interesting
/
riscv-openocd
mirror of https://github.com/riscv/riscv-openocd.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

gdb_server: improved gdb load performance 

by ack'ing memory writes immediately and reporting either
at next memory write or stepi/continue time. GDB will then
send off a new packet that is ready by the time the previous
packet has been written to target memory.

On faster adapters this can be as much as 10% improvement.

Signed-off-by: Øyvind Harboe <oyvind.harboe@zylin.com>
This commit is contained in:
Øyvind Harboe 2010-03-16 14:45:07 +01:00
parent 7e447043cd
commit 051e2c99ab
1 changed files with 37 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

45
src/server/gdb_server.c
View File

@ -2,7 +2,7 @@
* Copyright (C) 2005 by Dominic Rath *
* Dominic.Rath@gmx.de *
* *
* Copyright (C) 2007-2009 Øyvind Harboe *
* Copyright (C) 2007-2010 Øyvind Harboe *
* oyvind.harboe@zylin.com *
* *
* Copyright (C) 2008 by Spencer Oliver *
@ -61,7 +61,12 @@ struct gdb_connection
bool sync; /* set flag to true if you want the next stepi to return immediately.
allowing GDB to pick up a fresh set of register values from the target
without modifying the target state. */
/* We delay reporting memory write errors until next step/continue or memory
* write. This improves performance of gdb load significantly as the GDB packet
* can be replied immediately and a new GDB packet will be ready without delay
* (ca. 10% or so...).
*/
bool mem_write_error;
};
@ -821,6 +826,7 @@ static int gdb_new_connection(struct connection *connection)
gdb_connection->busy = 0;
gdb_connection->noack_mode = 0;
gdb_connection->sync = true;
gdb_connection->mem_write_error = false;
/* send ACK to GDB for debug request */
gdb_write(connection, "+", 1);
@ -1361,7 +1367,7 @@ static int gdb_write_memory_binary_packet(struct connection *connection,
uint32_t addr = 0;
uint32_t len = 0;
int retval;
int retval = ERROR_OK;
/* skip command character */
packet++;
@ -1382,14 +1388,18 @@ static int gdb_write_memory_binary_packet(struct connection *connection,
return ERROR_SERVER_REMOTE_CLOSED;
}
retval = ERROR_OK;
if (len)
{
LOG_DEBUG("addr: 0x%8.8" PRIx32 ", len: 0x%8.8" PRIx32 "", addr, len);
struct gdb_connection *gdb_connection = connection->priv;
retval = target_write_buffer(target, addr, len, (uint8_t*)separator);
if (gdb_connection->mem_write_error)
{
retval = ERROR_FAIL;
/* now that we have reported the memory write error, we can clear the condition */
gdb_connection->mem_write_error = false;
}
/* By replying the packet *immediately* GDB will send us a new packet
* while we write the last one to the target.
*/
if (retval == ERROR_OK)
{
gdb_put_packet(connection, "OK", 2);
@ -1400,6 +1410,17 @@ static int gdb_write_memory_binary_packet(struct connection *connection,
return retval;
}
if (len)
{
LOG_DEBUG("addr: 0x%8.8" PRIx32 ", len: 0x%8.8" PRIx32 "", addr, len);
retval = target_write_buffer(target, addr, len, (uint8_t*)separator);
if (retval != ERROR_OK)
{
gdb_connection->mem_write_error = true;
}
}
return ERROR_OK;
}
@ -2211,6 +2232,14 @@ static int gdb_input_inner(struct connection *connection)
struct gdb_connection *gdb_con = connection->priv;
log_add_callback(gdb_log_callback, connection);
if (gdb_con->mem_write_error)
{
LOG_ERROR("Memory write failure!");
/* now that we have reported the memory write error, we can clear the condition */
gdb_con->mem_write_error = false;
}
bool nostep = false;
bool already_running = false;
if (target->state == TARGET_RUNNING)