Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Remove implicit narrowing conversions from zlib #91245

Merged
merged 2 commits into from
Aug 30, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions src/mono/CMakeLists.txt
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,9 @@ if (MSVC)
if(EXISTS ${CLR_SOURCELINK_FILE_PATH})
add_link_options("/sourcelink:${CLR_SOURCELINK_FILE_PATH}")
endif()

# FIXME: Remove the line below when https://github.com/dotnet/runtime/issues/91249 is fixed.
add_compile_options($<$<COMPILE_LANGUAGE:C,CXX>:/wd4244>) # conversion from 'type1' to 'type2', possible loss of data
endif(MSVC)

set(CROSS_ROOTFS $ENV{ROOTFS_DIR})
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,75 @@
From edabaf799fd071a328e0adb743a98628df6649f0 Mon Sep 17 00:00:00 2001
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why is a patch file needed in a PR?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It makes it easier to separate our customizations so when we import a new upstream version we can just re-apply the patch.

From: Levi Broderick <levib@microsoft.com>
Date: Mon, 28 Aug 2023 15:26:38 -0700
Subject: [PATCH] Make zlib-intel compile clean against C4244 clang equivalent
is "implicit-int-conversion" warning

The change to deflate.c is legal because 'len' has an upper bound of
MAX_STORED, which means it fits cleanly into a 16-bit integer. So
writing out 2x 8-bit values will not result in data loss.

The change to trees.c is legal because within this loop, 'count' is
intended to have an upper bound of 138, with the target assignment
only executing if 'count' is bounded by 4. Neither the 'count' local
in isolation nor the addition that's part of the target line is
expected to result in integer overflow. But even if it did, that's a
matter for a different warning code and doesn't impact the correctness
of the narrowing cast being considered here.

The change to slide_sse.c is legal because 'w_size' is limited to
1 << 15 (see deflateInit2_ in deflate.c), so this fits cleanly into
a 16-bit value.
---
src/native/external/zlib-intel/deflate.c | 8 ++++----
src/native/external/zlib-intel/slide_sse.c | 2 +-
src/native/external/zlib-intel/trees.c | 2 +-
3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/src/native/external/zlib-intel/deflate.c b/src/native/external/zlib-intel/deflate.c
index bd5e95774a6..108b1a187af 100644
--- a/src/native/external/zlib-intel/deflate.c
+++ b/src/native/external/zlib-intel/deflate.c
@@ -1553,10 +1553,10 @@ local block_state deflate_stored(s, flush)
_tr_stored_block(s, (char *)0, 0L, last);

/* Replace the lengths in the dummy stored block with len. */
- s->pending_buf[s->pending - 4] = len;
- s->pending_buf[s->pending - 3] = len >> 8;
- s->pending_buf[s->pending - 2] = ~len;
- s->pending_buf[s->pending - 1] = ~len >> 8;
+ s->pending_buf[s->pending - 4] = (Bytef)len;
+ s->pending_buf[s->pending - 3] = (Bytef)(len >> 8);
+ s->pending_buf[s->pending - 2] = (Bytef)~len;
+ s->pending_buf[s->pending - 1] = (Bytef)(~len >> 8);

/* Write the stored block header bytes. */
flush_pending(s->strm);
diff --git a/src/native/external/zlib-intel/slide_sse.c b/src/native/external/zlib-intel/slide_sse.c
index 342fd562dd1..eb74202c5a0 100644
--- a/src/native/external/zlib-intel/slide_sse.c
+++ b/src/native/external/zlib-intel/slide_sse.c
@@ -18,7 +18,7 @@ void slide_hash_sse(deflate_state *s)
unsigned n;
Posf *p;
uInt wsize = s->w_size;
- z_const __m128i xmm_wsize = _mm_set1_epi16(s->w_size);
+ z_const __m128i xmm_wsize = _mm_set1_epi16((short)s->w_size);

n = s->hash_size;
p = &s->head[n] - 8;
diff --git a/src/native/external/zlib-intel/trees.c b/src/native/external/zlib-intel/trees.c
index 35462a1313a..f78b7d8c63e 100644
--- a/src/native/external/zlib-intel/trees.c
+++ b/src/native/external/zlib-intel/trees.c
@@ -717,7 +717,7 @@ local void scan_tree(s, tree, max_code)
if (++count < max_count && curlen == nextlen) {
continue;
} else if (count < min_count) {
- s->bl_tree[curlen].Freq += count;
+ s->bl_tree[curlen].Freq += (ush)count;
} else if (curlen != 0) {
if (curlen != prevlen) s->bl_tree[curlen].Freq++;
s->bl_tree[REP_3_6].Freq++;
--
2.42.0.windows.1

Original file line number Diff line number Diff line change
@@ -0,0 +1,57 @@
From 86d96652ddd60f61dc7b0c94b601f6d156d34632 Mon Sep 17 00:00:00 2001
From: Levi Broderick <levib@microsoft.com>
Date: Mon, 28 Aug 2023 15:26:38 -0700
Subject: [PATCH] Make zlib compile clean against C4244 clang equivalent is
"implicit-int-conversion" warning

The change to deflate.c is legal because 'len' has an upper bound of
MAX_STORED, which means it fits cleanly into a 16-bit integer. So
writing out 2x 8-bit values will not result in data loss.

The change to trees.c is legal because within this loop, 'count' is
intended to have an upper bound of 138, with the target assignment
only executing if 'count' is bounded by 4. Neither the 'count' local
in isolation nor the addition that's part of the target line is
expected to result in integer overflow. But even if it did, that's a
matter for a different warning code and doesn't impact the correctness
of the narrowing cast being considered here.
---
src/native/external/zlib/deflate.c | 8 ++++----
src/native/external/zlib/trees.c | 2 +-
2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/src/native/external/zlib/deflate.c b/src/native/external/zlib/deflate.c
index d2e1106ef5d..b7636639754 100644
--- a/src/native/external/zlib/deflate.c
+++ b/src/native/external/zlib/deflate.c
@@ -1738,10 +1738,10 @@ local block_state deflate_stored(s, flush)
_tr_stored_block(s, (char *)0, 0L, last);

/* Replace the lengths in the dummy stored block with len. */
- s->pending_buf[s->pending - 4] = len;
- s->pending_buf[s->pending - 3] = len >> 8;
- s->pending_buf[s->pending - 2] = ~len;
- s->pending_buf[s->pending - 1] = ~len >> 8;
+ s->pending_buf[s->pending - 4] = (Bytef)len;
+ s->pending_buf[s->pending - 3] = (Bytef)(len >> 8);
+ s->pending_buf[s->pending - 2] = (Bytef)~len;
+ s->pending_buf[s->pending - 1] = (Bytef)(~len >> 8);

/* Write the stored block header bytes. */
flush_pending(s->strm);
diff --git a/src/native/external/zlib/trees.c b/src/native/external/zlib/trees.c
index 5f305c47221..8a3eec559e5 100644
--- a/src/native/external/zlib/trees.c
+++ b/src/native/external/zlib/trees.c
@@ -721,7 +721,7 @@ local void scan_tree(s, tree, max_code)
if (++count < max_count && curlen == nextlen) {
continue;
} else if (count < min_count) {
- s->bl_tree[curlen].Freq += count;
+ s->bl_tree[curlen].Freq += (ush)count;
} else if (curlen != 0) {
if (curlen != prevlen) s->bl_tree[curlen].Freq++;
s->bl_tree[REP_3_6].Freq++;
--
2.42.0.windows.1

2 changes: 2 additions & 0 deletions src/native/external/zlib-intel-version.txt
Original file line number Diff line number Diff line change
Expand Up @@ -12,3 +12,5 @@ copied into our repo in order to build. Since then, we've just updated only
those files in-place, ignoring other files in the intel/zlib repo. If new files
are introduced which are necessary for building the product, feel free to bring
those down as well.

We have also applied the custom patches under the patches/zlib-intel folder.
6 changes: 0 additions & 6 deletions src/native/external/zlib-intel.cmake
Original file line number Diff line number Diff line change
@@ -1,9 +1,3 @@
if(MSVC)
add_compile_options($<$<COMPILE_LANGUAGE:C,CXX>:/wd4244>) # conversion from 'type1' to 'type2', possible loss of data
else(CMAKE_C_COMPILER_ID MATCHES "Clang")
add_compile_options($<$<COMPILE_LANGUAGE:C,CXX>:-Wno-implicit-int-conversion>)
endif()

set(ZLIB_SOURCES_BASE
adler32.c
compress.c
Expand Down
8 changes: 4 additions & 4 deletions src/native/external/zlib-intel/deflate.c
Original file line number Diff line number Diff line change
Expand Up @@ -1553,10 +1553,10 @@ local block_state deflate_stored(s, flush)
_tr_stored_block(s, (char *)0, 0L, last);

/* Replace the lengths in the dummy stored block with len. */
s->pending_buf[s->pending - 4] = len;
s->pending_buf[s->pending - 3] = len >> 8;
s->pending_buf[s->pending - 2] = ~len;
s->pending_buf[s->pending - 1] = ~len >> 8;
s->pending_buf[s->pending - 4] = (Bytef)len;
danmoseley marked this conversation as resolved.
Show resolved Hide resolved
s->pending_buf[s->pending - 3] = (Bytef)(len >> 8);
s->pending_buf[s->pending - 2] = (Bytef)~len;
s->pending_buf[s->pending - 1] = (Bytef)(~len >> 8);

/* Write the stored block header bytes. */
flush_pending(s->strm);
Expand Down
2 changes: 1 addition & 1 deletion src/native/external/zlib-intel/slide_sse.c
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ void slide_hash_sse(deflate_state *s)
unsigned n;
Posf *p;
uInt wsize = s->w_size;
z_const __m128i xmm_wsize = _mm_set1_epi16(s->w_size);
z_const __m128i xmm_wsize = _mm_set1_epi16((short)s->w_size);

n = s->hash_size;
p = &s->head[n] - 8;
Expand Down
2 changes: 1 addition & 1 deletion src/native/external/zlib-intel/trees.c
Original file line number Diff line number Diff line change
Expand Up @@ -717,7 +717,7 @@ local void scan_tree(s, tree, max_code)
if (++count < max_count && curlen == nextlen) {
continue;
} else if (count < min_count) {
s->bl_tree[curlen].Freq += count;
s->bl_tree[curlen].Freq += (ush)count;
} else if (curlen != 0) {
if (curlen != prevlen) s->bl_tree[curlen].Freq++;
s->bl_tree[REP_3_6].Freq++;
Expand Down
2 changes: 2 additions & 0 deletions src/native/external/zlib-version.txt
Original file line number Diff line number Diff line change
Expand Up @@ -13,3 +13,5 @@ We have also cherry-picked into our local copy:
This patch only affects memLevel 9 compression. .NET doesn't currently use this
memLevel, but we'll take this patch out of an abundance of caution just in case
we enable this functionality in a future release.

We have also applied the custom patches under the patches/zlib folder.
6 changes: 0 additions & 6 deletions src/native/external/zlib.cmake
Original file line number Diff line number Diff line change
@@ -1,9 +1,3 @@
if(MSVC)
add_compile_options($<$<COMPILE_LANGUAGE:C,CXX>:/wd4244>) # conversion from 'type1' to 'type2', possible loss of data
else(CMAKE_C_COMPILER_ID MATCHES "Clang")
add_compile_options($<$<COMPILE_LANGUAGE:C,CXX>:-Wno-implicit-int-conversion>)
endif()

set(ZLIB_SOURCES_BASE
adler32.c
compress.c
Expand Down
8 changes: 4 additions & 4 deletions src/native/external/zlib/deflate.c
Original file line number Diff line number Diff line change
Expand Up @@ -1738,10 +1738,10 @@ local block_state deflate_stored(s, flush)
_tr_stored_block(s, (char *)0, 0L, last);

/* Replace the lengths in the dummy stored block with len. */
s->pending_buf[s->pending - 4] = len;
s->pending_buf[s->pending - 3] = len >> 8;
s->pending_buf[s->pending - 2] = ~len;
s->pending_buf[s->pending - 1] = ~len >> 8;
s->pending_buf[s->pending - 4] = (Bytef)len;
s->pending_buf[s->pending - 3] = (Bytef)(len >> 8);
s->pending_buf[s->pending - 2] = (Bytef)~len;
s->pending_buf[s->pending - 1] = (Bytef)(~len >> 8);

/* Write the stored block header bytes. */
flush_pending(s->strm);
Expand Down
2 changes: 1 addition & 1 deletion src/native/external/zlib/trees.c
Original file line number Diff line number Diff line change
Expand Up @@ -721,7 +721,7 @@ local void scan_tree(s, tree, max_code)
if (++count < max_count && curlen == nextlen) {
continue;
} else if (count < min_count) {
s->bl_tree[curlen].Freq += count;
s->bl_tree[curlen].Freq += (ush)count;
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What's ush? Having trouble finding its definition.

Copy link
Member

@akoeplinger akoeplinger Aug 30, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's from a typedef here:

typedef unsigned short ush;

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks. Oof, I really dislike cryptic c typedefs.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Agreed. It's trying to match the maintainer's style as closely as possible, which is why it could look strange from our eyes.

} else if (curlen != 0) {
if (curlen != prevlen) s->bl_tree[curlen].Freq++;
s->bl_tree[REP_3_6].Freq++;
Expand Down
Loading