diff options
| author | Denys Vlasenko <vda.linux@googlemail.com> | 2013-02-06 13:18:42 +0100 |
|---|---|---|
| committer | Denys Vlasenko <vda.linux@googlemail.com> | 2013-02-06 13:18:42 +0100 |
| commit | 8d4ca0c8cd81329ad730ab52b1e0653c1bbea803 (patch) | |
| tree | f601b5d6bda9b037bc72cb9efba5de7399a26dfc /strace.c | |
| parent | 88eafd81ab061ff217c0b56f02b07c16cbd29ff4 (diff) | |
| download | strace-8d4ca0c8cd81329ad730ab52b1e0653c1bbea803.tar.gz strace-8d4ca0c8cd81329ad730ab52b1e0653c1bbea803.tar.bz2 strace-8d4ca0c8cd81329ad730ab52b1e0653c1bbea803.tar.xz | |
Shortcut tests for fork/exec syscalls
This change should speed up strace by a tiny bit.
More importantly, it makes it much more clear that
fork and exec fixups are not necessary for any reasonably
recent kernels. IOW: syscall_fixup_for_fork_exec() and its callees
are all dead code.
* defs.h: Declare new need_fork_exec_workarounds flag variable.
* strace.c: Define need_fork_exec_workarounds flag variable.
(test_ptrace_setoptions_followfork): Return 0/1 as success/fail indicator.
(test_ptrace_setoptions_for_all): Likewise.
(init): Set need_fork_exec_workarounds to TRUE if needed.
* syscall.c: Rename internal_syscall() to syscall_fixup_for_fork_exec().
(trace_syscall_entering): Call syscall_fixup_for_fork_exec() only if
need_fork_exec_workarounds == TRUE.
(trace_syscall_exiting): Likewise.
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Diffstat (limited to 'strace.c')
| -rw-r--r-- | strace.c | 23 |
1 files changed, 16 insertions, 7 deletions
@@ -65,6 +65,7 @@ cflag_t cflag = CFLAG_NONE; unsigned int followfork = 0; unsigned int ptrace_setoptions = 0; unsigned int xflag = 0; +bool need_fork_exec_workarounds = 0; bool debug_flag = 0; bool Tflag = 0; bool qflag = 0; @@ -1165,7 +1166,7 @@ startup_child(char **argv) * First fork a new child, call ptrace with PTRACE_SETOPTIONS on it, * and then see which options are supported by the kernel. */ -static void +static int test_ptrace_setoptions_followfork(void) { int pid, expected_grandchild = 0, found_grandchild = 0; @@ -1258,10 +1259,11 @@ test_ptrace_setoptions_followfork(void) if (debug_flag) fprintf(stderr, "ptrace_setoptions = %#x\n", ptrace_setoptions); - return; + return 0; } error_msg("Test for PTRACE_O_TRACECLONE failed, " "giving up using this feature."); + return 1; } /* @@ -1278,7 +1280,7 @@ test_ptrace_setoptions_followfork(void) * int $0x80 * (compile with: "gcc -nostartfiles -nostdlib -o int3 int3.S") */ -static void +static int test_ptrace_setoptions_for_all(void) { const unsigned int test_options = PTRACE_O_TRACESYSGOOD | @@ -1287,8 +1289,9 @@ test_ptrace_setoptions_for_all(void) int it_worked = 0; /* this fork test doesn't work on no-mmu systems */ + /* FIXME: isn't it better to assume we *succeed*? */ if (strace_vforked) - return; + return 1; pid = fork(); if (pid < 0) @@ -1357,11 +1360,12 @@ test_ptrace_setoptions_for_all(void) if (debug_flag) fprintf(stderr, "ptrace_setoptions = %#x\n", ptrace_setoptions); - return; + return 0; } error_msg("Test for PTRACE_O_TRACESYSGOOD failed, " "giving up using this feature."); + return 1; } #ifdef USE_SEIZE @@ -1654,9 +1658,14 @@ init(int argc, char *argv[]) run_gid = getgid(); } + /* + * On any reasonably recent Linux kernel (circa about 2.5.46) + * need_fork_exec_workarounds should stay 0 after these tests: + */ + /*need_fork_exec_workarounds = 0; - already is */ if (followfork) - test_ptrace_setoptions_followfork(); - test_ptrace_setoptions_for_all(); + need_fork_exec_workarounds = test_ptrace_setoptions_followfork(); + need_fork_exec_workarounds |= test_ptrace_setoptions_for_all(); test_ptrace_seize(); /* Check if they want to redirect the output. */ |