public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Luis Machado <luis.machado@arm.com>
To: Thiago Jung Bauermann <thiago.bauermann@linaro.org>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH v2 2/6] [gdbserver/aarch64] sme2: Enable SME2 support in gdbserver
Date: Wed, 30 Aug 2023 13:54:30 +0100	[thread overview]
Message-ID: <c5fc5433-53a0-afed-d267-296bf3fb02dc@arm.com> (raw)
In-Reply-To: <87jzthl1uv.fsf@linaro.org>

On 8/27/23 02:18, Thiago Jung Bauermann wrote:
> 
> Luis Machado <luis.machado@arm.com> writes:
> 
>> This patch teaches gdbserver about the SME2 and the ZT0 register.
>>
>> Since most of the code used by gdbserver for SME2 is shared with gdb, this
>> is a rather small patch that reuses most of the code put in place for native
>> AArch64 Linux.
>>
>> Validated under Fast Models.
>> ---
>>  gdbserver/linux-aarch64-low.cc | 63 ++++++++++++++++++++++++++++++++++
>>  1 file changed, 63 insertions(+)
>>
>> diff --git a/gdbserver/linux-aarch64-low.cc b/gdbserver/linux-aarch64-low.cc
>> index 95688480c3d..783053e359b 100644
>> --- a/gdbserver/linux-aarch64-low.cc
>> +++ b/gdbserver/linux-aarch64-low.cc
>> @@ -811,6 +811,52 @@ aarch64_za_regs_copy_from_regcache (struct regcache *regcache, void *buf)
>>    memcpy (buf, za_state.data (), za_state.size ());
>>  }
>>  
>> +/* Wrapper for aarch64_zt_regs_copy_to_reg_buf, to help copying NT_ARM_ZT
>> +   state from the thread (BUF) to the register cache.  */
>> +
>> +static void
>> +aarch64_zt_regs_copy_to_regcache (struct regcache *regcache,
>> +				  ATTRIBUTE_UNUSED const void *buf)
>> +{
>> +  /* BUF is unused here since we collect the data straight from a ptrace
>> +     request, therefore bypassing gdbserver's own call to ptrace.  */
>> +  int tid = lwpid_of (current_thread);
>> +
>> +  gdb::optional<int> zt_regnum
>> +    = find_regno_no_throw (regcache->tdesc, "zt0");
>> +
>> +  gdb_assert (zt_regnum.has_value ());
> 
> If this code is changed to use find_regno(), this gdb_assert() becomes
> unnecessary.
> 

Indeed. Fixed now.

>> +
>> +  /* Update the register cache.  aarch64_zt_regs_copy_to_reg_buf handles
>> +     fetching the NT_ARM_ZT state from thread TID.  */
>> +  aarch64_zt_regs_copy_to_reg_buf (tid, regcache, *zt_regnum);
>> +}
>> +
>> +/* Wrapper for aarch64_zt_regs_copy_from_reg_buf, to help copying NT_ARM_ZT
>> +   state from the register cache to the thread (BUF).  */
>> +
>> +static void
>> +aarch64_zt_regs_copy_from_regcache (struct regcache *regcache, void *buf)
>> +{
>> +  int tid = lwpid_of (current_thread);
>> +
>> +  gdb::optional<int> zt_regnum
>> +    = find_regno_no_throw (regcache->tdesc, "za");
>> +
>> +  gdb_assert (zt_regnum.has_value ());
> 
> Same comment about find_regno().
> 

Fixed.

>> +
>> +  /* Update the thread NT_ARM_ZT state.  aarch64_zt_regs_copy_from_reg_buf
>> +     handles writing the ZT state back to thread TID.  */
>> +  aarch64_zt_regs_copy_from_reg_buf (tid, regcache, *zt_regnum);
>> +
>> +  /* We need to return the expected data in BUF, so copy whatever the kernel
>> +     already has to BUF.  */
>> +
>> +  /* Obtain a dump of NT_ARM_ZT from ptrace.  */
>> +  gdb::byte_vector zt_state = aarch64_fetch_zt_regset (tid);
>> +  memcpy (buf, zt_state.data (), zt_state.size ());
>> +}
>> +
>>  /* Array containing all the possible register sets for AArch64/Linux.  During
>>     architecture setup, these will be checked against the HWCAP/HWCAP2 bits for
>>     validity and enabled/disabled accordingly.
>> @@ -838,6 +884,11 @@ static struct regset_info aarch64_regsets[] =
>>      0, EXTENDED_REGS,
>>      aarch64_za_regs_copy_from_regcache, aarch64_za_regs_copy_to_regcache
>>    },
>> +  /* Scalable Matrix Extension 2 (SME2) ZT registers.  */
>> +  { PTRACE_GETREGSET, PTRACE_SETREGSET, NT_ARM_ZT,
>> +    0, EXTENDED_REGS,
>> +    aarch64_zt_regs_copy_from_regcache, aarch64_zt_regs_copy_to_regcache
>> +  },
>>    /* PAC registers.  */
>>    { PTRACE_GETREGSET, PTRACE_SETREGSET, NT_ARM_PAC_MASK,
>>      0, OPTIONAL_REGS,
>> @@ -909,6 +960,10 @@ aarch64_adjust_register_sets (const struct aarch64_features &features)
>>  	  if (features.svq > 0)
>>  	    regset->size = ZA_PT_SIZE (features.svq);
>>  	  break;
>> +	case NT_ARM_ZT:
>> +	  if (features.sme2)
>> +	    regset->size = 64;
> 
> Suggest using AARCH64_SME2_ZT0_SIZE here.
> 

Yeah. I think I wrote this before I defined the size. Changed now.

>> +	  break;
>>  	default:
>>  	  gdb_assert_not_reached ("Unknown register set found.");
>>  	}
>> @@ -947,6 +1002,14 @@ aarch64_target::low_arch_setup ()
>>        if (linux_get_hwcap2 (pid, 8) & HWCAP2_SME)
>>  	features.svq = aarch64_za_get_svq (tid);
>>  
>> +      /* Scalable Matrix Extension 2 feature check.  */
>> +      CORE_ADDR hwcap2 = linux_get_hwcap2 (pid, 8);
>> +      if ((hwcap2 & HWCAP2_SME2) || (hwcap2 & HWCAP2_SME2P1))
>> +	{
>> +	  /* Make sure ptrace supports NT_ARM_ZT.  */
>> +	  features.sme2 = supports_zt_registers (tid);
>> +	}
>> +
>>        current_process ()->tdesc = aarch64_linux_read_description (features);
>>  
>>        /* Adjust the register sets we should use for this particular set of
> 
> Reviewed-by: Thiago Jung Bauermann <thiago.bauermann@linaro.org>
> 


  reply	other threads:[~2023-08-30 12:54 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-22 11:21 [PATCH v2 0/6] SME2 support for AArch64 gdb/gdbserver on Linux Luis Machado
2023-08-22 11:21 ` [PATCH v2 1/6] [gdb/aarch64] sme2: Enable SME2 for AArch64 gdb " Luis Machado
2023-08-27  1:16   ` Thiago Jung Bauermann
2023-08-30 12:53     ` Luis Machado
2023-08-22 11:21 ` [PATCH v2 2/6] [gdbserver/aarch64] sme2: Enable SME2 support in gdbserver Luis Machado
2023-08-27  1:18   ` Thiago Jung Bauermann
2023-08-30 12:54     ` Luis Machado [this message]
2023-08-22 11:21 ` [PATCH v2 3/6] [gdb/aarch64] sme2: signal frame support Luis Machado
2023-08-27  1:19   ` Thiago Jung Bauermann
2023-08-22 11:21 ` [PATCH v2 4/6] [gdb/aarch64] sme2: Core file support for ZT register set Luis Machado
2023-08-27  1:21   ` Thiago Jung Bauermann
2023-08-30 12:56     ` Luis Machado
2023-08-22 11:21 ` [PATCH v2 5/6] [gdb/testsuite] sme2: Extend SME tests to include SME2 Luis Machado
2023-08-27  1:23   ` Thiago Jung Bauermann
2023-08-30 12:56     ` Luis Machado
2023-08-22 11:21 ` [PATCH v2 6/6] [gdb/docs] sme2: Document SME2 registers and features Luis Machado
2023-08-22 11:34   ` Eli Zaretskii
2023-08-22 11:45     ` Luis Machado
2023-08-27  1:27   ` Thiago Jung Bauermann

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=c5fc5433-53a0-afed-d267-296bf3fb02dc@arm.com \
    --to=luis.machado@arm.com \
    --cc=gdb-patches@sourceware.org \
    --cc=thiago.bauermann@linaro.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 read-only IMAP folder(s) and NNTP newsgroup(s).