Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
 help / color / mirror / Atom feed
From: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: [FFmpeg-devel] [PATCH 1/2] avcodec/exr: Remove write-only gamma_table
Date: Wed, 5 Mar 2025 14:51:39 +0100
Message-ID: <AS8P250MB0744CBD15DBC1DBAF39C2F1D8FCB2@AS8P250MB0744.EURP250.PROD.OUTLOOK.COM> (raw)

[-- Attachment #1: Type: text/plain, Size: 102 bytes --]

Patches attached. None of which are meant to address the fate failure
mentioned by Martin.

- Andreas

[-- Attachment #2: 0001-avcodec-exr-Remove-write-only-gamma_table.patch --]
[-- Type: text/x-patch, Size: 2331 bytes --]

From d1a59c3fa3101192256bf67484b0cc9d5f2967b2 Mon Sep 17 00:00:00 2001
From: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Date: Wed, 5 Mar 2025 13:57:16 +0100
Subject: [PATCH 1/2] avcodec/exr: Remove write-only gamma_table

Forgotten in 0e917389fe73c932049635d947bba076f1709589.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
---
 libavcodec/exr.c | 31 -------------------------------
 1 file changed, 31 deletions(-)

diff --git a/libavcodec/exr.c b/libavcodec/exr.c
index 7e8020a0ff..8b673046c1 100644
--- a/libavcodec/exr.c
+++ b/libavcodec/exr.c
@@ -190,7 +190,6 @@ typedef struct EXRContext {
 
     enum AVColorTransferCharacteristic apply_trc_type;
     float gamma;
-    union av_intfloat32 gamma_table[65536];
 
     uint8_t *offset_table;
 
@@ -2240,10 +2239,6 @@ static int decode_frame(AVCodecContext *avctx, AVFrame *picture,
 static av_cold int decode_init(AVCodecContext *avctx)
 {
     EXRContext *s = avctx->priv_data;
-    uint32_t i;
-    union av_intfloat32 t;
-    float one_gamma = 1.0f / s->gamma;
-    av_csp_trc_function trc_func = NULL;
 
     ff_init_half2float_tables(&s->h2f_tables);
 
@@ -2255,32 +2250,6 @@ static av_cold int decode_init(AVCodecContext *avctx)
     ff_bswapdsp_init(&s->bbdsp);
 #endif
 
-    trc_func = av_csp_trc_func_from_id(s->apply_trc_type);
-    if (trc_func) {
-        for (i = 0; i < 65536; ++i) {
-            t.i = half2float(i, &s->h2f_tables);
-            t.f = trc_func(t.f);
-            s->gamma_table[i] = t;
-        }
-    } else {
-        if (one_gamma > 0.9999f && one_gamma < 1.0001f) {
-            for (i = 0; i < 65536; ++i) {
-                s->gamma_table[i].i = half2float(i, &s->h2f_tables);
-            }
-        } else {
-            for (i = 0; i < 65536; ++i) {
-                t.i = half2float(i, &s->h2f_tables);
-                /* If negative value we reuse half value */
-                if (t.f <= 0.0f) {
-                    s->gamma_table[i] = t;
-                } else {
-                    t.f = powf(t.f, one_gamma);
-                    s->gamma_table[i] = t;
-                }
-            }
-        }
-    }
-
     // allocate thread data, used for non EXR_RAW compression types
     s->thread_data = av_calloc(avctx->thread_count, sizeof(*s->thread_data));
     if (!s->thread_data)
-- 
2.45.2


[-- Attachment #3: 0002-avcodec-exr-Fix-potential-effective-type-violation.patch --]
[-- Type: text/x-patch, Size: 3393 bytes --]

From a0f2d79445366fda3b485e701c84e69d11148259 Mon Sep 17 00:00:00 2001
From: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Date: Wed, 5 Mar 2025 14:19:13 +0100
Subject: [PATCH 2/2] avcodec/exr: Fix potential effective-type violation

Storing the values via a union of an uint32_t and a float makes
said union the effective type of the destination. This means that
it may only be read via such a union which is of course not what
our users do/expect. So store the values via AV_WN32A instead
which disables effective type analysis (for compilers that perform it).

This also fixes a -Wdeclaration-after-statement warning
introduced in 0e917389fe73c932049635d947bba076f1709589.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
---
 libavcodec/exr.c | 29 ++++++++++++-----------------
 1 file changed, 12 insertions(+), 17 deletions(-)

diff --git a/libavcodec/exr.c b/libavcodec/exr.c
index 8b673046c1..1c753edb9f 100644
--- a/libavcodec/exr.c
+++ b/libavcodec/exr.c
@@ -1383,34 +1383,29 @@ static int decode_block(AVCodecContext *avctx, void *tdata,
                     s->compression == EXR_DWAA ||
                     s->compression == EXR_DWAB) {
                     // 32-bit
-                    union av_intfloat32 *ptr_x;
+                    uint8_t *ptr_x = ptr;
 
                     src = channel_buffer[c];
-                    ptr_x = (union av_intfloat32 *)ptr;
 
                     // Zero out the start if xmin is not 0
                     memset(ptr_x, 0, bxmin);
-                    ptr_x += window_xoffset;
+                    ptr_x += 4 * window_xoffset;
 
-                    union av_intfloat32 t;
                     if (trc_func && c < 3) {
-                        for (x = 0; x < xsize; x++) {
-                            t.i = bytestream_get_le32(&src);
-                            t.f = trc_func(t.f);
-                            *ptr_x++ = t;
+                        for (int x = 0; x < xsize; x++, ptr_x += 4) {
+                            float f = av_int2float(bytestream_get_le32(&src));
+                            AV_WN32A(ptr_x, av_float2int(trc_func(f)));
                         }
                     } else if (one_gamma != 1.f) {
-                        for (x = 0; x < xsize; x++) {
-                            t.i = bytestream_get_le32(&src);
-                            if (t.f > 0.0f && c < 3)  /* avoid negative values */
-                                t.f = powf(t.f, one_gamma);
-                            *ptr_x++ = t;
+                        for (int x = 0; x < xsize; x++, ptr_x += 4) {
+                            float f = av_int2float(bytestream_get_le32(&src));
+                            if (f > 0.0f && c < 3)  /* avoid negative values */
+                                f = powf(f, one_gamma);
+                            AV_WN32A(ptr_x, av_float2int(f));
                         }
                     } else {
-                        for (x = 0; x < xsize; x++) {
-                            t.i = bytestream_get_le32(&src);
-                            *ptr_x++ = t;
-                        }
+                        for (int x = 0; x < xsize; x++, ptr_x += 4)
+                            AV_WN32A(ptr_x, bytestream_get_le32(&src));
                     }
                     memset(ptr_x, 0, axmax);
                 } else if (s->pixel_type == EXR_HALF) {
-- 
2.45.2


[-- Attachment #4: Type: text/plain, Size: 251 bytes --]

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".

             reply	other threads:[~2025-03-05 13:51 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-05 13:51 Andreas Rheinhardt [this message]
2025-03-05 16:07 ` Michael Niedermayer

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=AS8P250MB0744CBD15DBC1DBAF39C2F1D8FCB2@AS8P250MB0744.EURP250.PROD.OUTLOOK.COM \
    --to=andreas.rheinhardt@outlook.com \
    --cc=ffmpeg-devel@ffmpeg.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

Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \
		ffmpegdev@gitmailbox.com
	public-inbox-index ffmpegdev

Example config snippet for mirrors.


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git