SHM_ANON not be defined? You've already restricted the function to freebsd13+ (presumably so that shm_open2() is in libc.a).
Quoted 13 lines
bsd-user/freebsd/os-misc.h +13
-0+#define SHM_PATH (p) (p) == SHM_ANON ? (p) : path(p) +if (pathptr == (uintptr_t )SHM_ANON ) { + upath =SHM_ANON ; + } else +#else +#define SHM_PATH (p) path(p) +#endif + { + upath =lock_user_string (pathptr); +if (upath ==NULL ) { +return -TARGET_EFAULT ; + } + }
bsd-user/freebsd/os-misc.h line 61
Quoted 10 lines
bsd-user/freebsd/os-misc.h +10
-0+ + uname =NULL ; +if (nameptr !=0 ) { + uname =lock_user_string (nameptr); +if (uname ==NULL ) { +unlock_user (upath, pathptr,0 ); +return -TARGET_EFAULT ; + } + } + ret = get_errno(shm_open2(SHM_PATH(upath),
There is no need for SHM_PATH because you've already handled the condition while setting upath above.
r~
Re: [PATCH 05/22] Implement shm_open2(2) system call
Quoted 25 lines · Richard Henderson
Quoted 14 lines · Karim Taha
From: Kyle Evans <[email protected]>
Signed-off-by Kyle Evans <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/freebsd/os-misc.h | 52 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 13 +++++++++ 2 files changed, 65 insertions(+)bsd-user/freebsd/os-misc.h +0
-0bsd-user/freebsd/os-misc.h line 32
Quoted 9 lines
bsd-user/freebsd/os-misc.h +6
-0@@ -24,5 +24,57 @@#include <sys/random.h> #include <sched.h> +int shm_open2 (const char *path,int flags,mode_t mode,int shmflags, +const char *); + +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 +/* shm_open2(2) */ +static inline abi_long do_freebsd_shm_open2(abi_ulong pathptr,
abi_ulongflags,
Quoted 11 lines
bsd-user/freebsd/os-misc.h line 38
Quoted 6 lines
bsd-user/freebsd/os-misc.h +6
-0+ abi_long mode, abi_ulong shmflags, abi_ulong nameptr) +{ + int ret; +void *uname, *upath; + +#ifdef SHM_ANON Why would
SHM_ANONnot be defined? You've already restricted the function to freebsd13+
(presumably so thatshm_open2()is in libc.a).
Good catch Richard. It's like this in the bsd-user fork, but after talking it over with Kyle, we're
sure that's due to a cut and past from shm_open when this was implemented. We should change
it before we submit it upstream absolutely. It's always defined on FreeBSD and that will make
this code simpler.
Warner
Quoted 32 lines
Quoted 13 lines
bsd-user/freebsd/os-misc.h +13
-0+#define SHM_PATH (p) (p) == SHM_ANON ? (p) : path(p) +if (pathptr == (uintptr_t )SHM_ANON ) { + upath =SHM_ANON ; + } else +#else +#define SHM_PATH (p) path(p) +#endif + { + upath =lock_user_string (pathptr); +if (upath ==NULL ) { +return -TARGET_EFAULT ; + } + }bsd-user/freebsd/os-misc.h line 61
Quoted 10 lines
bsd-user/freebsd/os-misc.h +10
-0+ + uname =NULL ; +if (nameptr !=0 ) { + uname =lock_user_string (nameptr); +if (uname ==NULL ) { +unlock_user (upath, pathptr,0 ); +return -TARGET_EFAULT ; + } + } + ret = get_errno(shm_open2(SHM_PATH(upath),There is no need for
SHM_PATHbecause you've already handled the condition while setting
upath above.r~
PATCH 6/22 Implement shm_rename(2) system call
In reply to Karim TahaFrom: Kyle Evans <[email protected]>
--- bsd-user/freebsd/os-misc.h | 24 ++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 6 ++++++ 2 files changed, 30 insertions(+)
bsd-user/freebsd/os-misc.h +24 -0
@@ -76,5 +76,29 @@ static inline abi_long do_freebsd_shm_open2(abi_ulong pathptr, abi_ulong flags,#undef SHM_PATH#endif /* __FreeBSD_version >= 1300048 */ +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300049 +/* shm_rename(2) */ +static inline abi_long do_freebsd_shm_rename (abi_ulong fromptr,abi_ulong toptr, +abi_ulong flags) +{ +int ret; +void *ufrom, *uto; + + ufrom =lock_user_string (fromptr); +if (ufrom ==NULL ) { +return -TARGET_EFAULT ; + } + uto =lock_user_string (toptr); +if (uto ==NULL ) { +unlock_user (ufrom, fromptr,0 ); +return -TARGET_EFAULT ; + } + ret =get_errno (shm_rename (ufrom, uto, flags)); +unlock_user (ufrom, fromptr,0 ); +unlock_user (uto, toptr,0 ); + +return ret; +} +#endif /* __FreeBSD_version >= 1300049 */ #endif /* OS_MISC_H */
bsd-user/freebsd/os-syscall.c +6 -0
@@ -495,6 +495,12 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1,break ;#endif +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300049 +case TARGET_FREEBSD_NR_shm_rename:/* shm_rename(2) */ + ret =do_freebsd_shm_rename (arg1, arg2, arg3); +break ; +#endif +/* * sys{ctl, arch, call} */
Signature · 1 line
2.40.0
Re: [PATCH 06/22] Implement shm_rename(2) system call
Quoted 9 lines · Karim Taha
From: Kyle Evans <[email protected]>
Signed-off-by Kyle Evans <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/freebsd/os-misc.h | 24 ++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 6 ++++++ 2 files changed, 30 insertions(+)
The FreeBSD 13 ifdef below still is needed since FreeBSD 12 is still supported by the
project until the end of the year.
Quoted 58 lines
bsd-user/freebsd/os-misc.h +24
-0@@ -76,5 +76,29 @@ static inline abi_long do_freebsd_shm_open2(abi_ulong pathptr, abi_ulong flags,#undef SHM_PATH#endif /* __FreeBSD_version >= 1300048 */ +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300049 +/* shm_rename(2) */ +static inline abi_long do_freebsd_shm_rename (abi_ulong fromptr,abi_ulong toptr, +abi_ulong flags) +{ +int ret; +void *ufrom, *uto; + + ufrom =lock_user_string (fromptr); +if (ufrom ==NULL ) { +return -TARGET_EFAULT ; + } + uto =lock_user_string (toptr); +if (uto ==NULL ) { +unlock_user (ufrom, fromptr,0 ); +return -TARGET_EFAULT ; + } + ret =get_errno (shm_rename (ufrom, uto, flags)); +unlock_user (ufrom, fromptr,0 ); +unlock_user (uto, toptr,0 ); + +return ret; +} +#endif /* __FreeBSD_version >= 1300049 */ #endif /*
OS_MISC_H*/bsd-user/freebsd/os-syscall.c +6
-0@@ -495,6 +495,12 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, break;#endif +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300049 +case TARGET_FREEBSD_NR_shm_rename:/* shm_rename(2) */ + ret =do_freebsd_shm_rename (arg1, arg2, arg3); +break ; +#endif +/* * sys{ctl, arch, call} */--
2.40.0
Re: [PATCH 06/22] Implement shm_rename(2) system call
In reply to Karim TahaQuoted 8 lines · Karim Taha
From: Kyle Evans<[email protected]>
Signed-off-by Kyle Evans<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/freebsd/os-misc.h | 24 ++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 6 ++++++ 2 files changed, 30 insertions(+)
r~
PATCH 7/22 Add bsd-mem.c to meson.build
In reply to Karim Taha--- bsd-user/bsd-mem.c | 0 bsd-user/meson.build | 1 + 2 files changed, 1 insertion(+) create mode 100644 bsd-user/bsd-mem.c
bsd-user/bsd-mem.c new file +0 -0
bsd-user/meson.build +1 -0
@@ -7,6 +7,7 @@ bsd_user_ss = ss.source_set() common_user_inc += include_directories('include') bsd_user_ss.add(files( + 'bsd-mem.c', 'bsdload.c', 'elfload.c', 'main.c',
Signature · 1 line
2.40.0
Re: [PATCH 07/22] Add bsd-mem.c to meson.build
Quoted 6 lines · Karim Taha
--- bsd-user/bsd-mem.c | 0 bsd-user/meson.build | 1 + 2 files changed, 1 insertion(+) create mode 100644 bsd-user/bsd-mem.c
Quoted 19 lines
bsd-user/bsd-mem.c new file +0
-0bsd-user/meson.build +1
-1@@ -7,6 +7,7 @@ bsd_user_ss = ss.source_set() common_user_inc += include_directories('include') bsd_user_ss.add(files( + 'bsd-mem.c', 'bsdload.c', 'elfload.c', 'main.c', --2.40.0
Re: [PATCH 07/22] Add bsd-mem.c to meson.build
In reply to Karim TahaQuoted 14 lines · Karim Taha
--- bsd-user/bsd-mem.c | 0 bsd-user/meson.build | 1 + 2 files changed, 1 insertion(+)create mode 100644
bsd-user/bsd-mem.cbsd-user/bsd-mem.c new file +0
-0bsd-user/meson.build +0
-0
bsd-user/meson.build bsd_user_ss = ss.source_set() · line 13
Quoted 7 lines
bsd-user/meson.build +1
-0@@ -7,6 +7,7 @@ bsd_user_ss = ss.source_set() common_user_inc += include_directories('include') bsd_user_ss.add(files( + 'bsd-mem.c', 'bsdload.c', 'elfload.c', 'main.c',
r~
PATCH 8/22 Implement target_set_brk function in bsd-mem.c instead of os-syscall.c
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.c | 38 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ---- 2 files changed, 38 insertions(+), 4 deletions(-)
bsd-user/bsd-mem.c +38 -0
@@ -0,0 +1,38 @@ +/* +* memory management system conversion routines +* +* Copyright (c) 2013 Stacey D. Son +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>. +*/ +#include "qemu/osdep.h" + +#include <sys/ipc.h> +#include <sys/shm.h> + +#include "qemu.h" +#include "qemu-bsd.h" + +struct bsd_shm_regions bsd_shm_regions[N_BSD_SHM_REGIONS ]; + +abi_ulong bsd_target_brk; +abi_ulong bsd_target_original_brk; +abi_ulong brk_page; + +void target_set_brk (abi_ulong new_brk) +{ + + bsd_target_original_brk = bsd_target_brk =HOST_PAGE_ALIGN (new_brk); + brk_page =HOST_PAGE_ALIGN (bsd_target_brk); +}
bsd-user/freebsd/os-syscall.c +0 -4
@@ -59,10 +59,6 @@ safe_syscall3(ssize_t, writev, int, fd, const struct iovec *, iov, int, iovcnt); safe_syscall4(ssize_t, pwritev, int, fd, const struct iovec *, iov, int, iovcnt, off_t, offset); -void target_set_brk (abi_ulong new_brk) -{ -} -/* * errno conversion. */
Signature · 1 line
2.40.0
Re: [PATCH 08/22] Implement target_set_brk function in bsd-mem.c instead of os-syscall.c
Quoted 10 lines · Karim Taha
From: Stacey Son <[email protected]>
Co-authored-by Mikaël Urankar <[email protected]>Signed-off-by Mikaël Urankar <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.c | 38 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ---- 2 files changed, 38 insertions(+), 4 deletions(-)
but see below
Quoted 19 lines
bsd-user/bsd-mem.c +13
-0@@ -0,0 +1,38 @@ +/* +* memory management system conversion routines +* +* Copyright (c) 2013 Stacey D. Son +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
bsd-user/bsd-mem.c line 23
Quoted 10 lines
bsd-user/bsd-mem.c +10
-0+* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>. +*/ +#include "qemu/osdep.h" + +#include <sys/ipc.h> +#include <sys/shm.h> +
I'll let others comment if this is a good thing here or it needs to be elsewhere.
I can't recall what the project's rules are for system headers on code that runs on a limited subset of systems. For code that can run anywhere, I know that the preference is to put this in osdep.h, but in this case I'm unsure.
Warner
Quoted 36 lines
bsd-user/bsd-mem.c +15
-0+#include "qemu.h" +#include "qemu-bsd.h" + +struct bsd_shm_regions bsd_shm_regions[N_BSD_SHM_REGIONS ]; + +abi_ulong bsd_target_brk; +abi_ulong bsd_target_original_brk; +abi_ulong brk_page; + +void target_set_brk (abi_ulong new_brk) +{ + + bsd_target_original_brk = bsd_target_brk =HOST_PAGE_ALIGN (new_brk); + brk_page =HOST_PAGE_ALIGN (bsd_target_brk); +}bsd-user/freebsd/os-syscall.c +0
-4@@ -59,10 +59,6 @@ safe_syscall3(ssize_t, writev, int, fd, const struct iovec *, iov, int, iovcnt); safe_syscall4(ssize_t, pwritev, int, fd, const struct iovec *, iov, int, iovcnt, off_t, offset); -void target_set_brk (abi_ulong new_brk) -{ -} -/* * errno conversion.
*/
--
2.40.0
Re: [PATCH 08/22] Implement target_set_brk function in bsd-mem.c instead of os-syscall.c
In reply to Karim TahaQuoted 32 lines · Karim Taha
From: Stacey Son <[email protected]>
Co-authored-by Mikaël Urankar <[email protected]>Signed-off-by Mikaël Urankar <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.c | 38 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ---- 2 files changed, 38 insertions(+), 4 deletions(-)bsd-user/bsd-mem.c +17
-0@@ -0,0 +1,38 @@ +/* +* memory management system conversion routines +* +* Copyright (c) 2013 Stacey D. Son +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>.
bsd-user/bsd-mem.c line 27
Quoted 10 lines
bsd-user/bsd-mem.c +10
-0+*/ +#include "qemu/osdep.h" + +#include <sys/ipc.h> +#include <sys/shm.h> + +#include "qemu.h" +#include "qemu-bsd.h" + +struct bsd_shm_regions bsd_shm_regions[N_BSD_SHM_REGIONS ];
This and the ipc/shm includes don't belong with this patch.
bsd-user/bsd-mem.c line 35
Quoted 8 lines
bsd-user/bsd-mem.c +8
-0+ +abi_ulong bsd_target_brk; +abi_ulong bsd_target_original_brk; +abi_ulong brk_page; + +void target_set_brk (abi_ulong new_brk) +{ +
Blank line.
bsd-user/bsd-mem.c line 37
Quoted 2 lines
bsd-user/bsd-mem.c +2
-0+ bsd_target_original_brk = bsd_target_brk =HOST_PAGE_ALIGN (new_brk); + brk_page =HOST_PAGE_ALIGN (bsd_target_brk);
I encourage you to use TARGET_PAGE_ALIGN instead.
This will match changes made in linux-user during the 8.1 cycle.
Otherwise,
r~
PATCH 9/22 Implement ipc_perm conversion between host and target.
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.c | 41 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+)
bsd-user/bsd-mem.c +41 -0
@@ -36,3 +36,44 @@ void target_set_brk(abi_ulong new_brk) bsd_target_original_brk = bsd_target_brk =HOST_PAGE_ALIGN (new_brk); brk_page =HOST_PAGE_ALIGN (bsd_target_brk); } + +abi_long target_to_host_ipc_perm (struct ipc_perm *host_ip, +abi_ulong target_addr) +{ +struct target_ipc_perm *target_ip; + +if (!lock_user_struct (VERIFY_READ , target_ip, target_addr,1 )) { +return -TARGET_EFAULT ; + } +__get_user (host_ip->cuid, &target_ip->cuid); +__get_user (host_ip->cgid, &target_ip->cgid); +__get_user (host_ip->uid, &target_ip->uid); +__get_user (host_ip->gid, &target_ip->gid); +__get_user (host_ip->mode, &target_ip->mode); +__get_user (host_ip->seq, &target_ip->seq); +__get_user (host_ip->key, &target_ip->key); +unlock_user_struct (target_ip, target_addr,0 ); + +return 0 ; +} + +abi_long host_to_target_ipc_perm (abi_ulong target_addr, +struct ipc_perm *host_ip) +{ +struct target_ipc_perm *target_ip; + +if (!lock_user_struct (VERIFY_WRITE , target_ip, target_addr,0 )) { +return -TARGET_EFAULT ; + } +__put_user (host_ip->cuid, &target_ip->cuid); +__put_user (host_ip->cgid, &target_ip->cgid); +__put_user (host_ip->uid, &target_ip->uid); +__put_user (host_ip->gid, &target_ip->gid); +__put_user (host_ip->mode, &target_ip->mode); +__put_user (host_ip->seq, &target_ip->seq); +__put_user (host_ip->key, &target_ip->key); +unlock_user_struct (target_ip, target_addr,1 ); + +return 0 ; +} +
Signature · 1 line
2.40.0
Re: [PATCH 09/22] Implement ipc_perm conversion between host and target.
Quoted 8 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.c | 41 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+)
Quoted 53 lines
bsd-user/bsd-mem.c +41
-1@@ -36,3 +36,44 @@ void target_set_brk(abi_ulong new_brk) bsd_target_original_brk = bsd_target_brk =HOST_PAGE_ALIGN (new_brk); brk_page =HOST_PAGE_ALIGN (bsd_target_brk); } + +abi_long target_to_host_ipc_perm (struct ipc_perm *host_ip, +abi_ulong target_addr) +{ +struct target_ipc_perm *target_ip; + +if (!lock_user_struct (VERIFY_READ , target_ip, target_addr,1 )) { +return -TARGET_EFAULT ; + } +__get_user (host_ip->cuid, &target_ip->cuid); +__get_user (host_ip->cgid, &target_ip->cgid); +__get_user (host_ip->uid, &target_ip->uid); +__get_user (host_ip->gid, &target_ip->gid); +__get_user (host_ip->mode, &target_ip->mode); +__get_user (host_ip->seq, &target_ip->seq); +__get_user (host_ip->key, &target_ip->key); +unlock_user_struct (target_ip, target_addr,0 ); + +return 0 ; +} + +abi_long host_to_target_ipc_perm (abi_ulong target_addr, +struct ipc_perm *host_ip) +{ +struct target_ipc_perm *target_ip; + +if (!lock_user_struct (VERIFY_WRITE , target_ip, target_addr,0 )) { +return -TARGET_EFAULT ; + } +__put_user (host_ip->cuid, &target_ip->cuid); +__put_user (host_ip->cgid, &target_ip->cgid); +__put_user (host_ip->uid, &target_ip->uid); +__put_user (host_ip->gid, &target_ip->gid); +__put_user (host_ip->mode, &target_ip->mode); +__put_user (host_ip->seq, &target_ip->seq); +__put_user (host_ip->key, &target_ip->key); +unlock_user_struct (target_ip, target_addr,1 ); + +return 0 ; +} + --2.40.0
Re: [PATCH 09/22] Implement ipc_perm conversion between host and target.
In reply to Karim TahaQuoted 7 lines · Karim Taha
From: Stacey Son<[email protected]>
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/bsd-mem.c | 41 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+)
r~
PATCH 10/22 Implement shmid_ds conversion between host and target.
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+)
bsd-user/bsd-mem.c +46 -0
@@ -77,3 +77,49 @@ abi_long host_to_target_ipc_perm(abi_ulong target_addr,return 0 ; } +abi_long target_to_host_shmid_ds (struct shmid_ds *host_sd, +abi_ulong target_addr) +{ +struct target_shmid_ds *target_sd; + +if (!lock_user_struct (VERIFY_READ , target_sd, target_addr,1 )) { +return -TARGET_EFAULT ; + } +if (target_to_host_ipc_perm (&(host_sd->shm_perm), target_addr)) { +return -TARGET_EFAULT ; + } +__get_user (host_sd->shm_segsz, &target_sd->shm_segsz); +__get_user (host_sd->shm_lpid, &target_sd->shm_lpid); +__get_user (host_sd->shm_cpid, &target_sd->shm_cpid); +__get_user (host_sd->shm_nattch, &target_sd->shm_nattch); +__get_user (host_sd->shm_atime, &target_sd->shm_atime); +__get_user (host_sd->shm_dtime, &target_sd->shm_dtime); +__get_user (host_sd->shm_ctime, &target_sd->shm_ctime); +unlock_user_struct (target_sd, target_addr,0 ); + +return 0 ; +} + +abi_long host_to_target_shmid_ds (abi_ulong target_addr, +struct shmid_ds *host_sd) +{ +struct target_shmid_ds *target_sd; + +if (!lock_user_struct (VERIFY_WRITE , target_sd, target_addr,0 )) { +return -TARGET_EFAULT ; + } +if (host_to_target_ipc_perm (target_addr, &(host_sd->shm_perm))) { +return -TARGET_EFAULT ; + } +__put_user (host_sd->shm_segsz, &target_sd->shm_segsz); +__put_user (host_sd->shm_lpid, &target_sd->shm_lpid); +__put_user (host_sd->shm_cpid, &target_sd->shm_cpid); +__put_user (host_sd->shm_nattch, &target_sd->shm_nattch); +__put_user (host_sd->shm_atime, &target_sd->shm_atime); +__put_user (host_sd->shm_dtime, &target_sd->shm_dtime); +__put_user (host_sd->shm_ctime, &target_sd->shm_ctime); +unlock_user_struct (target_sd, target_addr,1 ); + +return 0 ; +} +
Signature · 1 line
2.40.0
Re: [PATCH 10/22] Implement shmid_ds conversion between host and target.
Quoted 8 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.c | 46 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+)
bsd-user/bsd-mem.c +0 -0
Quoted 57 lines
index
e69250cc0d..4446c94725100644bsd-user/bsd-mem.c +46
-1@@ -77,3 +77,49 @@ abi_long host_to_target_ipc_perm(abi_ulong target_addr,return 0 ; } +abi_long target_to_host_shmid_ds (struct shmid_ds *host_sd, +abi_ulong target_addr) +{ +struct target_shmid_ds *target_sd; + +if (!lock_user_struct (VERIFY_READ , target_sd, target_addr,1 )) { +return -TARGET_EFAULT ; + } +if (target_to_host_ipc_perm (&(host_sd->shm_perm), target_addr)) { +return -TARGET_EFAULT ; + } +__get_user (host_sd->shm_segsz, &target_sd->shm_segsz); +__get_user (host_sd->shm_lpid, &target_sd->shm_lpid); +__get_user (host_sd->shm_cpid, &target_sd->shm_cpid); +__get_user (host_sd->shm_nattch, &target_sd->shm_nattch); +__get_user (host_sd->shm_atime, &target_sd->shm_atime); +__get_user (host_sd->shm_dtime, &target_sd->shm_dtime); +__get_user (host_sd->shm_ctime, &target_sd->shm_ctime); +unlock_user_struct (target_sd, target_addr,0 ); + +return 0 ; +} + +abi_long host_to_target_shmid_ds (abi_ulong target_addr, +struct shmid_ds *host_sd) +{ +struct target_shmid_ds *target_sd; + +if (!lock_user_struct (VERIFY_WRITE , target_sd, target_addr,0 )) { +return -TARGET_EFAULT ; + } +if (host_to_target_ipc_perm (target_addr, &(host_sd->shm_perm))) { +return -TARGET_EFAULT ; + } +__put_user (host_sd->shm_segsz, &target_sd->shm_segsz); +__put_user (host_sd->shm_lpid, &target_sd->shm_lpid); +__put_user (host_sd->shm_cpid, &target_sd->shm_cpid); +__put_user (host_sd->shm_nattch, &target_sd->shm_nattch); +__put_user (host_sd->shm_atime, &target_sd->shm_atime); +__put_user (host_sd->shm_dtime, &target_sd->shm_dtime); +__put_user (host_sd->shm_ctime, &target_sd->shm_ctime); +unlock_user_struct (target_sd, target_addr,1 ); + +return 0 ; +} + --2.40.0
Re: [PATCH 10/22] Implement shmid_ds conversion between host and target.
In reply to Karim Tahabsd-user/bsd-mem.c abi_long host_to_target_ipc_perm(abi_ulong target_addr, · line 113
Quoted 6 lines · Karim Taha
bsd-user/bsd-mem.c +6
-0+if (!lock_user_struct (VERIFY_WRITE , target_sd, target_addr,0 )) { +return -TARGET_EFAULT ; + } +if (host_to_target_ipc_perm (target_addr, &(host_sd->shm_perm))) { +return -TARGET_EFAULT ; + }
While it works, ideally you wouldn't double-lock a memory range, once here and once in host_to_target_ipc_perm. You could split out the middle of the function as host_to_target_ipc_perm__locked.
Anyway,
r~
Re: [PATCH 10/22] Implement shmid_ds conversion between host and target.
Quoted 11 lines · Richard Henderson
bsd-user/bsd-mem.c abi_long host_to_target_ipc_perm(abi_ulong target_addr, · line 113
Quoted 6 lines · Karim Taha
bsd-user/bsd-mem.c +6
-0+if (!lock_user_struct (VERIFY_WRITE , target_sd, target_addr,0 )) { +return -TARGET_EFAULT ; + } +if (host_to_target_ipc_perm (target_addr, &(host_sd->shm_perm))) { +return -TARGET_EFAULT ; + }While it works, ideally you wouldn't double-lock a memory range, once here and once in
host_to_target_ipc_perm. You could split out the middle of the function ashost_to_target_ipc_perm__locked.
Hi Richard,
Can you please verify the correctness of the following refactoring?
void host_to_target_ipc_perm__locked(abi_ulong target_addr,
struct ipc_perm *host_ip)
{
struct target_ipc_perm *target_ip = g2h_untagged(target_addr);
__put_user(host_ip->cuid, &target_ip->cuid);
__put_user(host_ip->cgid, &target_ip->cgid);
__put_user(host_ip->uid, &target_ip->uid);
__put_user(host_ip->gid, &target_ip->gid);
__put_user(host_ip->mode, &target_ip->mode);
__put_user(host_ip->seq, &target_ip->seq);
__put_user(host_ip->key, &target_ip->key);
}
abi_long host_to_target_shmid_ds(abi_ulong target_addr,
struct shmid_ds *host_sd)
{
struct target_shmid_ds *target_sd;
target_sd = lock_user(VERIFY_WRITE, target_addr, sizeof(*target_sd), 0);
if (!target_sd){
return -TARGET_EFAULT;
}
host_to_target_ipc_perm__locked(target_addr, &(host_sd->shm_perm));
__put_user(host_sd->shm_segsz, &target_sd->shm_segsz);
__put_user(host_sd->shm_lpid, &target_sd->shm_lpid);
__put_user(host_sd->shm_cpid, &target_sd->shm_cpid);
__put_user(host_sd->shm_nattch, &target_sd->shm_nattch);
__put_user(host_sd->shm_atime, &target_sd->shm_atime);
__put_user(host_sd->shm_dtime, &target_sd->shm_dtime);
__put_user(host_sd->shm_ctime, &target_sd->shm_ctime);
unlock_user_struct(target_sd, target_addr, 1);
return 0;
}As far as I understood the page_check_range function, defined at accel/tcg/user-exec.c::523:
-The locked range is (target_addr, target_addr + sizeof(target_ipc_perm) -1) in case of
host_to_target_ipc_perm function.
-The locked range is (target_addr, taregt_addr + sizeof(target_shmid_ds) -1) in case of
host_to_target_shmid_ds function.
Since host_to_target_shmid_ds struct has larger size, in the original code, is the sucess of the first lock guarantees the sucess of the second?
If I got it wrong, please elaborate further.
If I'm correct, do you think I should call g2h_untagged in host_to_target_ipc_perm__locked directly, or should I receive it as a paremeter?
--
Kariiem Taha
Re: [PATCH 10/22] Implement shmid_ds conversion between host and target.
Quoted 32 lines · Kariiem Taha
Quoted 11 lines · Richard Henderson
bsd-user/bsd-mem.c abi_long host_to_target_ipc_perm(abi_ulong target_addr, · line 113
Quoted 6 lines · Karim Taha
bsd-user/bsd-mem.c +6
-0+if (!lock_user_struct (VERIFY_WRITE , target_sd, target_addr,0 )) { +return -TARGET_EFAULT ; + } +if (host_to_target_ipc_perm (target_addr, &(host_sd->shm_perm))) { +return -TARGET_EFAULT ; + }While it works, ideally you wouldn't double-lock a memory range, once here and once in
host_to_target_ipc_perm. You could split out the middle of the function ashost_to_target_ipc_perm__locked.Hi Richard,
Can you please verify the correctness of the following refactoring?
void host_to_target_ipc_perm__locked(abi_ulong target_addr, struct ipc_perm *host_ip) { struct target_ipc_perm *target_ip = g2h_untagged(target_addr); __put_user(host_ip->cuid, &target_ip->cuid); __put_user(host_ip->cgid, &target_ip->cgid); __put_user(host_ip->uid, &target_ip->uid); __put_user(host_ip->gid, &target_ip->gid); __put_user(host_ip->mode, &target_ip->mode); __put_user(host_ip->seq, &target_ip->seq); __put_user(host_ip->key, &target_ip->key); } abi_long host_to_target_shmid_ds(abi_ulong target_addr, struct shmid_ds *host_sd)
Quoted 8 lines
{
struct target_shmid_ds *target_sd;
target_sd = lock_user(VERIFY_WRITE, target_addr, sizeof(*target_sd), 0);
if (!target_sd){
return -TARGET_EFAULT;
}
host_to_target_ipc_perm__locked(target_addr, &(host_sd->shm_perm));No. You'd pass &target_sd->shm_perm, not target_addr, and you don't use g2h at all.
r~
PATCH 11/22 Introduce bsd-mem.h to the source tree
In reply to Karim TahaFrom: Stacey Son <[email protected]>
To preserve the copyright notice and help with the 'Author' info for subsequent changes to the file.
--- bsd-user/bsd-mem.h | 65 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 1 + 2 files changed, 66 insertions(+) create mode 100644 bsd-user/bsd-mem.h
bsd-user/bsd-mem.h new file +65 -0
@@ -0,0 +1,65 @@ +/* +* memory management system call shims and definitions +* +* Copyright (c) 2013-15 Stacey D. Son +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>. +*/ + +/* +* Copyright (c) 1982, 1986, 1993 +* The Regents of the University of California. All rights reserved. +* +* Redistribution and use in source and binary forms, with or without +* modification, are permitted provided that the following conditions +* are met: +* 1. Redistributions of source code must retain the above copyright +* notice, this list of conditions and the following disclaimer. +* 2. Redistributions in binary form must reproduce the above copyright +* notice, this list of conditions and the following disclaimer in the +* documentation and/or other materials provided with the distribution. +* 4. Neither the name of the University nor the names of its contributors +* may be used to endorse or promote products derived from this software +* without specific prior written permission. +* +* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND +* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE +* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL +* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS +* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) +* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT +* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY +* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF +* SUCH DAMAGE. +*/ + +#ifndef BSD_USER_BSD_MEM_H +#define BSD_USER_BSD_MEM_H + +#include <sys/types.h> +#include <sys/ipc.h> +#include <sys/mman.h> +#include <sys/shm.h> +#include <fcntl.h> + +#include "qemu-bsd.h" + +extern struct bsd_shm_regions bsd_shm_regions[]; +extern abi_ulong bsd_target_brk; +extern abi_ulong bsd_target_original_brk; +extern abi_ulong brk_page; + +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +1 -0
@@ -35,6 +35,7 @@/* BSD independent syscall shims */ #include "bsd-file.h" +#include "bsd-mem.h" #include "bsd-proc.h" /* *BSD dependent syscall shims */
Signature · 1 line
2.40.0
Re: [PATCH 11/22] Introduce bsd-mem.h to the source tree
Quoted 5 lines · Karim Taha
From: Stacey Son <[email protected]>
To preserve the copyright notice and help with the 'Author' info for subsequent changes to the file.
I'd drop the first 'To' in this message. It's clearer without it what is meant.
Quoted 8 lines
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 65 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 1 + 2 files changed, 66 insertions(+) create mode 100644 bsd-user/bsd-mem.h
otherwise
Warner
Quoted 93 lines
bsd-user/bsd-mem.h new file +33
-0@@ -0,0 +1,65 @@ +/* +* memory management system call shims and definitions +* +* Copyright (c) 2013-15 Stacey D. Son +* +* This program is free software; you can redistribute it and/or modify +* it under the terms of the GNU General Public License as published by +* the Free Software Foundation; either version 2 of the License, or +* (at your option) any later version. +* +* This program is distributed in the hope that it will be useful, +* but WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +* GNU General Public License for more details. +* +* You should have received a copy of the GNU General Public License +* along with this program; if not, see <http://www.gnu.org/licenses/>. +*/ + +/* +* Copyright (c) 1982, 1986, 1993 +* The Regents of the University of California. All rights reserved. +* +* Redistribution and use in source and binary forms, with or without +* modification, are permitted provided that the following conditions +* are met: +* 1. Redistributions of source code must retain the above copyright +* notice, this list of conditions and the following disclaimer. +* 2. Redistributions in binary form must reproduce the above copyright +* notice, this list of conditions and the following disclaimer in the +* documentation and/or other materials provided with the distribution. +* 4. Neither the name of the University nor the names of its contributors +* may be used to endorse or promote products derived from this software
+ * without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */bsd-user/bsd-mem.h new file +18
-0+ +#ifndef BSD_USER_BSD_MEM_H +#define BSD_USER_BSD_MEM_H + +#include <sys/types.h> +#include <sys/ipc.h> +#include <sys/mman.h> +#include <sys/shm.h> +#include <fcntl.h> + +#include "qemu-bsd.h" + +extern struct bsd_shm_regions bsd_shm_regions[]; +extern abi_ulong bsd_target_brk; +extern abi_ulong bsd_target_original_brk; +extern abi_ulong brk_page; + +#endif /* BSD_USER_BSD_MEM_H */ bsd-user/freebsd/os-syscall.c +1
-1@@ -35,6 +35,7 @@/* BSD independent syscall shims */ #include "bsd-file.h" +#include "bsd-mem.h" #include "bsd-proc.h" /* *BSD dependent syscall shims */ --2.40.0
Re: [PATCH 11/22] Introduce bsd-mem.h to the source tree
In reply to Karim TahaQuoted 5 lines · Karim Taha
From: Stacey Son<[email protected]>
To preserve the copyright notice and help with the 'Author' info for subsequent changes to the file.
Quoted 7 lines
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/bsd-mem.h | 65 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 1 + 2 files changed, 66 insertions(+)create mode 100644
bsd-user/bsd-mem.h
r~
PATCH 12/22 Implement mmap(2) and munmap(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 20 ++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 9 +++++++++ 2 files changed, 29 insertions(+)
bsd-user/bsd-mem.h +20 -0
@@ -62,4 +62,24 @@ extern abi_ulong bsd_target_brk;extern abi_ulong bsd_target_original_brk;extern abi_ulong brk_page; +/* mmap(2) */ +static inline abi_long do_bsd_mmap (void *cpu_env,abi_long arg1,abi_long arg2, +abi_long arg3,abi_long arg4,abi_long arg5,abi_long arg6,abi_long arg7, +abi_long arg8) +{ +if (regpairs_aligned (cpu_env) !=0 ) { + arg6 = arg7; + arg7 = arg8; + } +return get_errno (target_mmap (arg1, arg2, arg3, +target_to_host_bitmask (arg4, mmap_flags_tbl), + arg5,target_arg64 (arg6, arg7))); +} + +/* munmap(2) */ +static inline abi_long do_bsd_munmap (abi_long arg1,abi_long arg2) +{ +return get_errno (target_munmap (arg1, arg2)); +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +9 -0
@@ -486,6 +486,15 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1,/* * Memory management system calls. */ +case TARGET_FREEBSD_NR_mmap:/* mmap(2) */ + ret =do_bsd_mmap (cpu_env, arg1, arg2, arg3, arg4, arg5, arg6, arg7, + arg8); +break ; + +case TARGET_FREEBSD_NR_munmap:/* munmap(2) */ + ret =do_bsd_munmap (arg1, arg2); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret =do_freebsd_shm_open2 (arg1, arg2, arg3, arg4, arg5);
Signature · 1 line
2.40.0
Re: [PATCH 12/22] Implement mmap(2) and munmap(2)
Quoted 9 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 20 ++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 9 +++++++++ 2 files changed, 29 insertions(+)
Quoted 58 lines
bsd-user/bsd-mem.h +3
-0@@ -62,4 +62,24 @@ extern abi_ulong bsd_target_brk;extern abi_ulong bsd_target_original_brk;extern abi_ulong brk_page; +/* mmap(2) */ +static inline abi_long do_bsd_mmap (void *cpu_env,abi_long arg1,abi_long arg2, + abi_long arg3, abi_long arg4, abi_long arg5, abi_long arg6, abi_longarg7,
bsd-user/bsd-mem.h +8
-0+ abi_long arg8) +{ +if (regpairs_aligned (cpu_env) !=0 ) { + arg6 = arg7; + arg7 = arg8; + } + return get_errno(target_mmap(arg1, arg2, arg3, + target_to_host_bitmask(arg4,
mmap_flags_tbl),bsd-user/bsd-mem.h +9
-0+ arg5, target_arg64(arg6, arg7))); +} + +/* munmap(2) */ +static inline abi_long do_bsd_munmap (abi_long arg1,abi_long arg2) +{ +return get_errno (target_munmap (arg1, arg2)); +} +#endif /* BSD_USER_BSD_MEM_H */ bsd-user/freebsd/os-syscall.c +9
-0@@ -486,6 +486,15 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1,/* * Memory management system calls. */ + case TARGET_FREEBSD_NR_mmap: /* mmap(2) */ + ret =do_bsd_mmap (cpu_env, arg1, arg2, arg3, arg4, arg5, arg6, arg7, + arg8); +break ; + +case TARGET_FREEBSD_NR_munmap:/* munmap(2) */ + ret =do_bsd_munmap (arg1, arg2); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2: /* shm_open2(2) */ ret = do_freebsd_shm_open2(arg1, arg2, arg3, arg4, arg5);--
2.40.0
Re: [PATCH 12/22] Implement mmap(2) and munmap(2)
In reply to Karim Tahabsd-user/freebsd/os-syscall.c static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, · line 491
Quoted 3 lines · Karim Taha
bsd-user/freebsd/os-syscall.c +3
-0+case TARGET_FREEBSD_NR_mmap:/* mmap(2) */ + ret =do_bsd_mmap (cpu_env, arg1, arg2, arg3, arg4, arg5, arg6, arg7, + arg8);
Indentation. Align after (.
r~
PATCH 13/22 Implement mprotect(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 7 +++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 11 insertions(+)
bsd-user/bsd-mem.h +7 -0
@@ -82,4 +82,11 @@ static inline abi_long do_bsd_munmap(abi_long arg1, abi_long arg2) return get_errno(target_munmap(arg1, arg2)); } +/* mprotect(2) */ +static inline abi_long do_bsd_mprotect (abi_long arg1,abi_long arg2, +abi_long arg3) +{ +return get_errno (target_mprotect (arg1, arg2, arg3)); +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +4 -0
@@ -495,6 +495,10 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_munmap (arg1, arg2);break ; +case TARGET_FREEBSD_NR_mprotect:/* mprotect(2) */ + ret =do_bsd_mprotect (arg1, arg2, arg3); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret =do_freebsd_shm_open2 (arg1, arg2, arg3, arg4, arg5);
Signature · 1 line
2.40.0
Re: [PATCH 13/22] Implement mprotect(2)
Quoted 9 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 7 +++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 11 insertions(+)
Quoted 37 lines
bsd-user/bsd-mem.h +7
-0@@ -82,4 +82,11 @@ static inline abi_long do_bsd_munmap(abi_long arg1, abi_long arg2)return get_errno (target_munmap (arg1, arg2)); } +/* mprotect(2) */ +static inline abi_long do_bsd_mprotect (abi_long arg1,abi_long arg2, +abi_long arg3) +{ +return get_errno (target_mprotect (arg1, arg2, arg3)); +} +#endif /*
BSD_USER_BSD_MEM_H*/bsd-user/freebsd/os-syscall.c +4
-0@@ -495,6 +495,10 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_munmap (arg1, arg2);break ; +case TARGET_FREEBSD_NR_mprotect:/* mprotect(2) */ + ret =do_bsd_mprotect (arg1, arg2, arg3); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret = do_freebsd_shm_open2(arg1, arg2, arg3, arg4, arg5);--
2.40.0
Re: [PATCH 13/22] Implement mprotect(2)
In reply to Karim TahaQuoted 8 lines · Karim Taha
From: Stacey Son<[email protected]>
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/bsd-mem.h | 7 +++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 11 insertions(+)
r~
PATCH 14/22 Implement msync(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 11 +++++++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 15 insertions(+)
bsd-user/bsd-mem.h +11 -0
@@ -89,4 +89,15 @@ static inline abi_long do_bsd_mprotect(abi_long arg1, abi_long arg2, return get_errno(target_mprotect(arg1, arg2, arg3)); } +/* msync(2) */ +static inline abi_long do_bsd_msync (abi_long addr,abi_long len,abi_long flags) +{ +if (!access_ok (VERIFY_WRITE , addr, len)) { +/* XXX Should be EFAULT, but FreeBSD seems to get this wrong. */ +return -TARGET_ENOMEM ; + } + +return get_errno (msync (g2h_untagged (addr), len, flags)); +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +4 -0
@@ -499,6 +499,10 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_mprotect (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_msync:/* msync(2) */ + ret =do_bsd_msync (arg1, arg2, arg3); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret =do_freebsd_shm_open2 (arg1, arg2, arg3, arg4, arg5);
Signature · 1 line
2.40.0
Re: [PATCH 14/22] Implement msync(2)
Quoted 4 lines · Karim Taha
From: Stacey Son <[email protected]>
Co-authored-by Kyle Evans <[email protected]>
Quoted 8 lines
Signed-off-by Stacey Son <[email protected]>Signed-off-by Kyle Evans <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 11 +++++++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 15 insertions(+)
but see below
Quoted 13 lines
bsd-user/bsd-mem.h +2
-0@@ -89,4 +89,15 @@ static inline abi_long do_bsd_mprotect(abi_long arg1,abi_long arg2, return get_errno(target_mprotect(arg1, arg2, arg3)); } +/* msync(2) */ +static inline abi_long do_bsd_msync(abi_long addr, abi_long len, abi_longflags)
bsd-user/bsd-mem.h static inline abi_long do_bsd_mprotect(abi_long arg1, abi_long arg2, · line 96
Quoted 3 lines
bsd-user/bsd-mem.h +3
-0+{ +if (!access_ok (VERIFY_WRITE , addr, len)) { +/* XXX Should be EFAULT, but FreeBSD seems to get this wrong. */
I'd reword this here (and in the bsd-user fork):
/* It seems odd, but POSIX wants this to be ENOMEM */
since it isn't FreeBSD getting it wrong, and this matches the behavior that's documented for Linux as well (though in less detail). POSIX open group appears to document ENOMEM for this.
Quoted 27 lines
bsd-user/bsd-mem.h +6
-0+ return -TARGET_ENOMEM; + } + +return get_errno (msync (g2h_untagged (addr), len, flags)); +} +#endif /* BSD_USER_BSD_MEM_H */ bsd-user/freebsd/os-syscall.c +4
-0@@ -499,6 +499,10 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_mprotect (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_msync:/* msync(2) */ + ret =do_bsd_msync (arg1, arg2, arg3); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret = do_freebsd_shm_open2(arg1, arg2, arg3, arg4, arg5);--
2.40.0
Re: [PATCH 14/22] Implement msync(2)
In reply to Karim Tahabsd-user/bsd-mem.h static inline abi_long do_bsd_mprotect(abi_long arg1, abi_long arg2, · line 95
Quoted 3 lines · Karim Taha
bsd-user/bsd-mem.h +3
-0+static inline abi_long do_bsd_msync (abi_long addr,abi_long len,abi_long flags) +{ +if (!access_ok (VERIFY_WRITE , addr, len)) {
I think this check is wrong. There's nothing in the kernel that requires writability, or even that the entire range be mapped.
I think you want guest_range_valid_untagged to simply check that the bounds are ok.
With that,
r~
PATCH 15/22 Implement mlock(2), munlock(2), mlockall(2), munlockall(2), madvise(2), minherit(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 44 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 24 +++++++++++++++++++ 2 files changed, 68 insertions(+)
bsd-user/bsd-mem.h +44 -0
@@ -100,4 +100,48 @@ static inline abi_long do_bsd_msync(abi_long addr, abi_long len, abi_long flags)return get_errno (msync (g2h_untagged (addr), len, flags)); } +/* mlock(2) */ +static inline abi_long do_bsd_mlock (abi_long arg1,abi_long arg2) +{ +return get_errno (mlock (g2h_untagged (arg1), arg2)); +} + +/* munlock(2) */ +static inline abi_long do_bsd_munlock (abi_long arg1,abi_long arg2) +{ +return get_errno (munlock (g2h_untagged (arg1), arg2)); +} + +/* mlockall(2) */ +static inline abi_long do_bsd_mlockall (abi_long arg1) +{ +return get_errno (mlockall (arg1)); +} + +/* munlockall(2) */ +static inline abi_long do_bsd_munlockall (void ) +{ +return get_errno (munlockall ()); +} + +/* madvise(2) */ +static inline abi_long do_bsd_madvise (abi_long arg1,abi_long arg2, +abi_long arg3) +{ +/* +* A straight passthrough may not be safe because qemu sometimes +* turns private file-backed mapping into anonymous mappings. This +* will break MADV_DONTNEED. This is a hint, so ignoring and returing +* success is ok. +*/ +return get_errno (0 ); +} + +/* minherit(2) */ +static inline abi_long do_bsd_minherit (abi_long addr,abi_long len, +abi_long inherit) +{ +return get_errno (minherit (g2h_untagged (addr), len, inherit)); +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +24 -0
@@ -503,6 +503,30 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_msync (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_mlock:/* mlock(2) */ + ret =do_bsd_mlock (arg1, arg2); +break ; + +case TARGET_FREEBSD_NR_munlock:/* munlock(2) */ + ret =do_bsd_munlock (arg1, arg2); +break ; + +case TARGET_FREEBSD_NR_mlockall:/* mlockall(2) */ + ret =do_bsd_mlockall (arg1); +break ; + +case TARGET_FREEBSD_NR_munlockall:/* munlockall(2) */ + ret =do_bsd_munlockall (); +break ; + +case TARGET_FREEBSD_NR_madvise:/* madvise(2) */ + ret =do_bsd_madvise (arg1, arg2, arg3); +break ; + +case TARGET_FREEBSD_NR_minherit:/* minherit(2) */ + ret =do_bsd_minherit (arg1, arg2, arg3); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret =do_freebsd_shm_open2 (arg1, arg2, arg3, arg4, arg5);
Signature · 1 line
2.40.0
Re: [PATCH 15/22] Implement mlock(2), munlock(2), mlockall(2), munlockall(2), madvise(2), minherit(2)
Quoted 44 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 44 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 24 +++++++++++++++++++ 2 files changed, 68 insertions(+)bsd-user/bsd-mem.h +25
-0@@ -100,4 +100,48 @@ static inline abi_long do_bsd_msync(abi_long addr, abi_long len, abi_long flags) return get_errno(msync(g2h_untagged(addr), len, flags)); } +/* mlock(2) */ +static inline abi_long do_bsd_mlock (abi_long arg1,abi_long arg2) +{ +return get_errno (mlock (g2h_untagged (arg1), arg2)); +} + +/* munlock(2) */ +static inline abi_long do_bsd_munlock (abi_long arg1,abi_long arg2) +{ +return get_errno (munlock (g2h_untagged (arg1), arg2)); +} + +/* mlockall(2) */ +static inline abi_long do_bsd_mlockall (abi_long arg1) +{ +return get_errno (mlockall (arg1)); +} + +/* munlockall(2) */ +static inline abi_long do_bsd_munlockall (void ) +{ +return get_errno (munlockall ()); +} + +/* madvise(2) */
bsd-user/bsd-mem.h static inline abi_long do_bsd_msync(abi_long addr, · line 138
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+static inline abi_long do_bsd_madvise (abi_long arg1,abi_long arg2, +abi_long arg3) +{ +/* +* A straight passthrough may not be safe because qemu sometimes +* turns private file-backed mapping into anonymous mappings. This +* will break MADV_DONTNEED. This is a hint, so ignoring and returing +* success is ok. +*/ +return get_errno (0 );
This looks like it was copied from an early linux-user implementation, and that seems to have been fixed to no longer cause problems. Can someone that knows about the linux-user history here comment?
Warner
Quoted 50 lines
bsd-user/bsd-mem.h +9
-0+} + +/* minherit(2) */ +static inline abi_long do_bsd_minherit (abi_long addr,abi_long len, +abi_long inherit) +{ +return get_errno (minherit (g2h_untagged (addr), len, inherit)); +} +#endif /*
BSD_USER_BSD_MEM_H*/bsd-user/freebsd/os-syscall.c +24
-0@@ -503,6 +503,30 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_msync (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_mlock:/* mlock(2) */ + ret =do_bsd_mlock (arg1, arg2); +break ; + +case TARGET_FREEBSD_NR_munlock:/* munlock(2) */ + ret =do_bsd_munlock (arg1, arg2); +break ; + +case TARGET_FREEBSD_NR_mlockall:/* mlockall(2) */ + ret =do_bsd_mlockall (arg1); +break ; + +case TARGET_FREEBSD_NR_munlockall:/* munlockall(2) */ + ret =do_bsd_munlockall (); +break ; + +case TARGET_FREEBSD_NR_madvise:/* madvise(2) */ + ret =do_bsd_madvise (arg1, arg2, arg3); +break ; + +case TARGET_FREEBSD_NR_minherit:/* minherit(2) */ + ret =do_bsd_minherit (arg1, arg2, arg3); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret = do_freebsd_shm_open2(arg1, arg2, arg3, arg4, arg5);--
2.40.0
Re: [PATCH 15/22] Implement mlock(2), munlock(2), mlockall(2), munlockall(2), madvise(2), minherit(2)
Quoted 13 lines · Warner Losh
+/* madvise(2) */
+static inlineabi_longdo_bsd_madvise(abi_long arg1, abi_long arg2, + abi_long arg3)
+{
+ /*
+ * A straight passthrough may not be safe because qemu sometimes + * turns private file-backed mapping into anonymous mappings. This + * will break MADV_DONTNEED. This is a hint, so ignoring and returing + * success is ok.
+ */
+ returnget_errno(0);
Quoted 3 lines
This looks like it was copied from an early linux-user implementation, and that seems to have been fixed to no longer cause problems. Can someone that knows about the linux-user history here comment?
We now track pages that are "passthrough" and ok for DONTNEED etc.
case MADV_DONTNEED:
if (page_check_range(start, len, PAGE_PASSTHROUGH)) {
ret = get_errno(madvise(g2h_untagged(start), len, advice));
if ((advice == MADV_DONTNEED) && (ret == 0)) {
page_reset_target_data(start, start + len - 1);
}
}It's still not ideal, but it's something.
r~
Re: [PATCH 15/22] Implement mlock(2), munlock(2), mlockall(2), munlockall(2), madvise(2), minherit(2)
In reply to Karim Tahabsd-user/bsd-mem.h static inline abi_long do_bsd_msync(abi_long addr, abi_long len, abi_long flags) · line 113
Quoted 10 lines · Karim Taha
bsd-user/bsd-mem.h +10
-0+static inline abi_long do_bsd_mlock (abi_long arg1,abi_long arg2) +{ +return get_errno (mlock (g2h_untagged (arg1), arg2)); +} + +/* munlock(2) */ +static inline abi_long do_bsd_munlock (abi_long arg1,abi_long arg2) +{ +return get_errno (munlock (g2h_untagged (arg1), arg2)); +}
I think these two need guest_range_valid_untagged.
Otherwise,
r~
PATCH 16/22 Implement mincore(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 23 +++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 27 insertions(+)
bsd-user/bsd-mem.h +23 -0
@@ -144,4 +144,27 @@ static inline abi_long do_bsd_minherit(abi_long addr, abi_long len, return get_errno(minherit(g2h_untagged(addr), len, inherit)); } +/* mincore(2) */ +static inline abi_long do_bsd_mincore (abi_ulong target_addr,abi_ulong len, +abi_ulong target_vec) +{ +abi_long ret; +void *p, *a; + + a =lock_user (VERIFY_WRITE , target_addr, len,0 ); +if (a ==NULL ) { +return -TARGET_EFAULT ; + } + p =lock_user_string (target_vec); +if (p ==NULL ) { +unlock_user (a, target_addr,0 ); +return -TARGET_EFAULT ; + } + ret =get_errno (mincore (a, len, p)); +unlock_user (p, target_vec, ret); +unlock_user (a, target_addr,0 ); + +return ret; +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +4 -0
@@ -527,6 +527,10 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_minherit (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_mincore:/* mincore(2) */ + ret =do_bsd_mincore (arg1, arg2, arg3); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret =do_freebsd_shm_open2 (arg1, arg2, arg3, arg4, arg5);
Signature · 1 line
2.40.0
Re: [PATCH 16/22] Implement mincore(2)
Quoted 9 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 23 +++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 27 insertions(+)
Re: [PATCH 16/22] Implement mincore(2)
In reply to Karim TahaQuoted 11 lines · Karim Taha
bsd-user/bsd-mem.h +11
-0+/* mincore(2) */ +static inline abi_long do_bsd_mincore (abi_ulong target_addr,abi_ulong len, +abi_ulong target_vec) +{ +abi_long ret; +void *p, *a; + + a =lock_user (VERIFY_WRITE , target_addr, len,0 ); +if (a ==NULL ) { +return -TARGET_EFAULT ; + }
bsd-user/bsd-mem.h static inline abi_long do_bsd_minherit(abi_long addr, abi_long len, · line 167
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+ p =lock_user_string (target_vec); +if (p ==NULL ) { +unlock_user (a, target_addr,0 ); +return -TARGET_EFAULT ; + } + ret =get_errno (mincore (a, len, p)); +unlock_user (p, target_vec, ret); +unlock_user (a, target_addr,0 ); + +return ret;
This is wrong.
(1) VERIFY_WRITE is incorrect. Here you need a combination of guest_range_valid_untagged and page_check_range(addr, len, PAGE_VALID).
(2) vec is not a string, it is an array of size DIV_ROUND_UP(len, TARGET_PAGE_SIZE). For that, you do want lock_user(VERIFY_WRITE).
r~
PATCH 17/22 Implement do_obreak function
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 85 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 7 +++ 2 files changed, 92 insertions(+)
bsd-user/bsd-mem.h +85 -0
@@ -167,4 +167,89 @@ static inline abi_long do_bsd_mincore(abi_ulong target_addr, abi_ulong len,return ret; } +#ifdef DO_DEBUG +#define DEBUGF_BRK(message, args...) \ + do { fprintf(stderr, (message), ## args); } while (0) +#else +#define DEBUGF_BRK(message, args...) +#endif + +/* do_brk() must return target values and target errnos. */ +static inline abi_long do_obreak (abi_ulong new_brk) +{ +abi_long mapped_addr; +int new_alloc_size; + +DEBUGF_BRK ("do_brk(" TARGET_ABI_FMT_lx") -> " , new_brk); + +if (!new_brk) { +DEBUGF_BRK (TARGET_ABI_FMT_lx" (!new_brk)\n" , bsd_target_brk); +return bsd_target_brk; + } +if (new_brk < bsd_target_original_brk) { +DEBUGF_BRK (TARGET_ABI_FMT_lx" (new_brk < bsd_target_original_brk)\n" , + bsd_target_brk); +return bsd_target_brk; + } + +/* +* If the new brk is less than the highest page reserved to the target heap +* allocation, set it and we're almost done... +*/ +if (new_brk <= brk_page) { +/* +* Heap contents are initialized to zero, as for anonymous mapped pages. +*/ +if (new_brk > bsd_target_brk) { +memset (g2h_untagged (bsd_target_brk),0 , new_brk - bsd_target_brk); + } + bsd_target_brk = new_brk; +DEBUGF_BRK (TARGET_ABI_FMT_lx" (new_brk <= brk_page)\n" , + bsd_target_brk); +return bsd_target_brk; + } + +/* +* We need to allocate more memory after the brk... Note that we don't use +* MAP_FIXED because that will map over the top of any existing mapping +* (like the one with the host libc or qemu itself); instead we treat +* "mapped but at wrong address" as a failure and unmap again. +*/ + new_alloc_size =HOST_PAGE_ALIGN (new_brk - brk_page); + mapped_addr =get_errno (target_mmap (brk_page, new_alloc_size, +PROT_READ |PROT_WRITE , +MAP_ANON |MAP_PRIVATE ,-1 ,0 )); + +if (mapped_addr == brk_page) { +/* +* Heap contents are initialized to zero, as for anonymous mapped pages. +* Technically the new pages are already initialized to zero since they +* *are* anonymous mapped pages, however we have to take care with the +* contents that come from the remaining part of the previous page: it +* may contains garbage data due to a previous heap usage (grown then +* shrunken). +*/ +memset (g2h_untagged (bsd_target_brk),0 , brk_page - bsd_target_brk); + + bsd_target_brk = new_brk; + brk_page =HOST_PAGE_ALIGN (bsd_target_brk); +DEBUGF_BRK (TARGET_ABI_FMT_lx" (mapped_addr == brk_page)\n" , + bsd_target_brk); +return bsd_target_brk; + }else if (mapped_addr !=-1 ) { +/* +* Mapped but at wrong address, meaning there wasn't actually enough +* space for this brk. +*/ +target_munmap (mapped_addr, new_alloc_size); + mapped_addr =-1 ; +DEBUGF_BRK (TARGET_ABI_FMT_lx" (mapped_addr != -1)\n" , bsd_target_brk); + }else { +DEBUGF_BRK (TARGET_ABI_FMT_lx" (otherwise)\n" , bsd_target_brk); + } + +/* For everything else, return the previous break. */ +return bsd_target_brk; +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +7 -0
@@ -543,6 +543,13 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1,break ;#endif +/* +* Misc +*/ +case TARGET_FREEBSD_NR_break: + ret =do_obreak (arg1); +break ; +/* * sys{ctl, arch, call} */
Signature · 1 line
2.40.0
Re: [PATCH 17/22] Implement do_obreak function
Quoted 4 lines · Karim Taha
From: Stacey Son <[email protected]>
Co-authored-by Mikaël Urankar <[email protected]>
Quoted 8 lines
Signed-off-by Mikaël Urankar <[email protected]>Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 85 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 7 +++ 2 files changed, 92 insertions(+)
emacs works with this, iirc, and it's super picky about brk being pedantically correct.
This also doesn't match current linux-user. Perhaps the caution it uses is no longer
needed? I think it's another area we should invite the linux-user maintainers to
comment.
Warner
Quoted 132 lines
bsd-user/bsd-mem.h +21
-0@@ -167,4 +167,89 @@ static inline abi_long do_bsd_mincore(abi_ulong target_addr, abi_ulong len, return ret; } +#ifdef DO_DEBUG +#define DEBUGF_BRK(message, args...) \ + do { fprintf(stderr, (message), ## args); } while (0) +#else +#define DEBUGF_BRK(message, args...) +#endif + +/* do_brk() must return target values and target errnos. */ +static inline abi_long do_obreak(abi_ulong new_brk) +{ + abi_long mapped_addr; + int new_alloc_size; + + DEBUGF_BRK("do_brk(" TARGET_ABI_FMT_lx ") -> ", new_brk); + + if (!new_brk) { + DEBUGF_BRK(TARGET_ABI_FMT_lx " (!new_brk)\n", bsd_target_brk); + return bsd_target_brk; + } + if (new_brk < bsd_target_original_brk) { + DEBUGF_BRK(TARGET_ABI_FMT_lx " (new_brk <bsd_target_original_brk)\n",
bsd-user/bsd-mem.h +6
-0+ bsd_target_brk); +return bsd_target_brk; + } + +/* +* If the new brk is less than the highest page reserved to the target heap
bsd-user/bsd-mem.h +5
-0+* allocation, set it and we're almost done... +*/ +if (new_brk <= brk_page) { +/* +* Heap contents are initialized to zero, as for anonymous mapped pages.
bsd-user/bsd-mem.h +3
-0+*/ +if (new_brk > bsd_target_brk) { + memset(g2h_untagged(bsd_target_brk), 0, new_brk -
bsd_target_brk);bsd-user/bsd-mem.h +9
-0+ } + bsd_target_brk = new_brk; +DEBUGF_BRK (TARGET_ABI_FMT_lx" (new_brk <= brk_page)\n" , + bsd_target_brk); +return bsd_target_brk; + } + +/* +* We need to allocate more memory after the brk... Note that we don't use
+ *MAP_FIXEDbecause that will map over the top of any existing
mappingbsd-user/bsd-mem.h +11
-0+* (like the one with the host libc or qemu itself); instead we treat +* "mapped but at wrong address" as a failure and unmap again. +*/ + new_alloc_size =HOST_PAGE_ALIGN (new_brk - brk_page); + mapped_addr =get_errno (target_mmap (brk_page, new_alloc_size, +PROT_READ |PROT_WRITE , +MAP_ANON |MAP_PRIVATE ,-1 ,0 )); + +if (mapped_addr == brk_page) { +/* +* Heap contents are initialized to zero, as for anonymous mapped pages.
+ * Technically the new pages are already initialized to zero
since they
+ * *are* anonymous mapped pages, however we have to take care
with the
+ * contents that come from the remaining part of the previous
page: it
+ * may contains garbage data due to a previous heap usage (grown
thenbsd-user/bsd-mem.h +3
-0+* shrunken). +*/ + memset(g2h_untagged(bsd_target_brk), 0, brk_page -
bsd_target_brk);bsd-user/bsd-mem.h +9
-0+ + bsd_target_brk = new_brk; + brk_page =HOST_PAGE_ALIGN (bsd_target_brk); +DEBUGF_BRK (TARGET_ABI_FMT_lx" (mapped_addr == brk_page)\n" , + bsd_target_brk); +return bsd_target_brk; + }else if (mapped_addr !=-1 ) { +/* +* Mapped but at wrong address, meaning there wasn't actually enough
bsd-user/bsd-mem.h +5
-0+* space for this brk. +*/ +target_munmap (mapped_addr, new_alloc_size); + mapped_addr =-1 ; + DEBUGF_BRK(TARGET_ABI_FMT_lx " (mapped_addr != -1)\n",
bsd_target_brk);bsd-user/bsd-mem.h +8
-0+ } else { +DEBUGF_BRK (TARGET_ABI_FMT_lx" (otherwise)\n" , bsd_target_brk); + } + +/* For everything else, return the previous break. */ +return bsd_target_brk; +} +#endif /* BSD_USER_BSD_MEM_H */ bsd-user/freebsd/os-syscall.c +7
-0@@ -543,6 +543,13 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, break;#endif +/* +* Misc +*/ +case TARGET_FREEBSD_NR_break: + ret =do_obreak (arg1); +break ; +/* * sys{ctl, arch, call} */--
2.40.0
Re: [PATCH 17/22] Implement do_obreak function
In reply to Karim TahaQuoted 17 lines · Karim Taha
From: Stacey Son <[email protected]>
Co-authored-by Mikaël Urankar <[email protected]>Signed-off-by Mikaël Urankar <[email protected]>Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 85 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 7 +++ 2 files changed, 92 insertions(+)bsd-user/bsd-mem.h +0
-0
bsd-user/bsd-mem.h static inline abi_long do_bsd_mincore(abi_ulong target_addr, abi_ulong len, · line 175
Quoted 9 lines
bsd-user/bsd-mem.h +6
-0@@ -167,4 +167,89 @@ static inline abi_long do_bsd_mincore(abi_ulong target_addr, abi_ulong len,return ret; } +#ifdef DO_DEBUG +#define DEBUGF_BRK(message, args...) \ + do { fprintf(stderr, (message), ## args); } while (0) +#else +#define DEBUGF_BRK(message, args...) +#endif
We are trying to get rid of all of this old-style debugging. We are replacing them with tracepoints.
Quoted 8 lines
bsd-user/bsd-mem.h +8
-0+ +/* do_brk() must return target values and target errnos. */ +static inline abi_long do_obreak (abi_ulong new_brk) +{ +abi_long mapped_addr; +int new_alloc_size; + +DEBUGF_BRK ("do_brk(" TARGET_ABI_FMT_lx") -> " , new_brk);
bsd-user/bsd-mem.h static inline abi_long do_bsd_mincore(abi_ulong target_addr, abi_ulong len, · line 193
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+ +if (!new_brk) { +DEBUGF_BRK (TARGET_ABI_FMT_lx" (!new_brk)\n" , bsd_target_brk); +return bsd_target_brk; + } +if (new_brk < bsd_target_original_brk) { +DEBUGF_BRK (TARGET_ABI_FMT_lx" (new_brk < bsd_target_original_brk)\n" , + bsd_target_brk); +return bsd_target_brk; + }
These two cases may be combined, since original_brk will never be 0.
Quoted 15 lines
bsd-user/bsd-mem.h +15
-0+ +/* +* If the new brk is less than the highest page reserved to the target heap +* allocation, set it and we're almost done... +*/ +if (new_brk <= brk_page) { +/* +* Heap contents are initialized to zero, as for anonymous mapped pages. +*/ +if (new_brk > bsd_target_brk) { +memset (g2h_untagged (bsd_target_brk),0 , new_brk - bsd_target_brk); + } + bsd_target_brk = new_brk; +DEBUGF_BRK (TARGET_ABI_FMT_lx" (new_brk <= brk_page)\n" , + bsd_target_brk);
bsd-user/bsd-mem.h static inline abi_long do_bsd_mincore(abi_ulong target_addr, abi_ulong len, · line 218
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+return bsd_target_brk; + } + +/* +* We need to allocate more memory after the brk... Note that we don't use +* MAP_FIXED because that will map over the top of any existing mapping +* (like the one with the host libc or qemu itself); instead we treat +* "mapped but at wrong address" as a failure and unmap again. +*/ + new_alloc_size =HOST_PAGE_ALIGN (new_brk - brk_page);
We removed all HOST_PAGE_ALIGN from brk in linux-user this cycle.
Quoted 21 lines
bsd-user/bsd-mem.h +21
-0+ mapped_addr =get_errno (target_mmap (brk_page, new_alloc_size, +PROT_READ |PROT_WRITE , +MAP_ANON |MAP_PRIVATE ,-1 ,0 )); + +if (mapped_addr == brk_page) { +/* +* Heap contents are initialized to zero, as for anonymous mapped pages. +* Technically the new pages are already initialized to zero since they +* *are* anonymous mapped pages, however we have to take care with the +* contents that come from the remaining part of the previous page: it +* may contains garbage data due to a previous heap usage (grown then +* shrunken). +*/ +memset (g2h_untagged (bsd_target_brk),0 , brk_page - bsd_target_brk); + + bsd_target_brk = new_brk; + brk_page =HOST_PAGE_ALIGN (bsd_target_brk); +DEBUGF_BRK (TARGET_ABI_FMT_lx" (mapped_addr == brk_page)\n" , + bsd_target_brk); +return bsd_target_brk; + }else if (mapped_addr !=-1 ) {
bsd-user/bsd-mem.h static inline abi_long do_bsd_mincore(abi_ulong target_addr, abi_ulong len, · line 249
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+/* +* Mapped but at wrong address, meaning there wasn't actually enough +* space for this brk. +*/ +target_munmap (mapped_addr, new_alloc_size); + mapped_addr =-1 ; +DEBUGF_BRK (TARGET_ABI_FMT_lx" (mapped_addr != -1)\n" , bsd_target_brk); + }else { +DEBUGF_BRK (TARGET_ABI_FMT_lx" (otherwise)\n" , bsd_target_brk); + }
It may require additional work in target_mmap, but you probably want to use MAP_FIXED | MAP_EXCL here. On the linux-user side we changed to use MAP_FIXED_NOREPLACE, which is approximately the same thing.
r~
PATCH 18/22 Implement shm_open(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 27 +++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 31 insertions(+)
bsd-user/bsd-mem.h +27 -0
@@ -252,4 +252,31 @@ static inline abi_long do_obreak(abi_ulong new_brk) return bsd_target_brk; } +/* shm_open(2) */ +static inline abi_long do_bsd_shm_open (abi_ulong arg1,abi_long arg2, +abi_long arg3) +{ +int ret; +void *p; + +#define SHM_PATH (p) ((p) == SHM_ANON ? (p) : path(p)) +if (arg1 == (uintptr_t )SHM_ANON ) { + p =SHM_ANON ; + }else { + p =lock_user_string (arg1); +if (p ==NULL ) { +return -TARGET_EFAULT ; + } + } + ret =get_errno (shm_open (SHM_PATH (p), +target_to_host_bitmask (arg2, fcntl_flags_tbl), arg3)); + +if (p !=SHM_ANON ) { +unlock_user (p, arg1,0 ); + } + +return ret; +} +#undef SHM_PATH +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +4 -0
@@ -531,6 +531,10 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_mincore (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_freebsd12_shm_open:/* shm_open(2) */ + ret =do_bsd_shm_open (arg1, arg2, arg3); +break ; +#if defined(__FreeBSD_version) && __FreeBSD_version >=1300048 case TARGET_FREEBSD_NR_shm_open2:/* shm_open2(2) */ ret =do_freebsd_shm_open2 (arg1, arg2, arg3, arg4, arg5);
Signature · 1 line
2.40.0
Re: [PATCH 18/22] Implement shm_open(2)
Quoted 4 lines · Karim Taha
From: Stacey Son <[email protected]>
Co-authored-by Kyle Evans <[email protected]>
Quoted 8 lines
Signed-off-by Stacey Son <[email protected]>Signed-off-by Kyle Evans <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 27 +++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 31 insertions(+)
Re: [PATCH 18/22] Implement shm_open(2)
In reply to Karim Tahabsd-user/bsd-mem.h static inline abi_long do_obreak(abi_ulong new_brk) · line 271
Quoted 10 lines · Karim Taha
bsd-user/bsd-mem.h +10
-0+#define SHM_PATH (p) ((p) == SHM_ANON ? (p) : path(p)) +if (arg1 == (uintptr_t )SHM_ANON ) { + p =SHM_ANON ; + }else { + p =lock_user_string (arg1); +if (p ==NULL ) { +return -TARGET_EFAULT ; + } + } + ret = get_errno(shm_open(SHM_PATH(p),
Again, SHM_PATH is not needed, because the condition is handled by this IF.
r~
Re: [PATCH 18/22] Implement shm_open(2)
Quoted 12 lines · Richard Henderson
Quoted 10 lines · Karim Taha
+#define
SHM_PATH(p)((p) ==SHM_ANON? (p) : path(p)) + if (arg1 == (uintptr_t)SHM_ANON) {
+ p =SHM_ANON;
+ } else {
+ p =lock_user_string(arg1);
+ if (p == NULL) {
+ return -TARGET_EFAULT;
+ }
+ }
+ ret = get_errno(shm_open(SHM_PATH(p),
Quoted 1 line
Again,
SHM_PATHis not needed, because the condition is handled by this IF.
Oh, no, that's something different.
But path() is wrong for shm_open, because it's not a path in the regular file system.
r~
PATCH 19/22 Implement shm_unlink(2) and shmget(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 23 +++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 8 ++++++++ 2 files changed, 31 insertions(+)
bsd-user/bsd-mem.h +23 -0
@@ -279,4 +279,27 @@ static inline abi_long do_bsd_shm_open(abi_ulong arg1, abi_long arg2, }#undef SHM_PATH +/* shm_unlink(2) */ +static inline abi_long do_bsd_shm_unlink (abi_ulong arg1) +{ +int ret; +void *p; + + p =lock_user_string (arg1); +if (p ==NULL ) { +return -TARGET_EFAULT ; + } + ret =get_errno (shm_unlink (p));/* XXX path(p)? */ +unlock_user (p, arg1,0 ); + +return ret; +} + +/* shmget(2) */ +static inline abi_long do_bsd_shmget (abi_long arg1,abi_ulong arg2, +abi_long arg3) +{ +return get_errno (shmget (arg1, arg2, arg3)); +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +8 -0
@@ -547,6 +547,14 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1,break ;#endif +case TARGET_FREEBSD_NR_shm_unlink:/* shm_unlink(2) */ + ret =do_bsd_shm_unlink (arg1); +break ; + +case TARGET_FREEBSD_NR_shmget:/* shmget(2) */ + ret =do_bsd_shmget (arg1, arg2, arg3); +break ; +/* * Misc */
Signature · 1 line
2.40.0
Re: [PATCH 19/22] Implement shm_unlink(2) and shmget(2)
Quoted 9 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 23 +++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 8 ++++++++ 2 files changed, 31 insertions(+)
Re: [PATCH 19/22] Implement shm_unlink(2) and shmget(2)
In reply to Karim TahaQuoted 8 lines · Karim Taha
From: Stacey Son<[email protected]>
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/bsd-mem.h | 23 +++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 8 ++++++++ 2 files changed, 31 insertions(+)
r~
Re: [PATCH 19/22] Implement shm_unlink(2) and shmget(2)
In reply to Karim TahaQuoted 8 lines · Karim Taha
From: Stacey Son<[email protected]>
Signed-off-by Stacey Son<[email protected]>Signed-off-by Karim Taha<[email protected]>--- bsd-user/bsd-mem.h | 23 +++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 8 ++++++++ 2 files changed, 31 insertions(+)
r~
PATCH 20/22 Implement shmctl(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 33 +++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 37 insertions(+)
bsd-user/bsd-mem.h +33 -0
@@ -302,4 +302,37 @@ static inline abi_long do_bsd_shmget(abi_long arg1, abi_ulong arg2, return get_errno(shmget(arg1, arg2, arg3)); } +/* shmctl(2) */ +static inline abi_long do_bsd_shmctl (abi_long shmid,abi_long cmd, +abi_ulong buff) +{ +struct shmid_ds dsarg; +abi_long ret = -TARGET_EINVAL ; + + cmd &=0xff ; + +switch (cmd) { +case IPC_STAT : +case IPC_SET : +if (target_to_host_shmid_ds (&dsarg, buff)) { +return -TARGET_EFAULT ; + } + ret =get_errno (shmctl (shmid, cmd, &dsarg)); +if (host_to_target_shmid_ds (buff, &dsarg)) { +return -TARGET_EFAULT ; + } +break ; + +case IPC_RMID : + ret =get_errno (shmctl (shmid, cmd,NULL )); +break ; + +default : + ret = -TARGET_EINVAL ; +break ; + } + +return ret; +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +4 -0
@@ -555,6 +555,10 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_shmget (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_shmctl:/* shmctl(2) */ + ret =do_bsd_shmctl (arg1, arg2, arg3); +break ; +/* * Misc */
Signature · 1 line
2.40.0
Re: [PATCH 20/22] Implement shmctl(2)
Quoted 9 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 33 +++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 4 ++++ 2 files changed, 37 insertions(+)
Re: [PATCH 20/22] Implement shmctl(2)
In reply to Karim Tahabsd-user/bsd-mem.h static inline abi_long do_bsd_shmget(abi_long arg1, abi_ulong arg2, · line 324
Quoted 11 lines · Karim Taha
bsd-user/bsd-mem.h +11
-0+switch (cmd) { +case IPC_STAT : +case IPC_SET : +if (target_to_host_shmid_ds (&dsarg, buff)) { +return -TARGET_EFAULT ; + } + ret =get_errno (shmctl (shmid, cmd, &dsarg)); +if (host_to_target_shmid_ds (buff, &dsarg)) { +return -TARGET_EFAULT ; + } +break ;
IPC_STAT treats buff as output, IPC_SET treats buff as input, so these cases can't be combined.
r~
Re: [PATCH 20/22] Implement shmctl(2)
Quoted 13 lines · Richard Henderson
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmget(abi_long arg1, abi_ulong arg2, · line 324
Quoted 11 lines · Karim Taha
bsd-user/bsd-mem.h +11
-0+switch (cmd) { +case IPC_STAT : +case IPC_SET : +if (target_to_host_shmid_ds (&dsarg, buff)) { +return -TARGET_EFAULT ; + } + ret =get_errno (shmctl (shmid, cmd, &dsarg)); +if (host_to_target_shmid_ds (buff, &dsarg)) { +return -TARGET_EFAULT ; + } +break ;
Quoted 5 lines
IPC_STATtreats buff as output,IPC_SETtreats buff as input, so these cases can't be combined.r~
I think they can be combined:
1- we marshal the struct buff from target to host 2- call shmctl
3- marshal the results back.
--
Karim Taha
Re: [PATCH 20/22] Implement shmctl(2)
Quoted 19 lines · Karim Taha
Quoted 17 lines · Richard Henderson
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmget(abi_long arg1, abi_ulong arg2, · line 324
Quoted 11 lines · Karim Taha
bsd-user/bsd-mem.h +11
-0+switch (cmd) { +case IPC_STAT : +case IPC_SET : +if (target_to_host_shmid_ds (&dsarg, buff)) { +return -TARGET_EFAULT ; + } + ret =get_errno (shmctl (shmid, cmd, &dsarg)); +if (host_to_target_shmid_ds (buff, &dsarg)) { +return -TARGET_EFAULT ; + } +break ;
IPC_STATtreats buff as output,IPC_SETtreats buff as input, so these cases can't be combined.
Quoted 6 lines
Quoted 1 line
r~
I think they can be combined:
1- we marshal the structbufffrom target to host 2- callshmctl
3- marshal the results back.
No. For IPC_SET, the target page need not be writable, as you assume here.
r~
PATCH 21/22 Implement shmat(2) and shmdt(2)
In reply to Karim TahaFrom: Stacey Son <[email protected]>
--- bsd-user/bsd-mem.h | 72 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 8 ++++ 2 files changed, 80 insertions(+)
bsd-user/bsd-mem.h +72 -0
@@ -335,4 +335,76 @@ static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd,return ret; } +/* shmat(2) */ +static inline abi_long do_bsd_shmat (int shmid,abi_ulong shmaddr,int shmflg) +{ +abi_ulong raddr; +abi_long ret; +void *host_raddr; +struct shmid_ds shm_info; +int i; + +/* Find out the length of the shared memory segment. */ + ret =get_errno (shmctl (shmid,IPC_STAT , &shm_info)); +if (is_error (ret)) { +/* Can't get the length */ +return ret; + } + +mmap_lock (); + +if (shmaddr) { + host_raddr =shmat (shmid, (void *)g2h_untagged (shmaddr), shmflg); + }else { +abi_ulong mmap_start; + + mmap_start =mmap_find_vma (0 , shm_info.shm_segsz); + +if (mmap_start ==-1 ) { + errno =ENOMEM ; + host_raddr = (void *)-1 ; + }else { + host_raddr =shmat (shmid,g2h_untagged (mmap_start), + shmflg);/* | SHM_REMAP XXX WHY? */ + } + } + +if (host_raddr == (void *)-1 ) { +mmap_unlock (); +return get_errno ((long )host_raddr); + } + raddr =h2g ((unsigned long )host_raddr); + +page_set_flags (raddr, raddr + shm_info.shm_segsz, +PAGE_VALID |PAGE_READ | ((shmflg &SHM_RDONLY ) ?0 :PAGE_WRITE )); + +for (i =0 ; i <N_BSD_SHM_REGIONS ; i++) { +if (bsd_shm_regions[i].start ==0 ) { + bsd_shm_regions[i].start = raddr; + bsd_shm_regions[i].size = shm_info.shm_segsz; +break ; + } + } + +mmap_unlock (); +return raddr; +} + +/* shmdt(2) */ +static inline abi_long do_bsd_shmdt (abi_ulong shmaddr) +{ +int i; + +for (i =0 ; i <N_BSD_SHM_REGIONS ; ++i) { +if (bsd_shm_regions[i].start == shmaddr) { + bsd_shm_regions[i].start =0 ; +page_set_flags (shmaddr, + shmaddr + bsd_shm_regions[i].size,0 ); +break ; + } + } + +return get_errno (shmdt (g2h_untagged (shmaddr))); +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +8 -0
@@ -559,6 +559,14 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_shmctl (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_shmat:/* shmat(2) */ + ret =do_bsd_shmat (arg1, arg2, arg3); +break ; + +case TARGET_FREEBSD_NR_shmdt:/* shmdt(2) */ + ret =do_bsd_shmdt (arg1); +break ; +/* * Misc */
Signature · 1 line
2.40.0
Re: [PATCH 21/22] Implement shmat(2) and shmdt(2)
Quoted 41 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 72 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 8 ++++ 2 files changed, 80 insertions(+)bsd-user/bsd-mem.h +2
-0@@ -335,4 +335,76 @@ static inline abi_long do_bsd_shmctl(abi_long shmid,abi_long cmd, return ret; } +/* shmat(2) */ +static inline abi_long do_bsd_shmat(int shmid, abi_ulong shmaddr, intshmflg)
bsd-user/bsd-mem.h +19
-0+{ +abi_ulong raddr; +abi_long ret; +void *host_raddr; +struct shmid_ds shm_info; +int i; + +/* Find out the length of the shared memory segment. */ + ret =get_errno (shmctl (shmid,IPC_STAT , &shm_info)); +if (is_error (ret)) { +/* Can't get the length */ +return ret; + } + +mmap_lock (); + +if (shmaddr) { + host_raddr =shmat (shmid, (void *)g2h_untagged (shmaddr), shmflg); + }else {
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd, · line 368
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+abi_ulong mmap_start; + + mmap_start =mmap_find_vma (0 , shm_info.shm_segsz); + +if (mmap_start ==-1 ) { + errno =ENOMEM ; + host_raddr = (void *)-1 ; + }else { + host_raddr =shmat (shmid,g2h_untagged (mmap_start), + shmflg);/* | SHM_REMAP XXX WHY? */
I was all set to hit reviewed by on this, but this has me curious. Kyle (or anybody else) do
you know the back story here. git blame is less than helpful.
Warner
Quoted 67 lines
bsd-user/bsd-mem.h +11
-0+ } + } + +if (host_raddr == (void *)-1 ) { +mmap_unlock (); +return get_errno ((long )host_raddr); + } + raddr =h2g ((unsigned long )host_raddr); + + page_set_flags(raddr, raddr + shm_info.shm_segsz, + PAGE_VALID | PAGE_READ | ((shmflg & SHM_RDONLY) ? 0 :
PAGE_WRITE));bsd-user/bsd-mem.h +30
-0+ +for (i =0 ; i <N_BSD_SHM_REGIONS ; i++) { +if (bsd_shm_regions[i].start ==0 ) { + bsd_shm_regions[i].start = raddr; + bsd_shm_regions[i].size = shm_info.shm_segsz; +break ; + } + } + +mmap_unlock (); + return raddr; +} + +/* shmdt(2) */ +static inline abi_long do_bsd_shmdt (abi_ulong shmaddr) +{ +int i; + +for (i =0 ; i <N_BSD_SHM_REGIONS ; ++i) { +if (bsd_shm_regions[i].start == shmaddr) { + bsd_shm_regions[i].start =0 ; +page_set_flags (shmaddr, + shmaddr + bsd_shm_regions[i].size,0 ); +break ; + } + } + +return get_errno (shmdt (g2h_untagged (shmaddr))); +} +#endif /* BSD_USER_BSD_MEM_H */ bsd-user/freebsd/os-syscall.c +8
-0@@ -559,6 +559,14 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_shmctl (arg1, arg2, arg3);break ; +case TARGET_FREEBSD_NR_shmat:/* shmat(2) */ + ret =do_bsd_shmat (arg1, arg2, arg3); +break ; + +case TARGET_FREEBSD_NR_shmdt:/* shmdt(2) */ + ret =do_bsd_shmdt (arg1); +break ; +/* * Misc */--
2.40.0
Re: [PATCH 21/22] Implement shmat(2) and shmdt(2)
In reply to Karim TahaQuoted 27 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 72 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 8 ++++ 2 files changed, 80 insertions(+)bsd-user/bsd-mem.h +10
-0@@ -335,4 +335,76 @@ static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd,return ret; } +/* shmat(2) */ +static inline abi_long do_bsd_shmat (int shmid,abi_ulong shmaddr,int shmflg) +{ +abi_ulong raddr; +abi_long ret; +void *host_raddr; +struct shmid_ds shm_info; +int i; + +/* Find out the length of the shared memory segment. */
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd, · line 357
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+ ret =get_errno (shmctl (shmid,IPC_STAT , &shm_info)); +if (is_error (ret)) { +/* Can't get the length */ +return ret; + } + +mmap_lock (); + +if (shmaddr) { + host_raddr =shmat (shmid, (void *)g2h_untagged (shmaddr), shmflg);
Missing
if (!guest_range_valid_untagged(shmaddr, shm_info.shm_segsz)) {
return -TARGET_EINVAL;
}bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd, · line 368
Quoted 11 lines
bsd-user/bsd-mem.h +11
-0+ }else { +abi_ulong mmap_start; + + mmap_start =mmap_find_vma (0 , shm_info.shm_segsz); + +if (mmap_start ==-1 ) { + errno =ENOMEM ; + host_raddr = (void *)-1 ; + }else { + host_raddr =shmat (shmid,g2h_untagged (mmap_start), + shmflg);/* | SHM_REMAP XXX WHY? */
With reserved_va, the entire guest address space is mapped PROT_NONE so that it is reserved, so that the kernel does not use it for something else. You need the SHM_REMAP to replace the reservation mapping.
Quoted 6 lines
bsd-user/bsd-mem.h +6
-0+/* shmdt(2) */ +static inline abi_long do_bsd_shmdt (abi_ulong shmaddr) +{ +int i; + +for (i =0 ; i <N_BSD_SHM_REGIONS ; ++i) {
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd, · line 408
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+if (bsd_shm_regions[i].start == shmaddr) { + bsd_shm_regions[i].start =0 ; +page_set_flags (shmaddr, + shmaddr + bsd_shm_regions[i].size,0 ); +break ; + } + } + +return get_errno (shmdt (g2h_untagged (shmaddr))); +}
Hmm, bug with linux-user as well, because here we should re-establish the reserved_va reservation.
Also, we should not be using a fixed sized array. Nothing good happens when the array fills up.
r~
Re: [PATCH 21/22] Implement shmat(2) and shmdt(2)
Quoted 78 lines · Richard Henderson
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long
Quoted 14 lines · Karim Taha
From: Stacey Son <[email protected]>
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 72 +++++++++++++++++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 8 ++++ 2 files changed, 80 insertions(+)bsd-user/bsd-mem.h +0
-0@@ -335,4 +335,76 @@ static inline abi_long do_bsd_shmctl(abi_longshmid,
abi_longcmd,bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd,
Quoted 5 lines
bsd-user/bsd-mem.h +2
-0return ret; } +/* shmat(2) */ +static inline abi_long do_bsd_shmat(int shmid, abi_ulong shmaddr, intshmflg)
Quoted 8 lines
bsd-user/bsd-mem.h +8
-0+{ +abi_ulong raddr; +abi_long ret; +void *host_raddr; +struct shmid_ds shm_info; +int i; + +/* Find out the length of the shared memory segment. */ bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd,
Quoted 10 lines
bsd-user/bsd-mem.h +10
-0+ ret =get_errno (shmctl (shmid,IPC_STAT , &shm_info)); +if (is_error (ret)) { +/* Can't get the length */ +return ret; + } + +mmap_lock (); + +if (shmaddr) { + host_raddr = shmat(shmid, (void *)g2h_untagged(shmaddr),shmflg);
Missing
if (!guest_range_valid_untagged(shmaddr, shm_info.shm_segsz)) { return -TARGET_EINVAL; }bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd, · line 368
Quoted 11 lines
bsd-user/bsd-mem.h +11
-0+ }else { +abi_ulong mmap_start; + + mmap_start =mmap_find_vma (0 , shm_info.shm_segsz); + +if (mmap_start ==-1 ) { + errno =ENOMEM ; + host_raddr = (void *)-1 ; + }else { + host_raddr =shmat (shmid,g2h_untagged (mmap_start), + shmflg);/* | SHM_REMAP XXX WHY? */ With
reserved_va, the entire guest address space is mappedPROT_NONEso that it is
reserved, so that the kernel does not use it for something else. You need theSHM_REMAP
to replace the reservation mapping.Quoted 12 lines
bsd-user/bsd-mem.h +12
-0+/* shmdt(2) */ +static inline abi_long do_bsd_shmdt (abi_ulong shmaddr) +{ +int i; + +for (i =0 ; i <N_BSD_SHM_REGIONS ; ++i) { +if (bsd_shm_regions[i].start == shmaddr) { + bsd_shm_regions[i].start =0 ; +page_set_flags (shmaddr, + shmaddr + bsd_shm_regions[i].size,0 ); +break ; + }
Quoted 9 lines
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmctl(abi_long shmid, abi_long cmd, · line 408
Quoted 4 lines
bsd-user/bsd-mem.h +4
-0+ } + +return get_errno (shmdt (g2h_untagged (shmaddr))); +}Hmm, bug with linux-user as well, because here we should re-establish the
reserved_va
reservation.
... of the shared memory region we just detached? Right?
Quoted 4 lines
Also, we should not be using a fixed sized array. Nothing good happens when the array
fills up.
File this as https://github.com/qemu-bsd-user/qemu-bsd-user/issues/47 so we don't forget.
It's good enough for the moment since the programs we've seen have a very limited number
of segments... but longer term, it should be dynamic.
Warner
Re: [PATCH 21/22] Implement shmat(2) and shmdt(2)
Quoted 5 lines · Warner Losh
Hmm, bug with linux-user as well, because here we should re-establish the
reserved_vareservation.... of the shared memory region we just detached? Right?
Correct.
On a related note, on FreeBSD is there any practical difference between
PROT_NONE, MAP_ANON
and
PROT_NONE, MAP_GUARD
for large memory regions?
I ask since FreeBSD doesn't have MAP_NORESERVE, which Linux uses to avoid allocation of gigabytes.
r~
Re: [PATCH 21/22] Implement shmat(2) and shmdt(2)
Quoted 10 lines · Richard Henderson
Quoted 1 line · Warner Losh
Hmm, bug with linux-user as well, because here we should
re-establish the
reserved_vaQuoted 4 lines
reservation.
... of the shared memory region we just detached? Right?
Correct.
Quoted 8 lines
On a related note, on FreeBSD is there any practical difference between
PROT_NONE,MAP_ANON
and
PROT_NONE,MAP_GUARDfor large memory regions?
They do different things. MAP_ANON maps the memory without a backing device. This means it allocates the VA space right away, but lazily allocates
the backing pages as the pages are dirtied.
MAP_GUARD creates the VA mapping, but never maps any pages to those pages (well, until it's remapped). Any read/write/exec access to MAP_GUARD pages results in a SIGSEGV.
Quoted 4 lines
I ask since FreeBSD doesn't have
MAP_NORESERVE, which Linux uses to avoid allocation of
gigabytes
Yea. It sounds like MAP_NORESERVE is what FreeBSD's default behavior is: We don't
allocate backing store in the swap areas until there's memory pressure. You can safely
allocate GB of space with MAP_ANON and get similar behavior to the MAP_NORESERVE.
MAP_GUARD could be used if you wanted to reserve the VA space, but didn't want to assign
anything to the VA space until later.
As a practical matter, they both consume about the same resources until the MAP_ANON
region starts to get populated with data...
With PROT_NONE, I think they would have the same effect. If it is to be a backing store for
something like malloc, then MAP_ANON would be best. If you are replacing it with a lot of
things, like a mix of files, devices and/or anon memory, then MAP_GUARD and replace it
with MAP_FIXED later. Most likely you'll want MAP_GUARD to reserve the area, and thenMAP_FIXED to use it for mmap'd memory, shared memory, executable pages, etc.
Does that tell you what you need to know?
Warner
Quoted 2 lines
r~
Re: [PATCH 21/22] Implement shmat(2) and shmdt(2)
Quoted 10 lines · Warner Losh
As a practical matter, they both consume about the same resources until the
MAP_ANONregion starts to get populated with data...With
PROT_NONE, I think they would have the same effect. If it is to be a backing store for something like malloc, thenMAP_ANONwould be best. If you are replacing it with a lot of things, like a mix of files, devices and/or anon memory, thenMAP_GUARDand replace it withMAP_FIXEDlater. Most likely you'll wantMAP_GUARDto reserve the area, and thenMAP_FIXEDto use it for mmap'd memory, shared memory, executable pages, etc.Does that tell you what you need to know?
Yes.
The reserved_va area is replaced with a mix of files, anon, etc, based on whatever the guest requires. So it might be reasonable to adjust bsd-user/mmap.c to use MAP_GUARD for managing the reserved_va area instead of MAP_ANON. No rush, of course.
r~
PATCH 22/22 Add stubs for vadvise(), sbrk() and sstk()
In reply to Karim TahaFrom: Stacey Son <[email protected]>
The above system calls are not supported by qemu.
--- bsd-user/bsd-mem.h | 21 +++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 12 ++++++++++++ 2 files changed, 33 insertions(+)
bsd-user/bsd-mem.h +21 -0
@@ -407,4 +407,25 @@ static inline abi_long do_bsd_shmdt(abi_ulong shmaddr) return get_errno(shmdt(g2h_untagged(shmaddr))); } +static inline abi_long do_bsd_vadvise (void ) +{ +/* See sys_ovadvise() in vm_unix.c */ +qemu_log ("qemu: Unsupported syscall vadvise()\n" ); +return -TARGET_ENOSYS ; +} + +static inline abi_long do_bsd_sbrk (void ) +{ +/* see sys_sbrk() in vm_mmap.c */ +qemu_log ("qemu: Unsupported syscall sbrk()\n" ); +return -TARGET_ENOSYS ; +} + +static inline abi_long do_bsd_sstk (void ) +{ +/* see sys_sstk() in vm_mmap.c */ +qemu_log ("qemu: Unsupported syscall sstk()\n" ); +return -TARGET_ENOSYS ; +} +#endif /* BSD_USER_BSD_MEM_H */
bsd-user/freebsd/os-syscall.c +12 -0
@@ -567,6 +567,18 @@ static abi_long freebsd_syscall(void *cpu_env, int num, abi_long arg1, ret =do_bsd_shmdt (arg1);break ; +case TARGET_FREEBSD_NR_freebsd11_vadvise: + ret =do_bsd_vadvise (); +break ; + +case TARGET_FREEBSD_NR_sbrk: + ret =do_bsd_sbrk (); +break ; + +case TARGET_FREEBSD_NR_sstk: + ret =do_bsd_sstk (); +break ; +/* * Misc */
Signature · 1 line
2.40.0
Re: [PATCH 22/22] Add stubs for vadvise(), sbrk() and sstk()
Quoted 11 lines · Karim Taha
From: Stacey Son <[email protected]>
The above system calls are not supported by qemu.
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 21 +++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 12 ++++++++++++ 2 files changed, 33 insertions(+)
Re: [PATCH 22/22] Add stubs for vadvise(), sbrk() and sstk()
In reply to Karim TahaQuoted 16 lines · Karim Taha
From: Stacey Son <[email protected]>
The above system calls are not supported by qemu.
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 21 +++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 12 ++++++++++++ 2 files changed, 33 insertions(+)bsd-user/bsd-mem.h +0
-0
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmdt(abi_ulong shmaddr) · line 415
Quoted 9 lines
bsd-user/bsd-mem.h +6
-0@@ -407,4 +407,25 @@ static inline abi_long do_bsd_shmdt(abi_ulong shmaddr)return get_errno (shmdt (g2h_untagged (shmaddr))); } +static inline abi_long do_bsd_vadvise (void ) +{ +/* See sys_ovadvise() in vm_unix.c */ +qemu_log ("qemu: Unsupported syscall vadvise()\n" ); +return -TARGET_ENOSYS ; +}
I see EINVAL not ENOSYS.
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmdt(abi_ulong shmaddr) · line 429
Quoted 13 lines
bsd-user/bsd-mem.h +13
-0+static inline abi_long do_bsd_sbrk (void ) +{ +/* see sys_sbrk() in vm_mmap.c */ +qemu_log ("qemu: Unsupported syscall sbrk()\n" ); +return -TARGET_ENOSYS ; +} + +static inline abi_long do_bsd_sstk (void ) +{ +/* see sys_sstk() in vm_mmap.c */ +qemu_log ("qemu: Unsupported syscall sstk()\n" ); +return -TARGET_ENOSYS ; +}
I see EOPNOTSUPP not ENOSYS.
I don't see any point in logging these.
r~
Re: [PATCH 22/22] Add stubs for vadvise(), sbrk() and sstk()
Quoted 21 lines · Richard Henderson
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmdt(abi_ulong
Quoted 16 lines · Karim Taha
From: Stacey Son <[email protected]>
The above system calls are not supported by qemu.
Signed-off-by Stacey Son <[email protected]>Signed-off-by Karim Taha <[email protected]>--- bsd-user/bsd-mem.h | 21 +++++++++++++++++++++ bsd-user/freebsd/os-syscall.c | 12 ++++++++++++ 2 files changed, 33 insertions(+)bsd-user/bsd-mem.h +0
-0@@ -407,4 +407,25 @@ static inline abi_long do_bsd_shmdt(abi_ulongshmaddr)
Quoted 3 lines
bsd-user/bsd-mem.h +0
-0return get_errno (shmdt (g2h_untagged (shmaddr))); }
Quoted 9 lines
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmdt(abi_ulong shmaddr) · line 415
Quoted 6 lines
bsd-user/bsd-mem.h +6
-0+static inline abi_long do_bsd_vadvise (void ) +{ +/* See sys_ovadvise() in vm_unix.c */ +qemu_log ("qemu: Unsupported syscall vadvise()\n" ); +return -TARGET_ENOSYS ; +}I see EINVAL not ENOSYS.
When the system call isn't present at all, it's ENOSYS + SIGSYS (I have patches to implement this
that developers love, but users hate it since many programs cope OK when an error is returned for
obscure system calls). When it is present, it's implemented as EINVAL. So maybe the right thing
here is to remove the log and just return EINVAL for the implementation.
Quoted 7 lines
Quoted 7 lines
bsd-user/bsd-mem.h +7
-0+static inline abi_long do_bsd_sbrk (void ) +{ +/* see sys_sbrk() in vm_mmap.c */ +qemu_log ("qemu: Unsupported syscall sbrk()\n" ); +return -TARGET_ENOSYS ; +} +
Quoted 9 lines
bsd-user/bsd-mem.h static inline abi_long do_bsd_shmdt(abi_ulong shmaddr) · line 429
Quoted 6 lines
bsd-user/bsd-mem.h +6
-0+static inline abi_long do_bsd_sstk (void ) +{ +/* see sys_sstk() in vm_mmap.c */ +qemu_log ("qemu: Unsupported syscall sstk()\n" ); +return -TARGET_ENOSYS ; +}I see EOPNOTSUPP not ENOSYS.
Same comment as above: we should just return EOPNOSUPP and call it implemented.
Quoted 2 lines
I don't see any point in logging these.
Yea, that's a general pattern that we have upstream, but there's a catch-all 'default' case that does
the logging with the right mask.
Warner