From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id E9BBB2F7477; Tue, 12 Aug 2025 17:40:52 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1755020453; cv=none; b=h45z+hNyRgwiTo3knKYy4xOpvpIoCdO/mxAFM5DpLnyyCE4/XtRBECXebFO+rAcyWkGe2Zw/VO5RmX6f5ZZ2As+9RzleZQinLfrIeuJDTt0daVjw8fG0UbMvKsDs+bsYN5n74zex/Xi2Lhlvxo1cRe8eBw1Ss/bjXv0RqDQM0l0= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1755020453; c=relaxed/simple; bh=9W9XsEOL3I9XOZqVzJBJgNwsmolxySWL2FPNI4RaaJE=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=KFofSkyicRm5LIQcaFhEMys4Cr89JulRN2L2s+GDRQuQxYs7Jjo0u16ahJge1i54683igdBCDYeyI3Tx1mMMCTrPuiiJcC0TPdnKo1hQJo/QEfBdOGQEHKoU51C9OEfX0s2mRKLF1DLMcZDv9DR9T0/B9ejwTYfgqaoozv+Qr6Y= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=Z3qpZxwI; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="Z3qpZxwI" Received: by smtp.kernel.org (Postfix) with ESMTPSA id EE2F6C4CEF0; Tue, 12 Aug 2025 17:40:51 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1755020452; bh=9W9XsEOL3I9XOZqVzJBJgNwsmolxySWL2FPNI4RaaJE=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=Z3qpZxwITWql6gueUgHRngzNs/CarfzXuJ1wbaTRYJoFlvYe4EcYlccv6yWGgPZk5 eZ26S/YW+E7sV12kFZE62SuXXdkPwXAWMwa6O/UWViNRIp1ka9GBvzPoEN4nSf+jsM QuI8k8HAP16yqrVUNhkfk51Oxc5pbuezkpVZ+aQ0= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Jan Kiszka , Stefan Kerkmann , Gao Xiang Subject: [PATCH 6.1 058/253] erofs: address D-cache aliasing Date: Tue, 12 Aug 2025 19:27:26 +0200 Message-ID: <20250812172951.195444718@linuxfoundation.org> X-Mailer: git-send-email 2.50.1 In-Reply-To: <20250812172948.675299901@linuxfoundation.org> References: <20250812172948.675299901@linuxfoundation.org> User-Agent: quilt/0.68 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: patches@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 6.1-stable review patch. If anyone has any objections, please let me know. ------------------ From: Gao Xiang commit 27917e8194f91dffd8b4825350c63cb68e98ce58 upstream. Flush the D-cache before unlocking folios for compressed inodes, as they are dirtied during decompression. Avoid calling flush_dcache_folio() on every CPU write, since it's more like playing whack-a-mole without real benefit. It has no impact on x86 and arm64/risc-v: on x86, flush_dcache_folio() is a no-op, and on arm64/risc-v, PG_dcache_clean (PG_arch_1) is clear for new page cache folios. However, certain ARM boards are affected, as reported. Fixes: 3883a79abd02 ("staging: erofs: introduce VLE decompression support") Closes: https://lore.kernel.org/r/c1e51e16-6cc6-49d0-a63e-4e9ff6c4dd53@pengutronix.de Closes: https://lore.kernel.org/r/38d43fae-1182-4155-9c5b-ffc7382d9917@siemens.com Tested-by: Jan Kiszka Tested-by: Stefan Kerkmann Link: https://lore.kernel.org/r/20250709034614.2780117-2-hsiangkao@linux.alibaba.com Signed-off-by: Gao Xiang Signed-off-by: Greg Kroah-Hartman --- fs/erofs/decompressor.c | 6 ++---- fs/erofs/zdata.c | 32 +++++++++++++++++++------------- 2 files changed, 21 insertions(+), 17 deletions(-) --- a/fs/erofs/decompressor.c +++ b/fs/erofs/decompressor.c @@ -342,14 +342,12 @@ static int z_erofs_transform_plain(struc if (outpages > inpages) { DBG_BUGON(!rq->out[outpages - 1]); - if (rq->out[outpages - 1] != rq->in[inpages - 1]) { + if (rq->out[outpages - 1] != rq->in[inpages - 1]) memcpy_to_page(rq->out[outpages - 1], 0, src + (interlaced_offset ? 0 : righthalf), lefthalf); - } else if (!interlaced_offset) { + else if (!interlaced_offset) memmove(src, src + righthalf, lefthalf); - flush_dcache_page(rq->in[inpages - 1]); - } } kunmap_local(src); return 0; --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -123,9 +123,11 @@ static inline unsigned int z_erofs_pclus /* * bit 30: I/O error occurred on this page + * bit 29: CPU has dirty data in D-cache (needs aliasing handling); * bit 0 - 29: remaining parts to complete this page */ -#define Z_EROFS_PAGE_EIO (1 << 30) +#define Z_EROFS_ONLINEPAGE_EIO 30 +#define Z_EROFS_ONLINEPAGE_DIRTY 29 static inline void z_erofs_onlinepage_init(struct page *page) { @@ -144,7 +146,7 @@ static inline void z_erofs_onlinepage_sp atomic_inc((atomic_t *)&page->private); } -static void z_erofs_onlinepage_endio(struct page *page, int err) +static void z_erofs_onlinepage_end(struct page *page, int err, bool dirty) { int orig, v; @@ -152,16 +154,20 @@ static void z_erofs_onlinepage_endio(str do { orig = atomic_read((atomic_t *)&page->private); - v = (orig - 1) | (err ? Z_EROFS_PAGE_EIO : 0); + DBG_BUGON(orig <= 0); + v = dirty << Z_EROFS_ONLINEPAGE_DIRTY; + v |= (orig - 1) | (!!err << Z_EROFS_ONLINEPAGE_EIO); } while (atomic_cmpxchg((atomic_t *)&page->private, orig, v) != orig); - if (!(v & ~Z_EROFS_PAGE_EIO)) { - set_page_private(page, 0); - ClearPagePrivate(page); - if (!(v & Z_EROFS_PAGE_EIO)) - SetPageUptodate(page); - unlock_page(page); - } + if (v & (BIT(Z_EROFS_ONLINEPAGE_DIRTY) - 1)) + return; + set_page_private(page, 0); + ClearPagePrivate(page); + if (v & BIT(Z_EROFS_ONLINEPAGE_DIRTY)) + flush_dcache_page(page); + if (!(v & BIT(Z_EROFS_ONLINEPAGE_EIO))) + SetPageUptodate(page); + unlock_page(page); } #define Z_EROFS_ONSTACK_PAGES 32 @@ -925,7 +931,7 @@ next_part: goto repeat; out: - z_erofs_onlinepage_endio(page, err); + z_erofs_onlinepage_end(page, err, false); return err; } @@ -1028,7 +1034,7 @@ static void z_erofs_fill_other_copies(st cur += len; } kunmap_local(dst); - z_erofs_onlinepage_endio(bvi->bvec.page, err); + z_erofs_onlinepage_end(bvi->bvec.page, err, true); list_del(p); kfree(bvi); } @@ -1196,7 +1202,7 @@ out: /* recycle all individual short-lived pages */ if (z_erofs_put_shortlivedpage(be->pagepool, page)) continue; - z_erofs_onlinepage_endio(page, err); + z_erofs_onlinepage_end(page, err, true); } if (be->decompressed_pages != be->onstack_pages)