{"id":26025,"date":"2026-09-18T04:32:14","date_gmt":"2026-09-17T20:32:14","guid":{"rendered":"https:\/\/www.54iter.cn\/?vuln=cve-2026-93168"},"modified":"2026-09-18T04:32:14","modified_gmt":"2026-09-17T20:32:14","slug":"cve-2026-93168","status":"publish","type":"vuln","link":"https:\/\/www.54iter.cn\/?vuln=cve-2026-93168","title":{"rendered":"CVE-2026-93168 &#8211; In the Linux kernel, the following vulnerability has been resolved:\n\ndmaengine: xilinx_dma"},"content":{"rendered":"<p>In the Linux kernel, the following vulnerability has been resolved:<\/p>\n<p>dmaengine: xilinx_dma: Fix CPU stall in xilinx_dma_poll_timeout<\/p>\n<p>Currently when calling xilinx_dma_poll_timeout with delay_us=0 and a<br \/>\ncondition that is never fulfilled, the CPU busy-waits for prolonged time<br \/>\nand the timeout triggers only with a massive delay causing a CPU stall.<\/p>\n<p>This happens due to a huge underestimation of wall clock time in<br \/>\npoll_timeout_us_atomic. Commit 7349a69cf312 (&#8220;iopoll: Do not use<br \/>\ntimekeeping in read_poll_timeout_atomic()&#8221;) changed the behavior to no<br \/>\nlonger use ktime_get at the expense of underestimation of wall clock<br \/>\ntime which appears to be very large for delay_us=0. Instead of timing<br \/>\nout after approximately XILINX_DMA_LOOP_COUNT microseconds, the timeout<br \/>\ntakes XILINX_DMA_LOOP_COUNT * 1000 * (time that the overhead of the for<br \/>\nloop in poll_timeout_us_atomic takes) which is in the range of several<br \/>\nminutes for XILINX_DMA_LOOP_COUNT=1000000. Fix this by using a non-zero<br \/>\nvalue for delay_us. Use delay_us=10 to keep the delay in the hot path of<br \/>\nstarting DMA transfers minimal but still avoid CPU stalls in case of<br \/>\nunexpected hardware failures.<\/p>\n<p>One-off measurement with delay_us=0 causes the cpu to busy wait around 7<br \/>\nminutes in the timeout case. After applying this patch with delay_us=10<br \/>\nthe measured timeout was 1053428 microseconds which is roughly<br \/>\nequivalent to the expected 1000000 microseconds specified in<br \/>\nXILINX_DMA_LOOP_COUNT.<\/p>\n<p>Add a constant XILINX_DMA_POLL_DELAY_US for delay_us value.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>In the Linux kernel, the following vulnerability has been resolved:<\/p>\n<p>dmaengine: xilinx_dma: Fix CPU stall in xilinx_dma_poll_timeout<\/p>\n<p>Currently when calling xilinx_dma_poll_timeout with delay_us=0 and<\/p>\n","protected":false},"template":"","meta":[],"categories":[],"tags":[],"class_list":["post-26025","vuln","type-vuln","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln\/26025","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/vuln"}],"about":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=\/wp\/v2\/types\/vuln"}],"wp:attachment":[{"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=26025"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=26025"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.54iter.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=26025"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}