Jens, Yeah, I checked the multimeter and that's how I knew I'd gotten my configuration wrong. I checked the scope to see what signaling I was getting. That was how I figured out what was happening. I may still be getting it wrong because I set the delay to zero and never put it back. So the device might be having problems.
I've been using this board for a while as a JTAG dongle and it has worked for many devices, and openocd has worked fairly well. Andreas, good to know. I was thinking, is there any way the two resets could be combined into one command to the chip to eliminate the time from sysreset being added to ntrst? Basically so they end up being treated independently? Or maybe it's supposed to be somewhat linked? Maybe the core sends ntrst then nsrst, making that difficult? It's not a big deal, but I can see cases where you wouldn't want JTAG reset extended by the width of sysrst. Most likely sent by Siri. > On Nov 6, 2013, at 6:41 PM, [email protected] wrote: > > Send OpenOCD-devel mailing list submissions to > [email protected] > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.sourceforge.net/lists/listinfo/openocd-devel > or, via email, send a message with subject or body 'help' to > [email protected] > > You can reach the person managing the list at > [email protected] > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of OpenOCD-devel digest..." > > > Today's Topics: > > 1. Re: [PATCH]: 04cc505 Add tcl configurations for Altera Soc > devices (Franck Jullien) > 2. Re: [PATCH]: 04cc505 Add tcl configurations for Altera Soc > devices (Franck Jullien) > 3. [PATCH]: 90512bd Correct argument numbering in trace command > ([email protected]) > 4. Re: CMSIS-DAP support (Spencer Oliver) > 5. [PATCH]: a8623fb Retire jtag_add_dr_out ([email protected]) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 5 Nov 2013 21:09:27 +0100 > From: Franck Jullien <[email protected]> > Subject: Re: [OpenOCD-devel] [PATCH]: 04cc505 Add tcl configurations > for Altera Soc devices > To: Brad Riensche <[email protected]>, OpenOCD ML > <[email protected]> > Message-ID: > <CAJfOKBxq=mohle3d0_yzvvm9uh_3okfeaxtmmnbyedurj-r...@mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > Ok, it works with the old USB-Blaster. > >> Debug: 227 210 arm_adi_v5.c:665 dap_syssec_kinetis_mdmap(): id doesn't >> match 24770002 != 0x001C0000 > > I was worried about this but it seems normal. > >> Debug: 228 257 adi_v5_jtag.c:271 jtagdp_transaction_endcheck(): jtag-dp: >> CTRL/STAT error, 0xf0000003 >> Debug: 229 257 adi_v5_jtag.c:291 jtagdp_transaction_endcheck(): MEM-AP >> Cached values: ap_bank 0x0, ap_csw 0xa2000020, ap_tar 0x0 >> Error: 230 257 adi_v5_jtag.c:295 jtagdp_transaction_endcheck(): JTAG-DP >> OVERRUN - check clock, memaccess, or reduce jtag speed >> Debug: 231 278 adi_v5_jtag.c:315 jtagdp_transaction_endcheck(): jtag-dp: >> CTRL/STAT 0xf0000001 >> Error: 232 324 adi_v5_jtag.c:331 jtagdp_transaction_endcheck(): MEM_AP_CSW >> 0x2800060, MEM_AP_TAR 0x0 >> Debug: 233 324 openocd.c:147 handle_init_command(): target examination failed > > This is my problem. The USB-Blaster II JTAG clock is much higher than > the old one (25MHz vs 6MHz). > I'll try to slow it down (it's not that easy bec > > > > ------------------------------ > > Message: 2 > Date: Tue, 5 Nov 2013 21:10:21 +0100 > From: Franck Jullien <[email protected]> > Subject: Re: [OpenOCD-devel] [PATCH]: 04cc505 Add tcl configurations > for Altera Soc devices > To: Brad Riensche <[email protected]>, OpenOCD ML > <[email protected]> > Message-ID: > <cajfokbyy6r4mrnmfnxxquj1jss4gr8tpt-6dyicd71eoyrf...@mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > 2013/11/5 Franck Jullien <[email protected]>: >> Ok, it works with the old USB-Blaster. >> >>> Debug: 227 210 arm_adi_v5.c:665 dap_syssec_kinetis_mdmap(): id doesn't >>> match 24770002 != 0x001C0000 >> >> I was worried about this but it seems normal. >> >>> Debug: 228 257 adi_v5_jtag.c:271 jtagdp_transaction_endcheck(): jtag-dp: >>> CTRL/STAT error, 0xf0000003 >>> Debug: 229 257 adi_v5_jtag.c:291 jtagdp_transaction_endcheck(): MEM-AP >>> Cached values: ap_bank 0x0, ap_csw 0xa2000020, ap_tar 0x0 >>> Error: 230 257 adi_v5_jtag.c:295 jtagdp_transaction_endcheck(): JTAG-DP >>> OVERRUN - check clock, memaccess, or reduce jtag speed >>> Debug: 231 278 adi_v5_jtag.c:315 jtagdp_transaction_endcheck(): jtag-dp: >>> CTRL/STAT 0xf0000001 >>> Error: 232 324 adi_v5_jtag.c:331 jtagdp_transaction_endcheck(): MEM_AP_CSW >>> 0x2800060, MEM_AP_TAR 0x0 >>> Debug: 233 324 openocd.c:147 handle_init_command(): target examination >>> failed >> >> This is my problem. The USB-Blaster II JTAG clock is much higher than >> the old one (25MHz vs 6MHz). >> I'll try to slow it down (it's not that easy bec > > Oups... > > ...because of the USB Blaster protocol. > > I'll keep you informed. > > Franck. > > > > ------------------------------ > > Message: 3 > Date: Tue, 5 Nov 2013 22:28:32 +0000 (UTC) > From: [email protected] > Subject: [OpenOCD-devel] [PATCH]: 90512bd Correct argument numbering > in trace command > To: [email protected] > Message-ID: <[email protected]> > Content-Type: text/plain; charset="utf-8" > > This is an automated email from Gerrit. > > Alan Bowman ([email protected]) just uploaded a new patch set to > Gerrit, which you can find at http://openocd.zylin.com/1800 > > -- gerrit > > commit 90512bdaf3970bdccaf225b01df202c5e064f5eb > Author: Alan Bowman <[email protected]> > Date: Tue Nov 5 22:25:51 2013 +0000 > > Correct argument numbering in trace command > > The trace command was opening a file named with the source frequency. > This change correctly passes the filename that is specified. > > Change-Id: Ia2eb8eda0e1e0f4f44b05c3a0ce7bef3bda51446 > Signed-off-by: Alan Bowman <[email protected]> > > diff --git a/src/jtag/hla/hla_interface.c b/src/jtag/hla/hla_interface.c > index ae33f54..9193133 100644 > --- a/src/jtag/hla/hla_interface.c > +++ b/src/jtag/hla/hla_interface.c > @@ -243,7 +243,7 @@ COMMAND_HANDLER(interface_handle_trace_command) > } > > if (CMD_ARGC == 2) { > - f = fopen(CMD_ARGV[0], "a"); > + f = fopen(CMD_ARGV[1], "a"); > if (!f) > return ERROR_COMMAND_SYNTAX_ERROR; > } > > -- > > > > ------------------------------ > > Message: 4 > Date: Wed, 6 Nov 2013 13:59:56 +0000 > From: Spencer Oliver <[email protected]> > Subject: Re: [OpenOCD-devel] CMSIS-DAP support > To: OpenOCD Devel <[email protected]> > Message-ID: > <CAK=H08Rr5Ug4zxkWX-bXDNxotq7h_-qv2R2=1g0fj0-_a-d...@mail.gmail.com> > Content-Type: text/plain; charset=UTF-8 > >> On 3 November 2013 19:14, Peter Stuge <[email protected]> wrote: >> Spencer Oliver wrote: >>> Shall we treat HIDAPI as a submodule or let the user worry about >>> this like we currently do with libusb. >> >> HIDAPI is packaged at least in some distributions now. Detect it >> as needed, ideally using pkg-config, and document that it's a >> requirement for CMSIS-DAP support. > > Even though the current patch on gerrit is a submodule I am also in > favour of leaving it to the user. > > Cheers > Spen > > > > ------------------------------ > > Message: 5 > Date: Wed, 6 Nov 2013 23:41:15 +0000 (UTC) > From: [email protected] > Subject: [OpenOCD-devel] [PATCH]: a8623fb Retire jtag_add_dr_out > To: [email protected] > Message-ID: <[email protected]> > Content-Type: text/plain; charset="utf-8" > > This is an automated email from Gerrit. > > Andreas Fritiofson ([email protected]) just uploaded a new patch > set to Gerrit, which you can find at http://openocd.zylin.com/1801 > > -- gerrit > > commit a8623fb0f1f390d8050d16cdaa8634610ca3f4f4 > Author: Andreas Fritiofson <[email protected]> > Date: Thu Nov 7 00:40:50 2013 +0100 > > Retire jtag_add_dr_out > > The out only version of jtag_add_dr_scan smells like a bogus optimization > that complicates the minidriver API for questionable gain. > > The function was only used by four old ARM targets. Rewrite the callers > to use the generic function and remove all implementations. > > Change-Id: I13b643687ee8ed6bc9b6336e7096c34f40ea96af > Signed-off-by: Andreas Fritiofson <[email protected]> > > diff --git a/src/jtag/drivers/driver.c b/src/jtag/drivers/driver.c > index 3b04aca..75ec115 100644 > --- a/src/jtag/drivers/driver.c > +++ b/src/jtag/drivers/driver.c > @@ -193,92 +193,6 @@ int interface_jtag_add_dr_scan(struct jtag_tap *active, > int in_num_fields, > return ERROR_OK; > } > > -/** > - * Generate a DR SCAN using the array of output values passed to the function > - * > - * This function assumes that the parameter target_tap specifies the one TAP > - * that is not bypassed. All other TAPs must be bypassed and the function > will > - * generate a dummy 1bit field for them. > - * > - * For the target_tap a sequence of output-only fields will be generated > where > - * each field has the size num_bits and the field's values are taken from > - * the array value. > - * > - * The bypass status of TAPs is set by jtag_add_ir_scan(). > - * > - */ > -void interface_jtag_add_dr_out(struct jtag_tap *target_tap, > - int in_num_fields, > - const int *num_bits, > - const uint32_t *value, > - tap_state_t end_state) > -{ > - /* count devices in bypass */ > - > - size_t bypass_devices = 0; > - > - for (struct jtag_tap *tap = jtag_tap_next_enabled(NULL); tap != NULL; > tap = jtag_tap_next_enabled(tap)) { > - if (tap->bypass) > - bypass_devices++; > - } > - > - > - struct jtag_command *cmd = cmd_queue_alloc(sizeof(struct jtag_command)); > - struct scan_command *scan = cmd_queue_alloc(sizeof(struct scan_command)); > - struct scan_field *out_fields = cmd_queue_alloc((in_num_fields + > bypass_devices) * sizeof(struct scan_field)); > - > - jtag_queue_command(cmd); > - > - cmd->type = JTAG_SCAN; > - cmd->cmd.scan = scan; > - > - scan->ir_scan = false; > - scan->num_fields = in_num_fields + bypass_devices; > - scan->fields = out_fields; > - scan->end_state = end_state; > - > - > - bool target_tap_match = false; > - > - struct scan_field *field = out_fields; /* keep track where we insert > data */ > - > - /* loop over all enabled TAPs */ > - > - for (struct jtag_tap *tap = jtag_tap_next_enabled(NULL); tap != NULL; > tap = jtag_tap_next_enabled(tap)) { > - /* if TAP is not bypassed insert matching input fields */ > - > - if (!tap->bypass) { > - assert(tap == target_tap); /* target_tap must match the one not > bypassed TAP */ > - > - target_tap_match = true; > - > - for (int j = 0; j < in_num_fields; j++) { > - uint8_t out_value[4]; > - size_t scan_size = num_bits[j]; > - buf_set_u32(out_value, 0, scan_size, value[j]); > - > - field->num_bits = scan_size; > - field->out_value = buf_cpy(out_value, > cmd_queue_alloc(DIV_ROUND_UP(scan_size, 8)), scan_size); > - field->in_value = NULL; > - > - field++; > - } > - } > - > - /* if a TAP is bypassed, generated a dummy bit*/ > - else { > - > - field->num_bits = 1; > - field->out_value = NULL; > - field->in_value = NULL; > - > - field++; > - } > - } > - > - assert(target_tap_match); /* target_tap should be enabled and not > bypassed */ > -} > - > static int jtag_add_plain_scan(int num_bits, const uint8_t *out_bits, > uint8_t *in_bits, tap_state_t state, bool ir_scan) > { > @@ -496,21 +410,6 @@ void interface_jtag_add_callback(jtag_callback1_t > callback, jtag_callback_data_t > jtag_add_callback4(jtag_convert_to_callback4, data0, > (jtag_callback_data_t)callback, 0, 0); > } > > -/* A minidriver can use use an inline versions of this API level fn */ > -void jtag_add_dr_out(struct jtag_tap *tap, > - int num_fields, const int *num_bits, const uint32_t *value, > - tap_state_t end_state) > -{ > - assert(end_state != TAP_RESET); > - assert(end_state != TAP_INVALID); > - > - cmd_queue_cur_state = end_state; > - > - interface_jtag_add_dr_out(tap, > - num_fields, num_bits, value, > - end_state); > -} > - > void jtag_add_callback(jtag_callback1_t f, jtag_callback_data_t data0) > { > interface_jtag_add_callback(f, data0); > diff --git a/src/jtag/drivers/minidriver_imp.h > b/src/jtag/drivers/minidriver_imp.h > index 1b5072b..b81368b 100644 > --- a/src/jtag/drivers/minidriver_imp.h > +++ b/src/jtag/drivers/minidriver_imp.h > @@ -29,21 +29,12 @@ static inline void > interface_jtag_add_scan_check_alloc(struct scan_field *field) > field->in_value = cmd_queue_alloc(num_bytes); > } > > -void interface_jtag_add_dr_out(struct jtag_tap *tap, > - int num_fields, const int *num_bits, const uint32_t > *value, > - tap_state_t end_state); > - > void interface_jtag_add_callback(jtag_callback1_t f, jtag_callback_data_t > data0); > > void interface_jtag_add_callback4(jtag_callback_t f, jtag_callback_data_t > data0, > jtag_callback_data_t data1, jtag_callback_data_t data2, > jtag_callback_data_t data3); > > -void jtag_add_dr_out(struct jtag_tap *tap, > - int num_fields, const int *num_bits, const uint32_t *value, > - tap_state_t end_state); > - > - > void jtag_add_callback4(jtag_callback_t f, jtag_callback_data_t data0, > jtag_callback_data_t data1, jtag_callback_data_t data2, > jtag_callback_data_t data3); > diff --git a/src/jtag/jtag.h b/src/jtag/jtag.h > index 0fe3a4a..eda4ccd 100644 > --- a/src/jtag/jtag.h > +++ b/src/jtag/jtag.h > @@ -610,33 +610,6 @@ void jtag_sleep(uint32_t us); > #define ERROR_JTAG_INIT_SOFT_FAIL (-110) > > /** > - * jtag_add_dr_out() is a version of jtag_add_dr_scan() which > - * only scans data out. It operates on 32 bit integers instead > - * of 8 bit, which makes it a better impedance match with > - * the calling code which often operate on 32 bit integers. > - * > - * Current or end_state can not be TAP_RESET. end_state can be TAP_INVALID > - * > - * num_bits[i] is the number of bits to clock out from value[i] LSB first. > - * > - * If the device is in bypass, then that is an error condition in > - * the caller code that is not detected by this fn, whereas > - * jtag_add_dr_scan() does detect it. Similarly if the device is not in > - * bypass, data must be passed to it. > - * > - * If anything fails, then jtag_error will be set and jtag_execute() will > - * return an error. There is no way to determine if there was a failure > - * during this function call. > - * > - * This is an inline fn to speed up embedded hosts. Also note that > - * interface_jtag_add_dr_out() can be a *small* inline function for > - * embedded hosts. > - * > - * There is no jtag_add_dr_outin() version of this fn that also allows > - * clocking data back in. Patches gladly accepted! > - */ > - > -/** > * Set the current JTAG core execution error, unless one was set > * by a previous call previously. Driver or application code must > * use jtag_error_clear to reset jtag_error once this routine has been > diff --git a/src/jtag/minidriver.h b/src/jtag/minidriver.h > index e85d9fb..0de8709 100644 > --- a/src/jtag/minidriver.h > +++ b/src/jtag/minidriver.h > @@ -39,7 +39,6 @@ > * define static inline versions of them): > * - jtag_add_callback > * - jtag_add_callback4 > - * - interface_jtag_add_dr_out > * > * The following core functions are declared in this file for use by > * the minidriver and do @b not need to be defined by an implementation: > diff --git a/src/jtag/minidriver/minidriver_imp.h > b/src/jtag/minidriver/minidriver_imp.h > index f056f01..8ad992c 100644 > --- a/src/jtag/minidriver/minidriver_imp.h > +++ b/src/jtag/minidriver/minidriver_imp.h > @@ -24,17 +24,6 @@ > > #include <jtag/jtag_minidriver.h> > > -static inline void jtag_add_dr_out(struct jtag_tap *tap, > - int num_fields, const int *num_bits, const uint32_t *value, > - tap_state_t end_state) > -{ > - cmd_queue_cur_state = end_state; > - > - interface_jtag_add_dr_out(tap, > - num_fields, num_bits, value, > - end_state); > -} > - > #define jtag_add_callback(callback, in) interface_jtag_add_callback(callback, > in) > > #define jtag_add_callback4(callback, in, data1, data2, data3) \ > diff --git a/src/jtag/minidummy/jtag_minidriver.h > b/src/jtag/minidummy/jtag_minidriver.h > index ef6a50a..452fbc8 100644 > --- a/src/jtag/minidummy/jtag_minidriver.h > +++ b/src/jtag/minidummy/jtag_minidriver.h > @@ -17,24 +17,6 @@ > * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. * > ***************************************************************************/ > > -static inline void interface_jtag_add_dr_out_core(struct jtag_tap *targettap, > - int num_fields, > - const int *num_bits, > - const uint32_t *value, > - enum tap_state end_state) > -{ > - /* synchronously do the operation here */ > -} > - > -static inline void interface_jtag_add_dr_out(struct jtag_tap *targettap, > - int num_fields, > - const int *num_bits, > - const uint32_t *value, > - enum tap_state end_state) > -{ > - /* synchronously do the operation here */ > -} > - > #define interface_jtag_add_callback(callback, in) callback(in) > > #define interface_jtag_add_callback4(callback, in, data1, data2, data3) \ > diff --git a/src/jtag/zy1000/jtag_minidriver.h > b/src/jtag/zy1000/jtag_minidriver.h > index 2ae31bc..d7fa437 100644 > --- a/src/jtag/zy1000/jtag_minidriver.h > +++ b/src/jtag/zy1000/jtag_minidriver.h > @@ -169,52 +169,6 @@ static inline void shiftValueInner(const enum tap_state > state, > #endif > } > > -static inline void interface_jtag_add_dr_out_core(struct jtag_tap > *target_tap, > - int num_fields, > - const int *num_bits, > - const uint32_t *value, > - enum tap_state end_state) > -{ > - enum tap_state pause_state = TAP_DRSHIFT; > - > - struct jtag_tap *tap, *nextTap; > - for (tap = jtag_tap_next_enabled(NULL); tap != NULL; tap = nextTap) { > - nextTap = jtag_tap_next_enabled(tap); > - if (nextTap == NULL) > - pause_state = end_state; > - if (tap == target_tap) { > - int j; > - for (j = 0; j < (num_fields-1); j++) > - shiftValueInner(TAP_DRSHIFT, TAP_DRSHIFT, num_bits[j], > value[j]); > - shiftValueInner(TAP_DRSHIFT, pause_state, num_bits[j], value[j]); > - } else { > - /* program the scan field to 1 bit length, and ignore it's value > */ > - shiftValueInner(TAP_DRSHIFT, pause_state, 1, 0); > - } > - } > -} > - > -static inline void interface_jtag_add_dr_out(struct jtag_tap *target_tap, > - int num_fields, > - const int *num_bits, > - const uint32_t *value, > - enum tap_state end_state) > -{ > - > - int singletap = (jtag_tap_next_enabled(jtag_tap_next_enabled(NULL)) == > NULL); > - if ((singletap) && (num_fields == 3)) { > - /* used by embeddedice_write_reg_inner() */ > - shiftValueInner(TAP_DRSHIFT, TAP_DRSHIFT, num_bits[0], value[0]); > - shiftValueInner(TAP_DRSHIFT, TAP_DRSHIFT, num_bits[1], value[1]); > - shiftValueInner(TAP_DRSHIFT, end_state, num_bits[2], value[2]); > - } else if ((singletap) && (num_fields == 2)) { > - /* used by arm7 code */ > - shiftValueInner(TAP_DRSHIFT, TAP_DRSHIFT, num_bits[0], value[0]); > - shiftValueInner(TAP_DRSHIFT, end_state, num_bits[1], value[1]); > - } else > - interface_jtag_add_dr_out_core(target_tap, num_fields, num_bits, > value, end_state); > -} > - > #if BUILD_ZY1000_MASTER > #define interface_jtag_add_callback(callback, in) callback(in) > #define interface_jtag_add_callback4(callback, in, data1, data2, \ > diff --git a/src/jtag/zy1000/zy1000.c b/src/jtag/zy1000/zy1000.c > index bcf4638..16c3e5d 100644 > --- a/src/jtag/zy1000/zy1000.c > +++ b/src/jtag/zy1000/zy1000.c > @@ -631,16 +631,6 @@ static void jtag_pre_post_bits(struct jtag_tap *tap, int > *pre, int *post) > *post = post_bits; > } > > -#if 0 > -static const int embeddedice_num_bits[] = {32, 6}; > - uint32_t values[2]; > - > - values[0] = value; > - values[1] = (1 << 5) | reg_addr; > - > - jtag_add_dr_out(tap, 2, embeddedice_num_bits, values, TAP_IDLE); > -#endif > - > void embeddedice_write_dcc(struct jtag_tap *tap, > int reg_addr, > const uint8_t *buffer, > @@ -700,8 +690,7 @@ int arm11_run_instr_data_to_core_noack_inner(struct > jtag_tap *tap, > uint32_t, const uint32_t *, size_t); > return arm11_run_instr_data_to_core_noack_inner_default(tap, opcode, > data, count); > } else { > - static const int bits[] = {32, 2}; > - uint32_t values[] = {0, 0}; > + static const uint8_t zero; > > /* FIX!!!!!! the target_write_memory() API started this nasty problem > * with unaligned uint32_t * pointers... */ > @@ -710,7 +699,7 @@ int arm11_run_instr_data_to_core_noack_inner(struct > jtag_tap *tap, > while (--count > 0) { > #if 1 > /* Danger! This code doesn't update cmd_queue_cur_state, so > - * invoking jtag_add_pathmove() before jtag_add_dr_out() after > + * invoking jtag_add_pathmove() before jtag_add_dr_scan() after > * this loop would fail! > */ > shiftValueInner(TAP_DRSHIFT, TAP_DRSHIFT, pre_bits, 0); > @@ -750,15 +739,15 @@ int arm11_run_instr_data_to_core_noack_inner(struct > jtag_tap *tap, > TAP_DRSELECT, TAP_DRCAPTURE, TAP_DRSHIFT > }; > > - values[0] = *t++; > - values[0] |= (*t++<<8); > - values[0] |= (*t++<<16); > - values[0] |= (*t++<<24); > + struct scan_field fields[2] = { > + { .num_bits = 32, .out_value = t }, > + { .num_bits = 2, .out_value = &zero }, > + }; > + t += 4; > > - jtag_add_dr_out(tap, > + jtag_add_dr_scan(tap, > 2, > - bits, > - values, > + fields, > TAP_IDLE); > > > jtag_add_pathmove(ARRAY_SIZE(arm11_MOVE_DRPAUSE_IDLE_DRPAUSE_with_delay), > @@ -766,18 +755,17 @@ int arm11_run_instr_data_to_core_noack_inner(struct > jtag_tap *tap, > #endif > } > > - values[0] = *t++; > - values[0] |= (*t++<<8); > - values[0] |= (*t++<<16); > - values[0] |= (*t++<<24); > + struct scan_field fields[2] = { > + { .num_bits = 32, .out_value = t }, > + { .num_bits = 2, .out_value = &zero }, > + }; > > /* This will happen on the last iteration updating cmd_queue_cur_state > * so we don't have to track it during the common code path > */ > - jtag_add_dr_out(tap, > + jtag_add_dr_scan(tap, > 2, > - bits, > - values, > + fields, > TAP_IDLE); > > return jtag_execute_queue(); > diff --git a/src/target/arm7tdmi.c b/src/target/arm7tdmi.c > index 3dc0013..807efad 100644 > --- a/src/target/arm7tdmi.c > +++ b/src/target/arm7tdmi.c > @@ -95,12 +95,18 @@ static const int arm7tdmi_num_bits[] = {1, 32}; > > static inline int arm7tdmi_clock_out_inner(struct arm_jtag *jtag_info, > uint32_t out, int breakpoint) > { > - uint32_t values[2] = {breakpoint, flip_u32(out, 32)}; > + uint8_t bp = breakpoint ? 1 : 0; > + uint8_t out_value[4]; > + buf_set_u32(out_value, 0, 32, flip_u32(out, 32)); > > - jtag_add_dr_out(jtag_info->tap, > + struct scan_field fields[2] = { > + { .num_bits = arm7tdmi_num_bits[0], .out_value = &bp }, > + { .num_bits = arm7tdmi_num_bits[1], .out_value = out_value }, > + }; > + > + jtag_add_dr_scan(jtag_info->tap, > 2, > - arm7tdmi_num_bits, > - values, > + fields, > TAP_DRPAUSE); > > jtag_add_runtest(0, TAP_DRPAUSE); > diff --git a/src/target/arm_jtag.c b/src/target/arm_jtag.c > index 1739a8a..6f70e85 100644 > --- a/src/target/arm_jtag.c > +++ b/src/target/arm_jtag.c > @@ -59,20 +59,18 @@ int arm_jtag_set_instr_inner(struct arm_jtag *jtag_info, > int arm_jtag_scann_inner(struct arm_jtag *jtag_info, uint32_t new_scan_chain, > tap_state_t end_state) > { > int retval = ERROR_OK; > - uint32_t values[1]; > - int num_bits[1]; > > - values[0] = new_scan_chain; > - num_bits[0] = jtag_info->scann_size; > + uint8_t out_value[4]; > + buf_set_u32(out_value, 0, jtag_info->scann_size, new_scan_chain); > + struct scan_field field = { .num_bits = jtag_info->scann_size, > .out_value = out_value, }; > > retval = arm_jtag_set_instr(jtag_info, jtag_info->scann_instr, NULL, > end_state); > if (retval != ERROR_OK) > return retval; > > - jtag_add_dr_out(jtag_info->tap, > + jtag_add_dr_scan(jtag_info->tap, > 1, > - num_bits, > - values, > + &field, > end_state); > > jtag_info->cur_scan_chain = new_scan_chain; > diff --git a/src/target/embeddedice.h b/src/target/embeddedice.h > index af28d38..f110e82 100644 > --- a/src/target/embeddedice.h > +++ b/src/target/embeddedice.h > @@ -112,13 +112,16 @@ int embeddedice_handshake(struct arm_jtag *jtag_info, > int hsbit, uint32_t timeou > */ > static inline void embeddedice_write_reg_inner(struct jtag_tap *tap, int > reg_addr, uint32_t value) > { > - static const int embeddedice_num_bits[] = {32, 6}; > - uint32_t values[2]; > + uint8_t out_reg_addr = (1 << 5) | reg_addr; > + uint8_t out_value[4]; > + buf_set_u32(out_value, 0, 32, value); > > - values[0] = value; > - values[1] = (1 << 5) | reg_addr; > + struct scan_field fields[2] = { > + { .num_bits = 32, .out_value = out_value }, > + { .num_bits = 6, .out_value = &out_reg_addr }, > + }; > > - jtag_add_dr_out(tap, 2, embeddedice_num_bits, values, TAP_IDLE); > + jtag_add_dr_scan(tap, 2, fields, TAP_IDLE); > } > > void embeddedice_write_dcc(struct jtag_tap *tap, int reg_addr, const uint8_t > *buffer, > diff --git a/src/target/xscale.c b/src/target/xscale.c > index 388c8eb..99f67af 100644 > --- a/src/target/xscale.c > +++ b/src/target/xscale.c > @@ -513,8 +513,6 @@ done: > static int xscale_send(struct target *target, const uint8_t *buffer, int > count, int size) > { > struct xscale_common *xscale = target_to_xscale(target); > - uint32_t t[3]; > - int bits[3]; > int retval; > int done_count = 0; > > @@ -522,37 +520,45 @@ static int xscale_send(struct target *target, const > uint8_t *buffer, int count, > XSCALE_DBGRX << xscale->xscale_variant, > TAP_IDLE); > > - bits[0] = 3; > - t[0] = 0; > - bits[1] = 32; > - t[2] = 1; > - bits[2] = 1; > + static const uint8_t t0; > + uint8_t t1[4]; > + static const uint8_t t2 = 1; > + struct scan_field fields[3] = { > + { .num_bits = 3, .out_value = &t0 }, > + { .num_bits = 32, .out_value = t1 }, > + { .num_bits = 1, .out_value = &t2 }, > + }; > + > int endianness = target->endianness; > while (done_count++ < count) { > + uint32_t t; > + > switch (size) { > case 4: > if (endianness == TARGET_LITTLE_ENDIAN) > - t[1] = le_to_h_u32(buffer); > + t = le_to_h_u32(buffer); > else > - t[1] = be_to_h_u32(buffer); > + t = be_to_h_u32(buffer); > break; > case 2: > if (endianness == TARGET_LITTLE_ENDIAN) > - t[1] = le_to_h_u16(buffer); > + t = le_to_h_u16(buffer); > else > - t[1] = be_to_h_u16(buffer); > + t = be_to_h_u16(buffer); > break; > case 1: > - t[1] = buffer[0]; > + t = buffer[0]; > break; > default: > LOG_ERROR("BUG: size neither 4, 2 nor 1"); > return ERROR_COMMAND_SYNTAX_ERROR; > } > - jtag_add_dr_out(target->tap, > + > + buf_set_u32(t1, 0, 32, t); > + > + jtag_add_dr_scan(target->tap, > 3, > - bits, > - t, > + fields, > TAP_IDLE); > buffer += size; > } > > -- > > > > ------------------------------ > > ------------------------------------------------------------------------------ > November Webinars for C, C++, Fortran Developers > Accelerate application performance with scalable programming models. Explore > techniques for threading, error checking, porting, and tuning. Get the most > from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk > > ------------------------------ > > _______________________________________________ > OpenOCD-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/openocd-devel > > > End of OpenOCD-devel Digest, Vol 25, Issue 7 > ******************************************** ------------------------------------------------------------------------------ November Webinars for C, C++, Fortran Developers Accelerate application performance with scalable programming models. Explore techniques for threading, error checking, porting, and tuning. Get the most from the latest Intel processors and coprocessors. See abstracts and register http://pubads.g.doubleclick.net/gampad/clk?id=60136231&iu=/4140/ostg.clktrk _______________________________________________ OpenOCD-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/openocd-devel
