All of lore.kernel.org
 help / color / mirror / Atom feed
From: frederic.recoules@univ-grenoble-alpes.fr
To: alsa-devel@alsa-project.org
Cc: "Frédéric Recoules" <frederic.recoules@orange.fr>
Subject: [PATCH v3 3/5] pcm_dmix assembly: fix the interface for safety (mmx)
Date: Wed,  6 May 2020 19:19:22 +0200	[thread overview]
Message-ID: <20200506171924.2644-3-frederic.recoules@univ-grenoble-alpes.fr> (raw)
In-Reply-To: <20200506171924.2644-1-frederic.recoules@univ-grenoble-alpes.fr>

From: Frédéric Recoules <frederic.recoules@orange.fr>

- add mm0 in the clobber list if the compiler
  is aware of the mmx technology;
- otherwise, add the mmx aliased x87 floating point
  registers in the clobbers;
- the configure now checks if the compiler is aware of
  the MMX technology.

  The compiler assumes none of the mmx or x87 registers are used
  by the function. If it chooses to store some data in them, they
  will be overwritten by the chunk.
  Recall that any mmx instruction invalidate the whole set of
  x87 floating point registers.

  Note: currently does not impact the binary output.

Signed-off-by: Frédéric Recoules <frederic.recoules@orange.fr>
---
 configure.ac              | 7 +++++++
 src/pcm/pcm_dmix_i386.h   | 6 ++++++
 src/pcm/pcm_dmix_x86_64.h | 6 ++++++
 3 files changed, 19 insertions(+)

diff --git a/configure.ac b/configure.ac
index 4b5ab662..1838e50b 100644
--- a/configure.ac
+++ b/configure.ac
@@ -516,6 +516,13 @@ if test -z "$gcc_have_atomics"; then
 fi
 AC_MSG_RESULT($gcc_have_atomics)
 
+dnl check mmx register for pcm_dmix_i386
+
+AC_TRY_LINK([],
+    [__asm__ volatile ("" : : : "mm0");],
+    [AC_DEFINE([HAVE_MMX], "1", [MMX technology is enabled])],
+    [])
+
 PCM_PLUGIN_LIST="copy linear route mulaw alaw adpcm rate plug multi shm file null empty share meter hooks lfloat ladspa dmix dshare dsnoop asym iec958 softvol extplug ioplug mmap_emul"
 
 build_pcm_plugin="no"
diff --git a/src/pcm/pcm_dmix_i386.h b/src/pcm/pcm_dmix_i386.h
index 3ea0737d..5c900bf1 100644
--- a/src/pcm/pcm_dmix_i386.h
+++ b/src/pcm/pcm_dmix_i386.h
@@ -237,6 +237,12 @@ static void MIX_AREAS_16_MMX(unsigned int size,
 		  [dst_step] "m" (dst_step),  [src_step] "m" (src_step),
 		  [sum_step] "m" (sum_step)
 		: "esi", "edi", "edx", "ecx", "eax", "memory", "cc"
+#ifdef HAVE_MMX
+		  , "mm0"
+#else
+		  , "st", "st(1)", "st(2)", "st(3)",
+		  "st(4)", "st(5)", "st(6)", "st(7)"
+#endif
 	);
 }
 
diff --git a/src/pcm/pcm_dmix_x86_64.h b/src/pcm/pcm_dmix_x86_64.h
index 7f711547..803d3b24 100644
--- a/src/pcm/pcm_dmix_x86_64.h
+++ b/src/pcm/pcm_dmix_x86_64.h
@@ -119,6 +119,12 @@ static void MIX_AREAS_16(unsigned int size,
 		  [dst_step] "m" (dst_step),  [src_step] "m" (src_step),
 		  [sum_step] "m" (sum_step)
 		: "rsi", "rdi", "edx", "ecx", "eax", "memory", "cc"
+#ifdef HAVE_MMX
+		  , "mm0"
+#else
+		  , "st", "st(1)", "st(2)", "st(3)",
+		  "st(4)", "st(5)", "st(6)", "st(7)"
+#endif
 	);
 }
 
-- 
2.17.1


  parent reply	other threads:[~2020-05-06 17:23 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-06 17:19 [PATCH v3 1/5] pcm_dmix assembly: change the token by symbolic names frederic.recoules
2020-05-06 17:19 ` [PATCH v3 2/5] pcm_dmix assembly: fix the interface for safety frederic.recoules
2020-05-06 17:19 ` frederic.recoules [this message]
2020-05-06 17:19 ` [PATCH v3 4/5] pcm_dmix assembly: refactor the PIC mode frederic.recoules
2020-05-06 17:19 ` [PATCH v3 5/5] pcm_dmix assembly: give more freedom over the constraints frederic.recoules
2020-05-06 20:36 ` [PATCH v3 1/5] pcm_dmix assembly: change the token by symbolic names Takashi Iwai
2020-05-14 17:33   ` FRÉDÉRIC RECOULES

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=20200506171924.2644-3-frederic.recoules@univ-grenoble-alpes.fr \
    --to=frederic.recoules@univ-grenoble-alpes.fr \
    --cc=alsa-devel@alsa-project.org \
    --cc=frederic.recoules@orange.fr \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.