From: Kyle Huey <me(a)kylehuey.com>
When management of the PKRU register was moved away from XSTATE, emulation
of PKRU's existence in XSTATE was added for APIs that read XSTATE, but not
for APIs that write XSTATE. This can be seen by running gdb and executing
`p $pkru`, `set $pkru = 42`, and `p $pkru`. On affected kernels (5.14+) the
write to the PKRU register (which gdb performs through ptrace) is ignored.
There are three relevant APIs: PTRACE_SETREGSET with NT_X86_XSTATE,
sigreturn, and KVM_SET_XSAVE. KVM_SET_XSAVE has its own special handling to
make PKRU writes take effect (in fpu_copy_uabi_to_guest_fpstate). Push that
down into copy_uabi_to_xstate and have PTRACE_SETREGSET with NT_X86_XSTATE
and sigreturn pass in pointers to the appropriate PKRU value.
This also adds code to initialize the PKRU value to the hardware init value
(namely 0) if the PKRU bit is not set in the XSTATE header to match XRSTOR.
This is a change to the current KVM_SET_XSAVE behavior.
Signed-off-by: Kyle Huey <me(a)kylehuey.com>
Cc: kvm(a)vger.kernel.org # For edge case behavior of KVM_SET_XSAVE
Cc: stable(a)vger.kernel.org # 5.14+
Fixes: e84ba47e313dbc097bf859bb6e4f9219883d5f78
---
arch/x86/kernel/fpu/core.c | 11 +----------
arch/x86/kernel/fpu/regset.c | 2 +-
arch/x86/kernel/fpu/signal.c | 2 +-
arch/x86/kernel/fpu/xstate.c | 26 +++++++++++++++++++++-----
arch/x86/kernel/fpu/xstate.h | 4 ++--
5 files changed, 26 insertions(+), 19 deletions(-)
diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c
index 0531d6a06df5..dfb79e2ee81f 100644
--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -406,16 +406,7 @@ int fpu_copy_uabi_to_guest_fpstate(struct fpu_guest *gfpu, const void *buf,
if (ustate->xsave.header.xfeatures & ~xcr0)
return -EINVAL;
- ret = copy_uabi_from_kernel_to_xstate(kstate, ustate);
- if (ret)
- return ret;
-
- /* Retrieve PKRU if not in init state */
- if (kstate->regs.xsave.header.xfeatures & XFEATURE_MASK_PKRU) {
- xpkru = get_xsave_addr(&kstate->regs.xsave, XFEATURE_PKRU);
- *vpkru = xpkru->pkru;
- }
- return 0;
+ return copy_uabi_from_kernel_to_xstate(kstate, ustate, vpkru);
}
EXPORT_SYMBOL_GPL(fpu_copy_uabi_to_guest_fpstate);
#endif /* CONFIG_KVM */
diff --git a/arch/x86/kernel/fpu/regset.c b/arch/x86/kernel/fpu/regset.c
index 75ffaef8c299..6d056b68f4ed 100644
--- a/arch/x86/kernel/fpu/regset.c
+++ b/arch/x86/kernel/fpu/regset.c
@@ -167,7 +167,7 @@ int xstateregs_set(struct task_struct *target, const struct user_regset *regset,
}
fpu_force_restore(fpu);
- ret = copy_uabi_from_kernel_to_xstate(fpu->fpstate, kbuf ?: tmpbuf);
+ ret = copy_uabi_from_kernel_to_xstate(fpu->fpstate, kbuf ?: tmpbuf, &target->thread.pkru);
out:
vfree(tmpbuf);
diff --git a/arch/x86/kernel/fpu/signal.c b/arch/x86/kernel/fpu/signal.c
index 91d4b6de58ab..558076dbde5b 100644
--- a/arch/x86/kernel/fpu/signal.c
+++ b/arch/x86/kernel/fpu/signal.c
@@ -396,7 +396,7 @@ static bool __fpu_restore_sig(void __user *buf, void __user *buf_fx,
fpregs = &fpu->fpstate->regs;
if (use_xsave() && !fx_only) {
- if (copy_sigframe_from_user_to_xstate(fpu->fpstate, buf_fx))
+ if (copy_sigframe_from_user_to_xstate(tsk, buf_fx))
return false;
} else {
if (__copy_from_user(&fpregs->fxsave, buf_fx,
diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c
index c8340156bfd2..1eea7af4afd9 100644
--- a/arch/x86/kernel/fpu/xstate.c
+++ b/arch/x86/kernel/fpu/xstate.c
@@ -1197,7 +1197,7 @@ static int copy_from_buffer(void *dst, unsigned int offset, unsigned int size,
static int copy_uabi_to_xstate(struct fpstate *fpstate, const void *kbuf,
- const void __user *ubuf)
+ const void __user *ubuf, u32 *pkru)
{
struct xregs_state *xsave = &fpstate->regs.xsave;
unsigned int offset, size;
@@ -1235,6 +1235,22 @@ static int copy_uabi_to_xstate(struct fpstate *fpstate, const void *kbuf,
for (i = 0; i < XFEATURE_MAX; i++) {
mask = BIT_ULL(i);
+ if (i == XFEATURE_PKRU) {
+ /*
+ * Retrieve PKRU if not in init state, otherwise
+ * initialize it.
+ */
+ if (hdr.xfeatures & mask) {
+ struct pkru_state xpkru = {0};
+
+ copy_from_buffer(&xpkru, xstate_offsets[i],
+ sizeof(xpkru), kbuf, ubuf);
+ *pkru = xpkru.pkru;
+ } else {
+ *pkru = 0;
+ }
+ }
+
if (hdr.xfeatures & mask) {
void *dst = __raw_xsave_addr(xsave, i);
@@ -1264,9 +1280,9 @@ static int copy_uabi_to_xstate(struct fpstate *fpstate, const void *kbuf,
* Convert from a ptrace standard-format kernel buffer to kernel XSAVE[S]
* format and copy to the target thread. Used by ptrace and KVM.
*/
-int copy_uabi_from_kernel_to_xstate(struct fpstate *fpstate, const void *kbuf)
+int copy_uabi_from_kernel_to_xstate(struct fpstate *fpstate, const void *kbuf, u32 *pkru)
{
- return copy_uabi_to_xstate(fpstate, kbuf, NULL);
+ return copy_uabi_to_xstate(fpstate, kbuf, NULL, pkru);
}
/*
@@ -1274,10 +1290,10 @@ int copy_uabi_from_kernel_to_xstate(struct fpstate *fpstate, const void *kbuf)
* XSAVE[S] format and copy to the target thread. This is called from the
* sigreturn() and rt_sigreturn() system calls.
*/
-int copy_sigframe_from_user_to_xstate(struct fpstate *fpstate,
+int copy_sigframe_from_user_to_xstate(struct task_struct *tsk,
const void __user *ubuf)
{
- return copy_uabi_to_xstate(fpstate, NULL, ubuf);
+ return copy_uabi_to_xstate(tsk->thread.fpu.fpstate, NULL, ubuf, &tsk->thread.pkru);
}
static bool validate_independent_components(u64 mask)
diff --git a/arch/x86/kernel/fpu/xstate.h b/arch/x86/kernel/fpu/xstate.h
index 5ad47031383b..a4ecb04d8d64 100644
--- a/arch/x86/kernel/fpu/xstate.h
+++ b/arch/x86/kernel/fpu/xstate.h
@@ -46,8 +46,8 @@ extern void __copy_xstate_to_uabi_buf(struct membuf to, struct fpstate *fpstate,
u32 pkru_val, enum xstate_copy_mode copy_mode);
extern void copy_xstate_to_uabi_buf(struct membuf to, struct task_struct *tsk,
enum xstate_copy_mode mode);
-extern int copy_uabi_from_kernel_to_xstate(struct fpstate *fpstate, const void *kbuf);
-extern int copy_sigframe_from_user_to_xstate(struct fpstate *fpstate, const void __user *ubuf);
+extern int copy_uabi_from_kernel_to_xstate(struct fpstate *fpstate, const void *kbuf, u32 *pkru);
+extern int copy_sigframe_from_user_to_xstate(struct task_struct *tsk, const void __user *ubuf);
extern void fpu__init_cpu_xstate(void);
--
2.37.0
The patch titled
Subject: mm: vmscan: fix extreme overreclaim and swap floods
has been added to the -mm mm-hotfixes-unstable branch. Its filename is
mm-vmscan-fix-extreme-overreclaim-and-swap-floods.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patche…
This patch will later appear in the mm-hotfixes-unstable branch at
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days
------------------------------------------------------
From: Johannes Weiner <hannes(a)cmpxchg.org>
Subject: mm: vmscan: fix extreme overreclaim and swap floods
Date: Tue, 2 Aug 2022 12:28:11 -0400
During proactive reclaim, we sometimes observe severe overreclaim, with
several thousand times more pages reclaimed than requested.
This trace was obtained from shrink_lruvec() during such an instance:
prio:0 anon_cost:1141521 file_cost:7767
nr_reclaimed:4387406 nr_to_reclaim:1047 (or_factor:4190)
nr=[7161123 345 578 1111]
While he reclaimer requested 4M, vmscan reclaimed close to 16G, most of it
by swapping. These requests take over a minute, during which the write()
to memory.reclaim is unkillably stuck inside the kernel.
Digging into the source, this is caused by the proportional reclaim
bailout logic. This code tries to resolve a fundamental conflict: to
reclaim roughly what was requested, while also aging all LRUs fairly and
in accordance to their size, swappiness, refault rates etc. The way it
attempts fairness is that once the reclaim goal has been reached, it stops
scanning the LRUs with the smaller remaining scan targets, and adjusts the
remainder of the bigger LRUs according to how much of the smaller LRUs was
scanned. It then finishes scanning that remainder regardless of the
reclaim goal.
This works fine if priority levels are low and the LRU lists are
comparable in size. However, in this instance, the cgroup that is
targeted by proactive reclaim has almost no files left - they've already
been squeezed out by proactive reclaim earlier - and the remaining anon
pages are hot. Anon rotations cause the priority level to drop to 0,
which results in reclaim targeting all of anon (a lot) and all of file
(almost nothing). By the time reclaim decides to bail, it has scanned
most or all of the file target, and therefor must also scan most or all of
the enormous anon target. This target is thousands of times larger than
the reclaim goal, thus causing the overreclaim.
The bailout code hasn't changed in years, why is this failing now? The
most likely explanations are two other recent changes in anon reclaim:
1. Before the series starting with commit 5df741963d52 ("mm: fix LRU
balancing effect of new transparent huge pages"), the VM was
overall relatively reluctant to swap at all, even if swap was
configured. This means the LRU balancing code didn't come into play
as often as it does now, and mostly in high pressure situations
where pronounced swap activity wouldn't be as surprising.
2. For historic reasons, shrink_lruvec() loops on the scan targets of
all LRU lists except the active anon one, meaning it would bail if
the only remaining pages to scan were active anon - even if there
were a lot of them.
Before the series starting with commit ccc5dc67340c ("mm/vmscan:
make active/inactive ratio as 1:1 for anon lru"), most anon pages
would live on the active LRU; the inactive one would contain only a
handful of preselected reclaim candidates. After the series, anon
gets aged similarly to file, and the inactive list is the default
for new anon pages as well, making it often the much bigger list.
As a result, the VM is now more likely to actually finish large
anon targets than before.
Change the code such that only one SWAP_CLUSTER_MAX-sized nudge toward the
larger LRU lists is made before bailing out on a met reclaim goal.
This fixes the extreme overreclaim problem.
Fairness is more subtle and harder to evaluate. No obvious misbehavior
was observed on the test workload, in any case. Conceptually, fairness
should primarily be a cumulative effect from regular, lower priority
scans. Once the VM is in trouble and needs to escalate scan targets to
make forward progress, fairness needs to take a backseat. This is also
acknowledged by the myriad exceptions in get_scan_count(). This patch
makes fairness decrease gradually, as it keeps fairness work static over
increasing priority levels with growing scan targets. This should make
more sense - although we may have to re-visit the exact values.
Link: https://lkml.kernel.org/r/20220802162811.39216-1-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes(a)cmpxchg.org>
Cc: Mel Gorman <mgorman(a)techsingularity.net>
Cc: Hugh Dickins <hughd(a)google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim(a)lge.com>
Cc: <stable(a)vger.kernel.org>
Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org>
---
mm/vmscan.c | 10 ++++------
1 file changed, 4 insertions(+), 6 deletions(-)
--- a/mm/vmscan.c~mm-vmscan-fix-extreme-overreclaim-and-swap-floods
+++ a/mm/vmscan.c
@@ -2897,8 +2897,8 @@ static void shrink_lruvec(struct lruvec
enum lru_list lru;
unsigned long nr_reclaimed = 0;
unsigned long nr_to_reclaim = sc->nr_to_reclaim;
+ bool proportional_reclaim;
struct blk_plug plug;
- bool scan_adjusted;
get_scan_count(lruvec, sc, nr);
@@ -2916,8 +2916,8 @@ static void shrink_lruvec(struct lruvec
* abort proportional reclaim if either the file or anon lru has already
* dropped to zero at the first pass.
*/
- scan_adjusted = (!cgroup_reclaim(sc) && !current_is_kswapd() &&
- sc->priority == DEF_PRIORITY);
+ proportional_reclaim = (!cgroup_reclaim(sc) && !current_is_kswapd() &&
+ sc->priority == DEF_PRIORITY);
blk_start_plug(&plug);
while (nr[LRU_INACTIVE_ANON] || nr[LRU_ACTIVE_FILE] ||
@@ -2937,7 +2937,7 @@ static void shrink_lruvec(struct lruvec
cond_resched();
- if (nr_reclaimed < nr_to_reclaim || scan_adjusted)
+ if (nr_reclaimed < nr_to_reclaim || proportional_reclaim)
continue;
/*
@@ -2988,8 +2988,6 @@ static void shrink_lruvec(struct lruvec
nr_scanned = targets[lru] - nr[lru];
nr[lru] = targets[lru] * (100 - percentage) / 100;
nr[lru] -= min(nr[lru], nr_scanned);
-
- scan_adjusted = true;
}
blk_finish_plug(&plug);
sc->nr_reclaimed += nr_reclaimed;
_
Patches currently in -mm which might be from hannes(a)cmpxchg.org are
mm-vmscan-fix-extreme-overreclaim-and-swap-floods.patch