Skip to contentSkip to conversation

RDMA/siw: Fix user page pinning accounting

PATCH In review Bernard Metzler 4 participants 5 messages Feb 1, 2023 – Feb 6, 2023

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 RDMA/siw: Fix user page pinning accounting

To avoid racing with other user memory reservations, immediately account full amount of pages to be pinned.

Fixes 2251334dcac9 ("rdma/siw: application buffer management")
---
 drivers/infiniband/sw/siw/siw_mem.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)
drivers/infiniband/sw/siw/siw_mem.c +12 -11
index b2b33dd3b4fa..f51ab2ccf151 100644
@@ -398,7 +398,7 @@ struct siw_umem *siw_umem_get(u64 start, u64 len, bool writable)
 
 	mlock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
 
-	if (num_pages + atomic64_read(&mm_s->pinned_vm) > mlock_limit) {
+	if (atomic64_add_return(num_pages, &mm_s->pinned_vm) > mlock_limit) {
 		rv = -ENOMEM;
 		goto out_sem_up;
 	}
@@ -411,30 +411,27 @@ struct siw_umem *siw_umem_get(u64 start, u64 len, bool writable)
 		goto out_sem_up;
 	}
 	for (i = 0; num_pages; i++) {
-		int got, nents = min_t(int, num_pages, PAGES_PER_CHUNK);
-
-		umem->page_chunk[i].plist =
+		int nents = min_t(int, num_pages, PAGES_PER_CHUNK);
+		struct page **plist =
 			kcalloc(nents, sizeof(struct page *), GFP_KERNEL);
-		if (!umem->page_chunk[i].plist) {
+
+		if (!plist) {
 			rv = -ENOMEM;
 			goto out_sem_up;
 		}
-		got = 0;
+		umem->page_chunk[i].plist = plist;
 		while (nents) {
-			struct page **plist = &umem->page_chunk[i].plist[got];
-
 			rv = pin_user_pages(first_page_va, nents, foll_flags,
 					    plist, NULL);
 			if (rv < 0)
 				goto out_sem_up;
 
 			umem->num_pages += rv;
-			atomic64_add(rv, &mm_s->pinned_vm);
 			first_page_va += rv * PAGE_SIZE;
+			plist += rv;
 			nents -= rv;
-			got += rv;
+			num_pages -= rv;
 		}
-		num_pages -= got;
 	}
 out_sem_up:
 	mmap_read_unlock(mm_s);
@@ -442,6 +439,10 @@ struct siw_umem *siw_umem_get(u64 start, u64 len, bool writable)
 	if (rv > 0)
 		return umem;
 
+	/* Adjust accounting for pages not pinned */
+	if (num_pages)
+		atomic64_sub(num_pages, &mm_s->pinned_vm);
+
 	siw_umem_release(umem, false);
 
 	return ERR_PTR(rv);
Signature · 1 line
2.32.0
Reviewed-by on v1 1/1
#1 lore raw

Thanks for cleaning this up, will make any potential follow on I do much easier. Feel free to add:

Quoted 73 lines · Bernard Metzler

To avoid racing with other user memory reservations, immediately account full amount of pages to be pinned.

Fixes 2251334dcac9 ("rdma/siw: application buffer management")
---
 drivers/infiniband/sw/siw/siw_mem.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)
drivers/infiniband/sw/siw/siw_mem.c +12 -11
index b2b33dd3b4fa..f51ab2ccf151 100644
@@ -398,7 +398,7 @@ struct siw_umem *siw_umem_get(u64 start, u64 len, bool writable)
 
 	mlock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
 
-	if (num_pages + atomic64_read(&mm_s->pinned_vm) > mlock_limit) {
+	if (atomic64_add_return(num_pages, &mm_s->pinned_vm) > mlock_limit) {
 		rv = -ENOMEM;
 		goto out_sem_up;
 	}
@@ -411,30 +411,27 @@ struct siw_umem *siw_umem_get(u64 start, u64 len, bool writable)
 		goto out_sem_up;
 	}
 	for (i = 0; num_pages; i++) {
-		int got, nents = min_t(int, num_pages, PAGES_PER_CHUNK);
-
-		umem->page_chunk[i].plist =
+		int nents = min_t(int, num_pages, PAGES_PER_CHUNK);
+		struct page **plist =
 			kcalloc(nents, sizeof(struct page *), GFP_KERNEL);
-		if (!umem->page_chunk[i].plist) {
+
+		if (!plist) {
 			rv = -ENOMEM;
 			goto out_sem_up;
 		}
-		got = 0;
+		umem->page_chunk[i].plist = plist;
 		while (nents) {
-			struct page **plist = &umem->page_chunk[i].plist[got];
-
 			rv = pin_user_pages(first_page_va, nents, foll_flags,
 					    plist, NULL);
 			if (rv < 0)
 				goto out_sem_up;
 
 			umem->num_pages += rv;
-			atomic64_add(rv, &mm_s->pinned_vm);
 			first_page_va += rv * PAGE_SIZE;
+			plist += rv;
 			nents -= rv;
-			got += rv;
+			num_pages -= rv;
 		}
-		num_pages -= got;
 	}
 out_sem_up:
 	mmap_read_unlock(mm_s);
@@ -442,6 +439,10 @@ struct siw_umem *siw_umem_get(u64 start, u64 len, bool writable)
 	if (rv > 0)
 		return umem;
 
+	/* Adjust accounting for pages not pinned */
+	if (num_pages)
+		atomic64_sub(num_pages, &mm_s->pinned_vm);
+
 	siw_umem_release(umem, false);
 
 	return ERR_PTR(rv);
#2 lore raw
Quoted 11 lines

-----Original Message-----
From: Alistair Popple <[email protected]>
Sent: Thursday, 2 February 2023 08:44
To: Bernard Metzler <[email protected]>
Cc: [email protected]; [email protected]; [email protected]
Subject: [EXTERNAL] Re: [PATCH] RDMA/siw: Fix user page pinning accounting

Quoted 5 lines

Thanks for cleaning this up, will make any potential follow on I do much easier. Feel free to add:

Thanks a lot! I'll resend for the last time (I hope), having you as a reviewer.

Many thanks,
Bernard.

maintainer
#3 lore raw
Quoted 11 lines · Bernard Metzler
Quoted 11 lines

-----Original Message-----
From: Alistair Popple <[email protected]>
Sent: Thursday, 2 February 2023 08:44
To: Bernard Metzler <[email protected]>
Cc: [email protected]; [email protected]; [email protected]
Subject: [EXTERNAL] Re: [PATCH] RDMA/siw: Fix user page pinning accounting

Quoted 8 lines
Quoted 5 lines

Thanks for cleaning this up, will make any potential follow on I do much easier. Feel free to add:

Thanks a lot! I'll resend for the last time (I hope), having you as a reviewer.

You don't need to resend for this. Patchworks catches all *-by tags and allows us to add them automatically.

However, it will be great if you write target rdma-next/rdma-rc in subject of the patch.

Thanks

Quoted 4 lines

Many thanks,
Bernard.

maintainer
#4 lore raw
Quoted 18 lines · Leon Romanovsky
Quoted 17 lines · Bernard Metzler
Quoted 16 lines

-----Original Message-----
From: Alistair Popple <[email protected]>
Sent: Thursday, 2 February 2023 08:44
To: Bernard Metzler <[email protected]>
Cc: [email protected]; [email protected]; [email protected]
Subject: [EXTERNAL] Re: [PATCH] RDMA/siw: Fix user page pinning accounting

Thanks for cleaning this up, will make any potential follow on I do much easier. Feel free to add:

Quoted 8 lines
Quoted 2 lines

Thanks a lot! I'll resend for the last time (I hope), having you as a reviewer.

You don't need to resend for this. Patchworks catches all *-by tags and allows us to add them automatically.

However, it will be great if you write target rdma-next/rdma-rc in subject of the patch.

And use version numbers..

Jason