On 9/3/25 19:57, Simon Marchi wrote:
> On 9/2/25 8:27 AM, Gopi Kumar Bulusu wrote:
>>  From 7f2bacc6fffa6830991da36c2044431237bacba1 Mon Sep 17 00:00:00 2001
>> From: Gopi Kumar Bulusu <gopi@sankhya.com>
>> Date: Tue, 12 Aug 2025 09:42:48 +0530
>> Subject: [PATCH v2] MicroBlaze: Add microblaze_get_next_pcs
>>
>> This patch enables software single stepping for gdbserver target
>>
>> * gdb/microblaze-tdep.c: Add microblaze_get_next_pcs
>>
>> Signed-off-by: David Holsgrove <david.holsgrove@petalogix.com>
>> Signed-off-by: Nathan Rossi <nathan.rossi@petalogix.com>
>> Signed-off-by: Mahesh Bodapati <mbodapat@xilinx.com>
>> Signed-off-by: Gopi Kumar Bulusu <gopi@sankhya.com>
>> ---
>>   gdb/microblaze-tdep.c | 91 +++++++++++++++++++++++++++++++++++++++++++
>>   1 file changed, 91 insertions(+)
>>
>> diff --git a/gdb/microblaze-tdep.c b/gdb/microblaze-tdep.c
>> index 7b58220871c..cacec8ba186 100644
>> --- a/gdb/microblaze-tdep.c
>> +++ b/gdb/microblaze-tdep.c
>> @@ -590,6 +590,95 @@ microblaze_stabs_argument_has_addr (struct gdbarch *gdbarch, struct type *type)
>>     return (type->length () == 16);
>>   }
>>   
>> +/* Return next pc values : next in sequence and/or branch/return target.  */
>> +
>> +static std::vector<CORE_ADDR>
>> +microblaze_get_next_pcs (regcache *regcache)
>> +{
>> +  CORE_ADDR pc = regcache_read_pc (regcache);
>> +  long insn = microblaze_fetch_instruction (pc);
>> +
>> +  enum microblaze_instr_type insn_type;
>> +  short delay_slots;
>> +  bool isunsignednum;
>> +
>> +  /* If the current instruction is an imm, look at the inst after.  */
>> +
>> +  get_insn_microblaze (insn, &isunsignednum, &insn_type, &delay_slots);
>> +
>> +  int imm;
>> +  bool immfound = false;
>> +
>> +  if (insn_type == immediate_inst)
>> +    {
>> +      int rd, ra, rb;
>> +      immfound = true;
>> +      microblaze_decode_insn (insn, &rd, &ra, &rb, &imm);
>> +      pc += INST_WORD_SIZE;
>> +      insn = microblaze_fetch_instruction (pc);
>> +      get_insn_microblaze (insn, &isunsignednum, &insn_type, &delay_slots);
>> +    }
>> +
>> +  std::optional<CORE_ADDR> next_pc, branch_or_return_pc;
>> +
>> +  /* Compute next instruction address - skip delay slots if any.  */
>> +
>> +  if (insn_type != return_inst)
>> +    next_pc = pc + INST_WORD_SIZE + (delay_slots * INST_WORD_SIZE);
>> +
>> +  microblaze_debug ("single-step insn_type=0x%x pc=0x%lx insn=0x%lx\n",
>> +                insn_type, pc, insn);
>
> You shouldn't need the trailing \n.  Apply elsewhere too.
>
>> +
>> +  /* Compute target instruction address for branch or return instruction.  */
>> +  if (insn_type == branch_inst || insn_type == return_inst)
>> +    {
>> +      int limm;
>> +      int lrd, lra, lrb;
>> +      long ra, rb;
>> +      bool targetvalid;
>> +      bool unconditionalbranch;
>
> Declare when first use, where possible (ra and rb).
>
> ra and rb should probably be of the type ULONGEST.
>
>> +
>> +      microblaze_decode_insn (insn, &lrd, &lra, &lrb, &limm);
>> +      ra = regcache_raw_get_unsigned (regcache, lra);
>> +      rb = regcache_raw_get_unsigned (regcache, lrb);
>> +
>> +      branch_or_return_pc = microblaze_get_target_address (insn, immfound,
>> +                  imm, pc, ra, rb, &targetvalid, &unconditionalbranch);
>
> Format like this:
>
>        branch_or_return_pc
>       = microblaze_get_target_address (insn, immfound,
>                                        imm, pc, ra, rb, &targetvalid,
>                                        &unconditionalbranch);
>
>> +
>> +      microblaze_debug (
>> +                  "single-step uncondbr=%d targetvalid=%d target=0x%lx\n",
>> +                  unconditionalbranch, targetvalid,
>> +                  branch_or_return_pc.value () );
>
> Format like this:
>
>        microblaze_debug ("single-step uncondbr=%d targetvalid=%d target=0x%lx",
>                       unconditionalbranch, targetvalid,
>                       branch_or_return_pc.value ());
>
>> +
>> +      /* Can't reach next address.  */
>> +      if (unconditionalbranch)
>> +    next_pc.reset ();
>> +
>> +      /* Can't reach a distinct (not here) target address.  */
>> +      if (! targetvalid || branch_or_return_pc == pc ||
>> +                  (next_pc && (branch_or_return_pc == next_pc)))
>> +      branch_or_return_pc.reset ();
>
> Format like this:
>
>        /* Can't reach a distinct (not here) target address.  */
>        if (!targetvalid
>         || branch_or_return_pc == pc
>         || (next_pc.has_value () && branch_or_return_pc == next_pc))
>       branch_or_return_pc.reset ();
>
>
>> +    } /* if (branch or return instruction).  */
>> +
>> +  /* Create next_pcs vector to return.  */
>> +
>> +  std::vector<CORE_ADDR> next_pcs;
>> +
>> +  if (next_pc)
>
> .has_value ()
>
>> +    {
>> +      next_pcs.push_back (next_pc.value () );
>
> We typically use `*next_pc` to access the value (apply elsewhere too).
>
> Remove the space before closing parenthesis (apply elsewhere too).
>
>> +      microblaze_debug ("push_back next_pc(0x%lx)\n", next_pc.value () );
>> +    }
>> +
>> +  if (branch_or_return_pc)
>> +    {
>> +      next_pcs.push_back ( branch_or_return_pc.value () );
>> +      microblaze_debug ("push_back branch_or_return_pc(0x%lx)\n",
>> +                  branch_or_return_pc.value ());
>
> Align the last line properly.
>
> The patch LGTM with those fixed.  Let me know if you need help pushing
> the patch.
>

This caused a buildbreaker on arm-linux with --enable-targets=all and
--enable-64-bit-bfd.

I've filed a PR ( https://sourceware.org/bugzilla/show_bug.cgi?id=33381 ). 

Thanks,
- Tom

> Approved-By: Simon Marchi <simon.marchi@efficios.com>
>
> Simon