From: Joe Komlodi <joe.komlodi@xilinx.com>
To: qemu-devel@nongnu.org
Cc: edgar.iglesias@xilinx.com
Subject: [PATCH v1 3/4] target/microblaze: gdb: Fix incorrect SReg reporting
Date: Mon, 11 May 2020 10:05:50 -0700 [thread overview]
Message-ID: <1589216751-106038-4-git-send-email-komlodi@xilinx.com> (raw)
In-Reply-To: <1589216751-106038-1-git-send-email-komlodi@xilinx.com>
SRegs used to be reported to GDB by iterating over the SRegs array,
however we do not store them in an order that allows them to be
reported to GDB in that way.
To fix this, a simple map is used to map the register GDB wants to its
location in the SRegs array.
Signed-off-by: Joe Komlodi <komlodi@xilinx.com>
---
target/microblaze/gdbstub.c | 59 +++++++++++++++++++++++++++++++++++++--------
1 file changed, 49 insertions(+), 10 deletions(-)
diff --git a/target/microblaze/gdbstub.c b/target/microblaze/gdbstub.c
index af29f00..485b717 100644
--- a/target/microblaze/gdbstub.c
+++ b/target/microblaze/gdbstub.c
@@ -25,6 +25,21 @@ int mb_cpu_gdb_read_register(CPUState *cs, GByteArray *mem_buf, int n)
{
MicroBlazeCPU *cpu = MICROBLAZE_CPU(cs);
CPUMBState *env = &cpu->env;
+ /*
+ * GDB expects SREGs in the following order:
+ * PC, MSR, EAR, ESR, FSR, BTR, EDR, PID, ZPR, TLBX, TLBSX, TLBLO, TLBHI.
+ * They aren't stored in this order, so make a map.
+ * PID, ZPR, TLBx, TLBsx, TLBLO, and TLBHI aren't modeled, so we don't
+ * map them to anything and return a value of 0 instead.
+ */
+ static const uint8_t sreg_map[6] = {
+ SR_PC,
+ SR_MSR,
+ SR_EAR,
+ SR_ESR,
+ SR_FSR,
+ SR_BTR
+ };
/*
* GDB expects registers to be reported in this order:
@@ -40,15 +55,16 @@ int mb_cpu_gdb_read_register(CPUState *cs, GByteArray *mem_buf, int n)
n -= 32;
switch (n) {
case 0 ... 5:
- return gdb_get_reg32(mem_buf, env->sregs[n]);
+ return gdb_get_reg32(mem_buf, env->sregs[sreg_map[n]]);
/* PVR12 is intentionally skipped */
case 6 ... 17:
n -= 6;
return gdb_get_reg32(mem_buf, env->pvr.regs[n]);
- case 18 ... 24:
- /* Add an offset of 6 to resume where we left off with SRegs */
- n = n - 18 + 6;
- return gdb_get_reg32(mem_buf, env->sregs[n]);
+ case 18:
+ return gdb_get_reg32(mem_buf, env->sregs[SR_EDR]);
+ /* Other SRegs aren't modeled, so report a value of 0 */
+ case 19 ... 24:
+ return gdb_get_reg32(mem_buf, 0);
case 25:
return gdb_get_reg32(mem_buf, env->slr);
case 26:
@@ -66,29 +82,52 @@ int mb_cpu_gdb_write_register(CPUState *cs, uint8_t *mem_buf, int n)
CPUMBState *env = &cpu->env;
uint32_t tmp;
+ /*
+ * GDB expects SREGs in the following order:
+ * PC, MSR, EAR, ESR, FSR, BTR, EDR, PID, ZPR, TLBX, TLBSX, TLBLO, TLBHI.
+ * They aren't stored in this order, so make a map.
+ * PID, ZPR, TLBx, TLBsx, TLBLO, and TLBHI aren't modeled, so we don't
+ * map them to anything.
+ */
+ static const uint8_t sreg_map[6] = {
+ SR_PC,
+ SR_MSR,
+ SR_EAR,
+ SR_ESR,
+ SR_FSR,
+ SR_BTR
+ };
+
if (n > cc->gdb_num_core_regs) {
return 0;
}
tmp = ldl_p(mem_buf);
+ /*
+ * GDB expects registers to be reported in this order:
+ * R0-R31
+ * PC-BTR
+ * PVR0-PVR11
+ * EDR-TLBHI
+ * SLR-SHR
+ */
if (n < 32) {
env->regs[n] = tmp;
} else {
n -= 32;
switch (n) {
case 0 ... 5:
- env->sregs[n] = tmp;
+ env->sregs[sreg_map[n]] = tmp;
break;
/* PVR12 is intentionally skipped */
case 6 ... 17:
n -= 6;
env->pvr.regs[n] = tmp;
break;
- case 18 ... 24:
- /* Add an offset of 6 to resume where we left off with SRegs */
- n = n - 18 + 6;
- env->sregs[n] = tmp;
+ /* Only EDR is modeled in these indeces, so ignore the rest */
+ case 18:
+ env->sregs[SR_EDR] = tmp;
break;
case 25:
env->slr = tmp;
--
2.7.4
next prev parent reply other threads:[~2020-05-11 17:22 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-11 17:05 [PATCH v1 0/4] target/microblaze: Add dynamic GDB XML and correct SReg reporting Joe Komlodi
2020-05-11 17:05 ` [PATCH v1 1/4] target/microblaze: gdb: Add dynamic GDB XML register support Joe Komlodi
2020-05-11 17:05 ` [PATCH v1 2/4] target/microblaze: gdb: Extend the number of registers presented to GDB Joe Komlodi
2020-05-11 17:05 ` Joe Komlodi [this message]
2020-05-11 17:05 ` [PATCH v1 4/4] target/microblaze: monitor: Increase the number of registers reported Joe Komlodi
2020-05-13 11:46 ` [PATCH v1 0/4] target/microblaze: Add dynamic GDB XML and correct SReg reporting Edgar E. Iglesias
2020-05-13 17:28 ` Edgar E. Iglesias
2020-05-13 17:56 ` Joe Komlodi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1589216751-106038-4-git-send-email-komlodi@xilinx.com \
--to=joe.komlodi@xilinx.com \
--cc=edgar.iglesias@xilinx.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).