[OpenOCD-commit] Main OpenOCD repository branch, master, updated. v0.10.0-1276-g37330f89d

[OpenOCD-Gerrit <ope...@us...>]

This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "Main OpenOCD repository".

The branch, master has been updated
       via  37330f89d789e5b0a74aa14f7a7bfcfb4260abff (commit)
       via  1fa66d3633862d824420fb90b5b63e41e415d70e (commit)
      from  ffc1ca4b91757911c53d5dc86cae55eb00d858d5 (commit)

Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.

- Log -----------------------------------------------------------------
commit 37330f89d789e5b0a74aa14f7a7bfcfb4260abff
Author: Antonio Borneo <bor...@gm...>
Date:   Wed May 27 00:03:45 2020 +0200

    target/cortex-m: enable C_DEBUGEN during examine
    
    Current code for Cortex M does not set C_DEBUGEN as soon as
    possible, (which means during target examine), but later-on either:
    1) at command 'halt' (e.g. for 'gdb-attach' event);
    2) at command 'soft_reset_halt';
    3) at commands 'reset', 'reset halt' or 'reset init';
    4) during polling, but only if the target:
        = enter in 'double fault', or
        = exit from a reset, or
        = halts (not possible if C_DEBUGEN is not set)
    
    Plus, if commands in 1) or 2) are executed before the very first
    poll of the target, the value of 'cortex_m->dcb_dhcsr' is used not
    initialized while writing it back in DCB_DHCSR.
    
    Another side effect of this situation is that it's possible to set
    a HW breakpoint with the target running, while C_DEBUGEN is not
    set. Accordingly to [1], C1.3.1 "Debug authentication":
        When DGBEN is LOW and DHCSR.S_HALT == 0:
        ...
        FPB breakpoints do not generate an entry to Debug state and,
        if no DebugMonitor exception is generated, will escalate to
        HardFault, Lockup, or be ignored.
    On STM32MP15x I get HW breakpoint ignored, while on STM32F411 I
    get HardFault.
    E.g. following these steps:
    - power-on a pre-flashed board that starts running the firmware;
    - connect openocd, without halting or resetting the board;
    - set a HW breakpoint to some address often executed;
    - wait, but the board doesn't halt ...;
    - type the command 'halt';
    - if the Cortex-M has HardFault it would be visible and the fault
      is at the breakpoint address;
    - if no HardFault then type the command 'resume';
    - wait and the board will finally halt at the HW breakpoint.
    
    A similar issue has been detected on Cortex-A code and fixed by
    commit bff87a7f28fb ("target/cortex_a: enable DSCR_HALT_DBG_MODE
    during examine").
    Follow the same approach and set C_DEBUGEN during examine.
    Also, initialize 'cortex_m->dcb_dhcsr' during examine.
    
    [1] ARM DDI 0403E "ARM v7-M Architecture Reference Manual"
    
    Change-Id: I5b0b23403634f7dfce38f104bba9f59c33eb3e99
    Signed-off-by: Antonio Borneo <bor...@gm...>
    Reviewed-on: http://openocd.zylin.com/5702
    Tested-by: jenkins
    Reviewed-by: Andreas Fritiofson <and...@gm...>
    Reviewed-by: Tarek BOCHKATI <tar...@gm...>
    Reviewed-by: Moritz Fischer <mo...@go...>

diff --git a/src/target/cortex_m.c b/src/target/cortex_m.c
index e540f8507..d9bee0e53 100644
--- a/src/target/cortex_m.c
+++ b/src/target/cortex_m.c
@@ -2230,6 +2230,19 @@ int cortex_m_examine(struct target *target)
 				armv7m->debug_ap->tar_autoincr_block = (1 << 10);
 		}
 
+		/* Enable debug requests */
+		retval = target_read_u32(target, DCB_DHCSR, &cortex_m->dcb_dhcsr);
+		if (retval != ERROR_OK)
+			return retval;
+		if (!(cortex_m->dcb_dhcsr & C_DEBUGEN)) {
+			uint32_t dhcsr = (cortex_m->dcb_dhcsr | C_DEBUGEN) & ~(C_HALT | C_STEP | C_MASKINTS);
+
+			retval = target_write_u32(target, DCB_DHCSR, DBGKEY | (dhcsr & 0x0000FFFFUL));
+			if (retval != ERROR_OK)
+				return retval;
+			cortex_m->dcb_dhcsr = dhcsr;
+		}
+
 		/* Configure trace modules */
 		retval = target_write_u32(target, DCB_DEMCR, TRCENA | armv7m->demcr);
 		if (retval != ERROR_OK)

commit 1fa66d3633862d824420fb90b5b63e41e415d70e
Author: Antonio Borneo <bor...@gm...>
Date:   Fri May 29 12:00:35 2020 +0200

    log: fix kept_alive() and report expired timeout
    
    The kept_alive() function is called to inform the keep-alive code
    that a keep-alive as been just kicked through some other path.
    But kept_alive() erroneously resets the timeout counter without
    checking if it has already expired, thus masking a potential
    timeout.
    
    Check if timeout counter has expired during kept_alive().
    While there, put the timeout values in macros and explicit the
    units in the timeout messages.
    
    Change-Id: Iaf6368b44e5b5352b1cc4e7efbb2368575dcfa08
    Signed-off-by: Antonio Borneo <bor...@gm...>
    Reviewed-on: http://openocd.zylin.com/5705
    Reviewed-by: <jon...@gm...>
    Tested-by: jenkins

diff --git a/src/helper/log.c b/src/helper/log.c
index 380f548f4..31122554e 100644
--- a/src/helper/log.c
+++ b/src/helper/log.c
@@ -401,25 +401,43 @@ char *alloc_printf(const char *format, ...)
  * fast when invoked more often than every 500ms.
  *
  */
+#define KEEP_ALIVE_KICK_TIME_MS  500
+#define KEEP_ALIVE_TIMEOUT_MS   1000
+
+static void gdb_timeout_warning(int64_t delta_time)
+{
+	extern int gdb_actual_connections;
+
+	if (gdb_actual_connections)
+		LOG_WARNING("keep_alive() was not invoked in the "
+			"%d ms timelimit. GDB alive packet not "
+			"sent! (%" PRId64 " ms). Workaround: increase "
+			"\"set remotetimeout\" in GDB",
+			KEEP_ALIVE_TIMEOUT_MS,
+			delta_time);
+	else
+		LOG_DEBUG("keep_alive() was not invoked in the "
+			"%d ms timelimit (%" PRId64 " ms). This may cause "
+			"trouble with GDB connections.",
+			KEEP_ALIVE_TIMEOUT_MS,
+			delta_time);
+}
+
 void keep_alive(void)
 {
 	current_time = timeval_ms();
-	if (current_time-last_time > 1000) {
-		extern int gdb_actual_connections;
-
-		if (gdb_actual_connections)
-			LOG_WARNING("keep_alive() was not invoked in the "
-				"1000ms timelimit. GDB alive packet not "
-				"sent! (%" PRId64 "). Workaround: increase "
-				"\"set remotetimeout\" in GDB",
-				current_time-last_time);
-		else
-			LOG_DEBUG("keep_alive() was not invoked in the "
-				"1000ms timelimit (%" PRId64 "). This may cause "
-				"trouble with GDB connections.",
-				current_time-last_time);
+
+	int64_t delta_time = current_time - last_time;
+
+	if (delta_time > KEEP_ALIVE_TIMEOUT_MS) {
+		last_time = current_time;
+
+		gdb_timeout_warning(delta_time);
 	}
-	if (current_time-last_time > 500) {
+
+	if (delta_time > KEEP_ALIVE_KICK_TIME_MS) {
+		last_time = current_time;
+
 		/* this will keep the GDB connection alive */
 		LOG_USER_N("%s", "");
 
@@ -430,8 +448,6 @@ void keep_alive(void)
 		 *
 		 * These functions should be invoked at a well defined spot in server.c
 		 */
-
-		last_time = current_time;
 	}
 }
 
@@ -439,7 +455,13 @@ void keep_alive(void)
 void kept_alive(void)
 {
 	current_time = timeval_ms();
+
+	int64_t delta_time = current_time - last_time;
+
 	last_time = current_time;
+
+	if (delta_time > KEEP_ALIVE_TIMEOUT_MS)
+		gdb_timeout_warning(delta_time);
 }
 
 /* if we sleep for extended periods of time, we must invoke keep_alive() intermittantly */

-----------------------------------------------------------------------

Summary of changes:
 src/helper/log.c      | 56 +++++++++++++++++++++++++++++++++++----------------
 src/target/cortex_m.c | 13 ++++++++++++
 2 files changed, 52 insertions(+), 17 deletions(-)


hooks/post-receive
-- 
Main OpenOCD repository