From 42132178bc8d6d602037be706572db00183cb014 Mon Sep 17 00:00:00 2001 From: eabdullin Date: Thu, 14 May 2026 14:28:05 +0300 Subject: [PATCH] Bump version to 6.12.0-124.56.3 net: skbuff: propagate shared-frag marker through frag-transfer helpers Replace 124.56.2's partial fix (which was a backport of Hyunwoo Kim's v1 patch covering only __pskb_copy_fclone() and skb_try_coalesce()) with a backport of the upstream v3 patch: https://lore.kernel.org/all/agW4vC0r8QOUKtRT@v4bel/ v3 also covers skb_shift() (the new hunk added in v2) and skb_gro_receive() / skb_gro_receive_list() (the audit follow-up suggested by Sultan Alsawaf in agVpIsaSherjHTYg@sultan-box). All five sites moved frag descriptors between skbs without propagating the SKBFL_SHARED_FRAG marker, so destinations could end up referencing externally-owned or page-cache-backed pages while reporting skb_has_shared_frag() as false. Combined with ESP-over-UDP and UDP GRO, or any nf_dup_ipv4 / xt_TEE caller, this lets an unprivileged user trigger in-place ESP decryption over root-owned page-cache pages (CVE-2026-46300, "Fragnesia"). --- ...-skbuff-propagate-shared-frag-marker.patch | 90 ++++++++++++++----- kernel.spec | 7 +- 2 files changed, 71 insertions(+), 26 deletions(-) diff --git a/1100-net-skbuff-propagate-shared-frag-marker.patch b/1100-net-skbuff-propagate-shared-frag-marker.patch index 03b8c5397..cbd4e37d2 100644 --- a/1100-net-skbuff-propagate-shared-frag-marker.patch +++ b/1100-net-skbuff-propagate-shared-frag-marker.patch @@ -1,27 +1,28 @@ -From: Andrew Lukoshko -Subject: [PATCH AlmaLinux 10] net: skbuff: propagate shared-frag marker through pskb_copy() +From: Eduard Abdullin +Subject: [PATCH AlmaLinux 10] net: skbuff: propagate shared-frag marker through frag-transfer helpers -Backport of upstream patch posted at -https://lore.kernel.org/all/agRfuVOeMI5pbHhY@v4bel/ -(sibling to the xfrm/esp shared-frag fix upstream commit f4c50a4034e6, -already merged into 6.12.0-124.56.1 via the c10s import). +Backport of upstream v3 patch posted at +https://lore.kernel.org/all/agW4vC0r8QOUKtRT@v4bel/ +(sibling to upstream commit f4c50a4034e6 "xfrm: esp: avoid in-place +decrypt on shared skb frags", already part of 6.12.0-124.56.1 via the +c10s import). -__pskb_copy_fclone() shallow-copies the source's frag descriptors and -bumps each page's refcount via skb_frag_ref(), then defers the rest -of the shinfo metadata to skb_copy_header(). That helper only carries -over gso_{size,segs,type} and never touches skb_shinfo()->flags, so -the destination skb keeps a reference to the same externally-owned or -page-cache-backed pages while reporting skb_has_shared_frag() as -false. +Three frag-transfer helpers (__pskb_copy_fclone(), skb_try_coalesce(), +and skb_shift()) and the GRO accumulator helpers (skb_gro_receive() +and skb_gro_receive_list()) fail to propagate the SKBFL_SHARED_FRAG +bit in skb_shinfo()->flags when moving frag descriptors from source +to destination. As a result, the destination skb keeps a reference +to the same externally-owned or page-cache-backed pages while +reporting skb_has_shared_frag() as false. The mismatch is harmful in any in-place writer that uses skb_has_shared_frag() to decide whether shared pages must be detoured through skb_cow_data(). ESP input is one such writer (esp4.c, -esp6.c), and a single nft 'dup to ' rule -- or any other -nf_dup_ipv4() / xt_TEE caller -- is enough to land a pskb_copy()'d -skb in esp_input() with the marker stripped, letting an unprivileged -user write into the page cache of a root-owned read-only file via -authencesn-ESN stray writes. +esp6.c). Combined with an nft 'dup to ' rule, an +nf_dup_ipv4() / xt_TEE caller, or ESP-over-UDP with UDP GRO, this +lets an unprivileged user write into the page cache of a root-owned +read-only file via authencesn-ESN stray writes (CVE-2026-46300, +"Fragnesia"). Set SKBFL_SHARED_FRAG on the destination whenever frag descriptors were actually moved from the source. skb_copy() and skb_copy_expand() @@ -29,18 +30,50 @@ share skb_copy_header() too but linearize all paged data into freshly allocated head storage and emerge with nr_frags == 0, so skb_has_shared_frag() returns false on its own; they need no change. -Verified to apply with `patch -p1 -F0` (no offset, no fuzz, no rejects) -against kernel-6.12.0-124.56.1.el10_1. +The same omission exists in skb_gro_receive() and +skb_gro_receive_list(). The former moves the incoming skb's frag +descriptors into the accumulator's last sub-skb via two paths plus a +"merge:" fallback that chains the whole skb onto frag_list; the +latter chains the incoming skb whole onto p's frag_list. Downstream +skb_segment() reads only skb_shinfo(p)->flags, and skb_segment_list() +reuses each sub-skb's shinfo as the nskb -- both p and lp must carry +the marker. Fixes: cef401de7be8 ("net: fix possible wrong checksum generation") Fixes: f4c50a4034e6 ("xfrm: esp: avoid in-place decrypt on shared skb frags") +Reported-by: Sultan Alsawaf Reported-by: William Bowling Reported-by: Hyunwoo Kim -Signed-off-by: Andrew Lukoshko +Signed-off-by: Eduard Abdullin --- - net/core/skbuff.c | 3 +++ - 1 file changed, 3 insertions(+) + net/core/gro.c | 4 ++++ + net/core/skbuff.c | 5 +++++ + 2 files changed, 9 insertions(+) +--- a/net/core/gro.c ++++ b/net/core/gro.c +@@ -213,10 +213,12 @@ int skb_gro_receive(struct sk_buff *p, struct sk_buff *skb) + p->data_len += len; + p->truesize += delta_truesize; + p->len += len; ++ skb_shinfo(p)->flags |= skbinfo->flags & SKBFL_SHARED_FRAG; + if (lp != p) { + lp->data_len += len; + lp->truesize += delta_truesize; + lp->len += len; ++ skb_shinfo(lp)->flags |= skbinfo->flags & SKBFL_SHARED_FRAG; + } + NAPI_GRO_CB(skb)->same_flow = 1; + return 0; +@@ -244,6 +246,8 @@ int skb_gro_receive_list(struct sk_buff *p, struct sk_buff *skb) + p->truesize += skb->truesize; + p->len += skb->len; + ++ skb_shinfo(p)->flags |= skb_shinfo(skb)->flags & SKBFL_SHARED_FRAG; ++ + NAPI_GRO_CB(skb)->same_flow = 1; + + return 0; --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2123,6 +2123,7 @@ struct sk_buff *__pskb_copy_fclone(struct sk_buff *skb, int headroom, @@ -51,7 +84,16 @@ Signed-off-by: Andrew Lukoshko } if (skb_has_frag_list(skb)) { -@@ -6028,6 +6029,8 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from, +@@ -4198,6 +4199,8 @@ int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen) + tgt->ip_summed = CHECKSUM_PARTIAL; + skb->ip_summed = CHECKSUM_PARTIAL; + ++ skb_shinfo(tgt)->flags |= skb_shinfo(skb)->flags & SKBFL_SHARED_FRAG; ++ + skb_len_add(skb, -shiftlen); + skb_len_add(tgt, shiftlen); + +@@ -6028,6 +6031,8 @@ bool skb_try_coalesce(struct sk_buff *to, struct sk_buff *from, from_shinfo->frags, from_shinfo->nr_frags * sizeof(skb_frag_t)); to_shinfo->nr_frags += from_shinfo->nr_frags; diff --git a/kernel.spec b/kernel.spec index d8830e770..d42060662 100644 --- a/kernel.spec +++ b/kernel.spec @@ -176,13 +176,13 @@ Summary: The Linux kernel %define specrpmversion 6.12.0 %define specversion 6.12.0 %define patchversion 6.12 -%define pkgrelease 124.56.2 +%define pkgrelease 124.56.3 %define kversion 6 %define tarfile_release 6.12.0-124.56.1.el10_1 # This is needed to do merge window version magic %define patchlevel 12 # This allows pkg_release to have configurable %%{?dist} tag -%define specrelease 124.56.2%{?buildid}%{?dist} +%define specrelease 124.56.3%{?buildid}%{?dist} # This defines the kabi tarball version %define kabiversion 6.12.0-124.56.1.el10_1 @@ -4379,6 +4379,9 @@ fi\ # # %changelog +* Thu May 14 2026 Eduard Abdullin - 6.12.0-124.56.3 +- net: skbuff: propagate shared-frag marker through frag-transfer helpers + * Wed May 13 2026 Andrew Lukoshko - 6.12.0-124.56.2 - net: skbuff: propagate shared-frag marker through pskb_copy()