public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Luis Machado <luis.machado@arm.com>
To: <gdb-patches@sourceware.org>
Subject: [PATCH v2 01/17] [gdb/aarch64] Fix register fetch/store order for native AArch64 Linux
Date: Fri, 19 May 2023 11:24:52 +0100	[thread overview]
Message-ID: <20230519102508.14020-2-luis.machado@arm.com> (raw)
In-Reply-To: <20230519102508.14020-1-luis.machado@arm.com>

I noticed we don't handle register reads/writes in the best way for native
AArch64 Linux.  Some registers are fetched/stored even if upper level code
told us to fetch a particular register number.

Fix this by being more strict about which registers we touch when
reading/writing them in the native AArch64 Linux layer.

There should be no user-visible changes due to this patch.

Regression-tested on aarch64-linux Ubuntu 22.04/20.04.
---
 gdb/aarch64-linux-nat.c | 60 ++++++++++++++++++++++-------------------
 1 file changed, 33 insertions(+), 27 deletions(-)

diff --git a/gdb/aarch64-linux-nat.c b/gdb/aarch64-linux-nat.c
index ecb2eeb9540..62656a7347b 100644
--- a/gdb/aarch64-linux-nat.c
+++ b/gdb/aarch64-linux-nat.c
@@ -504,6 +504,7 @@ aarch64_fetch_registers (struct regcache *regcache, int regno)
   aarch64_gdbarch_tdep *tdep
     = gdbarch_tdep<aarch64_gdbarch_tdep> (regcache->arch ());
 
+  /* Do we need to fetch all registers?  */
   if (regno == -1)
     {
       fetch_gregs_from_thread (regcache);
@@ -521,28 +522,28 @@ aarch64_fetch_registers (struct regcache *regcache, int regno)
       if (tdep->has_tls ())
 	fetch_tlsregs_from_thread (regcache);
     }
+  /* General purpose register?  */
   else if (regno < AARCH64_V0_REGNUM)
     fetch_gregs_from_thread (regcache);
-  else if (tdep->has_sve ())
+  /* SVE register?  */
+  else if (tdep->has_sve () && regno <= AARCH64_SVE_VG_REGNUM)
     fetch_sveregs_from_thread (regcache);
-  else
+  /* FPSIMD register?  */
+  else if (regno < AARCH64_FPCR_REGNUM)
     fetch_fpregs_from_thread (regcache);
-
-  if (tdep->has_pauth ())
-    {
-      if (regno == AARCH64_PAUTH_DMASK_REGNUM (tdep->pauth_reg_base)
-	  || regno == AARCH64_PAUTH_CMASK_REGNUM (tdep->pauth_reg_base))
-	fetch_pauth_masks_from_thread (regcache);
-    }
-
-  /* Fetch individual MTE registers.  */
-  if (tdep->has_mte ()
-      && (regno == tdep->mte_reg_base))
+  /* PAuth register?  */
+  else if (tdep->has_pauth ()
+	   && (regno == AARCH64_PAUTH_DMASK_REGNUM (tdep->pauth_reg_base)
+	       || regno == AARCH64_PAUTH_CMASK_REGNUM (tdep->pauth_reg_base)))
+    fetch_pauth_masks_from_thread (regcache);
+  /* MTE register?  */
+  else if (tdep->has_mte ()
+	   && (regno == tdep->mte_reg_base))
     fetch_mteregs_from_thread (regcache);
-
-  if (tdep->has_tls ()
-      && regno >= tdep->tls_regnum_base
-      && regno < tdep->tls_regnum_base + tdep->tls_register_count)
+  /* TLS register?  */
+  else if (tdep->has_tls ()
+	   && regno >= tdep->tls_regnum_base
+	   && regno < tdep->tls_regnum_base + tdep->tls_register_count)
     fetch_tlsregs_from_thread (regcache);
 }
 
@@ -592,6 +593,7 @@ aarch64_store_registers (struct regcache *regcache, int regno)
   aarch64_gdbarch_tdep *tdep
     = gdbarch_tdep<aarch64_gdbarch_tdep> (regcache->arch ());
 
+  /* Do we need to store all registers?  */
   if (regno == -1)
     {
       store_gregs_to_thread (regcache);
@@ -606,22 +608,26 @@ aarch64_store_registers (struct regcache *regcache, int regno)
       if (tdep->has_tls ())
 	store_tlsregs_to_thread (regcache);
     }
+  /* General purpose register?  */
   else if (regno < AARCH64_V0_REGNUM)
     store_gregs_to_thread (regcache);
-  else if (tdep->has_sve ())
+  /* SVE register?  */
+  else if (tdep->has_sve () && regno <= AARCH64_SVE_VG_REGNUM)
     store_sveregs_to_thread (regcache);
-  else
+  /* FPSIMD register?  */
+  else if (regno < AARCH64_FPCR_REGNUM)
     store_fpregs_to_thread (regcache);
-
-  /* Store MTE registers.  */
-  if (tdep->has_mte ()
-      && (regno == tdep->mte_reg_base))
+  /* MTE register?  */
+  else if (tdep->has_mte ()
+	   && (regno == tdep->mte_reg_base))
     store_mteregs_to_thread (regcache);
-
-  if (tdep->has_tls ()
-      && regno >= tdep->tls_regnum_base
-      && regno < tdep->tls_regnum_base + tdep->tls_register_count)
+  /* TLS register?  */
+  else if (tdep->has_tls ()
+	   && regno >= tdep->tls_regnum_base
+	   && regno < tdep->tls_regnum_base + tdep->tls_register_count)
     store_tlsregs_to_thread (regcache);
+
+  /* PAuth registers are read-only.  */
 }
 
 /* A version of the "store_registers" target_ops method used when running
-- 
2.25.1


  reply	other threads:[~2023-05-19 10:25 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-19 10:24 [PATCH v2 00/17] SME support for AArch64 gdb/gdbserver on Linux Luis Machado
2023-05-19 10:24 ` Luis Machado [this message]
2023-05-19 10:24 ` [PATCH v2 02/17] [gdb/aarch64] refactor: Rename SVE-specific files Luis Machado
2023-05-19 10:24 ` [PATCH v2 03/17] [gdb/gdbserver] refactor: Simplify SVE interface to read/write registers Luis Machado
2023-05-19 10:24 ` [PATCH v2 04/17] [gdb/aarch64] sve: Fix return command when using V registers in a SVE-enabled target Luis Machado
2023-05-19 10:24 ` [PATCH v2 05/17] [gdb/aarch64] sme: Enable SME registers and pseudo-registers Luis Machado
2023-05-19 10:24 ` [PATCH v2 06/17] [gdbserver/aarch64] refactor: Adjust expedited registers dynamically Luis Machado
2023-05-19 10:24 ` [PATCH v2 07/17] [gdbserver/aarch64] sme: Add support for SME Luis Machado
2023-05-19 10:24 ` [PATCH v2 08/17] [gdb/aarch64] sve: Fix signal frame z/v register restore Luis Machado
2023-05-19 10:25 ` [PATCH v2 09/17] [gdb/aarch64] sme: Signal frame support Luis Machado
2023-05-19 10:25 ` [PATCH v2 10/17] [gdb/aarch64] sme: Fixup sigframe gdbarch when vg/svg changes Luis Machado
2023-05-19 10:25 ` [PATCH v2 11/17] [gdb/aarch64] sme: Support TPIDR2 signal frame context Luis Machado
2023-05-19 10:25 ` [PATCH v2 12/17] [binutils/aarch64] sme: Core file support Luis Machado
2023-05-19 10:25 ` [PATCH v2 13/17] [gdb/generic] corefile/bug: Use thread-specific gdbarch when dumping register state to core files Luis Machado
2023-05-19 10:25 ` [PATCH v2 14/17] [gdb/generic] corefile/bug: Fixup (gcore) core file target description reading order Luis Machado
2023-05-19 10:25 ` [PATCH v2 15/17] [gdb/aarch64] sme: Core file support for Linux Luis Machado
2023-05-19 10:25 ` [PATCH v2 16/17] [gdb/testsuite] sme: Add SVE/SME testcases Luis Machado
2023-05-19 10:25 ` [PATCH v3 17/17] [gdb/docs] sme: Document SME registers and features Luis Machado
2023-05-19 11:20   ` Eli Zaretskii
2023-06-30 12:10     ` Luis Machado

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=20230519102508.14020-2-luis.machado@arm.com \
    --to=luis.machado@arm.com \
    --cc=gdb-patches@sourceware.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).