From nobody Thu Apr 2 20:22:04 2026 Received: from mail-pl1-f169.google.com (mail-pl1-f169.google.com [209.85.214.169]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 8D72838B7C5 for ; Thu, 26 Mar 2026 23:52:54 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.169 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1774569176; cv=none; b=se2QHJEuNSwXG9zJXHuoRB23nwHEphOj1jxXu7o5oioPRzj2belGnRZYO7VvHtaFpD09Em8Ei2fy0zfes0VQapHiAEWKHEOValEwtGSk9q5x6eZVp1W9eQDRYp5TTjtddd7YNtZDntKHvfl1ekis9bEY5wHhAFyk2hQwFr8FZAo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1774569176; c=relaxed/simple; bh=ORkpzv7iiCxiXOhQaUQdLjmDGsPudQzXaVUYF9lrn6U=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=egcbLVls/J1skVxTv7UY1cSBJvrNn19JYlTgXsynxbN9UMzM1Fh+bubAkOebVU0L4z1lLbuTbTiZmoyfzQu60Qi2GdK3esltyFk5z2QgTCVx0i0SNvQ/yJvoQKucaiOyKghYuzRsjGU81gDOSpdJOT59imWFDyxfAKG2qOowVkE= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=dama.to; spf=none smtp.mailfrom=dama.to; dkim=pass (2048-bit key) header.d=dama-to.20230601.gappssmtp.com header.i=@dama-to.20230601.gappssmtp.com header.b=EaVgK0r4; arc=none smtp.client-ip=209.85.214.169 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=dama.to Authentication-Results: smtp.subspace.kernel.org; spf=none smtp.mailfrom=dama.to Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=dama-to.20230601.gappssmtp.com header.i=@dama-to.20230601.gappssmtp.com header.b="EaVgK0r4" Received: by mail-pl1-f169.google.com with SMTP id d9443c01a7336-2b04e6a989eso9663885ad.3 for ; Thu, 26 Mar 2026 16:52:54 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=dama-to.20230601.gappssmtp.com; s=20230601; t=1774569174; x=1775173974; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=Blw7GY0uzkKurUX6ypQhQlLBgZS5L/EPKYGZQrzDQcs=; b=EaVgK0r4HyvU6sioqne1q3rSS+c2yk8KFt1ibXivQbdpxij4/5e/mxgbixlp7TX0nd zU9Pj5IdHN10tOAlaIvKR9waQAhv3gg94F+xUMRWXo/wdLC2Td9VAsuXOR2uX3Xmn8Eo BLb7sVH9ga0epiHNbzsHvTxpGgPD1HVOSpbsRYC0ERZtzJ6jHqcqq0AyoHqiqNA1dVAx 9OMhqDl+6BWYp3HnrEX6yrW2o14vHqmEHZ+wVJGp6cyNBeyO4tCkRQDX3+N5A7G85B2v R7571il9iY0bX72BoMk0LKFvHWOCCE64nAMcfmRWOGw7wt279okBpfWq+tTtSNP4YeLt qkkA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1774569174; x=1775173974; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to; bh=Blw7GY0uzkKurUX6ypQhQlLBgZS5L/EPKYGZQrzDQcs=; b=Jsa8YnyvYnWAw0pktrXp3pzfMF3XmxW38OIbTd9pvOof6ZZAGvxI6RindC1YSimdCX FlaRRECwUN4XfhP/kb+8VnEYTli24rHezJ5C22dkvB57TncwV6Ky61qUbVabbQqUeKOV M+bVNEwLfUWtD+NjSlhOnKas8llfWemnALOPiipowEF9hRw1yRdPEO7Ic7O6xoZqURKA f9748BAKXF//Mot5Zh5erN+74w+5/qeCTzbcuwjhaX35hJZMyuckuEa8AJI0U5XZJIQ9 Hh9hlmsz8t0CfiahqBL5Zvapm+6ojKH9dOsWUCw83Vo3nzoPcdjX96fxPiIq83OR1q9u mVBw== X-Forwarded-Encrypted: i=1; AJvYcCWeqOVGd7glCCLrJF2vq2xL9WcRnvDn/tZ7mWyI9tZwl/UrnHO/wihvuLy1KodiHPuyZ4CKHatUokovAwo=@vger.kernel.org X-Gm-Message-State: AOJu0YxVKw5ZGacmFWXdhKVinHp0rpgyJpXTdUL0Y64fSwsszqhsZfiq rVBHFYUGZ/2bKUdL3+t13866Ujl/csH01Ay9GT5Pb4BZAhAw65IBF53hMN4Yv/eKtyA= X-Gm-Gg: ATEYQzzHVhThQgi4KLuFV4fOZxEX9jPKs080o+i7PigIMHZpDnzd5HiL1Jrz6v6dirH LRUeVDU0aMPhVvYmfnZipWnlaqSnqtuRVxnhjzFO6u9qymD91q6zLuVgrrGS7rFqPooTqpXjGM7 L6nEjRjFH4xc+WvS7/BemDg8cMdJfL657pqJIB5SeYNRDsdTF5lx/eKIWtCE7vnB7TCc0D8p1pW WuRNOj9yAU5MaORzLWFyKcge3XjgFfRKs6XD39+5EUxaIqNtPWTNrc9qdjvA1Ja+aI8jyA3mVjg cuzFHO8RwFNtJuU5SQHwCOeR7/k+k5lzMHlz2JCrcJc+INrh5qnBlbZHVbTarJIWIN6Y187OA1Y Bi/GtKQr6CeS9T9p2pnyheqGS5H8hEtMD+XSEMoZsqfSPDilpFlAOpdwCHDH6WlcZl6bWCclV/n BLvyZp X-Received: by 2002:a17:902:ea0f:b0:2ae:57e6:616c with SMTP id d9443c01a7336-2b0cdc1eec0mr4725605ad.3.1774569173788; Thu, 26 Mar 2026 16:52:53 -0700 (PDT) Received: from localhost ([2a03:2880:2ff:59::]) by smtp.gmail.com with ESMTPSA id d9443c01a7336-2b0bc76b8acsm43309065ad.4.2026.03.26.16.52.53 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 26 Mar 2026 16:52:53 -0700 (PDT) From: Joe Damato To: netdev@vger.kernel.org, "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni Cc: andrew+netdev@lunn.ch, horms@kernel.org, michael.chan@broadcom.com, pavan.chebbi@broadcom.com, linux-kernel@vger.kernel.org, leon@kernel.org, Joe Damato Subject: [net-next v6 02/12] net: tso: Add tso_dma_map helpers Date: Thu, 26 Mar 2026 16:52:21 -0700 Message-ID: <20260326235238.2940471-3-joe@dama.to> X-Mailer: git-send-email 2.52.0 In-Reply-To: <20260326235238.2940471-1-joe@dama.to> References: <20260326235238.2940471-1-joe@dama.to> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="utf-8" Adds skb_frag_phys() to skbuff.h, returning the physical address of a paged fragment's data, which is used by the tso_dma_map helpers introduced in this commit described below: tso_dma_map_init(): DMA-maps the linear payload region and all frags upfront. Prefers the DMA IOVA API for a single contiguous mapping with one IOTLB sync; falls back to per-region dma_map_phys() otherwise. Returns 0 on success, cleans up partial mappings on failure. tso_dma_map_cleanup(): Handles both IOVA and fallback teardown paths. tso_dma_map_count(): counts how many descriptors the next N bytes of payload will need. Returns 1 if IOVA is used since the mapping is contiguous. tso_dma_map_next(): yields the next (dma_addr, chunk_len) pair. On the IOVA path, each segment is a single contiguous chunk. On the fallback path, indicates when a chunk starts a new DMA mapping so the driver can set dma_unmap_len on that descriptor for completion-time unmapping. Suggested-by: Jakub Kicinski Signed-off-by: Joe Damato --- v4: - Fix the kdoc for the TSO helpers. No functional changes. v3: - Added skb_frag_phys helper include/linux/skbuff.h. - Added tso_dma_map_use_iova() inline helper in tso.h. - Updated the helpers to use the DMA IOVA API and falls back to per-regi= on mapping instead. include/linux/skbuff.h | 11 ++ include/net/tso.h | 21 ++++ net/core/tso.c | 273 +++++++++++++++++++++++++++++++++++++++++ 3 files changed, 305 insertions(+) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 9cc98f850f1d..d8630eb366c5 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3758,6 +3758,17 @@ static inline void *skb_frag_address_safe(const skb_= frag_t *frag) return ptr + skb_frag_off(frag); } =20 +/** + * skb_frag_phys - gets the physical address of the data in a paged fragme= nt + * @frag: the paged fragment buffer + * + * Returns: the physical address of the data within @frag. + */ +static inline phys_addr_t skb_frag_phys(const skb_frag_t *frag) +{ + return page_to_phys(skb_frag_page(frag)) + skb_frag_off(frag); +} + /** * skb_frag_page_copy() - sets the page in a fragment from another fragment * @fragto: skb fragment where page is set diff --git a/include/net/tso.h b/include/net/tso.h index 8f8d9d74e873..f78a470a7277 100644 --- a/include/net/tso.h +++ b/include/net/tso.h @@ -68,4 +68,25 @@ struct tso_dma_map { } frags[MAX_SKB_FRAGS]; }; =20 +int tso_dma_map_init(struct tso_dma_map *map, struct device *dev, + const struct sk_buff *skb, unsigned int hdr_len); +void tso_dma_map_cleanup(struct tso_dma_map *map); +unsigned int tso_dma_map_count(struct tso_dma_map *map, unsigned int len); +bool tso_dma_map_next(struct tso_dma_map *map, dma_addr_t *addr, + unsigned int *chunk_len, unsigned int *mapping_len, + unsigned int seg_remaining); + +/** + * tso_dma_map_use_iova - check if this map used the DMA IOVA path + * @map: the map to check + * + * Return: true if the IOVA API was used for this mapping. When true, + * the driver must call tso_dma_map_cleanup() at completion time instead + * of doing per-region DMA unmaps. + */ +static inline bool tso_dma_map_use_iova(struct tso_dma_map *map) +{ + return dma_use_iova(&map->iova_state); +} + #endif /* _TSO_H */ diff --git a/net/core/tso.c b/net/core/tso.c index 6df997b9076e..8d3cfbd52e84 100644 --- a/net/core/tso.c +++ b/net/core/tso.c @@ -3,6 +3,7 @@ #include #include #include +#include #include =20 void tso_build_hdr(const struct sk_buff *skb, char *hdr, struct tso_t *tso, @@ -87,3 +88,275 @@ int tso_start(struct sk_buff *skb, struct tso_t *tso) return hdr_len; } EXPORT_SYMBOL(tso_start); + +static int tso_dma_iova_try(struct device *dev, struct tso_dma_map *map, + phys_addr_t phys, size_t linear_len, size_t total_len, + size_t *offset) +{ + const struct sk_buff *skb; + unsigned int nr_frags; + int i; + + if (!dma_iova_try_alloc(dev, &map->iova_state, phys, total_len)) + return 1; + + skb =3D map->skb; + nr_frags =3D skb_shinfo(skb)->nr_frags; + + if (linear_len) { + if (dma_iova_link(dev, &map->iova_state, + phys, *offset, linear_len, + DMA_TO_DEVICE, 0)) + goto iova_fail; + map->linear_len =3D linear_len; + *offset +=3D linear_len; + } + + for (i =3D 0; i < nr_frags; i++) { + skb_frag_t *frag =3D &skb_shinfo(skb)->frags[i]; + unsigned int frag_len =3D skb_frag_size(frag); + + if (dma_iova_link(dev, &map->iova_state, + skb_frag_phys(frag), *offset, + frag_len, DMA_TO_DEVICE, 0)) { + map->nr_frags =3D i; + goto iova_fail; + } + map->frags[i].len =3D frag_len; + *offset +=3D frag_len; + map->nr_frags =3D i + 1; + } + + if (dma_iova_sync(dev, &map->iova_state, 0, total_len)) + goto iova_fail; + + return 0; + +iova_fail: + dma_iova_destroy(dev, &map->iova_state, *offset, + DMA_TO_DEVICE, 0); + memset(&map->iova_state, 0, sizeof(map->iova_state)); + + /* reset map state */ + map->frag_idx =3D -1; + map->offset =3D 0; + map->linear_len =3D 0; + map->nr_frags =3D 0; + + return 1; +} + +/** + * tso_dma_map_init - DMA-map GSO payload regions + * @map: map struct to initialize + * @dev: device for DMA mapping + * @skb: the GSO skb + * @hdr_len: per-segment header length in bytes + * + * DMA-maps the linear payload (after headers) and all frags. + * Prefers the DMA IOVA API (one contiguous mapping, one IOTLB sync); + * falls back to per-region dma_map_phys() when IOVA is not available. + * Positions the iterator at byte 0 of the payload. + * + * Return: 0 on success, -ENOMEM on DMA mapping failure (partial mappings + * are cleaned up internally). + */ +int tso_dma_map_init(struct tso_dma_map *map, struct device *dev, + const struct sk_buff *skb, unsigned int hdr_len) +{ + unsigned int linear_len =3D skb_headlen(skb) - hdr_len; + unsigned int nr_frags =3D skb_shinfo(skb)->nr_frags; + size_t total_len =3D skb->len - hdr_len; + size_t offset =3D 0; + phys_addr_t phys; + int i; + + if (!total_len) + return 0; + + map->dev =3D dev; + map->skb =3D skb; + map->hdr_len =3D hdr_len; + map->frag_idx =3D -1; + map->offset =3D 0; + map->iova_offset =3D 0; + map->total_len =3D total_len; + map->linear_len =3D 0; + map->nr_frags =3D 0; + memset(&map->iova_state, 0, sizeof(map->iova_state)); + + if (linear_len) + phys =3D virt_to_phys(skb->data + hdr_len); + else + phys =3D skb_frag_phys(&skb_shinfo(skb)->frags[0]); + + if (tso_dma_iova_try(dev, map, phys, linear_len, total_len, &offset)) { + /* IOVA path failed, map state was reset. Fallback to + * per-region dma_map_phys() + */ + if (linear_len) { + map->linear_dma =3D dma_map_phys(dev, phys, linear_len, + DMA_TO_DEVICE, 0); + if (dma_mapping_error(dev, map->linear_dma)) + return -ENOMEM; + map->linear_len =3D linear_len; + } + + for (i =3D 0; i < nr_frags; i++) { + skb_frag_t *frag =3D &skb_shinfo(skb)->frags[i]; + unsigned int frag_len =3D skb_frag_size(frag); + + map->frags[i].len =3D frag_len; + map->frags[i].dma =3D dma_map_phys(dev, skb_frag_phys(frag), + frag_len, DMA_TO_DEVICE, 0); + if (dma_mapping_error(dev, map->frags[i].dma)) { + tso_dma_map_cleanup(map); + return -ENOMEM; + } + map->nr_frags =3D i + 1; + } + } + + if (linear_len =3D=3D 0 && nr_frags > 0) + map->frag_idx =3D 0; + + return 0; +} +EXPORT_SYMBOL(tso_dma_map_init); + +/** + * tso_dma_map_cleanup - unmap all DMA regions in a tso_dma_map + * @map: the map to clean up + * + * Handles both IOVA and fallback paths. For IOVA, calls + * dma_iova_destroy(). For fallback, unmaps each region individually. + */ +void tso_dma_map_cleanup(struct tso_dma_map *map) +{ + int i; + + if (dma_use_iova(&map->iova_state)) { + dma_iova_destroy(map->dev, &map->iova_state, map->total_len, + DMA_TO_DEVICE, 0); + memset(&map->iova_state, 0, sizeof(map->iova_state)); + map->linear_len =3D 0; + map->nr_frags =3D 0; + return; + } + + if (map->linear_len) + dma_unmap_phys(map->dev, map->linear_dma, map->linear_len, + DMA_TO_DEVICE, 0); + + for (i =3D 0; i < map->nr_frags; i++) + dma_unmap_phys(map->dev, map->frags[i].dma, map->frags[i].len, + DMA_TO_DEVICE, 0); + + map->linear_len =3D 0; + map->nr_frags =3D 0; +} +EXPORT_SYMBOL(tso_dma_map_cleanup); + +/** + * tso_dma_map_count - count descriptors for a payload range + * @map: the payload map + * @len: number of payload bytes in this segment + * + * Counts how many contiguous DMA region chunks the next @len bytes + * will span, without advancing the iterator. On the IOVA path this + * is always 1 (contiguous). On the fallback path, uses region sizes + * from the current position. + * + * Return: the number of descriptors needed for @len bytes of payload. + */ +unsigned int tso_dma_map_count(struct tso_dma_map *map, unsigned int len) +{ + unsigned int offset =3D map->offset; + int idx =3D map->frag_idx; + unsigned int count =3D 0; + + if (!len) + return 0; + + if (dma_use_iova(&map->iova_state)) + return 1; + + while (len > 0) { + unsigned int region_len, chunk; + + if (idx =3D=3D -1) + region_len =3D map->linear_len; + else + region_len =3D map->frags[idx].len; + + chunk =3D min(len, region_len - offset); + len -=3D chunk; + count++; + offset =3D 0; + idx++; + } + + return count; +} +EXPORT_SYMBOL(tso_dma_map_count); + +/** + * tso_dma_map_next - yield the next DMA address range + * @map: the payload map + * @addr: output DMA address + * @chunk_len: output chunk length + * @mapping_len: full DMA mapping length when this chunk starts a new + * mapping region, or 0 when continuing a previous one. + * On the IOVA path this is always 0 (driver must not + * do per-region unmaps; use tso_dma_map_cleanup instead). + * @seg_remaining: bytes left in current segment + * + * Yields the next (dma_addr, chunk_len) pair and advances the iterator. + * On the IOVA path, the entire payload is contiguous so each segment + * is always a single chunk. + * + * Return: true if a chunk was yielded, false when @seg_remaining is 0. + */ +bool tso_dma_map_next(struct tso_dma_map *map, dma_addr_t *addr, + unsigned int *chunk_len, unsigned int *mapping_len, + unsigned int seg_remaining) +{ + unsigned int region_len, chunk; + + if (!seg_remaining) + return false; + + /* IOVA path: contiguous DMA range, no region boundaries */ + if (dma_use_iova(&map->iova_state)) { + *addr =3D map->iova_state.addr + map->iova_offset; + *chunk_len =3D seg_remaining; + *mapping_len =3D 0; + map->iova_offset +=3D seg_remaining; + return true; + } + + /* Fallback path: per-region iteration */ + + if (map->frag_idx =3D=3D -1) { + region_len =3D map->linear_len; + chunk =3D min(seg_remaining, region_len - map->offset); + *addr =3D map->linear_dma + map->offset; + *mapping_len =3D (map->offset =3D=3D 0) ? region_len : 0; + } else { + region_len =3D map->frags[map->frag_idx].len; + chunk =3D min(seg_remaining, region_len - map->offset); + *addr =3D map->frags[map->frag_idx].dma + map->offset; + *mapping_len =3D (map->offset =3D=3D 0) ? region_len : 0; + } + + *chunk_len =3D chunk; + map->offset +=3D chunk; + + if (map->offset >=3D region_len) { + map->frag_idx++; + map->offset =3D 0; + } + + return true; +} +EXPORT_SYMBOL(tso_dma_map_next); --=20 2.52.0