public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Iain Sandoe <idsandoe@googlemail.com>
To: HAO CHEN GUI <guihaoc@linux.ibm.com>
Cc: Segher Boessenkool <segher@kernel.crashing.org>,
	Peter Bergner <bergner@linux.ibm.com>,
	GCC Patches <gcc-patches@gcc.gnu.org>,
	David Edelsohn <dje.gcc@gmail.com>
Subject: Re: [PATCH v2, rs6000] Change insn condition from TARGET_64BIT to TARGET_POWERPC64 for VSX scalar extract/insert instructions
Date: Tue, 13 Sep 2022 08:08:38 +0100	[thread overview]
Message-ID: <FD3F7834-D185-4686-848F-B6C814D0C1D9@googlemail.com> (raw)
In-Reply-To: <1287a1dc-4b11-ce65-e6e6-ac6e6fff88c1@linux.ibm.com>

Hi


> On 13 Sep 2022, at 03:34, HAO CHEN GUI via Gcc-patches <gcc-patches@gcc.gnu.org> wrote:

> On 7/9/2022 上午 1:19, Segher Boessenkool wrote:
>> make -k -j60 check RUNTESTFLAGS="--target_board=unix'{-m64,-m32,-m32/-mpowerpc64}'"
>> 
>> It is fine to not test -m32/-mpowerpc64 so often, and certaionly not
>> something I will ask everyone to always do :-)
> 
> IMO, if we add "-mpowerpc64" into dg-options, the "-m32/-mpowerpc64" will be tested
> automatically. It will increase the test coverage. So the concern is it increases test
> time?

Do you mean for the powerpc-target tests, or for the entire suite?
 - 50% increase is pretty significant to the entire suite.  It would probably only be acceptable on
the newer powerful machines.

The stategy would need to be a little more subtle than adding it everywhere  - it is not applicable
to 32b targets running on 32b hardware - however, it *is* potentially applicable to 32b targets
running on 64b hardware (I have one of those).

====

There are tricks you can play with dejagnu to automate this for your setup… by pointing the 
DEJAGNU environment to a script (or by using .dejagnurc in your home directory).

You can add customisation in this way, even to set different parameters per host.  This saves
me from remembering to put the RUNTESTFLAGS=… extras for the target on each test line.

— so here’s an example as I use it..

I have an environment/script setup like this (which I use to control what is done across a range of
machines with capabilities that vary by at least a factor of 10):

DEJAGNU=/path/to/gcc-runtest-site.exp

gcc-runtest-site.exp:

global target_list
global tool_timeout

case "$target_triplet" in {
   { "*-*-darwin[5-8]*" } {
	set target_list { "unix" }
	set tool_timeout 240
	puts "tool timeout 3mins"
   }
   { "*-*-darwin9*" } {
	set target_list { "unix{-m32,-m64}" }
	set tool_timeout 240
   }
   { "i?86-*-darwin10*" } {
	set target_list { "unix" }
   }
   { "powerpc-*-darwin10*" } {
	set target_list { "unix" }
   }
   { "*-*-darwin1[01234567]*" } {
	set target_list { "unix{-m64,-m32}" }
	set tool_timeout 120
	puts "gcc-runtest-site : tool timeout 2 mins"
   }
   { "*-*-darwin1[89]*" } {
	set target_list { "unix" }
	set tool_timeout 60
	puts "gcc-runtest-site : tool timeout 1 min"
   }
   { "x86_64-*-linux-gnu" } {
	set target_list { "unix{-m32,-m64}" }
	set tool_timeout 60
   }

<SNIP - you get the idea, I am sure>

   default {
	# Works for most unixy things and for darwin < 9.
       set target_list { "unix" }
   }
}

- sometimes I override this with specific values (especially for timeout) in ~/..dejagnurc on
  very slow boxes.

====

0.02GBP only,

cheers
Iain


      reply	other threads:[~2022-09-13  7:08 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-02  8:31 HAO CHEN GUI
2022-09-02 15:56 ` Segher Boessenkool
2022-09-05  6:36   ` HAO CHEN GUI
2022-09-06 17:19     ` Segher Boessenkool
2022-09-07 13:51       ` Paul A. Clarke
2022-09-07 14:25         ` Segher Boessenkool
2022-09-08  5:59           ` HAO CHEN GUI
2022-09-09 17:17             ` Segher Boessenkool
2022-09-13  2:13               ` HAO CHEN GUI
2022-09-13  2:34       ` HAO CHEN GUI
2022-09-13  7:08         ` Iain Sandoe [this message]

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=FD3F7834-D185-4686-848F-B6C814D0C1D9@googlemail.com \
    --to=idsandoe@googlemail.com \
    --cc=bergner@linux.ibm.com \
    --cc=dje.gcc@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=guihaoc@linux.ibm.com \
    --cc=segher@kernel.crashing.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).