From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from firstgate.proxmox.com (firstgate.proxmox.com [IPv6:2a01:7e0:0:424::9]) by lore.proxmox.com (Postfix) with ESMTPS id 8E8C41FF136 for ; Mon, 20 Apr 2026 18:22:31 +0200 (CEST) Received: from firstgate.proxmox.com (localhost [127.0.0.1]) by firstgate.proxmox.com (Proxmox) with ESMTP id 645309CDC; Mon, 20 Apr 2026 18:22:31 +0200 (CEST) From: Christian Ebner To: pbs-devel@lists.proxmox.com Subject: [PATCH proxmox-backup v4 25/30] sync: pull: extend encountered chunk by optional decrypted digest Date: Mon, 20 Apr 2026 18:15:28 +0200 Message-ID: <20260420161533.1055484-26-c.ebner@proxmox.com> X-Mailer: git-send-email 2.47.3 In-Reply-To: <20260420161533.1055484-1-c.ebner@proxmox.com> References: <20260420161533.1055484-1-c.ebner@proxmox.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Bm-Milter-Handled: 55990f41-d878-4baa-be0a-ee34c49e34d2 X-Bm-Transport-Timestamp: 1776701668704 X-SPAM-LEVEL: Spam detection results: 0 AWL 0.070 Adjusted score from AWL reputation of From: address BAYES_00 -1.9 Bayes spam probability is 0 to 1% DMARC_MISSING 0.1 Missing DMARC policy KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record SPF_PASS -0.001 SPF: sender matches SPF record Message-ID-Hash: YPNC4PJSIZQ5P224V35YWZEJXV3Z46GX X-Message-ID-Hash: YPNC4PJSIZQ5P224V35YWZEJXV3Z46GX X-MailFrom: c.ebner@proxmox.com X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; loop; banned-address; emergency; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header X-Mailman-Version: 3.3.10 Precedence: list List-Id: Proxmox Backup Server development discussion List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: For index files being decrypted during the pull, it is not enough to keep track of the processes source chunks, but the decrypted digest has to be known as well in order to rewrite the index file. Extend the encountered chunks such that this can be tracked as well. To not introduce clippy warnings and to keep the code readable, introduce the EncounteredChunksInfo struct as internal type for the hash map values. Signed-off-by: Christian Ebner --- src/server/pull.rs | 78 +++++++++++++++++++++++++++++++++------------- 1 file changed, 56 insertions(+), 22 deletions(-) diff --git a/src/server/pull.rs b/src/server/pull.rs index 174e50a91..1c1c9458e 100644 --- a/src/server/pull.rs +++ b/src/server/pull.rs @@ -176,7 +176,7 @@ async fn pull_index_chunks( .filter(|info| { let guard = encountered_chunks.lock().unwrap(); match guard.check_reusable(&info.digest) { - Some(touched) => !touched, // reusable and already touched, can always skip + Some(reusable) => !reusable.touched, // reusable and already touched, can always skip None => true, } }), @@ -212,22 +212,22 @@ async fn pull_index_chunks( { // limit guard scope let mut guard = encountered_chunks.lock().unwrap(); - if let Some(touched) = guard.check_reusable(&info.digest) { - if touched { + if let Some(reusable) = guard.check_reusable(&info.digest) { + if reusable.touched { return Ok::<_, Error>(()); } let chunk_exists = proxmox_async::runtime::block_in_place(|| { target.cond_touch_chunk(&info.digest, false) })?; if chunk_exists { - guard.mark_touched(&info.digest); + guard.mark_touched(&info.digest, None); //info!("chunk {} exists {}", pos, hex::encode(digest)); return Ok::<_, Error>(()); } } // mark before actually downloading the chunk, so this happens only once - guard.mark_reusable(&info.digest); - guard.mark_touched(&info.digest); + guard.mark_reusable(&info.digest, None); + guard.mark_touched(&info.digest, None); } //info!("sync {} chunk {}", pos, hex::encode(digest)); @@ -828,7 +828,7 @@ async fn pull_group( for pos in 0..index.index_count() { let chunk_info = index.chunk_info(pos).unwrap(); - reusable_chunks.mark_reusable(&chunk_info.digest); + reusable_chunks.mark_reusable(&chunk_info.digest, None); } } } @@ -1266,12 +1266,23 @@ async fn pull_ns( Ok((progress, sync_stats, errors)) } +struct EncounteredChunkInfo { + reusable: bool, + touched: bool, + decrypted_digest: Option<[u8; 32]>, +} + /// Store the state of encountered chunks, tracking if they can be reused for the /// index file currently being pulled and if the chunk has already been touched /// during this sync. struct EncounteredChunks { - // key: digest, value: (reusable, touched) - chunk_set: HashMap<[u8; 32], (bool, bool)>, + chunk_set: HashMap<[u8; 32], EncounteredChunkInfo>, +} + +/// Propertires of a reusable chunk +struct ReusableEncounteredChunk<'a> { + touched: bool, + decrypted_digest: Option<&'a [u8; 32]>, } impl EncounteredChunks { @@ -1284,41 +1295,64 @@ impl EncounteredChunks { /// Check if the current state allows to reuse this chunk and if so, /// if the chunk has already been touched. - fn check_reusable(&self, digest: &[u8; 32]) -> Option { - if let Some((reusable, touched)) = self.chunk_set.get(digest) { - if !reusable { + fn check_reusable(&self, digest: &[u8; 32]) -> Option> { + if let Some(chunk_info) = self.chunk_set.get(digest) { + if !chunk_info.reusable { None } else { - Some(*touched) + Some(ReusableEncounteredChunk { + touched: chunk_info.touched, + decrypted_digest: chunk_info.decrypted_digest.as_ref(), + }) } } else { None } } - /// Mark chunk as reusable, inserting it as un-touched if not present - fn mark_reusable(&mut self, digest: &[u8; 32]) { + /// Mark chunk as reusable, inserting it as un-touched if not present. + /// + /// If the mapping already contains the digest, set the decrypted digest only + /// if not already set previously. + fn mark_reusable(&mut self, digest: &[u8; 32], decrypted_digest: Option<[u8; 32]>) { match self.chunk_set.entry(*digest) { Entry::Occupied(mut occupied) => { - let (reusable, _touched) = occupied.get_mut(); - *reusable = true; + let chunk_info = occupied.get_mut(); + chunk_info.reusable = true; + if chunk_info.decrypted_digest.is_none() { + chunk_info.decrypted_digest = decrypted_digest; + } } Entry::Vacant(vacant) => { - vacant.insert((true, false)); + vacant.insert(EncounteredChunkInfo { + reusable: true, + touched: false, + decrypted_digest, + }); } } } /// Mark chunk as touched during this sync, inserting it as not reusable /// but touched if not present. - fn mark_touched(&mut self, digest: &[u8; 32]) { + /// + /// If the mapping already contains the digest, set the decrypted digest only + /// if not already set previously. + fn mark_touched(&mut self, digest: &[u8; 32], decrypted_digest: Option<[u8; 32]>) { match self.chunk_set.entry(*digest) { Entry::Occupied(mut occupied) => { - let (_reusable, touched) = occupied.get_mut(); - *touched = true; + let chunk_info = occupied.get_mut(); + chunk_info.touched = true; + if chunk_info.decrypted_digest.is_none() { + chunk_info.decrypted_digest = decrypted_digest; + } } Entry::Vacant(vacant) => { - vacant.insert((false, true)); + vacant.insert(EncounteredChunkInfo { + reusable: false, + touched: true, + decrypted_digest, + }); } } } -- 2.47.3