public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Andrew Burgess <aburgess@redhat.com>
To: "Maciej W. Rozycki" <macro@embecosm.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH 2/6] Avoid redundant operations in `fortran_array_walker'
Date: Wed, 15 Dec 2021 13:19:00 +0000	[thread overview]
Message-ID: <20211215131900.GJ175541@redhat.com> (raw)
In-Reply-To: <alpine.DEB.2.20.2112092225040.10183@tpp.orcam.me.uk>

* Maciej W. Rozycki <macro@embecosm.com> [2021-12-11 11:47:14 +0000]:

> Move inner dimension's element type determination outside the respective 
> loops in `fortran_array_walker'.  The operation is exactly the same with 
> each iteration, so there is no point in redoing it for each element and 
> while a smart compiler might be able to move it outside the loop it is 
> regardless a bad coding style.  No functional change.

LGTM.

Thanks,
Andrew


> ---
>  gdb/f-array-walker.h |    7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> gdb-fortran-array-walker-walk-type.diff
> Index: src/gdb/f-array-walker.h
> ===================================================================
> --- src.orig/gdb/f-array-walker.h
> +++ src/gdb/f-array-walker.h
> @@ -208,6 +208,8 @@ class fortran_array_walker
>  
>      if (nss != m_ndimensions)
>        {
> +	struct type *subarray_type = TYPE_TARGET_TYPE (check_typedef (type));
> +
>  	/* For dimensions other than the inner most, walk each element and
>  	   recurse while peeling off one more dimension of the array.  */
>  	for (LONGEST i = lowerbound;
> @@ -218,13 +220,13 @@ class fortran_array_walker
>  	    LONGEST new_offset = offset + calc.index_offset (i);
>  
>  	    /* Now print the lower dimension.  */
> -	    struct type *subarray_type
> -	      = TYPE_TARGET_TYPE (check_typedef (type));
>  	    walk_1 (nss + 1, subarray_type, new_offset, (i == upperbound));
>  	  }
>        }
>      else
>        {
> +	struct type *elt_type = check_typedef (TYPE_TARGET_TYPE (type));
> +
>  	/* For the inner most dimension of the array, process each element
>  	   within this dimension.  */
>  	for (LONGEST i = lowerbound;
> @@ -233,7 +235,6 @@ class fortran_array_walker
>  	  {
>  	    LONGEST elt_off = offset + calc.index_offset (i);
>  
> -	    struct type *elt_type = check_typedef (TYPE_TARGET_TYPE (type));
>  	    if (is_dynamic_type (elt_type))
>  	      {
>  		CORE_ADDR e_address = m_address + elt_off;
> 


  reply	other threads:[~2021-12-15 13:19 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-11 11:46 [PATCH 0/6] Make Fortran support respect more `set print' settings Maciej W. Rozycki
2021-12-11 11:47 ` [PATCH 1/6] Initialize `m_ndimensions' in the member initializer list Maciej W. Rozycki
2021-12-15 13:18   ` Andrew Burgess
2021-12-17 15:03     ` Maciej W. Rozycki
2021-12-11 11:47 ` [PATCH 2/6] Avoid redundant operations in `fortran_array_walker' Maciej W. Rozycki
2021-12-15 13:19   ` Andrew Burgess [this message]
2021-12-17 15:04     ` Maciej W. Rozycki
2021-12-11 11:47 ` [PATCH 3/6] Respect `set print repeats' with Fortran arrays Maciej W. Rozycki
2021-12-15 15:18   ` Andrew Burgess
2022-01-08 16:25     ` Maciej W. Rozycki
2021-12-11 11:47 ` [PATCH 4/6] Add `set print repeats' tests for C/C++ arrays Maciej W. Rozycki
2021-12-15 15:33   ` Andrew Burgess
2022-01-08 16:26     ` Maciej W. Rozycki
2021-12-11 11:47 ` [PATCH 5/6] Respect `set print array-indexes' with Fortran arrays Maciej W. Rozycki
2021-12-15 16:49   ` Andrew Burgess
2022-01-08 16:27     ` Maciej W. Rozycki
2021-12-11 11:48 ` [PATCH 6/6] Add `set print array-indexes' tests for C/C++ arrays Maciej W. Rozycki
2021-12-15 16:51   ` Andrew Burgess
2022-01-08 16:27     ` Maciej W. Rozycki

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=20211215131900.GJ175541@redhat.com \
    --to=aburgess@redhat.com \
    --cc=gdb-patches@sourceware.org \
    --cc=macro@embecosm.com \
    /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).