* [PATCH v2 1/2] vsock: Refactor vsock_*_getsockopt to resemble sock_getsockopt @ 2021-10-07 12:31 Richard Palethorpe 2021-10-07 12:31 ` [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout Richard Palethorpe 2021-10-07 12:51 ` [PATCH v2 1/2] vsock: Refactor vsock_*_getsockopt to resemble sock_getsockopt Arnd Bergmann 0 siblings, 2 replies; 6+ messages in thread From: Richard Palethorpe @ 2021-10-07 12:31 UTC (permalink / raw) To: Arnd Bergmann Cc: Richard Palethorpe, David S. Miller, Jakub Kicinski, Stefano Garzarella, Andra Paraschiv, Eric Dumazet, Arseny Krasnov, Willem de Bruijn, Deepa Dinamani, netdev, linux-kernel, Richard Palethorpe In preparation for sharing the implementation of sock_get_timeout. Signed-off-by: Richard Palethorpe <rpalethorpe@suse.com> Cc: Richard Palethorpe <rpalethorpe@richiejp.com> --- V1: https://lore.kernel.org/netdev/20211006074547.14724-1-rpalethorpe@suse.com/ V2: * Try to share more code with core/sock.c * Handle 64-bit timeout values in 32-bit net/vmw_vsock/af_vsock.c | 65 +++++++++++++++++----------------------- 1 file changed, 28 insertions(+), 37 deletions(-) diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 3e02cc3b24f8..97d56f6a4480 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -1648,68 +1648,59 @@ static int vsock_connectible_getsockopt(struct socket *sock, char __user *optval, int __user *optlen) { - int err; + struct sock *sk = sock->sk; + struct vsock_sock *vsk = vsock_sk(sk); + + union { + u64 val64; + struct __kernel_old_timeval tm; + } v; + + int lv = sizeof(v.val64); int len; - struct sock *sk; - struct vsock_sock *vsk; - u64 val; if (level != AF_VSOCK) return -ENOPROTOOPT; - err = get_user(len, optlen); - if (err != 0) - return err; - -#define COPY_OUT(_v) \ - do { \ - if (len < sizeof(_v)) \ - return -EINVAL; \ - \ - len = sizeof(_v); \ - if (copy_to_user(optval, &_v, len) != 0) \ - return -EFAULT; \ - \ - } while (0) + if (get_user(len, optlen)) + return -EFAULT; - err = 0; - sk = sock->sk; - vsk = vsock_sk(sk); + memset(&v, 0, sizeof(v)); switch (optname) { case SO_VM_SOCKETS_BUFFER_SIZE: - val = vsk->buffer_size; - COPY_OUT(val); + v.val64 = vsk->buffer_size; break; case SO_VM_SOCKETS_BUFFER_MAX_SIZE: - val = vsk->buffer_max_size; - COPY_OUT(val); + v.val64 = vsk->buffer_max_size; break; case SO_VM_SOCKETS_BUFFER_MIN_SIZE: - val = vsk->buffer_min_size; - COPY_OUT(val); + v.val64 = vsk->buffer_min_size; break; - case SO_VM_SOCKETS_CONNECT_TIMEOUT: { - struct __kernel_old_timeval tv; - tv.tv_sec = vsk->connect_timeout / HZ; - tv.tv_usec = + case SO_VM_SOCKETS_CONNECT_TIMEOUT: + lv = sizeof(v.tm); + v.tm.tv_sec = vsk->connect_timeout / HZ; + v.tm.tv_usec = (vsk->connect_timeout - - tv.tv_sec * HZ) * (1000000 / HZ); - COPY_OUT(tv); + v.tm.tv_sec * HZ) * (1000000 / HZ); break; - } + default: return -ENOPROTOOPT; } - err = put_user(len, optlen); - if (err != 0) + if (len < lv) + return -EINVAL; + if (len > lv) + len = lv; + if (copy_to_user(optval, &v, len)) return -EFAULT; -#undef COPY_OUT + if (put_user(len, optlen)) + return -EFAULT; return 0; } -- 2.33.0 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout 2021-10-07 12:31 [PATCH v2 1/2] vsock: Refactor vsock_*_getsockopt to resemble sock_getsockopt Richard Palethorpe @ 2021-10-07 12:31 ` Richard Palethorpe 2021-10-07 12:53 ` Arnd Bergmann 2021-10-07 16:04 ` Jakub Kicinski 2021-10-07 12:51 ` [PATCH v2 1/2] vsock: Refactor vsock_*_getsockopt to resemble sock_getsockopt Arnd Bergmann 1 sibling, 2 replies; 6+ messages in thread From: Richard Palethorpe @ 2021-10-07 12:31 UTC (permalink / raw) To: Arnd Bergmann Cc: Richard Palethorpe, David S. Miller, Jakub Kicinski, Stefano Garzarella, Andra Paraschiv, Eric Dumazet, Arseny Krasnov, Willem de Bruijn, Deepa Dinamani, netdev, linux-kernel, Richard Palethorpe Reuse the timeval compat code from core/sock to handle 32-bit and 64-bit timeval structures. Also introduce a new socket option define to allow using y2038 safe timeval under 32-bit. The existing behavior of sock_set_timeout and vsock's timeout setter differ when the time value is out of bounds. vsocks current behavior is retained at the expense of not being able to share the full implementation. This allows the LTP test vsock01 to pass under 32-bit compat mode. Fixes: fe0c72f3db11 ("socket: move compat timeout handling into sock.c") Signed-off-by: Richard Palethorpe <rpalethorpe@suse.com> Cc: Richard Palethorpe <rpalethorpe@richiejp.com> --- include/net/sock.h | 4 ++++ include/uapi/linux/vm_sockets.h | 13 +++++++++++- net/core/sock.c | 35 ++++++++++++++++++++++----------- net/vmw_vsock/af_vsock.c | 23 +++++++++++++--------- 4 files changed, 54 insertions(+), 21 deletions(-) diff --git a/include/net/sock.h b/include/net/sock.h index f23cb259b0e2..b93d0ac552c1 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2773,4 +2773,8 @@ void sock_set_sndtimeo(struct sock *sk, s64 secs); int sock_bind_add(struct sock *sk, struct sockaddr *addr, int addr_len); +int sock_get_timeout(long timeo, void *optval, bool old_timeval); +int sock_copy_user_timeval(struct __kernel_sock_timeval *tv, + sockptr_t optval, int optlen, bool old_timeval); + #endif /* _SOCK_H */ diff --git a/include/uapi/linux/vm_sockets.h b/include/uapi/linux/vm_sockets.h index 46918a1852d7..c60ca33eac59 100644 --- a/include/uapi/linux/vm_sockets.h +++ b/include/uapi/linux/vm_sockets.h @@ -64,7 +64,7 @@ * timeout for a STREAM socket. */ -#define SO_VM_SOCKETS_CONNECT_TIMEOUT 6 +#define SO_VM_SOCKETS_CONNECT_TIMEOUT_OLD 6 /* Option name for using non-blocking send/receive. Use as the option name * for setsockopt(3) or getsockopt(3) to set or get the non-blocking @@ -81,6 +81,17 @@ #define SO_VM_SOCKETS_NONBLOCK_TXRX 7 +#define SO_VM_SOCKETS_CONNECT_TIMEOUT_NEW 8 + +#if !defined(__KERNEL__) +#if __BITS_PER_LONG == 64 || (defined(__x86_64__) && defined(__ILP32__)) +#define SO_VM_SOCKETS_CONNECT_TIMEOUT SO_VM_SOCKETS_CONNECT_TIMEOUT_OLD +#else +#define SO_VM_SOCKETS_CONNECT_TIMEOUT \ + (sizeof(time_t) == sizeof(__kernel_long_t) ? SO_VM_SOCKETS_CONNECT_TIMEOUT_OLD : SO_VM_SOCKETS_CONNECT_TIMEOUT_NEW) +#endif +#endif + /* The vSocket equivalent of INADDR_ANY. This works for the svm_cid field of * sockaddr_vm and indicates the context ID of the current endpoint. */ diff --git a/net/core/sock.c b/net/core/sock.c index a3eea6e0b30a..a648fd3cc2ec 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -349,7 +349,7 @@ void sk_error_report(struct sock *sk) } EXPORT_SYMBOL(sk_error_report); -static int sock_get_timeout(long timeo, void *optval, bool old_timeval) +int sock_get_timeout(long timeo, void *optval, bool old_timeval) { struct __kernel_sock_timeval tv; @@ -378,12 +378,11 @@ static int sock_get_timeout(long timeo, void *optval, bool old_timeval) *(struct __kernel_sock_timeval *)optval = tv; return sizeof(tv); } +EXPORT_SYMBOL(sock_get_timeout); -static int sock_set_timeout(long *timeo_p, sockptr_t optval, int optlen, - bool old_timeval) +int sock_copy_user_timeval(struct __kernel_sock_timeval *tv, + sockptr_t optval, int optlen, bool old_timeval) { - struct __kernel_sock_timeval tv; - if (old_timeval && in_compat_syscall() && !COMPAT_USE_64BIT_TIME) { struct old_timeval32 tv32; @@ -392,8 +391,8 @@ static int sock_set_timeout(long *timeo_p, sockptr_t optval, int optlen, if (copy_from_sockptr(&tv32, optval, sizeof(tv32))) return -EFAULT; - tv.tv_sec = tv32.tv_sec; - tv.tv_usec = tv32.tv_usec; + tv->tv_sec = tv32.tv_sec; + tv->tv_usec = tv32.tv_usec; } else if (old_timeval) { struct __kernel_old_timeval old_tv; @@ -401,14 +400,28 @@ static int sock_set_timeout(long *timeo_p, sockptr_t optval, int optlen, return -EINVAL; if (copy_from_sockptr(&old_tv, optval, sizeof(old_tv))) return -EFAULT; - tv.tv_sec = old_tv.tv_sec; - tv.tv_usec = old_tv.tv_usec; + tv->tv_sec = old_tv.tv_sec; + tv->tv_usec = old_tv.tv_usec; } else { - if (optlen < sizeof(tv)) + if (optlen < sizeof(*tv)) return -EINVAL; - if (copy_from_sockptr(&tv, optval, sizeof(tv))) + if (copy_from_sockptr(tv, optval, sizeof(*tv))) return -EFAULT; } + + return 0; +} +EXPORT_SYMBOL(sock_copy_user_timeval); + +static int sock_set_timeout(long *timeo_p, sockptr_t optval, int optlen, + bool old_timeval) +{ + struct __kernel_sock_timeval tv; + int err = sock_copy_user_timeval(&tv, optval, optlen, old_timeval); + + if (err) + return err; + if (tv.tv_usec < 0 || tv.tv_usec >= USEC_PER_SEC) return -EDOM; diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 97d56f6a4480..1fa27cea5b2a 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -1614,9 +1614,14 @@ static int vsock_connectible_setsockopt(struct socket *sock, vsock_update_buffer_size(vsk, transport, vsk->buffer_size); break; - case SO_VM_SOCKETS_CONNECT_TIMEOUT: { - struct __kernel_old_timeval tv; - COPY_IN(tv); + case SO_VM_SOCKETS_CONNECT_TIMEOUT_NEW: + case SO_VM_SOCKETS_CONNECT_TIMEOUT_OLD: { + struct __kernel_sock_timeval tv; + + err = sock_copy_user_timeval(&tv, optval, optlen, + optname == SO_VM_SOCKETS_CONNECT_TIMEOUT_OLD); + if (err) + break; if (tv.tv_sec >= 0 && tv.tv_usec < USEC_PER_SEC && tv.tv_sec < (MAX_SCHEDULE_TIMEOUT / HZ - 1)) { vsk->connect_timeout = tv.tv_sec * HZ + @@ -1653,7 +1658,9 @@ static int vsock_connectible_getsockopt(struct socket *sock, union { u64 val64; + struct old_timeval32 tm32; struct __kernel_old_timeval tm; + struct __kernel_sock_timeval stm; } v; int lv = sizeof(v.val64); @@ -1680,12 +1687,10 @@ static int vsock_connectible_getsockopt(struct socket *sock, v.val64 = vsk->buffer_min_size; break; - case SO_VM_SOCKETS_CONNECT_TIMEOUT: - lv = sizeof(v.tm); - v.tm.tv_sec = vsk->connect_timeout / HZ; - v.tm.tv_usec = - (vsk->connect_timeout - - v.tm.tv_sec * HZ) * (1000000 / HZ); + case SO_VM_SOCKETS_CONNECT_TIMEOUT_NEW: + case SO_VM_SOCKETS_CONNECT_TIMEOUT_OLD: + lv = sock_get_timeout(vsk->connect_timeout, &v, + optname == SO_VM_SOCKETS_CONNECT_TIMEOUT_OLD); break; default: -- 2.33.0 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout 2021-10-07 12:31 ` [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout Richard Palethorpe @ 2021-10-07 12:53 ` Arnd Bergmann 2021-10-07 16:04 ` Jakub Kicinski 1 sibling, 0 replies; 6+ messages in thread From: Arnd Bergmann @ 2021-10-07 12:53 UTC (permalink / raw) To: Richard Palethorpe Cc: Arnd Bergmann, David S. Miller, Jakub Kicinski, Stefano Garzarella, Andra Paraschiv, Eric Dumazet, Arseny Krasnov, Willem de Bruijn, Deepa Dinamani, Networking, Linux Kernel Mailing List, Richard Palethorpe On Thu, Oct 7, 2021 at 2:34 PM Richard Palethorpe <rpalethorpe@suse.com> wrote: > > Reuse the timeval compat code from core/sock to handle 32-bit and > 64-bit timeval structures. Also introduce a new socket option define > to allow using y2038 safe timeval under 32-bit. > > The existing behavior of sock_set_timeout and vsock's timeout setter > differ when the time value is out of bounds. vsocks current behavior > is retained at the expense of not being able to share the full > implementation. > > This allows the LTP test vsock01 to pass under 32-bit compat mode. > > Fixes: fe0c72f3db11 ("socket: move compat timeout handling into sock.c") > Signed-off-by: Richard Palethorpe <rpalethorpe@suse.com> > Cc: Richard Palethorpe <rpalethorpe@richiejp.com> I had a look to make sure you've covered all the tricky corner cases, Reviewed-by: Arnd Bergmann <arnd@arndb.de> ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout 2021-10-07 12:31 ` [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout Richard Palethorpe 2021-10-07 12:53 ` Arnd Bergmann @ 2021-10-07 16:04 ` Jakub Kicinski 2021-10-07 17:14 ` Richard Palethorpe 1 sibling, 1 reply; 6+ messages in thread From: Jakub Kicinski @ 2021-10-07 16:04 UTC (permalink / raw) To: Richard Palethorpe Cc: Arnd Bergmann, David S. Miller, Stefano Garzarella, Andra Paraschiv, Eric Dumazet, Arseny Krasnov, Willem de Bruijn, Deepa Dinamani, netdev, linux-kernel, Richard Palethorpe On Thu, 7 Oct 2021 13:31:47 +0100 Richard Palethorpe wrote: > Reuse the timeval compat code from core/sock to handle 32-bit and > 64-bit timeval structures. Also introduce a new socket option define > to allow using y2038 safe timeval under 32-bit. > > The existing behavior of sock_set_timeout and vsock's timeout setter > differ when the time value is out of bounds. vsocks current behavior > is retained at the expense of not being able to share the full > implementation. > > This allows the LTP test vsock01 to pass under 32-bit compat mode. > > Fixes: fe0c72f3db11 ("socket: move compat timeout handling into sock.c") > Signed-off-by: Richard Palethorpe <rpalethorpe@suse.com> > Cc: Richard Palethorpe <rpalethorpe@richiejp.com> This breaks 32bit x86 build: ERROR: modpost: "__divdi3" [net/vmw_vsock/vsock.ko] undefined! If the 64 bit division is intention you need to use an appropriate helper. ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout 2021-10-07 16:04 ` Jakub Kicinski @ 2021-10-07 17:14 ` Richard Palethorpe 0 siblings, 0 replies; 6+ messages in thread From: Richard Palethorpe @ 2021-10-07 17:14 UTC (permalink / raw) To: Jakub Kicinski Cc: Arnd Bergmann, David S. Miller, Stefano Garzarella, Andra Paraschiv, Eric Dumazet, Arseny Krasnov, Willem de Bruijn, Deepa Dinamani, netdev, linux-kernel, Richard Palethorpe Hello Jakub, Jakub Kicinski <kuba@kernel.org> writes: > On Thu, 7 Oct 2021 13:31:47 +0100 Richard Palethorpe wrote: >> Reuse the timeval compat code from core/sock to handle 32-bit and >> 64-bit timeval structures. Also introduce a new socket option define >> to allow using y2038 safe timeval under 32-bit. >> >> The existing behavior of sock_set_timeout and vsock's timeout setter >> differ when the time value is out of bounds. vsocks current behavior >> is retained at the expense of not being able to share the full >> implementation. >> >> This allows the LTP test vsock01 to pass under 32-bit compat mode. >> >> Fixes: fe0c72f3db11 ("socket: move compat timeout handling into sock.c") >> Signed-off-by: Richard Palethorpe <rpalethorpe@suse.com> >> Cc: Richard Palethorpe <rpalethorpe@richiejp.com> > > This breaks 32bit x86 build: > > ERROR: modpost: "__divdi3" [net/vmw_vsock/vsock.ko] undefined! > > If the 64 bit division is intention you need to use an appropriate > helper. Ah, sorry, that's why sock.c casts usecs to unsigned long. I will do some more testing a reroll. -- Thank you, Richard. ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH v2 1/2] vsock: Refactor vsock_*_getsockopt to resemble sock_getsockopt 2021-10-07 12:31 [PATCH v2 1/2] vsock: Refactor vsock_*_getsockopt to resemble sock_getsockopt Richard Palethorpe 2021-10-07 12:31 ` [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout Richard Palethorpe @ 2021-10-07 12:51 ` Arnd Bergmann 1 sibling, 0 replies; 6+ messages in thread From: Arnd Bergmann @ 2021-10-07 12:51 UTC (permalink / raw) To: Richard Palethorpe Cc: Arnd Bergmann, David S. Miller, Jakub Kicinski, Stefano Garzarella, Andra Paraschiv, Eric Dumazet, Arseny Krasnov, Willem de Bruijn, Deepa Dinamani, Networking, Linux Kernel Mailing List, Richard Palethorpe On Thu, Oct 7, 2021 at 2:33 PM Richard Palethorpe <rpalethorpe@suse.com> wrote: > > In preparation for sharing the implementation of sock_get_timeout. > > Signed-off-by: Richard Palethorpe <rpalethorpe@suse.com> > Cc: Richard Palethorpe <rpalethorpe@richiejp.com> Looks good to me, both as a standalone cleanup and as a preparation for patch 2. Reviewed-by: Arnd Bergmann <arnd@arndb.de> ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2021-10-07 18:07 UTC | newest] Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-10-07 12:31 [PATCH v2 1/2] vsock: Refactor vsock_*_getsockopt to resemble sock_getsockopt Richard Palethorpe 2021-10-07 12:31 ` [PATCH v2 2/2] vsock: Enable y2038 safe timeval for timeout Richard Palethorpe 2021-10-07 12:53 ` Arnd Bergmann 2021-10-07 16:04 ` Jakub Kicinski 2021-10-07 17:14 ` Richard Palethorpe 2021-10-07 12:51 ` [PATCH v2 1/2] vsock: Refactor vsock_*_getsockopt to resemble sock_getsockopt Arnd Bergmann
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).