| Message ID | 20180816080724.25143-1-liwang@redhat.com |
|---|---|
| State | Changes Requested |
| Headers | show |
| Series | pty: fix some issues in pty02 | expand |
Hi! > 1. Add EXTPROC into lapi > > 2. Exit the test with TCONF if we get EINVAL from tcsetattr() > > 3. Giving newline('\n') to ptmx to avoid reading pts block on rhel6 > > 4. Using tcgetattr() to get attributes before re-setting it > Fix this error: > tst_test.c:1015: INFO: Timeout per run is 0h 50m 00s > pty02.c:42: BROK: tcsetattr() failed: EINVAL > > POSIX.1 General description: > Changes the attributes associated with a terminal. New attributes are > specified with a termios control structure. Programs should always > issue a tcgetattr() first, modify the desired fields, and then issue > a tcsetattr(). tcsetattr() should never be issued using a termios > structure that was not obtained using tcgetattr(). tcsetattr() should > use only a termios structure that was obtained by tcgetattr(). I've tried the test with this patch applied and I still got soft-lockup on unpatched kernel, so it seems like the reproducer works fine with these changes. > diff --git a/testcases/kernel/pty/pty02.c b/testcases/kernel/pty/pty02.c > index fd3d26b..8720407 100644 > --- a/testcases/kernel/pty/pty02.c > +++ b/testcases/kernel/pty/pty02.c > @@ -25,27 +25,42 @@ > */ > > #include <stdlib.h> > +#include <errno.h> > #include <termio.h> > > #include "tst_test.h" > +#include "lapi/termbits.h" > > static void do_test(void) > { > - struct termios io = { .c_lflag = EXTPROC | ICANON }; > + struct termios io; > int ptmx, pts; > char c = 'A'; > + char n = '\n'; > int nbytes; > > ptmx = SAFE_OPEN("/dev/ptmx", O_WRONLY); > > - if (tcsetattr(ptmx, TCSANOW, &io) != 0) > - tst_brk(TBROK | TERRNO, "tcsetattr() failed"); > + if (tcgetattr(ptmx, &io) != 0) > + tst_brk(TBROK | TERRNO, "tcgetattr() failed"); > + > + io.c_lflag = EXTPROC | ICANON; > + > + TEST(tcsetattr(ptmx, TCSANOW, &io)); > + if (TEST_RETURN == -1) { > + if (TEST_ERRNO == EINVAL) > + tst_res(TCONF, "tcsetattr(, , EXTPROC | ICANON) is not supported"); > + else > + tst_brk(TBROK | TERRNO, "tcsetattr() failed"); > + } We changed the TEST_RETURN to TST_RET and TEST_ERRNO to TST_ERR a few weeks ago, can you please update you git tree? > if (unlockpt(ptmx) != 0) > tst_brk(TBROK | TERRNO, "unlockpt() failed"); > > pts = SAFE_OPEN(ptsname(ptmx), O_RDONLY); > SAFE_WRITE(1, ptmx, &c, 1); > + /* giving newline('\n') to ptmx to avoid reading pts block */ > + SAFE_WRITE(1, ptmx, &n, 1); I wonder if we can just do SAFE_WRITE(1, ptmx, "A\n", 2) here instead? Also the comment is a bit misleading, I was thinking for a while about what is pts block and why do we need to avoid reading it. So maybe we should reword it as: /* write newline to ptmx to avoid read() on pts to block */ > SAFE_READ(1, pts, &c, 1); > > tst_res(TINFO, "Calling FIONREAD, this will hang in n_tty_ioctl() if the bug is present..."); > -- > 2.9.5 >
Cyril Hrubis <chrubis@suse.cz> wrote: > > I've tried the test with this patch applied and I still got soft-lockup > on unpatched kernel, so it seems like the reproducer works fine with > these changes. > > > > - if (tcsetattr(ptmx, TCSANOW, &io) != 0) > > - tst_brk(TBROK | TERRNO, "tcsetattr() failed"); > > + if (tcgetattr(ptmx, &io) != 0) > > + tst_brk(TBROK | TERRNO, "tcgetattr() failed"); > > + > > + io.c_lflag = EXTPROC | ICANON; > > + > > + TEST(tcsetattr(ptmx, TCSANOW, &io)); > > + if (TEST_RETURN == -1) { > > + if (TEST_ERRNO == EINVAL) > > + tst_res(TCONF, "tcsetattr(, , EXTPROC | ICANON) is > not supported"); > > + else > > + tst_brk(TBROK | TERRNO, "tcsetattr() failed"); > > + } > > We changed the TEST_RETURN to TST_RET and TEST_ERRNO to TST_ERR a few > weeks ago, can you please update you git tree? > Yes, sure. > > > if (unlockpt(ptmx) != 0) > > tst_brk(TBROK | TERRNO, "unlockpt() failed"); > > > > pts = SAFE_OPEN(ptsname(ptmx), O_RDONLY); > > SAFE_WRITE(1, ptmx, &c, 1); > > + /* giving newline('\n') to ptmx to avoid reading pts block */ > > + SAFE_WRITE(1, ptmx, &n, 1); > > I wonder if we can just do SAFE_WRITE(1, ptmx, "A\n", 2) here instead? > I think yes, and this is more tidy. > > Also the comment is a bit misleading, I was thinking for a while about > what is pts block and why do we need to avoid reading it. So maybe we > should reword it as: > > /* write newline to ptmx to avoid read() on pts to block */ > Ok, it will be updated in patch v2.
diff --git a/include/lapi/termbits.h b/include/lapi/termbits.h new file mode 100644 index 0000000..23ad5a7 --- /dev/null +++ b/include/lapi/termbits.h @@ -0,0 +1,25 @@ +/* + * Copyright (c) 2018 Linux Test Project + * + * 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 would 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, write the Free Software Foundation + */ + +#ifndef LAPI_TERMBITS_H__ +#define LAPI_TERMBITS_H__ + +#ifndef EXTPROC +# define EXTPROC 0200000 +#endif + +#endif diff --git a/testcases/kernel/pty/pty02.c b/testcases/kernel/pty/pty02.c index fd3d26b..8720407 100644 --- a/testcases/kernel/pty/pty02.c +++ b/testcases/kernel/pty/pty02.c @@ -25,27 +25,42 @@ */ #include <stdlib.h> +#include <errno.h> #include <termio.h> #include "tst_test.h" +#include "lapi/termbits.h" static void do_test(void) { - struct termios io = { .c_lflag = EXTPROC | ICANON }; + struct termios io; int ptmx, pts; char c = 'A'; + char n = '\n'; int nbytes; ptmx = SAFE_OPEN("/dev/ptmx", O_WRONLY); - if (tcsetattr(ptmx, TCSANOW, &io) != 0) - tst_brk(TBROK | TERRNO, "tcsetattr() failed"); + if (tcgetattr(ptmx, &io) != 0) + tst_brk(TBROK | TERRNO, "tcgetattr() failed"); + + io.c_lflag = EXTPROC | ICANON; + + TEST(tcsetattr(ptmx, TCSANOW, &io)); + if (TEST_RETURN == -1) { + if (TEST_ERRNO == EINVAL) + tst_res(TCONF, "tcsetattr(, , EXTPROC | ICANON) is not supported"); + else + tst_brk(TBROK | TERRNO, "tcsetattr() failed"); + } if (unlockpt(ptmx) != 0) tst_brk(TBROK | TERRNO, "unlockpt() failed"); pts = SAFE_OPEN(ptsname(ptmx), O_RDONLY); SAFE_WRITE(1, ptmx, &c, 1); + /* giving newline('\n') to ptmx to avoid reading pts block */ + SAFE_WRITE(1, ptmx, &n, 1); SAFE_READ(1, pts, &c, 1); tst_res(TINFO, "Calling FIONREAD, this will hang in n_tty_ioctl() if the bug is present...");
1. Add EXTPROC into lapi 2. Exit the test with TCONF if we get EINVAL from tcsetattr() 3. Giving newline('\n') to ptmx to avoid reading pts block on rhel6 4. Using tcgetattr() to get attributes before re-setting it Fix this error: tst_test.c:1015: INFO: Timeout per run is 0h 50m 00s pty02.c:42: BROK: tcsetattr() failed: EINVAL POSIX.1 General description: Changes the attributes associated with a terminal. New attributes are specified with a termios control structure. Programs should always issue a tcgetattr() first, modify the desired fields, and then issue a tcsetattr(). tcsetattr() should never be issued using a termios structure that was not obtained using tcgetattr(). tcsetattr() should use only a termios structure that was obtained by tcgetattr(). Signed-off-by: Li Wang <liwang@redhat.com> Cc: Eric Biggers <ebiggers@google.com> Cc: Cyril Hrubis <chrubis@suse.cz> Cc: Xiao Yang <yangx.jy@cn.fujitsu.com> Cc: Jinhui huang <huangjh.jy@cn.fujitsu.com> --- include/lapi/termbits.h | 25 +++++++++++++++++++++++++ testcases/kernel/pty/pty02.c | 21 ++++++++++++++++++--- 2 files changed, 43 insertions(+), 3 deletions(-) create mode 100644 include/lapi/termbits.h