Re: [PATCH v3] Harden gdb.base/step-over-syscall.exp

[Simon Marchi <simark@simark.ca>]

On 2020-01-27 1:13 p.m., Luis Machado wrote:
> @@ -75,39 +100,78 @@ proc setup { syscall } {
>      # Hit the breakpoint on $syscall for the second time.  In this time,
>      # the address of syscall insn and next insn of syscall are recorded.
>  
> -    gdb_test "display/i \$pc" ".*"
> +    # Check if the first instruction we stopped at is the syscall one.
> +    set syscall_insn_addr -1
> +    gdb_test_multiple "display/i \$pc" "fetch first stop pc" {
> +	-re "display/i .*: x/i .*=> ($hex) .*:.*$syscall_insn.*$gdb_prompt $" {
> +	    set insn_addr $expect_out(1,string)
>  
> -    # Single step until we see a syscall insn or we reach the
> -    # upper bound of loop iterations.
> -    set msg "find syscall insn in $syscall"
> -    set steps 0
> -    set max_steps 1000
> -    gdb_test_multiple "stepi" $msg {
> -	-re ".*$syscall_insn.*$gdb_prompt $" {
> -	    pass $msg
> +	    # Is the syscall number the correct one?
> +	    if {[syscall_number_matches $syscall]} {
> +		set syscall_insn_addr $insn_addr
> +	    }
> +	    pass $gdb_test_name
>  	}
> -	-re "x/i .*=>.*\r\n$gdb_prompt $" {
> -	    incr steps
> -	    if {$steps == $max_steps} {
> -		fail $msg
> -	    } else {
> -		send_gdb "stepi\n"
> -		exp_continue
> +	-re ".*$gdb_prompt $" {
> +	    pass $gdb_test_name
> +	}
> +    }
> +
> +    # If we are not at the syscall instruction yet, keep looking for it with
> +    # stepi commands.
> +    if {$syscall_insn_addr == -1} {
> +	# Single step until we see a syscall insn or we reach the
> +	# upper bound of loop iterations.
> +	set msg "find syscall insn in $syscall"

This can be a literal in the gdB_test_multiple call, we don't need `msg`
anymore.

The patch LGTM with that fixed.

Simon