[PATCH 03/20] migration: remove unreachble RDMA code in save_hook impl

Daniel P. Berrangé posted 20 patches 3 years, 1 month ago
There is a newer version of this series
[PATCH 03/20] migration: remove unreachble RDMA code in save_hook impl
Posted by Daniel P. Berrangé 3 years, 1 month ago
The QEMUFile 'save_hook' callback has a 'size_t size' parameter.

The RDMA impl of this has logic that takes different actions
depending on whether the value is zero or non-zero. It has
commented out logic that would have taken further actions
if the value was negative.

The only place where the 'save_hook' callback is invoked is
the ram_control_save_page() method, which passes 'size'
through from its caller. The only caller of this method is
in turn control_save_page(). This method unconditionally
passes the 'TARGET_PAGE_SIZE' constant for the 'size' parameter.

IOW, the only scenario for 'size' that can execute in the
qemu_rdma_save_page method is 'size > 0'. The remaining code
has been unreachable since RDMA support was first introduced
9 years ago.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
---
 migration/rdma.c | 120 +++++++++--------------------------------------
 1 file changed, 21 insertions(+), 99 deletions(-)

diff --git a/migration/rdma.c b/migration/rdma.c
index 672d1958a9..6e7756bee7 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
@@ -1486,34 +1486,6 @@ static uint64_t qemu_rdma_make_wrid(uint64_t wr_id, uint64_t index,
     return result;
 }
 
-/*
- * Set bit for unregistration in the next iteration.
- * We cannot transmit right here, but will unpin later.
- */
-static void qemu_rdma_signal_unregister(RDMAContext *rdma, uint64_t index,
-                                        uint64_t chunk, uint64_t wr_id)
-{
-    if (rdma->unregistrations[rdma->unregister_next] != 0) {
-        error_report("rdma migration: queue is full");
-    } else {
-        RDMALocalBlock *block = &(rdma->local_ram_blocks.block[index]);
-
-        if (!test_and_set_bit(chunk, block->unregister_bitmap)) {
-            trace_qemu_rdma_signal_unregister_append(chunk,
-                                                     rdma->unregister_next);
-
-            rdma->unregistrations[rdma->unregister_next++] =
-                    qemu_rdma_make_wrid(wr_id, index, chunk);
-
-            if (rdma->unregister_next == RDMA_SIGNALED_SEND_MAX) {
-                rdma->unregister_next = 0;
-            }
-        } else {
-            trace_qemu_rdma_signal_unregister_already(chunk);
-        }
-    }
-}
-
 /*
  * Consult the connection manager to see a work request
  * (of any kind) has completed.
@@ -3278,23 +3250,7 @@ qio_channel_rdma_shutdown(QIOChannel *ioc,
  *        Offset is an offset to be added to block_offset and used
  *        to also lookup the corresponding RAMBlock.
  *
- *    @size > 0 :
- *        Initiate an transfer this size.
- *
- *    @size == 0 :
- *        A 'hint' or 'advice' that means that we wish to speculatively
- *        and asynchronously unregister this memory. In this case, there is no
- *        guarantee that the unregister will actually happen, for example,
- *        if the memory is being actively transmitted. Additionally, the memory
- *        may be re-registered at any future time if a write within the same
- *        chunk was requested again, even if you attempted to unregister it
- *        here.
- *
- *    @size < 0 : TODO, not yet supported
- *        Unregister the memory NOW. This means that the caller does not
- *        expect there to be any future RDMA transfers and we just want to clean
- *        things up. This is used in case the upper layer owns the memory and
- *        cannot wait for qemu_fclose() to occur.
+ *    @size : Number of bytes to transfer
  *
  *    @bytes_sent : User-specificed pointer to indicate how many bytes were
  *                  sent. Usually, this will not be more than a few bytes of
@@ -3323,61 +3279,27 @@ static size_t qemu_rdma_save_page(QEMUFile *f, void *opaque,
 
     qemu_fflush(f);
 
-    if (size > 0) {
-        /*
-         * Add this page to the current 'chunk'. If the chunk
-         * is full, or the page doesn't belong to the current chunk,
-         * an actual RDMA write will occur and a new chunk will be formed.
-         */
-        ret = qemu_rdma_write(f, rdma, block_offset, offset, size);
-        if (ret < 0) {
-            error_report("rdma migration: write error! %d", ret);
-            goto err;
-        }
-
-        /*
-         * We always return 1 bytes because the RDMA
-         * protocol is completely asynchronous. We do not yet know
-         * whether an  identified chunk is zero or not because we're
-         * waiting for other pages to potentially be merged with
-         * the current chunk. So, we have to call qemu_update_position()
-         * later on when the actual write occurs.
-         */
-        if (bytes_sent) {
-            *bytes_sent = 1;
-        }
-    } else {
-        uint64_t index, chunk;
-
-        /* TODO: Change QEMUFileOps prototype to be signed: size_t => long
-        if (size < 0) {
-            ret = qemu_rdma_drain_cq(f, rdma);
-            if (ret < 0) {
-                fprintf(stderr, "rdma: failed to synchronously drain"
-                                " completion queue before unregistration.\n");
-                goto err;
-            }
-        }
-        */
-
-        ret = qemu_rdma_search_ram_block(rdma, block_offset,
-                                         offset, size, &index, &chunk);
-
-        if (ret) {
-            error_report("ram block search failed");
-            goto err;
-        }
-
-        qemu_rdma_signal_unregister(rdma, index, chunk, 0);
+    /*
+     * Add this page to the current 'chunk'. If the chunk
+     * is full, or the page doesn't belong to the current chunk,
+     * an actual RDMA write will occur and a new chunk will be formed.
+     */
+    ret = qemu_rdma_write(f, rdma, block_offset, offset, size);
+    if (ret < 0) {
+        error_report("rdma migration: write error! %d", ret);
+        goto err;
+    }
 
-        /*
-         * TODO: Synchronous, guaranteed unregistration (should not occur during
-         * fast-path). Otherwise, unregisters will process on the next call to
-         * qemu_rdma_drain_cq()
-        if (size < 0) {
-            qemu_rdma_unregister_waiting(rdma);
-        }
-        */
+    /*
+     * We always return 1 bytes because the RDMA
+     * protocol is completely asynchronous. We do not yet know
+     * whether an  identified chunk is zero or not because we're
+     * waiting for other pages to potentially be merged with
+     * the current chunk. So, we have to call qemu_update_position()
+     * later on when the actual write occurs.
+     */
+    if (bytes_sent) {
+        *bytes_sent = 1;
     }
 
     /*
-- 
2.36.1


Re: [PATCH 03/20] migration: remove unreachble RDMA code in save_hook impl
Posted by Dr. David Alan Gilbert 3 years, 1 month ago
* Daniel P. Berrangé (berrange@redhat.com) wrote:
> The QEMUFile 'save_hook' callback has a 'size_t size' parameter.
> 
> The RDMA impl of this has logic that takes different actions
> depending on whether the value is zero or non-zero. It has
> commented out logic that would have taken further actions
> if the value was negative.
> 
> The only place where the 'save_hook' callback is invoked is
> the ram_control_save_page() method, which passes 'size'
> through from its caller. The only caller of this method is
> in turn control_save_page(). This method unconditionally
> passes the 'TARGET_PAGE_SIZE' constant for the 'size' parameter.
> 
> IOW, the only scenario for 'size' that can execute in the
> qemu_rdma_save_page method is 'size > 0'. The remaining code
> has been unreachable since RDMA support was first introduced
> 9 years ago.
> 
> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>

Ah good; less rdma code!



Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

> ---
>  migration/rdma.c | 120 +++++++++--------------------------------------
>  1 file changed, 21 insertions(+), 99 deletions(-)
> 
> diff --git a/migration/rdma.c b/migration/rdma.c
> index 672d1958a9..6e7756bee7 100644
> --- a/migration/rdma.c
> +++ b/migration/rdma.c
> @@ -1486,34 +1486,6 @@ static uint64_t qemu_rdma_make_wrid(uint64_t wr_id, uint64_t index,
>      return result;
>  }
>  
> -/*
> - * Set bit for unregistration in the next iteration.
> - * We cannot transmit right here, but will unpin later.
> - */
> -static void qemu_rdma_signal_unregister(RDMAContext *rdma, uint64_t index,
> -                                        uint64_t chunk, uint64_t wr_id)
> -{
> -    if (rdma->unregistrations[rdma->unregister_next] != 0) {
> -        error_report("rdma migration: queue is full");
> -    } else {
> -        RDMALocalBlock *block = &(rdma->local_ram_blocks.block[index]);
> -
> -        if (!test_and_set_bit(chunk, block->unregister_bitmap)) {
> -            trace_qemu_rdma_signal_unregister_append(chunk,
> -                                                     rdma->unregister_next);
> -
> -            rdma->unregistrations[rdma->unregister_next++] =
> -                    qemu_rdma_make_wrid(wr_id, index, chunk);
> -
> -            if (rdma->unregister_next == RDMA_SIGNALED_SEND_MAX) {
> -                rdma->unregister_next = 0;
> -            }
> -        } else {
> -            trace_qemu_rdma_signal_unregister_already(chunk);
> -        }
> -    }
> -}
> -
>  /*
>   * Consult the connection manager to see a work request
>   * (of any kind) has completed.
> @@ -3278,23 +3250,7 @@ qio_channel_rdma_shutdown(QIOChannel *ioc,
>   *        Offset is an offset to be added to block_offset and used
>   *        to also lookup the corresponding RAMBlock.
>   *
> - *    @size > 0 :
> - *        Initiate an transfer this size.
> - *
> - *    @size == 0 :
> - *        A 'hint' or 'advice' that means that we wish to speculatively
> - *        and asynchronously unregister this memory. In this case, there is no
> - *        guarantee that the unregister will actually happen, for example,
> - *        if the memory is being actively transmitted. Additionally, the memory
> - *        may be re-registered at any future time if a write within the same
> - *        chunk was requested again, even if you attempted to unregister it
> - *        here.
> - *
> - *    @size < 0 : TODO, not yet supported
> - *        Unregister the memory NOW. This means that the caller does not
> - *        expect there to be any future RDMA transfers and we just want to clean
> - *        things up. This is used in case the upper layer owns the memory and
> - *        cannot wait for qemu_fclose() to occur.
> + *    @size : Number of bytes to transfer
>   *
>   *    @bytes_sent : User-specificed pointer to indicate how many bytes were
>   *                  sent. Usually, this will not be more than a few bytes of
> @@ -3323,61 +3279,27 @@ static size_t qemu_rdma_save_page(QEMUFile *f, void *opaque,
>  
>      qemu_fflush(f);
>  
> -    if (size > 0) {
> -        /*
> -         * Add this page to the current 'chunk'. If the chunk
> -         * is full, or the page doesn't belong to the current chunk,
> -         * an actual RDMA write will occur and a new chunk will be formed.
> -         */
> -        ret = qemu_rdma_write(f, rdma, block_offset, offset, size);
> -        if (ret < 0) {
> -            error_report("rdma migration: write error! %d", ret);
> -            goto err;
> -        }
> -
> -        /*
> -         * We always return 1 bytes because the RDMA
> -         * protocol is completely asynchronous. We do not yet know
> -         * whether an  identified chunk is zero or not because we're
> -         * waiting for other pages to potentially be merged with
> -         * the current chunk. So, we have to call qemu_update_position()
> -         * later on when the actual write occurs.
> -         */
> -        if (bytes_sent) {
> -            *bytes_sent = 1;
> -        }
> -    } else {
> -        uint64_t index, chunk;
> -
> -        /* TODO: Change QEMUFileOps prototype to be signed: size_t => long
> -        if (size < 0) {
> -            ret = qemu_rdma_drain_cq(f, rdma);
> -            if (ret < 0) {
> -                fprintf(stderr, "rdma: failed to synchronously drain"
> -                                " completion queue before unregistration.\n");
> -                goto err;
> -            }
> -        }
> -        */
> -
> -        ret = qemu_rdma_search_ram_block(rdma, block_offset,
> -                                         offset, size, &index, &chunk);
> -
> -        if (ret) {
> -            error_report("ram block search failed");
> -            goto err;
> -        }
> -
> -        qemu_rdma_signal_unregister(rdma, index, chunk, 0);
> +    /*
> +     * Add this page to the current 'chunk'. If the chunk
> +     * is full, or the page doesn't belong to the current chunk,
> +     * an actual RDMA write will occur and a new chunk will be formed.
> +     */
> +    ret = qemu_rdma_write(f, rdma, block_offset, offset, size);
> +    if (ret < 0) {
> +        error_report("rdma migration: write error! %d", ret);
> +        goto err;
> +    }
>  
> -        /*
> -         * TODO: Synchronous, guaranteed unregistration (should not occur during
> -         * fast-path). Otherwise, unregisters will process on the next call to
> -         * qemu_rdma_drain_cq()
> -        if (size < 0) {
> -            qemu_rdma_unregister_waiting(rdma);
> -        }
> -        */
> +    /*
> +     * We always return 1 bytes because the RDMA
> +     * protocol is completely asynchronous. We do not yet know
> +     * whether an  identified chunk is zero or not because we're
> +     * waiting for other pages to potentially be merged with
> +     * the current chunk. So, we have to call qemu_update_position()
> +     * later on when the actual write occurs.
> +     */
> +    if (bytes_sent) {
> +        *bytes_sent = 1;
>      }
>  
>      /*
> -- 
> 2.36.1
> 
-- 
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
Re: [PATCH 03/20] migration: remove unreachble RDMA code in save_hook impl
Posted by Eric Blake 3 years, 1 month ago
On Tue, May 24, 2022 at 12:02:18PM +0100, Daniel P. Berrangé wrote:
> The QEMUFile 'save_hook' callback has a 'size_t size' parameter.
> 
> The RDMA impl of this has logic that takes different actions
> depending on whether the value is zero or non-zero. It has
> commented out logic that would have taken further actions
> if the value was negative.
> 
> The only place where the 'save_hook' callback is invoked is
> the ram_control_save_page() method, which passes 'size'
> through from its caller. The only caller of this method is
> in turn control_save_page(). This method unconditionally
> passes the 'TARGET_PAGE_SIZE' constant for the 'size' parameter.
> 
> IOW, the only scenario for 'size' that can execute in the
> qemu_rdma_save_page method is 'size > 0'. The remaining code
> has been unreachable since RDMA support was first introduced
> 9 years ago.
> 
> Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
> ---
>  migration/rdma.c | 120 +++++++++--------------------------------------
>  1 file changed, 21 insertions(+), 99 deletions(-)
>

Reviewed-by: Eric Blake <eblake@redhat.com>

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org