From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: "Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
patches@lists.linux.dev,
"Per Sundström XP" <per.xp.sundstrom@ericsson.com>,
"Andrii Nakryiko" <andrii@kernel.org>,
"Daniel Borkmann" <daniel@iogearbox.net>,
"Sasha Levin" <sashal@kernel.org>
Subject: [PATCH 5.15 95/99] libbpf: Fix BTF-to-C converters padding logic
Date: Mon, 3 Apr 2023 16:09:58 +0200 [thread overview]
Message-ID: <20230403140406.904734982@linuxfoundation.org> (raw)
In-Reply-To: <20230403140356.079638751@linuxfoundation.org>
From: Andrii Nakryiko <andrii@kernel.org>
[ Upstream commit ea2ce1ba99aa6a60c8d8a706e3abadf3de372163 ]
Turns out that btf_dump API doesn't handle a bunch of tricky corner
cases, as reported by Per, and further discovered using his testing
Python script ([0]).
This patch revamps btf_dump's padding logic significantly, making it
more correct and also avoiding unnecessary explicit padding, where
compiler would pad naturally. This overall topic turned out to be very
tricky and subtle, there are lots of subtle corner cases. The comments
in the code tries to give some clues, but comments themselves are
supposed to be paired with good understanding of C alignment and padding
rules. Plus some experimentation to figure out subtle things like
whether `long :0;` means that struct is now forced to be long-aligned
(no, it's not, turns out).
Anyways, Per's script, while not completely correct in some known
situations, doesn't show any obvious cases where this logic breaks, so
this is a nice improvement over the previous state of this logic.
Some selftests had to be adjusted to accommodate better use of natural
alignment rules, eliminating some unnecessary padding, or changing it to
`type: 0;` alignment markers.
Note also that for when we are in between bitfields, we emit explicit
bit size, while otherwise we use `: 0`, this feels much more natural in
practice.
Next patch will add few more test cases, found through randomized Per's
script.
[0] https://lore.kernel.org/bpf/85f83c333f5355c8ac026f835b18d15060725fcb.camel@ericsson.com/
Reported-by: Per Sundström XP <per.xp.sundstrom@ericsson.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Link: https://lore.kernel.org/bpf/20221212211505.558851-6-andrii@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
tools/lib/bpf/btf_dump.c | 169 +++++++++++++-----
.../bpf/progs/btf_dump_test_case_bitfields.c | 2 +-
.../bpf/progs/btf_dump_test_case_padding.c | 58 ++++--
3 files changed, 164 insertions(+), 65 deletions(-)
diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index f620911ad3bb5..77fcc5883a954 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -817,6 +817,25 @@ static void btf_dump_emit_type(struct btf_dump *d, __u32 id, __u32 cont_id)
}
}
+static int btf_natural_align_of(const struct btf *btf, __u32 id)
+{
+ const struct btf_type *t = btf__type_by_id(btf, id);
+ int i, align, vlen;
+ const struct btf_member *m;
+
+ if (!btf_is_composite(t))
+ return btf__align_of(btf, id);
+
+ align = 1;
+ m = btf_members(t);
+ vlen = btf_vlen(t);
+ for (i = 0; i < vlen; i++, m++) {
+ align = max(align, btf__align_of(btf, m->type));
+ }
+
+ return align;
+}
+
static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
const struct btf_type *t)
{
@@ -824,16 +843,16 @@ static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
int align, i, bit_sz;
__u16 vlen;
- align = btf__align_of(btf, id);
- /* size of a non-packed struct has to be a multiple of its alignment*/
- if (align && t->size % align)
+ align = btf_natural_align_of(btf, id);
+ /* size of a non-packed struct has to be a multiple of its alignment */
+ if (align && (t->size % align) != 0)
return true;
m = btf_members(t);
vlen = btf_vlen(t);
/* all non-bitfield fields have to be naturally aligned */
for (i = 0; i < vlen; i++, m++) {
- align = btf__align_of(btf, m->type);
+ align = btf_natural_align_of(btf, m->type);
bit_sz = btf_member_bitfield_size(t, i);
if (align && bit_sz == 0 && m->offset % (8 * align) != 0)
return true;
@@ -846,44 +865,97 @@ static bool btf_is_struct_packed(const struct btf *btf, __u32 id,
return false;
}
-static int chip_away_bits(int total, int at_most)
-{
- return total % at_most ? : at_most;
-}
-
static void btf_dump_emit_bit_padding(const struct btf_dump *d,
- int cur_off, int m_off, int m_bit_sz,
- int align, int lvl)
+ int cur_off, int next_off, int next_align,
+ bool in_bitfield, int lvl)
{
- int off_diff = m_off - cur_off;
- int ptr_bits = d->ptr_sz * 8;
+ const struct {
+ const char *name;
+ int bits;
+ } pads[] = {
+ {"long", d->ptr_sz * 8}, {"int", 32}, {"short", 16}, {"char", 8}
+ };
+ int new_off, pad_bits, bits, i;
+ const char *pad_type;
+
+ if (cur_off >= next_off)
+ return; /* no gap */
+
+ /* For filling out padding we want to take advantage of
+ * natural alignment rules to minimize unnecessary explicit
+ * padding. First, we find the largest type (among long, int,
+ * short, or char) that can be used to force naturally aligned
+ * boundary. Once determined, we'll use such type to fill in
+ * the remaining padding gap. In some cases we can rely on
+ * compiler filling some gaps, but sometimes we need to force
+ * alignment to close natural alignment with markers like
+ * `long: 0` (this is always the case for bitfields). Note
+ * that even if struct itself has, let's say 4-byte alignment
+ * (i.e., it only uses up to int-aligned types), using `long:
+ * X;` explicit padding doesn't actually change struct's
+ * overall alignment requirements, but compiler does take into
+ * account that type's (long, in this example) natural
+ * alignment requirements when adding implicit padding. We use
+ * this fact heavily and don't worry about ruining correct
+ * struct alignment requirement.
+ */
+ for (i = 0; i < ARRAY_SIZE(pads); i++) {
+ pad_bits = pads[i].bits;
+ pad_type = pads[i].name;
- if (off_diff <= 0)
- /* no gap */
- return;
- if (m_bit_sz == 0 && off_diff < align * 8)
- /* natural padding will take care of a gap */
- return;
+ new_off = roundup(cur_off, pad_bits);
+ if (new_off <= next_off)
+ break;
+ }
- while (off_diff > 0) {
- const char *pad_type;
- int pad_bits;
-
- if (ptr_bits > 32 && off_diff > 32) {
- pad_type = "long";
- pad_bits = chip_away_bits(off_diff, ptr_bits);
- } else if (off_diff > 16) {
- pad_type = "int";
- pad_bits = chip_away_bits(off_diff, 32);
- } else if (off_diff > 8) {
- pad_type = "short";
- pad_bits = chip_away_bits(off_diff, 16);
- } else {
- pad_type = "char";
- pad_bits = chip_away_bits(off_diff, 8);
+ if (new_off > cur_off && new_off <= next_off) {
+ /* We need explicit `<type>: 0` aligning mark if next
+ * field is right on alignment offset and its
+ * alignment requirement is less strict than <type>'s
+ * alignment (so compiler won't naturally align to the
+ * offset we expect), or if subsequent `<type>: X`,
+ * will actually completely fit in the remaining hole,
+ * making compiler basically ignore `<type>: X`
+ * completely.
+ */
+ if (in_bitfield ||
+ (new_off == next_off && roundup(cur_off, next_align * 8) != new_off) ||
+ (new_off != next_off && next_off - new_off <= new_off - cur_off))
+ /* but for bitfields we'll emit explicit bit count */
+ btf_dump_printf(d, "\n%s%s: %d;", pfx(lvl), pad_type,
+ in_bitfield ? new_off - cur_off : 0);
+ cur_off = new_off;
+ }
+
+ /* Now we know we start at naturally aligned offset for a chosen
+ * padding type (long, int, short, or char), and so the rest is just
+ * a straightforward filling of remaining padding gap with full
+ * `<type>: sizeof(<type>);` markers, except for the last one, which
+ * might need smaller than sizeof(<type>) padding.
+ */
+ while (cur_off != next_off) {
+ bits = min(next_off - cur_off, pad_bits);
+ if (bits == pad_bits) {
+ btf_dump_printf(d, "\n%s%s: %d;", pfx(lvl), pad_type, pad_bits);
+ cur_off += bits;
+ continue;
+ }
+ /* For the remainder padding that doesn't cover entire
+ * pad_type bit length, we pick the smallest necessary type.
+ * This is pure aesthetics, we could have just used `long`,
+ * but having smallest necessary one communicates better the
+ * scale of the padding gap.
+ */
+ for (i = ARRAY_SIZE(pads) - 1; i >= 0; i--) {
+ pad_type = pads[i].name;
+ pad_bits = pads[i].bits;
+ if (pad_bits < bits)
+ continue;
+
+ btf_dump_printf(d, "\n%s%s: %d;", pfx(lvl), pad_type, bits);
+ cur_off += bits;
+ break;
}
- btf_dump_printf(d, "\n%s%s: %d;", pfx(lvl), pad_type, pad_bits);
- off_diff -= pad_bits;
}
}
@@ -903,9 +975,11 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
{
const struct btf_member *m = btf_members(t);
bool is_struct = btf_is_struct(t);
- int align, i, packed, off = 0;
+ bool packed, prev_bitfield = false;
+ int align, i, off = 0;
__u16 vlen = btf_vlen(t);
+ align = btf__align_of(d->btf, id);
packed = is_struct ? btf_is_struct_packed(d->btf, id, t) : 0;
btf_dump_printf(d, "%s%s%s {",
@@ -915,33 +989,36 @@ static void btf_dump_emit_struct_def(struct btf_dump *d,
for (i = 0; i < vlen; i++, m++) {
const char *fname;
- int m_off, m_sz;
+ int m_off, m_sz, m_align;
+ bool in_bitfield;
fname = btf_name_of(d, m->name_off);
m_sz = btf_member_bitfield_size(t, i);
m_off = btf_member_bit_offset(t, i);
- align = packed ? 1 : btf__align_of(d->btf, m->type);
+ m_align = packed ? 1 : btf__align_of(d->btf, m->type);
- btf_dump_emit_bit_padding(d, off, m_off, m_sz, align, lvl + 1);
+ in_bitfield = prev_bitfield && m_sz != 0;
+
+ btf_dump_emit_bit_padding(d, off, m_off, m_align, in_bitfield, lvl + 1);
btf_dump_printf(d, "\n%s", pfx(lvl + 1));
btf_dump_emit_type_decl(d, m->type, fname, lvl + 1);
if (m_sz) {
btf_dump_printf(d, ": %d", m_sz);
off = m_off + m_sz;
+ prev_bitfield = true;
} else {
m_sz = max((__s64)0, btf__resolve_size(d->btf, m->type));
off = m_off + m_sz * 8;
+ prev_bitfield = false;
}
+
btf_dump_printf(d, ";");
}
/* pad at the end, if necessary */
- if (is_struct) {
- align = packed ? 1 : btf__align_of(d->btf, id);
- btf_dump_emit_bit_padding(d, off, t->size * 8, 0, align,
- lvl + 1);
- }
+ if (is_struct)
+ btf_dump_emit_bit_padding(d, off, t->size * 8, align, false, lvl + 1);
if (vlen)
btf_dump_printf(d, "\n");
diff --git a/tools/testing/selftests/bpf/progs/btf_dump_test_case_bitfields.c b/tools/testing/selftests/bpf/progs/btf_dump_test_case_bitfields.c
index 8f44767a75fa5..22a7cd8fd9acf 100644
--- a/tools/testing/selftests/bpf/progs/btf_dump_test_case_bitfields.c
+++ b/tools/testing/selftests/bpf/progs/btf_dump_test_case_bitfields.c
@@ -53,7 +53,7 @@ struct bitfields_only_mixed_types {
*/
/* ------ END-EXPECTED-OUTPUT ------ */
struct bitfield_mixed_with_others {
- long: 4; /* char is enough as a backing field */
+ char: 4; /* char is enough as a backing field */
int a: 4;
/* 8-bit implicit padding */
short b; /* combined with previous bitfield */
diff --git a/tools/testing/selftests/bpf/progs/btf_dump_test_case_padding.c b/tools/testing/selftests/bpf/progs/btf_dump_test_case_padding.c
index db5458da61826..28833f2694e9c 100644
--- a/tools/testing/selftests/bpf/progs/btf_dump_test_case_padding.c
+++ b/tools/testing/selftests/bpf/progs/btf_dump_test_case_padding.c
@@ -19,7 +19,7 @@ struct padded_implicitly {
/*
*struct padded_explicitly {
* int a;
- * int: 32;
+ * long: 0;
* int b;
*};
*
@@ -28,41 +28,28 @@ struct padded_implicitly {
struct padded_explicitly {
int a;
- int: 1; /* algo will explicitly pad with full 32 bits here */
+ int: 1; /* algo will emit aligning `long: 0;` here */
int b;
};
/* ----- START-EXPECTED-OUTPUT ----- */
-/*
- *struct padded_a_lot {
- * int a;
- * long: 32;
- * long: 64;
- * long: 64;
- * int b;
- *};
- *
- */
-/* ------ END-EXPECTED-OUTPUT ------ */
-
struct padded_a_lot {
int a;
- /* 32 bit of implicit padding here, which algo will make explicit */
long: 64;
long: 64;
int b;
};
+/* ------ END-EXPECTED-OUTPUT ------ */
+
/* ----- START-EXPECTED-OUTPUT ----- */
/*
*struct padded_cache_line {
* int a;
- * long: 32;
* long: 64;
* long: 64;
* long: 64;
* int b;
- * long: 32;
* long: 64;
* long: 64;
* long: 64;
@@ -85,7 +72,7 @@ struct padded_cache_line {
*struct zone {
* int a;
* short b;
- * short: 16;
+ * long: 0;
* struct zone_padding __pad__;
*};
*
@@ -108,6 +95,39 @@ struct padding_wo_named_members {
long: 64;
};
+struct padding_weird_1 {
+ int a;
+ long: 64;
+ short: 16;
+ short b;
+};
+
+/* ------ END-EXPECTED-OUTPUT ------ */
+
+/* ----- START-EXPECTED-OUTPUT ----- */
+/*
+ *struct padding_weird_2 {
+ * long: 56;
+ * char a;
+ * long: 56;
+ * char b;
+ * char: 8;
+ *};
+ *
+ */
+/* ------ END-EXPECTED-OUTPUT ------ */
+struct padding_weird_2 {
+ int: 32; /* these paddings will be collapsed into `long: 56;` */
+ short: 16;
+ char: 8;
+ char a;
+ int: 32; /* these paddings will be collapsed into `long: 56;` */
+ short: 16;
+ char: 8;
+ char b;
+ char: 8;
+};
+
/* ------ END-EXPECTED-OUTPUT ------ */
int f(struct {
@@ -117,6 +137,8 @@ int f(struct {
struct padded_cache_line _4;
struct zone _5;
struct padding_wo_named_members _6;
+ struct padding_weird_1 _7;
+ struct padding_weird_2 _8;
} *_)
{
return 0;
--
2.39.2
next prev parent reply other threads:[~2023-04-03 14:35 UTC|newest]
Thread overview: 108+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-03 14:08 [PATCH 5.15 00/99] 5.15.106-rc1 review Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 01/99] fsverity: dont drop pagecache at end of FS_IOC_ENABLE_VERITY Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 02/99] usb: dwc3: gadget: move cmd_endtransfer to extra function Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 03/99] usb: dwc3: gadget: Add 1ms delay after end transfer command without IOC Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 04/99] kernel: kcsan: kcsan_test: build without structleak plugin Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 05/99] kcsan: avoid passing -g for test Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 06/99] ksmbd: dont terminate inactive sessions after a few seconds Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 07/99] bus: imx-weim: fix branch condition evaluates to a garbage value Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 08/99] xfrm: Zero padding when dumping algos and encap Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 09/99] ASoC: codecs: tx-macro: Fix for KASAN: slab-out-of-bounds Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 10/99] md: avoid signed overflow in slot_store() Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 11/99] x86/PVH: obtain VGA console info in Dom0 Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 12/99] net: hsr: Dont log netdev_err message on unknown prp dst node Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 13/99] ALSA: asihpi: check pao in control_message() Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 14/99] ALSA: hda/ca0132: fixup buffer overrun at tuning_ctl_set() Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 15/99] fbdev: tgafb: Fix potential divide by zero Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 16/99] sched_getaffinity: dont assume cpumask_size() is fully initialized Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 17/99] fbdev: nvidia: Fix potential divide by zero Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 18/99] fbdev: intelfb: " Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 19/99] fbdev: lxfb: " Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 20/99] fbdev: au1200fb: " Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 21/99] tools/power turbostat: Fix /dev/cpu_dma_latency warnings Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 22/99] tools/power turbostat: fix decoding of HWP_STATUS Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 23/99] tracing: Fix wrong return in kprobe_event_gen_test.c Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 24/99] ca8210: Fix unsigned mac_len comparison with zero in ca8210_skb_tx() Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 25/99] mips: bmips: BCM6358: disable RAC flush for TP1 Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 26/99] ALSA: usb-audio: Fix recursive locking at XRUN during syncing Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 27/99] platform/x86: think-lmi: add missing type attribute Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 28/99] platform/x86: think-lmi: use correct possible_values delimiters Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 29/99] platform/x86: think-lmi: only display possible_values if available Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 30/99] platform/x86: think-lmi: Add possible_values for ThinkStation Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 31/99] mtd: rawnand: meson: invalidate cache on polling ECC bit Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 32/99] SUNRPC: fix shutdown of NFS TCP client socket Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 33/99] sfc: ef10: dont overwrite offload features at NIC reset Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 34/99] scsi: megaraid_sas: Fix crash after a double completion Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 35/99] scsi: mpt3sas: Dont print sense pool info twice Greg Kroah-Hartman
2023-04-03 14:08 ` [PATCH 5.15 36/99] ptp_qoriq: fix memory leak in probe() Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 37/99] net: dsa: microchip: ksz8863_smi: fix bulk access Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 38/99] r8169: fix RTL8168H and RTL8107E rx crc error Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 39/99] regulator: Handle deferred clk Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 40/99] net/net_failover: fix txq exceeding warning Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 41/99] net: stmmac: dont reject VLANs when IFF_PROMISC is set Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 42/99] drm/i915/tc: Fix the ICL PHY ownership check in TC-cold state Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 43/99] platform/x86/intel/pmc: Alder Lake PCH slp_s0_residency fix Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 44/99] can: bcm: bcm_tx_setup(): fix KMSAN uninit-value in vfs_write Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 45/99] s390/vfio-ap: fix memory leak in vfio_ap device driver Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 46/99] loop: suppress uevents while reconfiguring the device Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 47/99] loop: LOOP_CONFIGURE: send uevents for partitions Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 48/99] net: mvpp2: classifier flow fix fragmentation flags Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 49/99] net: mvpp2: parser fix QinQ Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 50/99] net: mvpp2: parser fix PPPoE Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 51/99] smsc911x: avoid PHY being resumed when interface is not up Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 52/99] ice: add profile conflict check for AVF FDIR Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 53/99] ice: fix invalid check for empty list in ice_sched_assoc_vsi_to_agg() Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 54/99] ALSA: ymfpci: Create card with device-managed snd_devm_card_new() Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 55/99] ALSA: ymfpci: Fix BUG_ON in probe function Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 56/99] net: ipa: compute DMA pool size properly Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 57/99] i40e: fix registers dump after run ethtool adapter self test Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 58/99] bnxt_en: Fix reporting of test result in ethtool selftest Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 59/99] bnxt_en: Fix typo in PCI id to device description string mapping Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 60/99] bnxt_en: Add missing 200G link speed reporting Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 61/99] net: dsa: mv88e6xxx: Enable IGMP snooping on user ports only Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 62/99] net: ethernet: mtk_eth_soc: fix flow block refcounting logic Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 63/99] pinctrl: ocelot: Fix alt mode for ocelot Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 64/99] iommu/vt-d: Allow zero SAGAW if second-stage not supported Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 65/99] Input: alps - fix compatibility with -funsigned-char Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 66/99] Input: focaltech - use explicitly signed char type Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 67/99] cifs: prevent infinite recursion in CIFSGetDFSRefer() Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 68/99] cifs: fix DFS traversal oops without CONFIG_CIFS_DFS_UPCALL Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 69/99] Input: goodix - add Lenovo Yoga Book X90F to nine_bytes_report DMI table Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 70/99] btrfs: fix race between quota disable and quota assign ioctls Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 71/99] btrfs: scan device in non-exclusive mode Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 72/99] zonefs: Always invalidate last cached page on append write Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 73/99] can: j1939: prevent deadlock by moving j1939_sk_errqueue() Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 74/99] xen/netback: dont do grant copy across page boundary Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 75/99] net: phy: dp83869: fix default value for tx-/rx-internal-delay Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 76/99] pinctrl: amd: Disable and mask interrupts on resume Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 77/99] pinctrl: at91-pio4: fix domain name assignment Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 78/99] powerpc: Dont try to copy PPR for task with NULL pt_regs Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 79/99] NFSv4: Fix hangs when recovering open state after a server reboot Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 80/99] ALSA: hda/conexant: Partial revert of a quirk for Lenovo Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 81/99] ALSA: usb-audio: Fix regression on detection of Roland VS-100 Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 82/99] ALSA: hda/realtek: Add quirks for some Clevo laptops Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 83/99] ALSA: hda/realtek: Add quirk for Lenovo ZhaoYang CF4620Z Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 84/99] xtensa: fix KASAN report for show_stack Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 85/99] rcu: Fix rcu_torture_read ftrace event Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 86/99] drm/etnaviv: fix reference leak when mmaping imported buffer Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 87/99] drm/amd/display: Add DSC Support for Synaptics Cascaded MST Hub Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 88/99] KVM: arm64: Disable interrupts while walking userspace PTs Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 89/99] s390/uaccess: add missing earlyclobber annotations to __clear_user() Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 90/99] KVM: VMX: Move preemption timer <=> hrtimer dance to common x86 Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 91/99] KVM: x86: Inject #GP on x2APIC WRMSR that sets reserved bits 63:32 Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 92/99] KVM: x86: Purge "highest ISR" cache when updating APICv state Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 93/99] zonefs: Fix error message in zonefs_file_dio_append() Greg Kroah-Hartman
2023-04-03 14:09 ` [PATCH 5.15 94/99] selftests/bpf: Test btf dump for struct with padding only fields Greg Kroah-Hartman
2023-04-03 14:09 ` Greg Kroah-Hartman [this message]
2023-04-03 14:09 ` [PATCH 5.15 96/99] selftests/bpf: Add few corner cases to test padding handling of btf_dump Greg Kroah-Hartman
2023-04-03 14:10 ` [PATCH 5.15 97/99] libbpf: Fix btf_dumps packed struct determination Greg Kroah-Hartman
2023-04-03 14:10 ` [PATCH 5.15 98/99] hsr: ratelimit only when errors are printed Greg Kroah-Hartman
2023-04-03 14:10 ` [PATCH 5.15 99/99] x86/PVH: avoid 32-bit build warning when obtaining VGA console info Greg Kroah-Hartman
2023-04-03 22:59 ` [PATCH 5.15 00/99] 5.15.106-rc1 review Shuah Khan
2023-04-04 2:06 ` Florian Fainelli
2023-04-04 2:41 ` Bagas Sanjaya
2023-04-04 10:05 ` Naresh Kamboju
2023-04-04 10:55 ` Chris Paterson
2023-04-04 21:21 ` Guenter Roeck
2023-04-05 6:08 ` Harshit Mogalapalli
2023-04-05 8:49 ` Ron Economos
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=20230403140406.904734982@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=andrii@kernel.org \
--cc=daniel@iogearbox.net \
--cc=patches@lists.linux.dev \
--cc=per.xp.sundstrom@ericsson.com \
--cc=sashal@kernel.org \
--cc=stable@vger.kernel.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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox