From: Luis Machado <lgustavo@codesourcery.com>
To: Andreas Arnez <arnez@linux.vnet.ibm.com>
Cc: gdb-patches@sourceware.org, Ulrich.Weigand@de.ibm.com
Subject: Re: [RFA][PATCH v4 1/5] S/390 regmap rework
Date: Mon, 15 Jul 2013 14:46:00 -0000 [thread overview]
Message-ID: <51E40B28.3080605@codesourcery.com> (raw)
In-Reply-To: <87vc4rinoe.fsf@br87z6lw.de.ibm.com>
Hi,
A few nits and only nits.
On 07/03/2013 02:00 PM, Andreas Arnez wrote:
> S/390 regmap rework: Represent register maps in a less redundant and
> more readable way. Also remove some code repetition.
>
> /* GNU/Linux target descriptions. */
> extern struct target_desc *tdesc_s390_linux32;
> Index: gdb/gdb/s390-nat.c
> ===================================================================
> --- gdb.orig/gdb/s390-nat.c
> +++ gdb/gdb/s390-nat.c
> @@ -63,139 +63,129 @@ static int have_regset_system_call = 0;
>
> #define regmap_fpregset s390_regmap_fpregset
>
> -/* When debugging a 32-bit executable running under a 64-bit kernel,
> - we have to fix up the 64-bit registers we get from the kernel
> - to make them look like 32-bit registers. */
> +static void
> +s390_native_supply (struct regcache *regcache, const short *map,
> + const gdb_byte *regp)
> +{
> + for (; map[0] >= 0; map += 2)
> + regcache_raw_supply (regcache, map[1], regp + map[0]);
> +}
>
> static void
> -s390_native_supply (struct regcache *regcache, int regno,
> - const gdb_byte *regp, int *regmap)
> +s390_native_collect (const struct regcache *regcache, const short *map,
> + int regno, gdb_byte *regp)
> {
> - int offset = regmap[regno];
> + for (; map[0] >= 0; map += 2)
> + if (regno == -1 || regno == map[1])
> + regcache_raw_collect (regcache, map[1], regp + map[0]);
> +}
Maybe add a comment to these? Observing that...
> +
> +/* Fill GDB's register array with the general-purpose register values
> + in *REGP.
>
> + When debugging a 32-bit executable running under a 64-bit kernel,
> + we have to fix up the 64-bit registers we get from the kernel to
> + make them look like 32-bit registers. */
> +void
> +supply_gregset (struct regcache *regcache, const gregset_t *regp)
> +{
... you need a empty line between the comment of a function and the
declaration of that function.
> #ifdef __s390x__
> struct gdbarch *gdbarch = get_regcache_arch (regcache);
> - if (offset != -1 && gdbarch_ptr_bit (gdbarch) == 32)
> + if (gdbarch_ptr_bit (gdbarch) == 32)
> {
> enum bfd_endian byte_order = gdbarch_byte_order (gdbarch);
> + ULONGEST pswm = 0, pswa = 0;
> + gdb_byte buf[4];
> + const short *map;
>
> - if (regno == S390_PSWM_REGNUM)
> - {
> - ULONGEST pswm;
> - gdb_byte buf[4];
> -
> - pswm = extract_unsigned_integer (regp + regmap[S390_PSWM_REGNUM],
> - 8, byte_order);
> -
> - store_unsigned_integer (buf, 4, byte_order, (pswm >> 32) | 0x80000);
> - regcache_raw_supply (regcache, regno, buf);
> - return;
> - }
> -
> - if (regno == S390_PSWA_REGNUM)
> + for (map = regmap_gregset; map[0] >= 0; map += 2)
> {
> - ULONGEST pswm, pswa;
> - gdb_byte buf[4];
> + const gdb_byte *p = (const gdb_byte *) regp + map[0];
> + int regno = map[1];
>
> - pswa = extract_unsigned_integer (regp + regmap[S390_PSWA_REGNUM],
> - 8, byte_order);
> - pswm = extract_unsigned_integer (regp + regmap[S390_PSWM_REGNUM],
> - 8, byte_order);
> -
> - store_unsigned_integer (buf, 4, byte_order,
> - (pswa & 0x7fffffff) | (pswm & 0x80000000));
> - regcache_raw_supply (regcache, regno, buf);
> - return;
> + if (regno == S390_PSWM_REGNUM)
> + pswm = extract_unsigned_integer (p, 8, byte_order);
> + else if (regno == S390_PSWA_REGNUM)
> + pswa = extract_unsigned_integer (p, 8, byte_order);
> + else
> + {
> + if ((regno >= S390_R0_REGNUM && regno <= S390_R15_REGNUM)
> + || regno == S390_ORIG_R2_REGNUM)
> + p += 4;
> + regcache_raw_supply (regcache, regno, p);
> + }
> }
>
> - if ((regno >= S390_R0_REGNUM && regno <= S390_R15_REGNUM)
> - || regno == S390_ORIG_R2_REGNUM)
> - offset += 4;
> + store_unsigned_integer (buf, 4, byte_order, (pswm >> 32) | 0x80000);
> + regcache_raw_supply (regcache, S390_PSWM_REGNUM, buf);
> + store_unsigned_integer (buf, 4, byte_order,
> + (pswa & 0x7fffffff) | (pswm & 0x80000000));
> + regcache_raw_supply (regcache, S390_PSWA_REGNUM, buf);
> + return;
> }
> #endif
>
> - if (offset != -1)
> - regcache_raw_supply (regcache, regno, regp + offset);
> + s390_native_supply (regcache, regmap_gregset, (const gdb_byte *) regp);
> }
>
> -static void
> -s390_native_collect (const struct regcache *regcache, int regno,
> - gdb_byte *regp, int *regmap)
> +/* Fill register REGNO (if it is a general-purpose register) in
> + *REGP with the value in GDB's register array. If REGNO is -1,
> + do this for all registers. */
> +void
> +fill_gregset (const struct regcache *regcache, gregset_t *regp, int regno)
Another instance of a missing empty line after the comment.
next prev parent reply other threads:[~2013-07-15 14:46 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-03 16:57 [RFA][PATCH v4 0/5] Add TDB regset support Andreas Arnez
2013-07-03 17:00 ` [RFA][PATCH v4 1/5] S/390 regmap rework Andreas Arnez
2013-07-15 14:46 ` Luis Machado [this message]
2013-07-15 17:15 ` Andreas Arnez
2013-07-03 17:02 ` [RFA][PATCH v4 2/5] S/390: Add TDB regset Andreas Arnez
2013-07-03 19:25 ` Eli Zaretskii
2013-07-03 17:04 ` [RFA][PATCH v4 3/5] Dynamic core regset sections support Andreas Arnez
2013-07-03 17:11 ` [RFA][PATCH v4 4/5] S/390: Exploit dynamic core regset sections Andreas Arnez
2013-07-03 17:21 ` [RFA][PATCH v4 5/5] PowerPC: " Andreas Arnez
2013-07-08 15:44 ` [ping] [RFA][PATCH v4 0/5] Add TDB regset support Andreas Arnez
2013-07-15 8:49 ` [ping 2] " Andreas Arnez
2013-07-15 13:27 ` Luis Machado
2013-07-15 15:34 ` Andreas Arnez
2013-07-15 15:40 ` Luis Machado
2013-07-15 16:30 ` Andreas Arnez
2013-07-16 15:26 ` Andreas Arnez
2013-07-16 16:03 ` Ulrich Weigand
2013-07-16 17:06 ` Andreas Arnez
2013-07-15 15:58 ` Mark Kettenis
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=51E40B28.3080605@codesourcery.com \
--to=lgustavo@codesourcery.com \
--cc=Ulrich.Weigand@de.ibm.com \
--cc=arnez@linux.vnet.ibm.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