aboutsummaryrefslogtreecommitdiff
path: root/gdb/ser-tcp.c
diff options
context:
space:
mode:
authorSergio Durigan Junior <sergiodj@redhat.com>2018-05-18 01:29:24 -0400
committerSergio Durigan Junior <sergiodj@redhat.com>2018-07-11 19:41:31 -0400
commitc7ab0aef11d91b637bf091aa9176b8dc4aadee46 (patch)
tree8dd380d93603e4601912dbf08d8ce861098fadc3 /gdb/ser-tcp.c
parent4c7333b308f5178813745f40e641231efb1cb763 (diff)
downloadfsf-binutils-gdb-c7ab0aef11d91b637bf091aa9176b8dc4aadee46.zip
fsf-binutils-gdb-c7ab0aef11d91b637bf091aa9176b8dc4aadee46.tar.gz
fsf-binutils-gdb-c7ab0aef11d91b637bf091aa9176b8dc4aadee46.tar.bz2
Implement IPv6 support for GDB/gdbserver
This patch implements IPv6 support for both GDB and gdbserver. Based on my research, it is the fourth attempt to do that since 2006. Since I used ideas from all of the previous patches, I also added their authors's names on the ChangeLogs as a way to recognize their efforts. For reference sake, you can find the previous attempts at: https://sourceware.org/ml/gdb-patches/2006-09/msg00192.html https://sourceware.org/ml/gdb-patches/2014-02/msg00248.html https://sourceware.org/ml/gdb-patches/2016-02/msg00226.html The basic idea behind the patch is to start using the new 'getaddrinfo'/'getnameinfo' calls, which are responsible for translating names and addresses in a protocol-independent way. This means that if we ever have a new version of the IP protocol, we won't need to change the code again (or, at least, won't have to change the majority of the code). The function 'getaddrinfo' returns a linked list of possible addresses to connect to. Dealing with multiple addresses proved to be a hard task with the current TCP auto-retry mechanism implemented on ser-tcp:net_open. For example, when gdbserver listened only on an IPv4 socket: $ ./gdbserver --once 127.0.0.1:1234 ./a.out and GDB was instructed to try to connect to both IPv6 and IPv4 sockets: $ ./gdb -ex 'target extended-remote localhost:1234' ./a.out the user would notice a somewhat big delay before GDB was able to connect to the IPv4 socket. This happened because GDB was trying to connect to the IPv6 socket first, and had to wait until the connection timed out before it tried to connect to the IPv4 socket. For that reason, I had to rewrite the main loop and implement a new method for handling multiple connections. After some discussion, Pedro and I agreed on the following algorithm: 1) For each entry returned by 'getaddrinfo', we try to open a socket and connect to it. 2.a) If we have a successful 'connect', we just use that connection. 2.b) If we don't have a successfull 'connect', but if we've got a ECONNREFUSED (meaning the the connection was refused), we keep track of this fact by using a flag. 2.c) If we don't have a successfull 'connect', but if we've got a EINPROGRESS (meaning that the connection is in progress), we perform a 'select' call on the socket until we have a result (either a successful connection, or an error on the socket). 3) If tcp_auto_retry is true, and we haven't gotten a successful connection, and at least one of our attempts failed with ECONNREFUSED, then we wait a little bit (i.e., call 'wait_for_connect'), check to see if there was a timeout/interruption (in which case we bail out), and then go back to (1). After multiple tests, I was able to connect without delay on the scenario described above, and was also able to connect in all other types of scenarios. I also implemented some hostname parsing functions (along with their corresponding unit tests) which are used to help GDB and gdbserver to parse hostname strings provided by the user. These new functions are living inside common/netstuff.[ch]. I've had to do that since IPv6 introduces a new URL scheme, which defines that square brackets can be used to enclose the host part and differentiate it from the port (e.g., "[::1]:1234" means "host ::1, port 1234"). I spent some time thinking about a reasonable way to interpret what the user wants, and I came up with the following: - If the user has provided a prefix that doesn't specify the protocol version (i.e., "tcp:" or "udp:"), or if the user has not provided any prefix, don't make any assumptions (i.e., assume AF_UNSPEC when dealing with 'getaddrinfo') *unless* the host starts with "[" (in which case, assume it's an IPv6 host). - If the user has provided a prefix that does specify the protocol version (i.e., "tcp4:", "tcp6:", "udp4:" or "udp6:"), then respect that. This method doesn't follow strictly what RFC 2732 proposes (that literal IPv6 addresses should be provided enclosed in "[" and "]") because IPv6 addresses still can be provided without square brackets in our case, but since we have prefixes to specify protocol versions I think this is not an issue. Another thing worth mentioning is the new 'GDB_TEST_SOCKETHOST' testcase parameter, which makes it possible to specify the hostname (without the port) to be used when testing GDB and gdbserver. For example, to run IPv6 tests: $ make check-gdb RUNTESTFLAGS='GDB_TEST_SOCKETHOST=tcp6:[::1]' Or, to run IPv4 tests: $ make check-gdb RUNTESTFLAGS='GDB_TEST_SOCKETHOST=tcp4:127.0.0.1' This required a few changes on the gdbserver-base.exp, and also a minimal adjustment on gdb.server/run-without-local-binary.exp. Finally, I've implemented a new testcase, gdb.server/server-connect.exp, which is supposed to run on the native host and perform various "smoke tests" using different connection methods. This patch has been regression-tested on BuildBot and locally, and also built using a x86_64-w64-mingw32 GCC, and no problems were found. gdb/ChangeLog: 2018-07-11 Sergio Durigan Junior <sergiodj@redhat.com> Jan Kratochvil <jan.kratochvil@redhat.com> Paul Fertser <fercerpav@gmail.com> Tsutomu Seki <sekiriki@gmail.com> Pedro Alves <palves@redhat.com> * Makefile.in (SUBDIR_UNITTESTS_SRCS): Add 'unittests/parse-connection-spec-selftests.c'. (COMMON_SFILES): Add 'common/netstuff.c'. (HFILES_NO_SRCDIR): Add 'common/netstuff.h'. * NEWS (Changes since GDB 8.2): Mention IPv6 support. * common/netstuff.c: New file. * common/netstuff.h: New file. * ser-tcp.c: Include 'netstuff.h' and 'wspiapi.h'. (wait_for_connect): Update comment. New parameter 'gdb::optional<int> sock' instead of 'struct serial *scb'. Use 'sock' directly instead of 'scb->fd'. (try_connect): New function, with code from 'net_open'. (net_open): Rewrite main loop to deal with multiple sockets/addresses. Handle IPv6-style hostnames; implement support for IPv6 connections. * unittests/parse-connection-spec-selftests.c: New file. gdb/gdbserver/ChangeLog: 2018-07-11 Sergio Durigan Junior <sergiodj@redhat.com> Jan Kratochvil <jan.kratochvil@redhat.com> Paul Fertser <fercerpav@gmail.com> Tsutomu Seki <sekiriki@gmail.com> * Makefile.in (SFILES): Add '$(srcdir)/common/netstuff.c'. (OBS): Add 'common/netstuff.o'. (GDBREPLAY_OBS): Likewise. * gdbreplay.c: Include 'wspiapi.h' and 'netstuff.h'. (remote_open): Implement support for IPv6 connections. * remote-utils.c: Include 'netstuff.h', 'filestuff.h' and 'wspiapi.h'. (handle_accept_event): Accept connections from IPv6 sources. (remote_prepare): Handle IPv6-style hostnames; implement support for IPv6 connections. (remote_open): Implement support for printing connections from IPv6 sources. gdb/testsuite/ChangeLog: 2018-07-11 Sergio Durigan Junior <sergiodj@redhat.com> Jan Kratochvil <jan.kratochvil@redhat.com> Paul Fertser <fercerpav@gmail.com> Tsutomu Seki <sekiriki@gmail.com> * README (Testsuite Parameters): Mention new 'GDB_TEST_SOCKETHOST' parameter. * boards/native-extended-gdbserver.exp: Do not set 'sockethost' by default. * boards/native-gdbserver.exp: Likewise. * gdb.server/run-without-local-binary.exp: Improve regexp used for detecting when a remote debugging connection succeeds. * gdb.server/server-connect.exp: New file. * lib/gdbserver-support.exp (gdbserver_default_get_comm_port): Do not prefix the port number with ":". (gdbserver_start): New global GDB_TEST_SOCKETHOST. Implement support for detecting and using it. Add '$debughost_gdbserver' to the list of arguments used to start gdbserver. Handle case when gdbserver cannot resolve a network name. gdb/doc/ChangeLog: 2018-07-11 Sergio Durigan Junior <sergiodj@redhat.com> Jan Kratochvil <jan.kratochvil@redhat.com> Paul Fertser <fercerpav@gmail.com> Tsutomu Seki <sekiriki@gmail.com> * gdb.texinfo (Remote Connection Commands): Add explanation about new IPv6 support. Add new connection prefixes.
Diffstat (limited to 'gdb/ser-tcp.c')
-rw-r--r--gdb/ser-tcp.c296
1 files changed, 176 insertions, 120 deletions
diff --git a/gdb/ser-tcp.c b/gdb/ser-tcp.c
index 23ef3b0..8f165bc 100644
--- a/gdb/ser-tcp.c
+++ b/gdb/ser-tcp.c
@@ -25,6 +25,7 @@
#include "cli/cli-decode.h"
#include "cli/cli-setshow.h"
#include "filestuff.h"
+#include "netstuff.h"
#include <sys/types.h>
@@ -39,6 +40,7 @@
#ifdef USE_WIN32API
#include <winsock2.h>
+#include <wspiapi.h>
#ifndef ETIMEDOUT
#define ETIMEDOUT WSAETIMEDOUT
#endif
@@ -81,12 +83,13 @@ static unsigned int tcp_retry_limit = 15;
#define POLL_INTERVAL 5
-/* Helper function to wait a while. If SCB is non-null, wait on its
- file descriptor. Otherwise just wait on a timeout, updating *POLLS.
- Returns -1 on timeout or interrupt, otherwise the value of select. */
+/* Helper function to wait a while. If SOCK is not -1, wait on its
+ file descriptor. Otherwise just wait on a timeout, updating
+ *POLLS. Returns -1 on timeout or interrupt, otherwise the value of
+ select. */
static int
-wait_for_connect (struct serial *scb, unsigned int *polls)
+wait_for_connect (int sock, unsigned int *polls)
{
struct timeval t;
int n;
@@ -120,24 +123,24 @@ wait_for_connect (struct serial *scb, unsigned int *polls)
t.tv_usec = 0;
}
- if (scb)
+ if (sock >= 0)
{
fd_set rset, wset, eset;
FD_ZERO (&rset);
- FD_SET (scb->fd, &rset);
+ FD_SET (sock, &rset);
wset = rset;
eset = rset;
-
+
/* POSIX systems return connection success or failure by signalling
wset. Windows systems return success in wset and failure in
eset.
-
+
We must call select here, rather than gdb_select, because
the serial structure has not yet been initialized - the
MinGW select wrapper will not know that this FD refers
to a socket. */
- n = select (scb->fd + 1, &rset, &wset, &eset, &t);
+ n = select (sock + 1, &rset, &wset, &eset, &t);
}
else
/* Use gdb_select here, since we have no file descriptors, and on
@@ -153,80 +156,28 @@ wait_for_connect (struct serial *scb, unsigned int *polls)
return n;
}
-/* Open a tcp socket. */
+/* Try to connect to the host represented by AINFO. If the connection
+ succeeds, return its socket. Otherwise, return -1 and set ERRNO
+ accordingly. POLLS is used when 'connect' returns EINPROGRESS, and
+ we need to invoke 'wait_for_connect' to obtain the status. */
-int
-net_open (struct serial *scb, const char *name)
+static int
+try_connect (const struct addrinfo *ainfo, unsigned int *polls)
{
- char hostname[100];
- const char *port_str;
- int n, port, tmp;
- int use_udp;
- struct hostent *hostent;
- struct sockaddr_in sockaddr;
-#ifdef USE_WIN32API
- u_long ioarg;
-#else
- int ioarg;
-#endif
- unsigned int polls = 0;
-
- use_udp = 0;
- if (startswith (name, "udp:"))
- {
- use_udp = 1;
- name = name + 4;
- }
- else if (startswith (name, "tcp:"))
- name = name + 4;
-
- port_str = strchr (name, ':');
-
- if (!port_str)
- error (_("net_open: No colon in host name!")); /* Shouldn't ever
- happen. */
-
- tmp = std::min (port_str - name, (ptrdiff_t) sizeof hostname - 1);
- strncpy (hostname, name, tmp); /* Don't want colon. */
- hostname[tmp] = '\000'; /* Tie off host name. */
- port = atoi (port_str + 1);
-
- /* Default hostname is localhost. */
- if (!hostname[0])
- strcpy (hostname, "localhost");
-
- hostent = gethostbyname (hostname);
- if (!hostent)
- {
- fprintf_unfiltered (gdb_stderr, "%s: unknown host\n", hostname);
- errno = ENOENT;
- return -1;
- }
+ int sock = gdb_socket_cloexec (ainfo->ai_family, ainfo->ai_socktype,
+ ainfo->ai_protocol);
- sockaddr.sin_family = PF_INET;
- sockaddr.sin_port = htons (port);
- memcpy (&sockaddr.sin_addr.s_addr, hostent->h_addr,
- sizeof (struct in_addr));
-
- retry:
-
- if (use_udp)
- scb->fd = gdb_socket_cloexec (PF_INET, SOCK_DGRAM, 0);
- else
- scb->fd = gdb_socket_cloexec (PF_INET, SOCK_STREAM, 0);
-
- if (scb->fd == -1)
+ if (sock < 0)
return -1;
-
+
/* Set socket nonblocking. */
- ioarg = 1;
- ioctl (scb->fd, FIONBIO, &ioarg);
+ int ioarg = 1;
+
+ ioctl (sock, FIONBIO, &ioarg);
/* Use Non-blocking connect. connect() will return 0 if connected
already. */
- n = connect (scb->fd, (struct sockaddr *) &sockaddr, sizeof (sockaddr));
-
- if (n < 0)
+ if (connect (sock, ainfo->ai_addr, ainfo->ai_addrlen) < 0)
{
#ifdef USE_WIN32API
int err = WSAGetLastError();
@@ -234,21 +185,26 @@ net_open (struct serial *scb, const char *name)
int err = errno;
#endif
- /* Maybe we're waiting for the remote target to become ready to
- accept connections. */
- if (tcp_auto_retry
+ /* If we've got a "connection refused" error, just return
+ -1. The caller will know what to do. */
+ if (
#ifdef USE_WIN32API
- && err == WSAECONNREFUSED
+ err == WSAECONNREFUSED
#else
- && err == ECONNREFUSED
+ err == ECONNREFUSED
#endif
- && wait_for_connect (NULL, &polls) >= 0)
+ )
{
- close (scb->fd);
- goto retry;
+ close (sock);
+ errno = err;
+ return -1;
}
if (
+ /* Any other error (except EINPROGRESS) will be "swallowed"
+ here. We return without specifying a return value, and
+ set errno if the caller wants to inspect what
+ happened. */
#ifdef USE_WIN32API
/* Under Windows, calling "connect" with a non-blocking socket
results in WSAEWOULDBLOCK, not WSAEINPROGRESS. */
@@ -258,66 +214,166 @@ net_open (struct serial *scb, const char *name)
#endif
)
{
+ close (sock);
errno = err;
- net_close (scb);
return -1;
}
/* Looks like we need to wait for the connect. */
- do
- {
- n = wait_for_connect (scb, &polls);
- }
+ int n;
+
+ do
+ n = wait_for_connect (sock, polls);
while (n == 0);
+
if (n < 0)
{
- net_close (scb);
+ int saved_errno = errno;
+
+ /* A negative value here means that we either timed out or
+ got interrupted by the user. Just return. */
+ close (sock);
+ errno = saved_errno;
return -1;
}
}
/* Got something. Is it an error? */
- {
- int res, err;
- socklen_t len;
-
- len = sizeof (err);
- /* On Windows, the fourth parameter to getsockopt is a "char *";
- on UNIX systems it is generally "void *". The cast to "char *"
- is OK everywhere, since in C++ any data pointer type can be
- implicitly converted to "void *". */
- res = getsockopt (scb->fd, SOL_SOCKET, SO_ERROR, (char *) &err, &len);
- if (res < 0 || err)
- {
- /* Maybe the target still isn't ready to accept the connection. */
- if (tcp_auto_retry
+ int err;
+ socklen_t len = sizeof (err);
+
+ /* On Windows, the fourth parameter to getsockopt is a "char *";
+ on UNIX systems it is generally "void *". The cast to "char *"
+ is OK everywhere, since in C++ any data pointer type can be
+ implicitly converted to "void *". */
+ int ret = getsockopt (sock, SOL_SOCKET, SO_ERROR, (char *) &err, &len);
+
+ if (ret < 0)
+ {
+ int saved_errno = errno;
+
+ close (sock);
+ errno = saved_errno;
+ return -1;
+ }
+ else if (ret == 0 && err != 0)
+ {
+ close (sock);
+ errno = err;
+ return -1;
+ }
+
+ /* The connection succeeded. Return the socket. */
+ return sock;
+}
+
+/* Open a tcp socket. */
+
+int
+net_open (struct serial *scb, const char *name)
+{
+ struct addrinfo hint;
+ struct addrinfo *ainfo;
+
+ memset (&hint, 0, sizeof (hint));
+ /* Assume no prefix will be passed, therefore we should use
+ AF_UNSPEC. */
+ hint.ai_family = AF_UNSPEC;
+ hint.ai_socktype = SOCK_STREAM;
+ hint.ai_protocol = IPPROTO_TCP;
+
+ parsed_connection_spec parsed = parse_connection_spec (name, &hint);
+
+ if (parsed.port_str.empty ())
+ error (_("Missing port on hostname '%s'"), name);
+
+ int r = getaddrinfo (parsed.host_str.c_str (),
+ parsed.port_str.c_str (),
+ &hint, &ainfo);
+
+ if (r != 0)
+ {
+ fprintf_unfiltered (gdb_stderr, _("%s: cannot resolve name: %s\n"),
+ name, gai_strerror (r));
+ errno = ENOENT;
+ return -1;
+ }
+
+ scoped_free_addrinfo free_ainfo (ainfo);
+
+ /* Flag to indicate whether we've got a connection refused. It will
+ be true if any of the connections tried was refused. */
+ bool got_connrefused;
+ /* If a connection succeeeds, SUCCESS_AINFO will point to the
+ 'struct addrinfo' that succeed. */
+ struct addrinfo *success_ainfo = NULL;
+ unsigned int polls = 0;
+
+ /* Assume the worst. */
+ scb->fd = -1;
+
+ do
+ {
+ got_connrefused = false;
+
+ for (struct addrinfo *iter = ainfo; iter != NULL; iter = iter->ai_next)
+ {
+ /* Iterate over the list of possible addresses to connect
+ to. For each, we'll try to connect and see if it
+ succeeds. */
+ int sock = try_connect (iter, &polls);
+
+ if (sock >= 0)
+ {
+ /* We've gotten a successful connection. Save its
+ 'struct addrinfo', the socket, and break. */
+ success_ainfo = iter;
+ scb->fd = sock;
+ break;
+ }
+ else if (
#ifdef USE_WIN32API
- && err == WSAECONNREFUSED
+ errno == WSAECONNREFUSED
#else
- && err == ECONNREFUSED
+ errno == ECONNREFUSED
#endif
- && wait_for_connect (NULL, &polls) >= 0)
- {
- close (scb->fd);
- goto retry;
- }
- if (err)
- errno = err;
- net_close (scb);
- return -1;
- }
- }
+ )
+ got_connrefused = true;
+ }
+ }
+ /* Just retry if:
+
+ - tcp_auto_retry is true, and
+ - We haven't gotten a connection yet, and
+ - Any of our connection attempts returned with ECONNREFUSED, and
+ - wait_for_connect signals that we can keep going. */
+ while (tcp_auto_retry
+ && success_ainfo == NULL
+ && got_connrefused
+ && wait_for_connect (-1, &polls) >= 0);
+
+ if (success_ainfo == NULL)
+ {
+ net_close (scb);
+ return -1;
+ }
/* Turn off nonblocking. */
- ioarg = 0;
+#ifdef USE_WIN32API
+ u_long ioarg = 0;
+#else
+ int ioarg = 0;
+#endif
+
ioctl (scb->fd, FIONBIO, &ioarg);
- if (use_udp == 0)
+ if (success_ainfo->ai_socktype == IPPROTO_TCP)
{
/* Disable Nagle algorithm. Needed in some cases. */
- tmp = 1;
+ int tmp = 1;
+
setsockopt (scb->fd, IPPROTO_TCP, TCP_NODELAY,
- (char *)&tmp, sizeof (tmp));
+ (char *) &tmp, sizeof (tmp));
}
#ifdef SIGPIPE