data_type
large_stringclasses
3 values
source
large_stringclasses
29 values
code
large_stringlengths
98
49.4M
filepath
large_stringlengths
5
161
message
large_stringclasses
234 values
commit
large_stringclasses
234 values
subject
large_stringclasses
418 values
critique
large_stringlengths
101
1.26M
metadata
dict
lkml_critique
netdev
Hi Laurent, I'm testing the latest openwrt state and found an issue probably caused by your usb mtu limit patch :-) I'm talking about this one: 662dc80a5e86 ("usbnet: limit max_mtu based on device's hard_mtu") https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.12.74&id=662dc80a5e86b35bbf33...
null
null
null
commit 662dc80a5e86 breaks rmnet over usb
On Wed, 25 Feb 2026 08:19:46 +0100 Daniele Palmas wrote: What's the policy in ModemManager to change the rx_urb_size? Increase it to make sure it can hold some specific cmd / packet? I wonder if having rx_urb_size max of (mtu, 32k) would break anything. Since we're talking about rx buffer config the right API to con...
{ "author": "Jakub Kicinski <kuba@kernel.org>", "date": "Wed, 25 Feb 2026 17:01:55 -0800", "is_openbsd": false, "thread_id": "CAGRyCJFDKv+U4004bPKVGGq292nNiOxuovOc0magetRCJ8vNaQ@mail.gmail.com.mbox.gz" }
lkml_critique
netdev
Hi Laurent, I'm testing the latest openwrt state and found an issue probably caused by your usb mtu limit patch :-) I'm talking about this one: 662dc80a5e86 ("usbnet: limit max_mtu based on device's hard_mtu") https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.12.74&id=662dc80a5e86b35bbf33...
null
null
null
commit 662dc80a5e86 breaks rmnet over usb
Hello Jakub, Il giorno gio 26 feb 2026 alle ore 02:01 Jakub Kicinski <kuba@kernel.org> ha scritto: Typically the host sends a QMI request to the modem for setting the size of the maximum QMAP aggregated packets block. Then the modem replies with the maximum size it supports and that one is then set by the host throug...
{ "author": "Daniele Palmas <dnlplm@gmail.com>", "date": "Thu, 26 Feb 2026 09:26:11 +0100", "is_openbsd": false, "thread_id": "CAGRyCJFDKv+U4004bPKVGGq292nNiOxuovOc0magetRCJ8vNaQ@mail.gmail.com.mbox.gz" }
lkml_critique
netdev
Hi Laurent, I'm testing the latest openwrt state and found an issue probably caused by your usb mtu limit patch :-) I'm talking about this one: 662dc80a5e86 ("usbnet: limit max_mtu based on device's hard_mtu") https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.12.74&id=662dc80a5e86b35bbf33...
null
null
null
commit 662dc80a5e86 breaks rmnet over usb
On 2/26/26 09:26, Daniele Palmas wrote: I was thinking to something like that (see below), but I'm not really able to test it. If everyone thinks it's the path to follow, I can send a patch. Thanks, Laurent diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c index 3a4985b582cb..05acac10cd2b 100644 ...
{ "author": "Laurent Vivier <lvivier@redhat.com>", "date": "Thu, 26 Feb 2026 10:09:07 +0100", "is_openbsd": false, "thread_id": "CAGRyCJFDKv+U4004bPKVGGq292nNiOxuovOc0magetRCJ8vNaQ@mail.gmail.com.mbox.gz" }
lkml_critique
netdev
Hi Laurent, I'm testing the latest openwrt state and found an issue probably caused by your usb mtu limit patch :-) I'm talking about this one: 662dc80a5e86 ("usbnet: limit max_mtu based on device's hard_mtu") https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.12.74&id=662dc80a5e86b35bbf33...
null
null
null
commit 662dc80a5e86 breaks rmnet over usb
Il giorno gio 26 feb 2026 alle ore 10:09 Laurent Vivier <lvivier@redhat.com> ha scritto: I think I can test this by the end of the week and let you know. Thanks, Daniele
{ "author": "Daniele Palmas <dnlplm@gmail.com>", "date": "Thu, 26 Feb 2026 19:18:53 +0100", "is_openbsd": false, "thread_id": "CAGRyCJFDKv+U4004bPKVGGq292nNiOxuovOc0magetRCJ8vNaQ@mail.gmail.com.mbox.gz" }
lkml_critique
netdev
Hi Laurent, I'm testing the latest openwrt state and found an issue probably caused by your usb mtu limit patch :-) I'm talking about this one: 662dc80a5e86 ("usbnet: limit max_mtu based on device's hard_mtu") https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v6.12.74&id=662dc80a5e86b35bbf33...
null
null
null
commit 662dc80a5e86 breaks rmnet over usb
Hello, Il giorno gio 26 feb 2026 alle ore 19:18 Daniele Palmas <dnlplm@gmail.com> ha scritto: I've verified that without this patch I'm not able to set mtu > 1500 and ip -d link shows: 3: wwan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000 link/ether 16:7f:28:3c:c4:3e...
{ "author": "Daniele Palmas <dnlplm@gmail.com>", "date": "Fri, 27 Feb 2026 17:15:10 +0100", "is_openbsd": false, "thread_id": "CAGRyCJFDKv+U4004bPKVGGq292nNiOxuovOc0magetRCJ8vNaQ@mail.gmail.com.mbox.gz" }
lkml_critique
linux-mm
This series can be found on GitHub: https://github.com/dmatlack/linux/tree/liveupdate/vfio/cdev/v2 This series adds the base support to preserve a VFIO device file across a Live Update. "Base support" means that this allows userspace to safely preserve a VFIO device file with LIVEUPDATE_SESSION_PRESERVE_FD and retr...
null
null
null
[PATCH v2 00/22] vfio/pci: Base Live Update support for VFIO device files
On Thu, 26 Feb 2026 10:40:57 -0400 Jason Gunthorpe <jgg@nvidia.com> wrote: Not only fabric topology, but also routing. ACS overrides on the command line would need to be enforced between the original and kexec kernel such that IOMMU groups are deterministic. Thanks, Alex
{ "author": "Alex Williamson <alex@shazbot.org>", "date": "Fri, 27 Feb 2026 09:04:49 -0700", "is_openbsd": false, "thread_id": "20260227090449.2a23d06d@shazbot.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
Compaction uses compact_lock_irqsave(), which currently operates on a raw spinlock_t pointer so it can be used for both zone->lock and lruvec->lru_lock. Since zone lock operations are now wrapped, compact_lock_irqsave() can no longer directly operate on a spinlock_t when the lock belongs to a zone. Split the helper in...
{ "author": "Dmitry Ilvokhin <d@ilvokhin.com>", "date": "Thu, 26 Feb 2026 18:26:20 +0000", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
Zone lock contention can significantly impact allocation and reclaim latency, as it is a central synchronization point in the page allocator and reclaim paths. Improved visibility into its behavior is therefore important for diagnosing performance issues in memory-intensive workloads. On some production workloads at M...
{ "author": "Dmitry Ilvokhin <d@ilvokhin.com>", "date": "Thu, 26 Feb 2026 18:26:17 +0000", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
Replace direct zone lock acquire/release operations with the newly introduced wrappers. The changes are purely mechanical substitutions. No functional change intended. Locking semantics and ordering remain unchanged. The compaction path is left unchanged for now and will be handled separately in the following patch d...
{ "author": "Dmitry Ilvokhin <d@ilvokhin.com>", "date": "Thu, 26 Feb 2026 18:26:19 +0000", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
Add thin wrappers around zone lock acquire/release operations. This prepares the code for future tracepoint instrumentation without modifying individual call sites. Centralizing zone lock operations behind wrappers allows future instrumentation or debugging hooks to be added without touching all users. No functional ...
{ "author": "Dmitry Ilvokhin <d@ilvokhin.com>", "date": "Thu, 26 Feb 2026 18:26:18 +0000", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
Add tracepoint instrumentation to zone lock acquire/release operations via the previously introduced wrappers. The implementation follows the mmap_lock tracepoint pattern: a lightweight inline helper checks whether the tracepoint is enabled and calls into an out-of-line helper when tracing is active. When CONFIG_TRACI...
{ "author": "Dmitry Ilvokhin <d@ilvokhin.com>", "date": "Thu, 26 Feb 2026 18:26:22 +0000", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, Feb 26, 2026 at 06:26:20PM +0000, Dmitry Ilvokhin wrote: Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
{ "author": "Shakeel Butt <shakeel.butt@linux.dev>", "date": "Thu, 26 Feb 2026 11:07:47 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, Feb 26, 2026 at 06:26:21PM +0000, Dmitry Ilvokhin wrote: Acked-by: Shakeel Butt <shakeel.butt@linux.dev>
{ "author": "Shakeel Butt <shakeel.butt@linux.dev>", "date": "Thu, 26 Feb 2026 11:09:59 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, Feb 26, 2026 at 06:26:22PM +0000, Dmitry Ilvokhin wrote: One nit below other than that: Acked-by: Shakeel Butt <shakeel.butt@linux.dev> [...] No reason to not have these as EXPORT_SYMBOL_GPL (& below)
{ "author": "Shakeel Butt <shakeel.butt@linux.dev>", "date": "Thu, 26 Feb 2026 11:14:52 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, 26 Feb 2026 11:14:52 -0800 Shakeel Butt <shakeel.butt@linux.dev> wrote: Do we need the exports at all? include/linux/mmzone.h include/linux/zone_lock.h include/trace/events/zone_lock.h MAINTAINERS mm/compaction.c mm/internal.h mm/Makefile mm/memory_hotplug.c mm/mm_init.c mm/page_alloc.c mm/page_isolation.c m...
{ "author": "Andrew Morton <akpm@linux-foundation.org>", "date": "Thu, 26 Feb 2026 13:25:01 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, Feb 26, 2026 at 01:25:01PM -0800, Andrew Morton wrote: Very good point and we don't. I think this might just be copying the mmap_lock tracepoint wrappers which might need the exports as some drivers might be taking the mmap_lock. Dmitry, please confirm (test) and let us know.
{ "author": "Shakeel Butt <shakeel.butt@linux.dev>", "date": "Thu, 26 Feb 2026 13:31:48 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
Hi Dmitry, kernel test robot noticed the following build errors: [auto build test ERROR on linus/master] [also build test ERROR on v7.0-rc1 next-20260226] [cannot apply to akpm-mm/mm-everything rppt-memblock/for-next rppt-memblock/fixes] [If your patch is applied to the wrong git tree, kindly drop us a note. And when...
{ "author": "kernel test robot <lkp@intel.com>", "date": "Fri, 27 Feb 2026 05:48:05 +0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Fri, 27 Feb 2026 05:48:05 +0800 kernel test robot <lkp@intel.com> wrote: yep, thanks. And kernel/power/snapshot.c. I've added fixups.
{ "author": "Andrew Morton <akpm@linux-foundation.org>", "date": "Thu, 26 Feb 2026 14:08:46 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
Hi Dmitry, kernel test robot noticed the following build errors: [auto build test ERROR on linus/master] [also build test ERROR on v7.0-rc1 next-20260226] [cannot apply to akpm-mm/mm-everything rppt-memblock/for-next rppt-memblock/fixes] [If your patch is applied to the wrong git tree, kindly drop us a note. And when...
{ "author": "kernel test robot <lkp@intel.com>", "date": "Fri, 27 Feb 2026 07:13:01 +0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, 26 Feb 2026 18:26:18 +0000 Dmitry Ilvokhin <d@ilvokhin.com> wrote: I'm bit worried if I will think this as a file for another general locking, not the mm specific one. I hence think renaming it to more clearly saying the fact, say, mmzone_lock.h, might be less confusing. Or, putting things in mmzone.h might...
{ "author": "SeongJae Park <sj@kernel.org>", "date": "Thu, 26 Feb 2026 16:31:39 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
I should sent this together with the previous reply, but I forgot as usual, sorry. On Thu, 26 Feb 2026 18:26:18 +0000 Dmitry Ilvokhin <d@ilvokhin.com> wrote: [...] checkpatch.pl complains as below. Should be ok to ignore, but, may better to kindly make it silence? WARNING: Single statement macros should not us...
{ "author": "SeongJae Park <sj@kernel.org>", "date": "Thu, 26 Feb 2026 16:38:55 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, 26 Feb 2026 18:26:19 +0000 Dmitry Ilvokhin <d@ilvokhin.com> wrote: Reviewed-by: SeongJae Park <sj@kernel.org> Thanks, SJ [...]
{ "author": "SeongJae Park <sj@kernel.org>", "date": "Thu, 26 Feb 2026 16:40:02 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, 26 Feb 2026 18:26:20 +0000 Dmitry Ilvokhin <d@ilvokhin.com> wrote: Nit. Why don't you keep the indentation? My impression based on below output is that mm code prefer indenting __acquires(). $ git grep __acquires mm mm/compaction.c:__acquires(&zone->_lock) mm/compaction.c:__acquires(&lruvec->lr...
{ "author": "SeongJae Park <sj@kernel.org>", "date": "Thu, 26 Feb 2026 16:45:54 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, 26 Feb 2026 18:26:21 +0000 Dmitry Ilvokhin <d@ilvokhin.com> wrote: Acked-by: SeongJae Park <sj@kernel.org> Thanks, SJ [...]
{ "author": "SeongJae Park <sj@kernel.org>", "date": "Thu, 26 Feb 2026 16:49:23 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, 26 Feb 2026 16:38:55 -0800 SeongJae Park <sj@kernel.org> wrote: Hmm, why is this an issue? I know this is checkpatch and not you complaining about it, but I really think it's a useless complaint. I can see it better as a do { } while (0) because it is creating a "function" like feature but can't be inline d...
{ "author": "Steven Rostedt <rostedt@goodmis.org>", "date": "Thu, 26 Feb 2026 19:53:55 -0500", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, 26 Feb 2026 19:53:55 -0500 Steven Rostedt <rostedt@goodmis.org> wrote: Makes sense to me, thank you Steve :) Thanks, SJ [...]
{ "author": "SeongJae Park <sj@kernel.org>", "date": "Thu, 26 Feb 2026 17:01:01 -0800", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, Feb 26, 2026 at 01:31:48PM -0800, Shakeel Butt wrote: This is a good catch, thank you. I don't think we need EXPORT_SYMBOL() here. Just verified it locally. I'll remove it in v4.
{ "author": "Dmitry Ilvokhin <d@ilvokhin.com>", "date": "Fri, 27 Feb 2026 12:39:00 +0000", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, Feb 26, 2026 at 04:31:39PM -0800, SeongJae Park wrote: Thanks for the feedback, SJ. Good point. I agree the current name looks too generic. Putting it into mmzone.h would further overload that header, so renaming zone_lock.h to mmzone_lock.h seems like the clearest option. I'll make that change in v4.
{ "author": "Dmitry Ilvokhin <d@ilvokhin.com>", "date": "Fri, 27 Feb 2026 13:06:00 +0000", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
This intentionally breaks direct users of zone->lock at compile time so all call sites are converted to the zone lock wrappers. Without the rename, present and future out-of-tree code could continue using spin_lock(&zone->lock) and bypass the wrappers and tracing infrastructure. No functional change intended. Suggest...
null
null
null
[PATCH v3 4/5] mm: rename zone->lock to zone->_lock
On Thu, Feb 26, 2026 at 04:45:54PM -0800, SeongJae Park wrote: Thanks for spotting it, will be fixed in v4.
{ "author": "Dmitry Ilvokhin <d@ilvokhin.com>", "date": "Fri, 27 Feb 2026 14:46:24 +0000", "is_openbsd": false, "thread_id": "20260227010102.83789-1-sj@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
The CMA heap instantiation was initially developed by having the contiguous DMA code call into the CMA heap to create a new instance every time a reserved memory area is probed. Turning the CMA heap into a module would create a dependency of the kernel on a module, which doesn't work. Let's turn the logic around and ...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:40 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
As we try to enable dma-buf heaps, and the CMA one in particular, to compile as modules, we need to export dev_get_cma_area(). It's currently implemented as an inline function that returns either the content of device->cma_area or dma_contiguous_default_area. Thus, it means we need to export dma_contiguous_default_are...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:41 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
Now that dev_get_cma_area() is no longer inline, we don't have any user of dma_contiguous_default_area() outside of contiguous.c so we can make it static. Signed-off-by: Maxime Ripard <mripard@kernel.org> --- include/linux/dma-map-ops.h | 2 -- kernel/dma/contiguous.c | 2 +- 2 files changed, 1 insertion(+), 3 de...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:42 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
The CMA dma-buf heap uses the dev_get_cma_area() function to retrieve the default contiguous area. Now that this function is no longer inlined, and since we want to turn the CMA heap into a module, let's export it. Signed-off-by: Maxime Ripard <mripard@kernel.org> --- kernel/dma/contiguous.c | 1 + 1 file changed, 1...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:43 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
The CMA dma-buf heap uses cma_alloc() and cma_release() to allocate and free, respectively, its CMA buffers. However, these functions are not exported. Since we want to turn the CMA heap into a module, let's export them both. Reviewed-by: T.J. Mercier <tjmercier@google.com> Signed-off-by: Maxime Ripard <mripard@kerne...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:44 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
The CMA dma-buf heap uses the cma_get_name() function to get the name of the heap instance it's going to create. However, this function is not exported. Since we want to turn the CMA heap into a module, let's export it. Reviewed-by: T.J. Mercier <tjmercier@google.com> Signed-off-by: Maxime Ripard <mripard@kernel.org>...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:45 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
The mem_accounting kernel parameter is used by heaps to know if they should account allocations in their respective cgroup controllers. Since we're going to allow heaps to compile as modules, we need to export that variable. Reviewed-by: T.J. Mercier <tjmercier@google.com> Signed-off-by: Maxime Ripard <mripard@kernel...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:46 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
Now that all the symbols used by the CMA heap are exported, turning the CMA heap into a module becomes pretty easy: we just need to add the usual MODULE_* macros, import the proper namespaces and change the Kconfig symbol to a tristate. This heap won't be able to unload though, since we're missing a lot of infrastruct...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:47 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH v2 0/9] dma-buf: heaps: Turn heaps into modules
The system heap can be easily turned into a module by adding the usual MODULE_* macros, importing the proper namespaces and changing the Kconfig symbol to a tristate. This heap won't be able to unload though, since we're missing a lot of infrastructure to make it safe. Reviewed-by: T.J. Mercier <tjmercier@google.com>...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:15:48 +0100", "is_openbsd": false, "thread_id": "20260227-dma-buf-heaps-as-modules-v2-8-454aee7e06cc@kernel.org.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
folio_end_dropbehind() is called from folio_end_writeback(), which can run in IRQ context through buffer_head completion. Previously, when folio_end_dropbehind() detected !in_task(), it skipped the invalidation entirely. This meant that folios marked for dropbehind via RWF_DONTCACHE would remain in the page cache afte...
{ "author": "Tal Zussman <tz2294@columbia.edu>", "date": "Wed, 25 Feb 2026 17:40:56 -0500", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
Block device buffered reads and writes already pass through filemap_read() and iomap_file_buffered_write() respectively, both of which handle IOCB_DONTCACHE. Enable RWF_DONTCACHE for block device files by setting FOP_DONTCACHE in def_blk_fops. For CONFIG_BUFFER_HEAD paths, thread the kiocb through block_write_begin() ...
{ "author": "Tal Zussman <tz2294@columbia.edu>", "date": "Wed, 25 Feb 2026 17:40:57 -0500", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
On 2/25/26 3:40 PM, Tal Zussman wrote: How well does this scale? I did a patch basically the same as this, but not using a folio batch though. But the main sticking point was dropbehind_lock contention, to the point where I left it alone and thought "ok maybe we just do this when we're done with the awful buffer_head ...
{ "author": "Jens Axboe <axboe@kernel.dk>", "date": "Wed, 25 Feb 2026 15:52:41 -0700", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
On Wed, Feb 25, 2026 at 5:52 PM Jens Axboe <axboe@kernel.dk> wrote: Yep, that makes sense. I think a per-cpu folio_batch, spinlock, and work_struct would solve this (assuming that's what you meant by per-cpu lists) and would be simple enough to implement. I can put that together and send it tomorrow. I'll see if I can...
{ "author": "Tal Zussman <tz2294@columbia.edu>", "date": "Wed, 25 Feb 2026 20:38:29 -0500", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
On Wed, Feb 25, 2026 at 03:52:41PM -0700, Jens Axboe wrote: Back in 2021, I had Vishal look at switching the page cache from using hardirq-disabling locks to softirq-disabling locks [1]. Some of the feedback (which doesn't seem to be entirely findable on the lists ...) was that we'd be better off punting writeback co...
{ "author": "Matthew Wilcox <willy@infradead.org>", "date": "Thu, 26 Feb 2026 02:55:23 +0000", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
On 2/25/26 6:38 PM, Tal Zussman wrote: Was just looking for my patch as well... I don't think I ever posted it, because I didn't like it very much. It's probably sitting in my git tree somewhere. But it looks very much the same as yours, modulo the folio batching. One thing to keep in mind with per-cpu lists and the...
{ "author": "Jens Axboe <axboe@kernel.dk>", "date": "Wed, 25 Feb 2026 20:11:04 -0700", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
On 2/25/26 7:55 PM, Matthew Wilcox wrote: Perhaps, even though the punting tends to suck... One idea I toyed with but had to abandon due to fs freezeing was letting callers that process completions in task context anyway just do the necessary work at that time. There's literally nothing worse than having part of a com...
{ "author": "Jens Axboe <axboe@kernel.dk>", "date": "Wed, 25 Feb 2026 20:15:28 -0700", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
On Wed, Feb 25, 2026 at 08:15:28PM -0700, Jens Axboe wrote: I spoke a little hastily when I said "all write completions". What I really meant was something like: +++ b/block/bio.c @@ -1788,7 +1788,9 @@ void bio_endio(struct bio *bio) } #endif - if (bio->bi_end_io) + if (!in_task() && bio_flagge...
{ "author": "Matthew Wilcox <willy@infradead.org>", "date": "Thu, 26 Feb 2026 21:12:13 +0000", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
On Wed, Feb 25, 2026 at 03:52:41PM -0700, Jens Axboe wrote: Having per-cpu lists would be nice, but I'd really love to have them in iomap, as we have quite a few iomap features that would benefit from generic offload to user context on completion. Right now we only have code for that in XFS, and that's because the li...
{ "author": "Christoph Hellwig <hch@infradead.org>", "date": "Thu, 26 Feb 2026 14:04:09 -0800", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
Please don't change the prototoype for block_write_begin and thus cause churn for all these legacy file systems. Add a new block_write_begin_iocb, and use that in the block code and to implement block_write_begin. And avoid the overly long line there to keep the code readable. Note that you also need to cover the !C...
{ "author": "Christoph Hellwig <hch@infradead.org>", "date": "Thu, 26 Feb 2026 14:07:14 -0800", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
Add support for using RWF_DONTCACHE with block devices and other buffer_head-based I/O. Dropbehind pruning needs to be done in non-IRQ context, but block devices complete writeback in IRQ context. To fix this, we first defer dropbehind completion initiated from IRQ context by scheduling a work item on the system workq...
null
null
null
[PATCH RFC v2 0/2] block: enable RWF_DONTCACHE for block devices
On Thu, Feb 26, 2026 at 5:07 PM Christoph Hellwig <hch@infradead.org> wrote: Will do. I don't think there is a !CONFIG_BUFFER_HEAD case. The only user of block_write_begin_iocb() would be blkdev_write_begin(), which is only defined under CONFIG_BUFFER_HEAD. !CONFIG_BUFFER_HEAD paths use iomap which doesn't use it. A...
{ "author": "Tal Zussman <tz2294@columbia.edu>", "date": "Thu, 26 Feb 2026 19:44:38 -0500", "is_openbsd": false, "thread_id": "20260225-blk-dontcache-v2-0-70e7ac4f7108@columbia.edu.mbox.gz" }
lkml_critique
linux-mm
To avoid potential problems related to cpu/compiler optimizations around ->child_reaper, let's use WRITE_ONCE (additional to task_list lock) everywhere we write it and use READ_ONCE where we read it without explicit lock. Note: It also pairs with existing READ_ONCE with no lock in nsfs_fh_to_dentry(). Also let's add A...
null
null
null
[PATCH v4 1/4] pid_namespace: avoid optimization of accesses to ->child_reaper
First testcase "pidns_init_via_setns" checks that a process can become Pid 1 (init) in a new Pid namespace created via unshare() and joined via setns(). Second testcase "pidns_init_via_setns_set_tid" checks that during this process we can use clone3() + set_tid and set the pid in both the new and old pid namespaces (o...
{ "author": "Pavel Tikhomirov <ptikhomirov@virtuozzo.com>", "date": "Wed, 25 Feb 2026 14:32:26 +0100", "is_openbsd": false, "thread_id": "20260225133229.550302-1-ptikhomirov@virtuozzo.com.mbox.gz" }
lkml_critique
linux-mm
To avoid potential problems related to cpu/compiler optimizations around ->child_reaper, let's use WRITE_ONCE (additional to task_list lock) everywhere we write it and use READ_ONCE where we read it without explicit lock. Note: It also pairs with existing READ_ONCE with no lock in nsfs_fh_to_dentry(). Also let's add A...
null
null
null
[PATCH v4 1/4] pid_namespace: avoid optimization of accesses to ->child_reaper
This moves the condition (tid != 1 && !tmp->child_reaper) to after idr alloc, so it not only covers that first process in pid namespace has pid 1 in case of clone3(set_tid) requesting wrong pid, but also if idr itself gives wrong pid for some reason. This could've been the case before this patch, when creating first p...
{ "author": "Pavel Tikhomirov <ptikhomirov@virtuozzo.com>", "date": "Wed, 25 Feb 2026 14:32:24 +0100", "is_openbsd": false, "thread_id": "20260225133229.550302-1-ptikhomirov@virtuozzo.com.mbox.gz" }
lkml_critique
linux-mm
To avoid potential problems related to cpu/compiler optimizations around ->child_reaper, let's use WRITE_ONCE (additional to task_list lock) everywhere we write it and use READ_ONCE where we read it without explicit lock. Note: It also pairs with existing READ_ONCE with no lock in nsfs_fh_to_dentry(). Also let's add A...
null
null
null
[PATCH v4 1/4] pid_namespace: avoid optimization of accesses to ->child_reaper
This effectively gives us an ability to create the pid namespace init as a child of the process (setns-ed to the pid namespace) different to the process which created the pid namespace itself. Original problem: There is a cool set_tid feature in clone3() syscall, it allows you to create process with desired pids on m...
{ "author": "Pavel Tikhomirov <ptikhomirov@virtuozzo.com>", "date": "Wed, 25 Feb 2026 14:32:25 +0100", "is_openbsd": false, "thread_id": "20260225133229.550302-1-ptikhomirov@virtuozzo.com.mbox.gz" }
lkml_critique
linux-mm
To avoid potential problems related to cpu/compiler optimizations around ->child_reaper, let's use WRITE_ONCE (additional to task_list lock) everywhere we write it and use READ_ONCE where we read it without explicit lock. Note: It also pairs with existing READ_ONCE with no lock in nsfs_fh_to_dentry(). Also let's add A...
null
null
null
[PATCH v4 1/4] pid_namespace: avoid optimization of accesses to ->child_reaper
The first patch properly annotates accesses to ->child_reaper with _ONCE macroses, to protect unlocked accesses from possible cpu/compiler optimization problems. The second patch makes sure that the init is always a first process in the pid namespace, previously this was only checked for set_tid case. The third patch...
{ "author": "Pavel Tikhomirov <ptikhomirov@virtuozzo.com>", "date": "Wed, 25 Feb 2026 14:32:22 +0100", "is_openbsd": false, "thread_id": "20260225133229.550302-1-ptikhomirov@virtuozzo.com.mbox.gz" }
lkml_critique
linux-mm
To avoid potential problems related to cpu/compiler optimizations around ->child_reaper, let's use WRITE_ONCE (additional to task_list lock) everywhere we write it and use READ_ONCE where we read it without explicit lock. Note: It also pairs with existing READ_ONCE with no lock in nsfs_fh_to_dentry(). Also let's add A...
null
null
null
[PATCH v4 1/4] pid_namespace: avoid optimization of accesses to ->child_reaper
On Wed, Feb 25, 2026 at 5:33 AM Pavel Tikhomirov <ptikhomirov@virtuozzo.com> wrote: Acked-by: Andrei Vagin <avagin@google.com> Thanks, Andrei
{ "author": "Andrei Vagin <avagin@google.com>", "date": "Wed, 25 Feb 2026 10:38:08 -0800", "is_openbsd": false, "thread_id": "20260225133229.550302-1-ptikhomirov@virtuozzo.com.mbox.gz" }
lkml_critique
linux-mm
To avoid potential problems related to cpu/compiler optimizations around ->child_reaper, let's use WRITE_ONCE (additional to task_list lock) everywhere we write it and use READ_ONCE where we read it without explicit lock. Note: It also pairs with existing READ_ONCE with no lock in nsfs_fh_to_dentry(). Also let's add A...
null
null
null
[PATCH v4 1/4] pid_namespace: avoid optimization of accesses to ->child_reaper
Andrew, On top of Pavel's series but but doesn't depend on it. Can be applied before or after. Oleg. --- kernel/fork.c | 6 +++++- kernel/pid.c | 18 +++++++++++------- 2 files changed, 16 insertions(+), 8 deletions(-)
{ "author": "Oleg Nesterov <oleg@redhat.com>", "date": "Fri, 27 Feb 2026 13:02:03 +0100", "is_openbsd": false, "thread_id": "20260225133229.550302-1-ptikhomirov@virtuozzo.com.mbox.gz" }
lkml_critique
linux-mm
To avoid potential problems related to cpu/compiler optimizations around ->child_reaper, let's use WRITE_ONCE (additional to task_list lock) everywhere we write it and use READ_ONCE where we read it without explicit lock. Note: It also pairs with existing READ_ONCE with no lock in nsfs_fh_to_dentry(). Also let's add A...
null
null
null
[PATCH v4 1/4] pid_namespace: avoid optimization of accesses to ->child_reaper
Currently we allow only one attempt to create init in a new namespace. If the first fork() fails after alloc_pid() succeeds, free_pid() clears PIDNS_ADDING and thus disables further PID allocations. Nowadays this looks like an unnecessary limitation. The original reason to handle "case PIDNS_ADDING" in free_pid() is g...
{ "author": "Oleg Nesterov <oleg@redhat.com>", "date": "Fri, 27 Feb 2026 13:03:41 +0100", "is_openbsd": false, "thread_id": "20260225133229.550302-1-ptikhomirov@virtuozzo.com.mbox.gz" }
lkml_critique
linux-mm
To avoid potential problems related to cpu/compiler optimizations around ->child_reaper, let's use WRITE_ONCE (additional to task_list lock) everywhere we write it and use READ_ONCE where we read it without explicit lock. Note: It also pairs with existing READ_ONCE with no lock in nsfs_fh_to_dentry(). Also let's add A...
null
null
null
[PATCH v4 1/4] pid_namespace: avoid optimization of accesses to ->child_reaper
Both copy_process() and alloc_pid() do the same PIDNS_ADDING check. The reasons for these checks, and the fact that both are necessary, are not immediately obvious. Add the comments. Signed-off-by: Oleg Nesterov <oleg@redhat.com> --- kernel/fork.c | 6 +++++- kernel/pid.c | 5 +++++ 2 files changed, 10 insertions(+)...
{ "author": "Oleg Nesterov <oleg@redhat.com>", "date": "Fri, 27 Feb 2026 13:04:20 +0100", "is_openbsd": false, "thread_id": "20260225133229.550302-1-ptikhomirov@virtuozzo.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
From: Ye Bin <yebin10@huawei.com> This patch is prepare for support drop_caches for specify file system. Signed-off-by: Ye Bin <yebin10@huawei.com> --- include/linux/mm.h | 1 + mm/internal.h | 3 +++ mm/shrinker.c | 4 ++-- mm/vmscan.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++ 4 files...
{ "author": "Ye Bin <yebin@huaweicloud.com>", "date": "Fri, 27 Feb 2026 10:55:46 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
From: Ye Bin <yebin10@huawei.com> In order to better analyze the issue of file system uninstallation caused by kernel module opening files, it is necessary to perform dentry recycling on a single file system. But now, apart from global dentry recycling, it is not supported to do dentry recycling on a single file syste...
{ "author": "Ye Bin <yebin@huaweicloud.com>", "date": "Fri, 27 Feb 2026 10:55:45 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
From: Ye Bin <yebin10@huawei.com> In order to better analyze the issue of file system uninstallation caused by kernel module opening files, it is necessary to perform dentry recycling on a single file system. But now, apart from global dentry recycling, it is not supported to do dentry recycling on a single file syste...
{ "author": "Ye Bin <yebin@huaweicloud.com>", "date": "Fri, 27 Feb 2026 10:55:47 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
Would shrinker-debugfs satisfy your needs (See Documentation/admin-guide/mm/shrinker_debugfs.rst)? Thanks, Muchun
{ "author": "Muchun Song <muchun.song@linux.dev>", "date": "Fri, 27 Feb 2026 11:31:14 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
On 2026/2/27 11:31, Muchun Song wrote: Thank you for the reminder. The reclamation of dentries and nodes can meet my needs. However, the reclamation of the page cache alone does not satisfy my requirements. I have reviewed the code of shrinker_debugfs_scan_write() and found that it does not support batch deletion o...
{ "author": "\"yebin (H)\" <yebin10@huawei.com>", "date": "Fri, 27 Feb 2026 14:39:22 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
On 2/27/26 2:39 PM, yebin (H) wrote: Using shrinker-debugfs allows users to specify the size of a single reclaim cycle (nr_to_scan), which controls the strength of each reclaim cycle to adapt to different workloads. Can the new drop_fs_caches support a similar approach? Thanks, Qi
{ "author": "Qi Zheng <qi.zheng@linux.dev>", "date": "Fri, 27 Feb 2026 14:50:17 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
I don't really like that you're implementing another mechanism with duplicate functionality. If you'd like, you could write a script to iterate through them and execute it that way—I don't think that would be particularly inconvenient, would it? If the iteration operation of memcg is indeed quite cumbersome, I think ex...
{ "author": "Muchun Song <muchun.song@linux.dev>", "date": "Fri, 27 Feb 2026 14:55:04 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
On 2026/2/27 14:50, Qi Zheng wrote: "drop_fs_caches" is similar to "drop_caches," but it only operates on the specified file system. It does not support specifying the number of pages to scan (nr_to_scan).
{ "author": "\"yebin (H)\" <yebin10@huawei.com>", "date": "Fri, 27 Feb 2026 15:18:31 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
On 2026/2/27 14:55, Muchun Song wrote: The shrinker_debugfs can be extended to support node/memcg/fs granularity reclamation, similar to the extended function of echo " 0 - X" > count /echo " - 0 X" > count /echo " - - X" > count. This only solves the problem of reclaiming dentries/inode based on a single file syst...
{ "author": "\"yebin (H)\" <yebin10@huawei.com>", "date": "Fri, 27 Feb 2026 15:32:18 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
If the inode is evicted, the page cache is evicted as well. It cannot evict page cache alone. Why you want to evict cache alone?
{ "author": "Muchun Song <muchun.song@linux.dev>", "date": "Fri, 27 Feb 2026 15:45:02 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
On 2026/2/27 15:45, Muchun Song wrote: The condition for dentry/inode to be reclaimed is that there are no references to them. Therefore, relying on inode reclamation for page cache reclamation has limitations. Additionally, there is currently no usage statistics for the page cache based on a single file system. By com...
{ "author": "\"yebin (H)\" <yebin10@huawei.com>", "date": "Fri, 27 Feb 2026 16:17:24 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
What limit? I'm curious why dropping inodes doesn't show a noticeable difference in page cache usage before and after?
{ "author": "Muchun Song <muchun.song@linux.dev>", "date": "Fri, 27 Feb 2026 16:27:13 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
From: Ye Bin <yebin10@huawei.com> Add instructions for 'drop_fs_caches sysctl' sysctl in 'vm.rst'. Signed-off-by: Ye Bin <yebin10@huawei.com> --- Documentation/admin-guide/sysctl/vm.rst | 44 +++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/Documentation/admin-guide/sysctl/vm.rst b/Documentat...
null
null
null
[PATCH v3 3/3] Documentation: add instructions for using 'drop_fs_caches sysctl' sysctl
On 2026/2/27 16:27, Muchun Song wrote: If the file is occupied, the page cache cannot be reclaimed through inode reclamation.
{ "author": "\"yebin (H)\" <yebin10@huawei.com>", "date": "Fri, 27 Feb 2026 17:02:59 +0800", "is_openbsd": false, "thread_id": "20260227025548.2252380-4-yebin@huaweicloud.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
Hi All, Apologies I forgot to add the proper tag in the previous email so resending this. MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. P...
{ "author": "Kairui Song <ryncsn@gmail.com>", "date": "Fri, 20 Feb 2026 01:25:33 +0800", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
On Fri, Feb 20, 2026 at 01:25:33AM +0800, Kairui Song wrote: I would be very interested in discussing this topic as well. Are you referring to refaults on the page cache side, or swapins? Last time we evaluated MGLRU on Meta workloads, we noticed that it tends to do better with zswap, but worse with disk swap. It s...
{ "author": "Johannes Weiner <hannes@cmpxchg.org>", "date": "Fri, 20 Feb 2026 13:24:26 -0500", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
On Sat, Feb 21, 2026 at 2:24 AM Johannes Weiner <hannes@cmpxchg.org> wrote: Thanks, glad to hear that! A bit more than that. When there is no swap, MGLRU still performs worse in some workloads like MongoDB. From what I've noticed that's because the PID protection is a bit too passive, and there is a force protection...
{ "author": "Kairui Song <ryncsn@gmail.com>", "date": "Sat, 21 Feb 2026 14:03:43 +0800", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
On Thu, Feb 19, 2026 at 9:10 AM Kairui Song <ryncsn@gmail.com> wrote: There might be some overlap with this topic proposal: https://lore.kernel.org/all/cb0c0a0bfc7247cf85858eecf0db6eca@honor.com/ but either way I'm interested in participating, especially on the topics of regressions and reclaim behavior as it's very r...
{ "author": "Suren Baghdasaryan <surenb@google.com>", "date": "Tue, 24 Feb 2026 09:19:26 -0800", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
On Thu, Feb 19, 2026 at 9:26 AM Kairui Song <ryncsn@gmail.com> wrote: Hi Kairui, I would be very interested in joining this discussion at LSF/MM. We use MGLRU on Android. While the reduced CPU usage leads to power improvements for mobile devices, we've run into a few notable issues as well. Off the top of my head: ...
{ "author": "Kalesh Singh <kaleshsingh@google.com>", "date": "Wed, 25 Feb 2026 17:55:01 -0800", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
On Thu, Feb 26, 2026 at 9:55 AM Kalesh Singh <kaleshsingh@google.com> wrote: Hi Kelash, Glad to discuss this with you. Yes, this is one of the main issues for us too. Per our observation one cause for that is MGLRU's usage of flags like PG_workingset is different from active / inactive LRU, and flags like the PG_wo...
{ "author": "Kairui Song <ryncsn@gmail.com>", "date": "Thu, 26 Feb 2026 11:06:46 +0800", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
Hi Kairui, hi Kalesh, Yes, we’re interested in this work. We see file pages being under-protected in smartphone workload, and an LFU-like approach sounds promising to better promote and protect hot file pages. Kairui has shared the patches; we’ll backport them to our tree and report back once we have results f...
{ "author": "wangzicheng <wangzicheng@honor.com>", "date": "Thu, 26 Feb 2026 10:10:59 +0000", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
On Fri, Feb 20, 2026 at 01:25:33AM +0800, Kairui Song wrote: To my mind, the biggest problem with MGLRU is that Google dumped it on us and ran away. Commit 44958000bada claimed that it was now maintained and added three people as maintainers. In the six months since that commit, none of those three people have any c...
{ "author": "Matthew Wilcox <willy@infradead.org>", "date": "Thu, 26 Feb 2026 15:54:22 +0000", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
I guess not—MGLRU needs at least two generations to function, similar to active and inactive lists, meaning it requires two lists. You Zhao mentioned this in commit ec1c86b25f4b: "This protocol, AKA second chance, requires a minimum of two generations, hence MIN_NR_GENS." But I do feel the issue is that anon and file...
{ "author": "Barry Song <21cnbao@gmail.com>", "date": "Fri, 27 Feb 2026 11:30:13 +0800", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
Hi Matthew, Can we keep it for now? Kairui, Zicheng, and I are working on it. approach after applying a few vendor hooks on Android, such as forced aging and avoiding direct activation of read-ahead folios during page faults, among others. To be honest, performance was worse than active/inactive without those ...
{ "author": "Barry Song <21cnbao@gmail.com>", "date": "Fri, 27 Feb 2026 12:31:39 +0800", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
On Fri, 20 Feb 2026, Kairui Song wrote: I think this would be a very useful topic to discuss and I really like how this was framed in the context of what needs to be addressed so that MGLRU can be on a path to becoming the default implementation and we can eliminate two separate implementations. Yes, MGLRU can fo...
{ "author": "David Rientjes <rientjes@google.com>", "date": "Thu, 26 Feb 2026 23:11:11 -0800 (PST)", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi All, MGLRU has been introduced in the mainline for years, but we still have two LRUs today. There are many reasons MGLRU is still not the only LRU implementation in the kernel. And I've been looking at a few major issues here: 1. Page flag usage: MGLRU uses many more flags (3+ more) than Active/Inactive LRU. 2. R...
null
null
null
[LSF/MM/BPF] Improving MGLRU
On Fri, Feb 20, 2026 at 01:25:33AM +0800, Kairui Song wrote: Hi Kairui, I would be very interested in discussing this topic as well. In Linux desktop distributions, when the system rapidly enters low memory state, it is almost impossible to enter S4, the success rate only is 10%. When analyzing this issue, it was id...
{ "author": "Vernon Yang <vernon2gm@gmail.com>", "date": "Fri, 27 Feb 2026 18:29:38 +0800", "is_openbsd": false, "thread_id": "20260227043139.95115-1-21cnbao@gmail.com.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
The CMA heap instantiation was initially developed by having the contiguous DMA code call into the CMA heap to create a new instance every time a reserved memory area is probed. Turning the CMA heap into a module would create a dependency of the kernel on a module, which doesn't work. Let's turn the logic around and ...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Wed, 25 Feb 2026 17:41:49 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
The CMA dma-buf heap uses cma_alloc() and cma_release() to allocate and free, respectively, its CMA buffers. However, these functions are not exported. Since we want to turn the CMA heap into a module, let's export them both. Signed-off-by: Maxime Ripard <mripard@kernel.org> --- mm/cma.c | 2 ++ 1 file changed, 2 in...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Wed, 25 Feb 2026 17:41:50 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
The CMA dma-buf heap uses the cma_get_name() function to get the name of the heap instance it's going to create. However, this function is not exported. Since we want to turn the CMA heap into a module, let's export it. Signed-off-by: Maxime Ripard <mripard@kernel.org> --- mm/cma.c | 1 + 1 file changed, 1 insertion...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Wed, 25 Feb 2026 17:41:51 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
The CMA dma-buf heap uses the dev_get_cma_area() inline function that would either return the content of device.cma_area or the content of dma_contiguous_default_area. The latter holds a pointer to the default CMA region, and is stored in a public variable. However, that variable isn't exported which prevents to use d...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Wed, 25 Feb 2026 17:41:52 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
The mem_accounting kernel parameter is used by heaps to know if they should account allocations in their respective cgroup controllers. Since we're going to allow heaps to compile as modules, we need to export that variable. Signed-off-by: Maxime Ripard <mripard@kernel.org> --- drivers/dma-buf/dma-heap.c | 1 + 1 fi...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Wed, 25 Feb 2026 17:41:53 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
Now that all the symbols used by the CMA heap are exported, turning the CMA heap into a module becomes pretty easy: we just need to add the usual MODULE_* macros, import the proper namespaces and change the Kconfig symbol to a tristate. Signed-off-by: Maxime Ripard <mripard@kernel.org> --- drivers/dma-buf/heaps/Kconf...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Wed, 25 Feb 2026 17:41:54 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
The system heap can be easily turned into a module by adding the usual MODULE_* macros, importing the proper namespaces and changing the Kconfig symbol to a tristate. Signed-off-by: Maxime Ripard <mripard@kernel.org> --- drivers/dma-buf/heaps/Kconfig | 2 +- drivers/dma-buf/heaps/system_heap.c | 5 +++++ 2 file...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Wed, 25 Feb 2026 17:41:55 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On Wed, Feb 25, 2026 at 8:42 AM Maxime Ripard <mripard@kernel.org> wrote: So heaps-as-modules is common in the Android kernels, and was attempted to be upstreamed long ago: https://lore.kernel.org/lkml/20191025234834.28214-1-john.stultz@linaro.org/ And it got a fairly chilly reception, but maybe having the addition...
{ "author": "John Stultz <jstultz@google.com>", "date": "Wed, 25 Feb 2026 10:51:30 -0800", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On Wed, Feb 25, 2026 at 8:42 AM Maxime Ripard <mripard@kernel.org> wrote: Reviewed-by: T.J. Mercier <tjmercier@google.com> Thanks! I disabled this registration in the Android kernels where we build all dmabuf heaps as modules instead of built-in, so I can undo that if this is merged.
{ "author": "\"T.J. Mercier\" <tjmercier@google.com>", "date": "Wed, 25 Feb 2026 11:32:11 -0800", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On Wed, Feb 25, 2026 at 8:42 AM Maxime Ripard <mripard@kernel.org> wrote: Reviewed-by: T.J. Mercier <tjmercier@google.com>
{ "author": "\"T.J. Mercier\" <tjmercier@google.com>", "date": "Wed, 25 Feb 2026 11:34:13 -0800", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On Wed, Feb 25, 2026 at 8:42 AM Maxime Ripard <mripard@kernel.org> wrote: It'd be good to add a note that heap modules can't be unloaded (unless we add more code to verify there are no live heap allocations). With that: Reviewed-by: T.J. Mercier <tjmercier@google.com>
{ "author": "\"T.J. Mercier\" <tjmercier@google.com>", "date": "Wed, 25 Feb 2026 11:54:31 -0800", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On Wed, Feb 25, 2026 at 8:42 AM Maxime Ripard <mripard@kernel.org> wrote: Same comment about a note about the module being permanent. Reviewed-by: T.J. Mercier <tjmercier@google.com>
{ "author": "\"T.J. Mercier\" <tjmercier@google.com>", "date": "Wed, 25 Feb 2026 11:55:53 -0800", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
Hi John, Thanks for the review On Wed, Feb 25, 2026 at 10:51:30AM -0800, John Stultz wrote: I didn't know that Android was using heap as modules only, but I'd say that it's even more of a reason to upstream it then. I'm curious about this one though. It looks like you add refcounting, but never really get the refe...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Thu, 26 Feb 2026 11:18:08 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On 2/25/26 17:41, Maxime Ripard wrote: I'm wondering whether we want to restrict all these exports to the dma-buf module only using EXPORT_SYMBOL_FOR_MODULES(). Especially dma_contiguous_default_area() (patch #4), I am not sure whether we want arbitrary modules to mess with that. -- Cheers, David
{ "author": "\"David Hildenbrand (Arm)\" <david@kernel.org>", "date": "Thu, 26 Feb 2026 11:25:24 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
Hi David, On Thu, Feb 26, 2026 at 11:25:24AM +0100, David Hildenbrand (Arm) wrote: TIL about EXPORT_SYMBOL_FOR_MODULES, thanks. Yeah, I wasn't too fond about that one either. Alternatively, I guess we could turn dev_get_cma_area into a non-inlined function and export that instead? Or we could do both. Maxime
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Thu, 26 Feb 2026 11:38:45 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On Thu, Feb 26, 2026 at 2:38 AM Maxime Ripard <mripard@kernel.org> wrote: Ohh, ditto. I checked the history to see if dev_get_cma_area was converted to inline at some point for performance, but it has always been that way since 3.5. That'd be my only worry with un-inlining and exporting it. EXPORT_SYMBOL_FOR_MODULES...
{ "author": "\"T.J. Mercier\" <tjmercier@google.com>", "date": "Thu, 26 Feb 2026 08:58:51 -0800", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On Thu, Feb 26, 2026 at 2:18 AM Maxime Ripard <mripard@kernel.org> wrote: Oh, apologies I mixed this up. You can ignore that suggestion. In Android, once folks were familiar with thinking about dma-buf heaps, some (out of tree) drivers wanted to be able to internally allocate from a given heap (somewhat of a hold-ove...
{ "author": "John Stultz <jstultz@google.com>", "date": "Thu, 26 Feb 2026 10:03:21 -0800", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }
lkml_critique
linux-mm
Hi, The recent introduction of heaps in the optee driver [1] made possible the creation of heaps as modules. It's generally a good idea if possible, including for the already existing system and CMA heaps. The system one is pretty trivial, the CMA one is a bit more involved, especially since we have a call from kern...
null
null
null
[PATCH 0/7] dma-buf: heaps: Turn heaps into modules
On Thu, Feb 26, 2026 at 08:58:51AM -0800, T.J. Mercier wrote: I just realised the new tegra heap is also going to use cma_alloc: https://lore.kernel.org/dri-devel/20260122161009.3865888-7-thierry.reding@kernel.org/ I exported dev_get_cma_area() like we discussed, but kept EXPORT_SYMBOL for now in the new version. Ma...
{ "author": "Maxime Ripard <mripard@kernel.org>", "date": "Fri, 27 Feb 2026 14:16:44 +0100", "is_openbsd": false, "thread_id": "20260227-psychedelic-tireless-herring-0adfa9@houat.mbox.gz" }