[rpms/chromium] epel9-next: Backport upstream, fix Print preview almost fails and prevents printing

[rpms/chromium] epel9-next: Backport upstream, fix Print preview almost fails and prevents printing

[Than Ngo]

A new commit has been pushed.

Repo   : rpms/chromium
Branch : epel9-next
Commit : 09bc163b2e823a3d430e1c048ef50e30c2265385
Author : Than Ngo <than@redhat.com>
Date   : 2025-04-14T10:41:17+02:00
Stats  : +47/-0 in 2 file(s)
URL    : https://src.fedoraproject.org/rpms/chromium/c/09bc163b2e823a3d430e1c048ef50e30c2265385?branch=epel9-next

Log:
Backport upstream, fix Print preview almost fails and prevents printing

---
diff --git a/chromium-135-print-review-fail.patch b/chromium-135-print-review-fail.patch
new file mode 100644
index 0000000..478b3dc
--- /dev/null
+++ b/chromium-135-print-review-fail.patch
@@ -0,0 +1,44 @@
+Only call format_message when needed
+
+SkJpegDecoderMgr.cpp#print_message calls into `format_message` to print
+informational strings from the jpeg decoder. However, the informational
+strings are only output in developer builds where
+`SK_PRINT_CODEC_MESSAGES` is defined. As a result this call is usually
+just extra work which is then ignored. Move the call of `format_message`
+into the argument list of `SkCodecPrintf` so that the call is only
+performed if its result is going to be used.
+
+When Skia is compiled to use the system libjpeg-turbo the
+`format_message` function pointer is set by an external library to point
+at a function inside an external library and is then called by Skia.
+This can cause false positive reports [0] from CFI icall [1]. This
+change effectively solves this issue by never calling `format_message`
+in a CFI enabled build.
+
+[0] https://gitlab.archlinux.org/archlinux/packaging/packages/chromium/-/issues/13
+[1] https://clang.llvm.org/docs/ControlFlowIntegrity.html#indirect-function-call-checking
+
+Change-Id: I5a59459e1e87bf8cdb3d7e90481a115a2a77d6cf
+Reviewed-on: https://skia-review.googlesource.com/c/skia/+/961356
+Reviewed-by: Daniel Dilan <danieldilan@google.com>
+Commit-Queue: Ben Wagner <bungeman@google.com>
+
+diff --git a/third_party/skia/src/codec/SkJpegDecoderMgr.cpp b/third_party/skia/src/codec/SkJpegDecoderMgr.cpp
+index c905ee1..5ae9573 100644
+--- a/third_party/skia/src/codec/SkJpegDecoderMgr.cpp
++++ b/third_party/skia/src/codec/SkJpegDecoderMgr.cpp
+@@ -24,9 +24,11 @@
+  * Print information, warning, and error messages
+  */
+ static void print_message(const j_common_ptr info, const char caller[]) {
+-    char buffer[JMSG_LENGTH_MAX];
+-    info->err->format_message(info, buffer);
+-    SkCodecPrintf("libjpeg error %d <%s> from %s\n", info->err->msg_code, buffer, caller);
++    [[maybe_unused]] char buffer[JMSG_LENGTH_MAX];
++    SkCodecPrintf("libjpeg error %d <%s> from %s\n",
++                  info->err->msg_code,
++                  (info->err->format_message(info, buffer), buffer),
++                  caller);
+ }
+ 
+ /*

diff --git a/chromium.spec b/chromium.spec
index f2e1557..e99eaad 100644
--- a/chromium.spec
+++ b/chromium.spec
@@ -455,6 +455,8 @@ Patch502: flatpak-Expose-Widevine-into-the-sandbox.patch
 Patch600: chromium-135-add-cfi-suppressions-for-pipewire-functions.patch
 # https://chromium-review.googlesource.com/c/chromium/src/+/6445471
 Patch601: chromium-135-gperf.patch
+# https://skia-review.googlesource.com/c/skia/+/961356
+Patch602: chromium-135-print-review-fail.patch
 
 # Use chromium-latest.py to generate clean tarball from released build tarballs, found here:
 # http://build.chromium.org/buildbot/official/
@@ -1144,6 +1146,7 @@ Qt6 UI for chromium.
 # Upstream patches
 %patch -P600 -p1 -b .add-cfi-suppressions-for-pipewire-functions
 %patch -P601 -p1 -b .gperf-3.2
+%patch -P602 -p1 -b .print-review-fail
 
 # Change shebang in all relevant files in this directory and all subdirectories
 # See `man find` for how the `-exec command {} +` syntax works