Menu
▾
▴
openocd-commit
|
From: David B. <dbr...@us...> - 2009-10-13 10:21:45
|
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 1c2e48b000503f15abba9092e8a159e0914c587b (commit)
from 9aab763fa555f049f03a242114ade0d1978f4291 (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 1c2e48b000503f15abba9092e8a159e0914c587b
Author: David Brownell <dbr...@us...>
Date: Tue Oct 13 01:21:24 2009 -0700
xscale: stackframe corruption bugfix
Resolve a "FIX" comment; yes that was superfluous given that the
JTAG core does that check by default. It was also buggy since it
wrote to a stack frame that went away before the write happened!!
Other fixes: remove pointless malloc(); zero-init scan_field_t
values wherever they appear; whitespace scrub; spelling fix.
Signed-off-by: David Brownell <dbr...@us...>
diff --git a/src/target/xscale.c b/src/target/xscale.c
index 82a2c57..dd16b35 100644
--- a/src/target/xscale.c
+++ b/src/target/xscale.c
@@ -42,7 +42,7 @@
* Intel XScale® Core Developerâs Manual, January 2004
* Order Number: 273473-002
* This has a chapter detailing debug facilities, and punts some
- * details to chip-specific microarchitecture documentats.
+ * details to chip-specific microarchitecture documents.
*
* Hot-Debug for Intel XScale® Core Debug White Paper, May 2005
* Document Number: 273539-005
@@ -166,21 +166,15 @@ static int xscale_jtag_set_instr(jtag_tap_t *tap, uint32_t new_instr)
if (buf_get_u32(tap->cur_instr, 0, tap->ir_length) != new_instr)
{
scan_field_t field;
+ uint8_t scratch[4];
+ memset(&field, 0, sizeof field);
field.tap = tap;
field.num_bits = tap->ir_length;
- field.out_value = calloc(CEIL(field.num_bits, 8), 1);
+ field.out_value = scratch;
buf_set_u32(field.out_value, 0, field.num_bits, new_instr);
- uint8_t tmp[4];
- field.in_value = tmp;
-
jtag_add_ir_scan(1, &field, jtag_get_end_state());
-
- /* FIX!!!! isn't this check superfluous? verify_ircapture handles this? */
- jtag_check_value_mask(&field, tap->expected, tap->expected_mask);
-
- free(field.out_value);
}
return ERROR_OK;
@@ -190,9 +184,7 @@ static int xscale_read_dcsr(target_t *target)
{
armv4_5_common_t *armv4_5 = target->arch_info;
xscale_common_t *xscale = armv4_5->arch_info;
-
int retval;
-
scan_field_t fields[3];
uint8_t field0 = 0x0;
uint8_t field0_check_value = 0x2;
@@ -207,6 +199,8 @@ static int xscale_read_dcsr(target_t *target)
buf_set_u32(&field0, 1, 1, xscale->hold_rst);
buf_set_u32(&field0, 2, 1, xscale->external_debug_break);
+ memset(&fields, 0, sizeof fields);
+
fields[0].tap = target->tap;
fields[0].num_bits = 3;
fields[0].out_value = &field0;
@@ -215,7 +209,6 @@ static int xscale_read_dcsr(target_t *target)
fields[1].tap = target->tap;
fields[1].num_bits = 32;
- fields[1].out_value = NULL;
fields[1].in_value = xscale->reg_cache->reg_list[XSCALE_DCSR].value;
fields[2].tap = target->tap;
@@ -277,30 +270,24 @@ static int xscale_receive(target_t *target, uint32_t *buffer, int num_words)
uint8_t field2_check_mask = 0x1;
int words_done = 0;
int words_scheduled = 0;
-
int i;
path[0] = TAP_DRSELECT;
path[1] = TAP_DRCAPTURE;
path[2] = TAP_DRSHIFT;
+ memset(&fields, 0, sizeof fields);
+
fields[0].tap = target->tap;
fields[0].num_bits = 3;
- fields[0].out_value = NULL;
- fields[0].in_value = NULL;
fields[0].check_value = &field0_check_value;
fields[0].check_mask = &field0_check_mask;
fields[1].tap = target->tap;
fields[1].num_bits = 32;
- fields[1].out_value = NULL;
- fields[1].check_value = NULL;
- fields[1].check_mask = NULL;
fields[2].tap = target->tap;
fields[2].num_bits = 1;
- fields[2].out_value = NULL;
- fields[2].in_value = NULL;
fields[2].check_value = &field2_check_value;
fields[2].check_mask = &field2_check_mask;
@@ -377,10 +364,8 @@ static int xscale_read_tx(target_t *target, int consume)
xscale_common_t *xscale = armv4_5->arch_info;
tap_state_t path[3];
tap_state_t noconsume_path[6];
-
int retval;
struct timeval timeout, now;
-
scan_field_t fields[3];
uint8_t field0_in = 0x0;
uint8_t field0_check_value = 0x2;
@@ -403,19 +388,18 @@ static int xscale_read_tx(target_t *target, int consume)
noconsume_path[4] = TAP_DREXIT2;
noconsume_path[5] = TAP_DRSHIFT;
+ memset(&fields, 0, sizeof fields);
+
fields[0].tap = target->tap;
fields[0].num_bits = 3;
- fields[0].out_value = NULL;
fields[0].in_value = &field0_in;
fields[1].tap = target->tap;
fields[1].num_bits = 32;
- fields[1].out_value = NULL;
fields[1].in_value = xscale->reg_cache->reg_list[XSCALE_TX].value;
fields[2].tap = target->tap;
fields[2].num_bits = 1;
- fields[2].out_value = NULL;
uint8_t tmp;
fields[2].in_value = &tmp;
@@ -477,10 +461,8 @@ static int xscale_write_rx(target_t *target)
{
armv4_5_common_t *armv4_5 = target->arch_info;
xscale_common_t *xscale = armv4_5->arch_info;
-
int retval;
struct timeval timeout, now;
-
scan_field_t fields[3];
uint8_t field0_out = 0x0;
uint8_t field0_in = 0x0;
@@ -494,6 +476,8 @@ static int xscale_write_rx(target_t *target)
xscale_jtag_set_instr(target->tap, XSCALE_DBGRX);
+ memset(&fields, 0, sizeof fields);
+
fields[0].tap = target->tap;
fields[0].num_bits = 3;
fields[0].out_value = &field0_out;
@@ -502,7 +486,6 @@ static int xscale_write_rx(target_t *target)
fields[1].tap = target->tap;
fields[1].num_bits = 32;
fields[1].out_value = xscale->reg_cache->reg_list[XSCALE_RX].value;
- fields[1].in_value = NULL;
fields[2].tap = target->tap;
fields[2].num_bits = 1;
@@ -637,9 +620,7 @@ static int xscale_write_dcsr(target_t *target, int hold_rst, int ext_dbg_brk)
{
armv4_5_common_t *armv4_5 = target->arch_info;
xscale_common_t *xscale = armv4_5->arch_info;
-
int retval;
-
scan_field_t fields[3];
uint8_t field0 = 0x0;
uint8_t field0_check_value = 0x2;
@@ -660,6 +641,8 @@ static int xscale_write_dcsr(target_t *target, int hold_rst, int ext_dbg_brk)
buf_set_u32(&field0, 1, 1, xscale->hold_rst);
buf_set_u32(&field0, 2, 1, xscale->external_debug_break);
+ memset(&fields, 0, sizeof fields);
+
fields[0].tap = target->tap;
fields[0].num_bits = 3;
fields[0].out_value = &field0;
@@ -669,7 +652,6 @@ static int xscale_write_dcsr(target_t *target, int hold_rst, int ext_dbg_brk)
fields[1].tap = target->tap;
fields[1].num_bits = 32;
fields[1].out_value = xscale->reg_cache->reg_list[XSCALE_DCSR].value;
- fields[1].in_value = NULL;
fields[2].tap = target->tap;
fields[2].num_bits = 1;
@@ -728,15 +710,15 @@ static int xscale_load_ic(target_t *target, uint32_t va, uint32_t buffer[8])
/* virtual address of desired cache line */
buf_set_u32(packet, 0, 27, va >> 5);
+ memset(&fields, 0, sizeof fields);
+
fields[0].tap = target->tap;
fields[0].num_bits = 6;
fields[0].out_value = &cmd;
- fields[0].in_value = NULL;
fields[1].tap = target->tap;
fields[1].num_bits = 27;
fields[1].out_value = packet;
- fields[1].in_value = NULL;
jtag_add_dr_scan(2, fields, jtag_get_end_state());
@@ -776,15 +758,15 @@ static int xscale_invalidate_ic_line(target_t *target, uint32_t va)
/* virtual address of desired cache line */
buf_set_u32(packet, 0, 27, va >> 5);
+ memset(&fields, 0, sizeof fields);
+
fields[0].tap = target->tap;
fields[0].num_bits = 6;
fields[0].out_value = &cmd;
- fields[0].in_value = NULL;
fields[1].tap = target->tap;
fields[1].num_bits = 27;
fields[1].out_value = packet;
- fields[1].in_value = NULL;
jtag_add_dr_scan(2, fields, jtag_get_end_state());
-----------------------------------------------------------------------
Summary of changes:
src/target/xscale.c | 54 +++++++++++++++++----------------------------------
1 files changed, 18 insertions(+), 36 deletions(-)
hooks/post-receive
--
Main OpenOCD repository
|
|
From: Øyvind H. <oyv...@zy...> - 2009-10-13 11:05:57
|
I'm not crazy about the memset() w.r.t. performance of inner loops. If it turns up in profiling, I'll see about removing memset()'s, otherwise their fine by me. -- Øyvind Harboe http://www.zylin.com/zy1000.html ARM7 ARM9 ARM11 XScale Cortex JTAG debugger and flash programmer |
|
From: David B. <da...@pa...> - 2009-10-13 11:24:59
|
On Tuesday 13 October 2009, Øyvind Harboe wrote: > I'm not crazy about the memset() w.r.t. performance of inner loops. I wouldn't call those "inner loops"; or consider *knowing* the code is correct to ever be an issue. And in any case, GCC tends to use a builtin, which gives *VERY* tight code. Loop runs inside of one cacheline, etc. p.s. gcc -std=gnu99 -DHAVE_CONFIG_H -I. -I../.. -I../../src/helper -I../../src/jtag -I../../src/xsvf -I/usr/local/include -Wall -Wstrict-prototypes -Wformat-security -Wextra -Wno-unused-parameter -Wbad-function-cast -Wcast-align -Wredundant-decls -Werror -MT arm11.lo -MD -MP -MF .deps/arm11.Tpo -c arm11.c -o arm11.o cc1: warnings being treated as errors arm11_dbgtap.c: In function ‘arm11_run_instr_data_to_core_noack’: arm11_dbgtap.c:593: warning: format ‘%d’ expects type ‘int’, but argument 6 has type ‘size_t’ make[3]: *** [arm11_dbgtap.lo] Error 1 make[3]: *** Waiting for unfinished jobs.... |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X