Script 'mail_helper' called by obssrc
Hello community,

here is the log from the commit of package intel-ipsec-mb for openSUSE:Factory 
checked in at 2026-10-31 23:06:40
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/intel-ipsec-mb (Old)
 and      /work/SRC/openSUSE:Factory/.intel-ipsec-mb.new.1255 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Package is "intel-ipsec-mb"

Sat Oct 31 23:06:40 2026 rev:8 rq:1393977 version:3.0.1

Changes:
--------
--- /work/SRC/openSUSE:Factory/intel-ipsec-mb/intel-ipsec-mb.changes    
2026-05-29 18:08:12.509107274 +0200
+++ /work/SRC/openSUSE:Factory/.intel-ipsec-mb.new.1255/intel-ipsec-mb.changes  
2026-10-31 23:06:43.049824851 +0100
@@ -1,0 +2,66 @@
+Fri Oct 30 21:50:01 UTC 2026 - Tomasz Kantecki <[email protected]>
+
+- Added ipsecmb_v3_1.patch file fixing 15.7 x86_64 compilation errors and 
warnings
+- Added intel-ipsec-mb-rpmlintrc file to fix one rpmlint warning
+
+-------------------------------------------------------------------
+Fri Oct 30 21:42:41 UTC 2026 - Tomasz Kantecki <[email protected]>
+
+- Removed installation of the man page
+- Fixed rpmlint warning
+- Fixed compilation errors and warnings on 15.7 x86_64
+
+-------------------------------------------------------------------
+Fri Oct 23 15:10:36 UTC 2026 - Tomasz Kantecki <[email protected]>
+
+- Fixed spec file references to README and ReleaseNotes.txt
+
+-------------------------------------------------------------------
+Fri Oct 23 10:05:54 UTC 2026 - Tomasz Kantecki <[email protected]>
+
+- Update to v3.0.1:
+  * Build system: removed legacy Makefile and Windows `.mak` builds; CMake is 
now the only supported build flow.
+  * Build system: GCC and Clang builds now use the C11 standard.
+  * Licensing: replaced full BSD-3-Clause license text in source files with 
SPDX identifiers.
+  * OpenSSL: added `imb-provider`, an OpenSSL 3.x test provider supporting 
symmetric crypto, hash/HMAC and PQC operations, including asynchronous 
operation.
+  * Self-tests: library initialization now fails closed on self-test failure, 
preventing cryptographic operations on the affected manager until successful 
reinitialization.
+  * Architecture support: added AVX10 architecture support.
+  * AES: added AVX2 Type 2 implementations for AES-CFB, AES-CBC, AES-CMAC and 
AES-CCM.
+  * HMAC-SHA1: added variable tag sizes from 4 to 20 bytes.
+  * SHA3/SHAKE: added SHA3-224/256/384/512, SHAKE128/256 and HMAC-SHA3 
support, including optimized AVX2 and AVX512 implementations.
+  * Zero-length messages: extended support across HMAC and AES-CBC algorithms.
+  * 3GPP: added implementations of the new 256-bit confidentiality, integrity 
and combined algorithms based on AES, SNOW5G and ZUC.
+  * DES/3DES: added a new common SSE Type 1 DES block encryption 
implementation.
+  * PQC: added ML-DSA (FIPS 204) and ML-KEM (FIPS 203), including all 
standardized parameter sets.
+  * CRC: added AVX2 Type 2 implementations using 256-bit VPCLMULQDQ polynomial 
folding.
+  * Constant-time validation: added `CONSTANT_TIME_VALIDATION` build option 
for detecting secret-dependent branches and memory accesses using Valgrind.
+  * Wireless algorithms: removed bit-length APIs and non-byte-aligned message 
length/offset support.
+  * Direct APIs: removed direct API support for ZUC-EEA3/EIA3, 
SNOW3G-UEA2/UIA2, KASUMI-F8/F9 and CRC; these algorithms remain available 
through the job API.
+  * Removed AES-CTR bit-length, AES-CBCS 1:9, SNOW-V/AEAD SNOW-V, ZUC-EEA3-256 
and ZUC-EIA3-256 support.
+  * Removed QUIC API, custom cipher and custom hash support.
+  * API validation: added AES-CBC/AES-CFB message length checks and updated 
CFB APIs to return operation status.
+  * Test applications: added `imb-safe-check` for detection of sensitive data 
left in registers and memory.
+  * Test applications: added `imb-oob` for detecting out-of-bounds reads and 
writes using guard pages.
+  * Test applications: expanded SHA3, SHAKE, ML-DSA and ML-KEM coverage across 
`imb-kat`, `imb-xvalid`, `imb-acvp` and fuzz testing.
+  * Test applications: added ABI checking and improved zero-length message 
testing.
+  * Wycheproof: moved Project Wycheproof coverage into `imb-kat`, updated to 
the upstream v1 vector set and removed `imb-wycheproof`.
+  * Performance: added `imb-perf-cmp.py`, PQC benchmarking with 
`imb-speed-pqc`, SHA3/SHAKE and AVX10 support, and improved `imb-perf` 
stability.
+  * Performance: renamed `ipsec_perf_tool.py` to `imb-perf-tool.py` and 
enhanced core selection and algorithm filtering.
+  * Tools: added assembly formatting and coverage tools, plus source-file 
non-ASCII character detection and replacement.
+- Fixed issues:
+  * Fixed SM4-GCM plaintext/ciphertext address calculation (issue #159).
+  * Fixed SNOW3G and KASUMI JOB API destination pointer calculation (issue 
#162).
+  * Fixed SNOW-V-AEAD ciphertext address calculation (issue #161).
+  * Fixed MD5-HMAC mapping in the performance application.
+  * Fixed AES-GCM key structure alignment.
+  * Fixed DES/3DES-CBC plaintext/ciphertext output (issue #169).
+  * Fixed key pointer handling for AES-GCM, AES-GCM-SGL and SM4-GCM decryption.
+  * Fixed PON XGEM PLI validation that could result in out-of-bounds memory 
access.
+  * Fixed MinGW build issues (issue #185).
+  * Fixed Windows GCC/MinGW stack alignment issue affecting ZUC-NCA6.
+  * Fixed AVX10 architecture auto-detection.
+  * Fixed GHASH job API out-of-bounds read for tag lengths below 16 bytes.
+- More details at:
+  * [intel-ipsec-mb v3.0.0 
release](https://github.com/intel/intel-ipsec-mb/releases/tag/v3.0.0)
+
+-------------------------------------------------------------------

Old:
----
  intel-ipsec-mb-2.0.2.tar.gz

New:
----
  intel-ipsec-mb-3.0.1.tar.gz
  intel-ipsec-mb-rpmlintrc
  ipsecmb_v3_1.patch

----------(New B)----------
  New:
- Added ipsecmb_v3_1.patch file fixing 15.7 x86_64 compilation errors and 
warnings
- Added intel-ipsec-mb-rpmlintrc file to fix one rpmlint warning
----------(New E)----------

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Other differences:
------------------
++++++ intel-ipsec-mb.spec ++++++
--- /var/tmp/diff_new_pack.BhN39b/_old  2026-10-31 23:06:43.946861912 +0100
+++ /var/tmp/diff_new_pack.BhN39b/_new  2026-10-31 23:06:43.948861994 +0100
@@ -17,9 +17,9 @@
 #
 
 
-%global major        2
+%global major        3
 %global minor        0
-%global patch        2
+%global patch        1
 
 # GitHub properties
 %global githubver    %{major}.%{minor}.%{patch}
@@ -30,11 +30,13 @@
 Name:           intel-ipsec-mb
 Summary:        IPSec cryptography library optimized for Intel Architecture
 Release:        0
-Version:        2.0.2
-License:        BSD-3-Clause
+Version:        3.0.1
+License:        Apache-2.0 AND BSD-3-Clause
 Group:          Development/Libraries/C and C++
 URL:            https://github.com/intel/%{name}
 Source0:        
https://github.com/intel/%{name}/archive/v%{githubver}.tar.gz#/%{githubfull}.tar.gz
+Source1:        intel-ipsec-mb-rpmlintrc
+Patch1:         ipsecmb_v3_1.patch
 ExclusiveArch:  x86_64
 BuildRequires:  cmake
 BuildRequires:  gcc >= 4.8.3
@@ -65,28 +67,29 @@
 and primarily targeted at packet processing applications.
 
 %prep
-%autosetup -n %{name}-%{githubver}
+%autosetup -p1 -n %{name}-%{githubver}
 
 %build
-%cmake
+%cmake --no-warn-unused-cli
 %cmake_build
 
 %install
 %cmake_install
+# Do not install the library and development man pages
+rm -f %{buildroot}%{_mandir}/man7/libipsec-mb.7*
+rm -f %{buildroot}%{_mandir}/man7/libipsec-mb-dev.7*
 
 %check
 %ctest
 
 %files -n %{rpm_name}%{major}
 %license LICENSE
-%doc README ReleaseNotes.txt
+%doc README.md ReleaseNotes.md
 %{_libdir}/libIPSec_MB.so.%{version}
 %{_libdir}/libIPSec_MB.so.%{major}
-%{_mandir}/man7/libipsec-mb.7.gz
 
 %files devel
 %{_includedir}/intel-ipsec-mb.h
-%{_mandir}/man7/libipsec-mb-dev.7.gz
 %{_libdir}/libIPSec_MB.so
 
 %post  -n %{rpm_name}%{major} -p /sbin/ldconfig

++++++ intel-ipsec-mb-2.0.2.tar.gz -> intel-ipsec-mb-3.0.1.tar.gz ++++++
/work/SRC/openSUSE:Factory/intel-ipsec-mb/intel-ipsec-mb-2.0.2.tar.gz 
/work/SRC/openSUSE:Factory/.intel-ipsec-mb.new.1255/intel-ipsec-mb-3.0.1.tar.gz 
differ: char 12, line 1

++++++ intel-ipsec-mb-rpmlintrc ++++++
addFilter("intel-ipsec-mb-devel.*no-dependency-on")

++++++ ipsecmb_v3_1.patch ++++++
diff --git a/lib/openssl/crypto/sha/asm/keccak1600x4-avx512vl.pl 
b/lib/openssl/crypto/sha/asm/keccak1600x4-avx512vl.pl
index 8d80aed6..27fba7cc 100755
--- a/lib/openssl/crypto/sha/asm/keccak1600x4-avx512vl.pl
+++ b/lib/openssl/crypto/sha/asm/keccak1600x4-avx512vl.pl
@@ -95,7 +95,8 @@ $sf_size="856";       # 48 + 808 = 856 bytes
 # Emit an internal helper call used by one-shot wrappers.
 # - Win64: call the provided *_internal shim and bracket it with 32-byte
 #   shadow space so shim entry can use xlate-compatible [rsp+8]/[rsp+16].
-# - non-Win64: call the public API symbol (same base name without _internal).
+# - non-Win64: call the local .L_ entry label of the public API function
+#   (same base name without _internal).
 # The argument must be the shim/internal symbol name, e.g.
 #   SHA3_shake128_x4_inc_squeeze_avx512vl_internal
 sub call_internal {
@@ -110,8 +111,11 @@ sub call_internal {
     add     \$32, %rsp
 ___
 
+    # Call the file-local entry label rather than the global symbol so
+    # no PC32 relocation against a preemptible symbol is emitted
+    # (rejected by older linkers when building a shared object).
     return <<___;
-    call    $external_name
+    call    .L_$external_name
 ___
 }
 
diff --git a/perf/ipsec_perf.c b/perf/ipsec_perf.c
index 3ce86d96..dbf0fdbc 100644
--- a/perf/ipsec_perf.c
+++ b/perf/ipsec_perf.c
@@ -1774,7 +1774,7 @@ static inline void
 set_job_fields(IMB_JOB *job, uint8_t *p_buffer, imb_uint128_t *p_keys, const 
uint32_t i,
                const uint32_t index, const IMB_JOB *template)
 {
-        uint32_t list_idx;
+        uint32_t list_idx = 0;
 
         /* If IMIX testing is being done, set the buffer size to cipher and 
hash
          * going through the list of sizes precalculated */
@@ -1792,10 +1792,7 @@ set_job_fields(IMB_JOB *job, uint8_t *p_buffer, 
imb_uint128_t *p_keys, const uin
                 uint64_t *p_src = (uint64_t *) get_src_buffer(index, p_buffer);
 
                 job->src = (const uint8_t *) p_src;
-                if (imix_list_count != 0)
-                        p_src[0] = xgem_hdr_list[list_idx];
-                else
-                        p_src[0] = xgem_hdr_list[0];
+                p_src[0] = xgem_hdr_list[list_idx];
         } else
                 job->src = get_src_buffer(index, p_buffer);
 
diff --git a/test/oob-app/oob_test.c b/test/oob-app/oob_test.c
index e3517ed5..badbdd5a 100644
--- a/test/oob-app/oob_test.c
+++ b/test/oob-app/oob_test.c
@@ -2151,7 +2151,7 @@ test_ml_dsa_keyops_oob(struct pqc_test_ctx *t, IMB_ML_DSA 
*ctx, const struct ml_
                 /* volatile: live across the TEST_SETJMP() checkpoint below */
                 uint8_t *const volatile sk = oob_buf(&mem->privkey, 
tv->sk_len, t->oob);
                 uint8_t *const volatile pk = oob_buf(&mem->pubkey, tv->pk_len, 
t->oob);
-                int ok = 0;
+                volatile int ok = 0;
 
                 memcpy(sk, ref_sk, tv->sk_len);
 
@@ -2213,8 +2213,9 @@ test_ml_dsa_sign_verify_oob(struct pqc_test_ctx *t, 
IMB_ML_DSA *ctx,
                             const unsigned msg_len, const unsigned ctx_len, 
const int hedged,
                             const int msg_is_mu)
 {
-        uint8_t *msg = oob_buf(&mem->msg, msg_len, t->oob);
-        uint8_t *sig = oob_buf(&mem->sig, tv->sig_len, t->oob);
+        /* volatile: reassigned between TEST_SETJMP() checkpoints below */
+        uint8_t *volatile msg = oob_buf(&mem->msg, msg_len, t->oob);
+        uint8_t *volatile sig = oob_buf(&mem->sig, tv->sig_len, t->oob);
         uint8_t *cstr = oob_buf(&mem->ctxstr, ctx_len, t->oob);
         uint8_t *rnd = oob_buf(&mem->seed, IMB_ML_DSA_SIGN_RND_BYTES, t->oob);
         IMB_ML_DSA_SIGN_PARAMS sp;
@@ -2534,8 +2535,9 @@ test_ml_kem_encap_decap_oob(struct pqc_test_ctx *t, 
IMB_ML_KEM *ctx,
                             const int seeded)
 {
         uint8_t ref_ss[IMB_ML_KEM_SHARED_SECRET_BYTES];
-        uint8_t *ct = oob_buf(&mem->sig, tv->ct_len, t->oob);
-        uint8_t *ss = oob_buf(&mem->ss, IMB_ML_KEM_SHARED_SECRET_BYTES, 
t->oob);
+        /* volatile: reassigned between TEST_SETJMP() checkpoints below */
+        uint8_t *volatile ct = oob_buf(&mem->sig, tv->ct_len, t->oob);
+        uint8_t *volatile ss = oob_buf(&mem->ss, 
IMB_ML_KEM_SHARED_SECRET_BYTES, t->oob);
         uint8_t *const m = oob_buf(&mem->seed, IMB_ML_KEM_ENCAP_SEED_BYTES, 
t->oob);
         IMB_ML_KEM_ENCAP_PARAMS params;
 

Reply via email to